doe/nsf thermoelectric partnership project … · 300 350 400 450 500 550 0.10 0.15 0.20 0.25. zt....

1

PI: Joseph P. Heremans (Ohio State University)

co-PI’s: Mercouri Kanatzidis (Northwestern University) Guo-Quan Lu (Virginia Polytechnic Institute and State University)

Lon Bell (BSST) Dmitri Kossakovski (ZTPlus)

2012 DOE Annual Merit Review, Crystal City Marriott, 18 May 2012

Report up to 16 March 2012

DOE/NSF Thermoelectric Partnership Project SEEBECK

Saving Energy Effectively By Engaging in Collaborative research and sharing Knowledge

This presentation does not contain any proprietary, confidential, or otherwise restricted information

Project ID#: ACE068

2

Overview Timeline

Start: January 1, 2011 End: December 31, 2014 Percent complete: 31%

Budget Total (NSF+DOE): $1,453,532 DOE share: 50% (unverified) • Funding FY13: OSU+subcontracts: $249,301 NU: $128,000 VT: $118,05614,413

Barriers • Overall program barriers addressed: A, B, C, D • Barriers specific to thermoelectric generators: - High-ZT low-cost materials made from available elements - Thermal management - Interface resistances - Durability - Metrology

Partners Ohio State University (OSU, lead), Northwestern University (NU), Virginia Polytechnic Institute and State University (VT), ZT Plus & BSST as subcontractors to OSU

3

Objectives Project goal: Develop elements for a practical automotive exhaust waste-heat

recovery system that meets cost and durability requirements of the industry. Project objectives

1. Develop high-ZT low-cost materials made from available elements: • ZT>1.5 • materials that are not rare or toxic (Te, Tl): PbS, Mg2Sn, ZnSb

2. New thermal management strategy, not developed under this program, in use at Gentherm

3. Minimize electrical and thermal interface resistances: • Compliant, to accommodate thermal expansion • High thermal conductance across interface • High electrical conductance across interface

4. Metrology: • Materials characterization • Electrical and Thermal interface resistance measurements • Overall system performance measurements

5. Durability: • Compatible with automotive durability requirements

GROUP

• PbS: the cheapest thermoelectric

• Raising ZT of p-type PbS with endotaxial nanostructuring and valence-band offset engineering using CdS and ZnS

• Band alignment engineering between nanostructures and matrix is a key

path forward to increasing ZT • High performance in nanostructured p-type PbS (ZT~1.2-1.3 at 900 K):

This is a breakthrough in the performance of PbS

Objective 1: Materials

Approach 1: PbS, the cheapest thermoelectric (Northwestern)

GROUP

PbS and MS second phases: band alignment

Calculated band gap energy levels of the metal sulfides, PbS, CdS, ZnS, CaS and SrS, all in the NaCl structure. All values are in eV. (courtesy of Prof. C. Wolverton)

n-type

p-type

Band alignment engineering between nanostructures

PbS+1.0 at. % Bi2S3+1.0 at. % PbCl2 PbS + 1.0 at. % Sb2S3 + 1.0 at. % PbCl2

TEM: nanostructured PbS

GROUP

P-type PbS: κlat, μn and μp

p-type, CdS containing sample shows higher μ

n-type, all samples show similar μ

Similar reduce levels on lattice κ

GROUP

Zhao L.D. et al. JACS, 2012, in press.

ZT for PbS system

0.13eV

CdS

e e

phonons

PbS

PbS CdS

EgE’g

minimal valence band

offsetVB

(a)

(b)

ΔE

9

Approach 2: Resonant levels in Mg2Sn1-xSix (OSU) • Cheap , Environmentally friendly, & Abundantly available. • N-type High ZT reported in Mg2Sn1-xSix (ZT >1) - Zaitsev et al. Physical Review B 74, 045207 (2006) - Q. Zhang et al. Appl. Phys. Lett. 93, 102109 (2008). • P-type Few studies for P-type Mg2Sn1-xSix materials have been reported. • Candidate dopants identified by band structure calculations

(J. Tobola): Ag

10

Calculated Masses & Fermi surface: favorable

0.7mo -1mo for heavy band, 0.15mo -0.2mo for intermediate

10

Ag, Cu, In, Zn Doped Binary Mg2Sn

300 350 400 450 500 550

0.10

0.15

0.20

0.25

zT

Temperature (K)

Ag, n = 3.8*1019

100 200 300 400 500 6000

4

8

12

16

20Po

wer

Fac

tor

(µ W

cm

-1K

-2)

Mg2Sn Ag, n =3.8*1019

Ag, n =5.0*1019

Ag, n =8.0*1019

Ag, n =5.8*1019

Cu, n =0.27*1019

Cu, n =0.8*1019

Cu, n =1.1*1019

Zn, n =0.02*1019

Zn, n =0.015*1019

In, n =0.28*1019

In, n =1.5*1019

Temperature (K)

300 400 500 600 700 8003

4

5

6

7

8

9

Ther

mal

Con

duct

ivity

(W m

-1K

-1)

Mg2Sn Ag, n =3.8*1019

Ag, n =5*1019

Cu, n =0.8*1019

In, n =1.5*1019

Zn, n=0.02*1019

Temperature (K)

Binary Mg2Sn has 1. too high a thermal conductivity 2. too low a band gap => minority carriers appear at 400 K and decrease thermopower

11

Mg2Sn1-xSix X = 0.1, 0.05, doped with Ag

100 200 300 4001E16

1E17

1E18

1E19

1E20

Carri

er C

once

ntra

tion

(cm

-3)

Temperature (K)

Mg2Sn0.95Si0.05 :Ag(1%) Mg2Sn0.95Si0.05 :Ag(2%) Mg2Sn0.95Si0.05 :Ag(3%)

100 200 300 400-100

0

100

200

300

400

Seeb

eck (

µV/K

)

Temperature (K)

Mg2Sn0.95Si0.05

Ag, n = 5*1017

Ag, n = 7*1016

Ag, n = 2*1019

0 100 200 300 400

1E18

1E19

1E20

Mg2Sn0.9Si0.1 :Ag(2%) Mg2Sn0.9Si0.1 : Ag(5%)Ca

rrier

Con

cent

ratio

n (c

m-3)

Temperature (K)

100 200 300 400

-50

0

50

100

150

Seeb

eck (

µV/K

)

Tempereature (K)

Mg2Sn0.9Si0.1

Ag, n = 2*1018

Ag, n = 8*1019

Ongoing work

12

13

Approach 3: ZnSb (ZTPlus, OSU) Cheap , Environmentally friendly, & Abundantly available.

Peritectic melting

The peritectic melting of the phase generates impurities and degrades properties.

Usually extremely long post annealing times are required SPS allows rapid production of 99% clean ZnSb specimens (XRD).

14

29 300

1000

2000

3000

4000

5000

6000

7000

x=0.1x=0.05

Inte

nsity

2θ (deg)

x=0.025

Zn1-xMgxSb

20 30 40 500

1000

2000

3000

4000

5000

6000

7000

8000Zn1-xMgxSb

*

Inte

nsity

2θ (deg)

*x=0.1x=0.05x=0.025

Impurities appear at 10 % at. Mg substitution for Zn Bismuth substitution for antimony did not result in a single phase even at subsitution levels of 2% at.

Zn1-xMgxSb alloys (ZTPlus)

14

Can be alloyed with Mg (transport properties not yet studied)

Transport properties of representative ZnSb sample

0 200 400 600Temperature (K)

0

4

8

12

16R

esis

tivity

(m

Ωcm

)OSU ES-0991-1OSU ES-0991-2HT ES-0991-1HT ES-0991-2


40

80

120

160

200

240

280

Seeb

eck

Coe

ffici

ent (

µV/K

)

OSU ES-0991-1OSU ES-0991-2HT ES-0991-1HT ES-0991-2


0

2

4

6

Ther

mal

Con

duct

ivity

(W

/m⋅K

)0 200 400 600

Temperature (K)

0

0.2

0.4

0.6

0.8

1

ZT

• Data obtained both at ZTPlus and at OSU

• Glitch at 423 K

• Excellent ZT

15

High Temperature XRD on ZnSb

• Peaks position change with temperature (thermal expansion)

• Abnormal jump between 398 and 423K (more measurements needed)

20 30 40 502θ degree

Inte

nsity

a.

u.

373K398K423K448K

360 380 400 420 440 460Temperature (K)

28.6

28.8

29

29.2

29.4

2θ

Deg

ree

High Temperature XRD XRD(20~550)

Highest Peak Position vs T

16

17

Objective 3. Interfaces (Virginia Tech) Compliant, high electrical and thermal

conductance

New compliant thermoelectric device interconnects using nanosilver Problem addressed: interdiffusion between chalcogen atoms in thermoelectric material and silver in contacts Achievement this year: diffusion barrier for thermoelectric materials

18

Approach: Nanosilver interconnect packaging material

Pb-free, high-T die-attach solution:

Substrate

Device

Densificationby diffusion

Heat up

Cool down

Sintered joint

Organicthinner

Surfactant Ag nano-powder

Uniform Dispersion

Silver PastenanoTach®

+ +Binder

Thinner

Surfactant

Joint formed below 280oC; Melting at 960oC; Thermal conduct > 150

W/m-K.

Assembled TE device

10 mm

Ag Ti

Thermoelectric material

100nm 100nm

Schematic of Ti/Ag layer deposited on TE material

TE materials with Ti/Ag deposited as the coating layer

Diffusion barrier: vapor deposition of Ti/Ag interface layer on thermoelectric substrate

Objective: to metallize Bi2Te3-based thermoelectric material with diffusion barrier layer for bonding by nanosilver paste.

Light regions coated with Ti/Ag

19

Evaluation of film adhesion by scratch-test

As-coated sample Scratched by a diamond scriber, then peeled by

adhesive tape Enlarged view of the

tested region

The fact that the coating did not come off completely with the tape suggests that the adhesion is reasonably strong.

Work is underway to quantify the adhesion strength.

20

21

Objective 4. Metrology: thermal contact resistance measurements

Objective: to determine the thermal property of nanosilver-bonded interface from bonding strength measurement.

0.20.30.40.50.6

0 10 20 30 40Zth

(K/W

)

Heating time (ms)

nanoTach

Thermal impedance decreases drastically with increasing die-shearing strength;

A die-shear strength of at least 10 MPa is necessary for low thermal impedance.

Thermal impedance vs. die-shear strength

Observations:

22

23

Objective 5. Durability Ensure that this project incorporates automotive

durability standards.

1. Durability is built into every step of the design 2. Thermal stability of PbS’ ZT (backup slide only) 3. Extensive durability testing at Gentherm and ZTPlus Gentherm and ZTPlus, have state-of-the-art durability testing facilities

used in the development of automotive products. Achievement this year: thermal cycling resistance of newly developed materials reproducibility of newly developed materials

ZT ~ 1.1 @ 923 K ZT ~ 1.06 @ 923 K

M: normal melting B: Bridgman S: SPS BN coating

Good repeatable

Repeatability of n-type PbS, ZT=1.1

25

50 cycles between 30 to 370 oC: No degradation of properties. Improvement of the power factor. In-gradient testing 48 hours: sample chemically stable repeated properties

sample

350 oC

25 oC I= 70 A T-controlled Cu block, electrode

T-controlled Cu block, electrode

ZnSb (ZTPlus) figure of merit, thermal stability

THERMAL STABILITY TESTS

25

0 100 200 300 4000.0

0.2

0.4

0.6

0.8

1.0

ES-0984-1 ES-0984-2 ES-0984-3

ZT

Temperature (°C) ZTPlus Confidential

26

SUMMARY: FIVE objectives

1.New Materials • PbS record ZT achieved; literally dirt-cheap • Mg2(Si, Sn): Doping studies of p-type material • ZnSb: excellent ZT up to 0.85, repeatable

2. Gentherm thermal design used in prototypes (not developed under this program) 3. New Interface technologies: flexible Ag nanopaste

• Developed diffusion barrier layer 4. Metrology: • 1. Thermal contact resistance measurements, the role of die-attach

bond strength 5. Reliability: inherent in design

• 1.Data on reproducibility of PbS • 2. Data on cyclability of ZnSb

Collaboration is inherent, flow of materials from partner to partner.

Technical back-up slides

28

Project Phase Years 1 – 3 Task Description Time Period in years 1 2 3 1 Optimize thermoelectric properties of PbSe

1.1 Determine conditions for thermodynamic synthesis X

1.2 Develop bulk nanostructuring material by liquid encapsulation for Sb-PbSe, Bi-PbSe, Ga-PbSe, In-PbSe, As-PbSe X X X

1.2.1 Chemical characterization (XRD/SEM/EDX) X X X

1.2.2 Thermoelectric characterization: measuring, analyzing X X

1.3 Introduce resonant impurities (In, Al, Ga, Tl) X X X

1.3.1 Chemical characterization (XRD/SEM/EDX) X 1.3.2 Thermoelectric characterization: measuring, analyzing ← X X

1.4 Introduce 3d,4d or 5d elements and tune ← X X

1.4.1 Chemical characterization (XRD/SEM/EDX) ← X X

1.4.2 Thermoelectric characterization: measuring, analyzing ← X X

Milestones (bold = achieved, ← = pulled forward) Summary: record ZT’s achieved after year 1 => expand to cheaper alternative PbS

29

2 Optimize thermoelectric properties of Mg2X (Si,Sn,Pb)

2.1 Determine conditions for thermodynamic synthesis X

2.2 Develop bulk nanostructuring material by liquid encapsulation for Mg2Si1-xSnx/Y, Mg2Si1-xPbx/Y, Mg2Sn1-xPbx/Y with Y= W,Mo, Ta, Hf, Nb X X X

2.2.1 Chemical characterization (XRD/SEM/EDX) ← X X

2.2.2 Thermoelectric characterization: measuring, analyzing ← X X

2.3 Extend investigation with Y = stannides (Hf5Sn3, HfSn, La3Sn5, LaSn3, CoSn, FeSn) X

2.3.1 Chemical characterization (XRD/SEM/EDX) X

2.3.2 Thermoelectric characterization: measuring, analyzing

2.4 Extend investigation with Y = silicides (Co2Si, CoSi, CoSi2, NiSi2 (CaF2 type), FeSi, LiAlSi, ZrSi2, Zr2Si, Zr3Si, Hf2Si, Hf3Si2, WSi2, W5Si3, RuSi) X


2.4.2 Thermoelectric characterization: measuring, analyzing X

2.5 Introduce resonant level in Mg2Pb1-xXx X=Sb, Bi for n-types X X



2.6 Introduce resonant level in Mg2Pb1-xXx X=Ga, In for p-type or by substituting Mg by Na, Ag, X X

2.6.1 Chemical characterization (XRD/SEM/EDX) X X


30

3 Metallization of TE materials 3.1 For PbSe-based material (SPS) X 3.1.1 Develop blend of Fe/Sn or Pb chalcogenides X 3.1.2 Chemical characterization of intermetallics (SEM/EDX) X 3.1.3 Co-pressed the blend with PbSe by SPS X X

3.1.4 Chemical characterization of intermetallics (SEM/EDX) X X

3.1.5 Optimize densification properties X X

3.1.6 Measurement of the contact resistance X X X

3.1.7 Durability test (thermal cycling and chock resistance) X X

3.1.8 Explore other barriers of diffusion co-pressed by SPS ← X X

3.2 For PbSe-based material (PVD) X

3.2.1 Identify potential element (e.g. nitride or carbide) X

3.2.2 Development of the sputtering process X

3.2.3 Chemical characterization (SEM/EDX) X X

3.2.4 Measurement of the contact resistance X X


Milestones (bold = achieved, ← = pulled forward)

31

3 Metallization of TE materials

3.3 Develop a process for Mg2X (SPS) X X

3.3.1 Identify potential blend X X

3.3.2 Co-pressed the blend with Mg2X by SPS X X

3.3.3 Chemical characterization of intermetallics (SEM/EDX) X X

3.3.4 Optimize densification properties X X

3.3.5 Measurement of the contact resistance X


3.4 Develop a process for Mg2X (PVD) X

3.4.1 Identify potential element (e.g. nitride or carbide) X X

3.4.2 Development of the sputtering process X

3.4.3 Chemical characterization (SEM/EDX) X

3.4.4 Measurement of the contact resistance X

3.4.5 Durability test (thermal cycling and chock resistance) X

Milestones

32

4 Device interconnection (bonding element to heat spreader)

4.1 Chemical investigation of Ag diffusion in metallization (SEM/XPS) X

4.1.1 Influence of the amount of Ag X

4.1.2 Influence of coating gold on Ag joint X X

4.1.3 Measurement of the contact resistance X X

4.2 Study of other metals (M) ← X

4.2.1 Chemical investigation of M diffusion in metallization (SEM/XPS) ← X

4.2.2 Influence of the amount of M ← X

4.2.3 Measurement of the contact resistance X X X


33

5 Integration of material and interfaces into patented module 5.1 Chemical characterization (XRD/SEM/EDX) X

5.2 Measurement of the contact resistance X X

5.3 Thermoelectric characterization: measuring, analyzing x

5.4 Durability test x x x

6 Develop new module/heat exchanger design 6.1 Chemical characterization (XRD/SEM/EDX) X

6.2 Thermoelectric characterization: measuring, analyzing X X

6.3 Measurement of the contact resistance X X

6.4 Durability test X X


PbS with second phases without doping

n-type PbS with second phases

Second phases: Bi2S3, Sb2S3

PbS: the cheapest thermoelectric Band alignment engineering between nanostructures and matrix is a key path forward to increasing ZT High performance in nanostructured p-type PbS (ZT~1.2-1.3 at 900 K): This is a breakthrough in the performance of PbS

35

Both total and lattice κ were reduced by SrS inclusions

P-type Pb0.975Na0.025S-x%SrS

36

Pb0.975Na0.025S-3%CaS/SrS & thermal stability

37

Pb0.975Na0.025S+3%SrS shows ZT about 1.2 at 923K, Pb0.975Na0.025S+3%CaS shows ZT about 1.1 at 923K, both samples show excellent thermal stabilities after annealing treatments.

GROUP

Venkatasubramanian, R. et. al., Nature 413(2001)597.

SrTe

PbTe

Biswas, K. et. al., Nature Chem. 3(2011)160.

Concept adapted from PbTe work new ideas

38

doe/nsf thermoelectric partnership project … · 300 350 400 450 500 550 0.10 0.15 0.20 0.25. zt....

Documents