NSF-DOE Thermoelectrics Partnership:
Automotive Thermoelectric Modules with Scalable Thermo- and Electro-Mechanical Interfaces
Prof. Ken Goodson Department of Mechanical Engineering Stanford University
Dr. Boris Kozinsky Energy Modeling, Control, & Computation R. Bosch LLC
Prof. George Nolas Department of Physics University of South Florida
This presentation does not contain any proprietary, confidential, or otherwise restricted information
ACE067
1
• Start – January 2011 • End – December 2013 • ~40% complete
• Thermoelectric Device/System Packaging
• Component/System Durability
• Scaleup
• $1.22 Million (DOE+NSF) • FY12 Funding = $423K • Leveraging:
• ONR (FY09-11) • Fellowships (3 NSF, Sandia,
Stanford DARE)
Budget
Barriers (2.3.2)
• K.E. Goodson, Stanford (lead) • George Nolas, USF • Boris Kozinsky, Bosch
Partners
Overview Timeline
2
Relevance: Addressing Key Challenges for Thermoelectrics in Combustion Systems
…Low resistance interfaces that are stable under thermal cycling. …High-temperature TE materials that are stable and promise low-cost scaleup. …Characterization methods that include interfaces and correlate better with system performance.
hot side / heat exchanger
insulation / e.g. ceramic plate
conductor/ e.g. copper
insulation / e.g. ceramic plate
cold side / heat exchanger
conductor/ e.g. copper conductor/ e.g. copper
n p
thermalthermal
electrical& thermal
electrical& thermalthermalthermal
diffusion barrier
joining technology
contact resistanceHeat
Cooling
hot side / heat exchanger
insulation / e.g. ceramic plate
conductor/ e.g. copper
insulation / e.g. ceramic plate
cold side / heat exchanger
conductor/ e.g. copper conductor/ e.g. copper
n p
thermalthermal
electrical& thermal
electrical& thermalthermalthermal
diffusion barrier
joining technology
contact resistanceHeat
Cooling
600 °C
90 °C
Improvements in the intrinsic ZT of TE materials are proving to be very difficult to translate into efficient, reliable power recovery systems. Major needs include…
3
• Combustion TEG systems experience enormous interface stresses due to wide temperature spans.
• Thermal cycling degrades interface due to cracks, delamination, reflow, reducing efficiency.
• Our simulations show importance of thermodynamic stability (chemical reactivity, inter-solubility, etc.) and elastic modulus.
Relevance: Thermoelectric Interface Challenge
Technical Accomplishment 2011 (Bosch)
Gao, Goodson et al., Journal of Electronic Materials, 2010
50 μm50 μm
SEM45,000 Cycles
From our New Paper: Barako, Park, Marconnet, Asheghi, Goodson, “Infrared Imaging and Reliability Study of Thermoelectric Modules under Thermal Cycling,” Proceedings of ITHERM, San Diego, May, 2012.
Infrared
80oC
50oC
20oC
4
Research Objectives & Approach
OBJECTIVES
Develop, and assess the impact of, novel interface and material solutions for TEG systems of particular interest for Bosch.
Explore and integrate promising technologies including nanostructured interfaces, filled skutterudites, cold-side microfluidics.
Practical TE characterization including interface effects and thermal cycling.
APPROACH
Multiphysics simulations ranging from atomic to system scale.
Advanced materials development including CNT and metal nanowire TIMS, and high temperature thermoelectrics. Photothermal metrology including Pico/nanosecond, cross-sectional IR. MEMS-based mechanical characterization.
Removable backer
Nanostructured Film
Mid-temperature binder Adhesion layer
Panzer, Goodson, et al., Patent Pending (2007) Hu, Goodson, Fisher, et al., ASME JHT (2006) Won, Kenny, Fisher, et al., CARBON (2012)
Bosch Prototype TEG in exhaust system
5
Area (emphasis)
Specifics Source
Interfaces 100%
Nanostructured films & composites, metallic bonding Ab initio simulations and optimization
Stanford Bosch
Metrology 100%
(ZT)eff including interfaces, thermal cycling High temperature ZT
Stanford USF/NIST
Materials 100%
Filled skutterudites and half Heusler intermetallics Ab initio simulations for high-T optimization
USF Bosch
Durability 50%
In-situ thermal cycling tests, properties Interface analysis through SEM, XRD, EDS
Stanford Bosch
Heat sink 50%
Gas/liquid simulations using ANSYS-Fluent Novel cold HX using microfluidics, vapor venting
Bosch Stanford
System 50%
System specification, multiphysics code Evaluation of research impacts
Bosch Stanford
Research Approach
Additional Faculty & Staff beyond PIs Prof. Mehdi Asheghi, Stanford Mechanical Engineering Dr. Winnie Wong-Ng, NIST Functional Properties Group Dr. Yongkwan Dong, USF Department of Physics Stanford Students: Michael Barako, Yuan Gao (NSF Fellow), Lewis Hom (NSF Fellow), Saniya Leblanc (Sandia Fellow), Woosung Park, Amy Marconnet (NSF Fellow), Sri Lingamneni, and Antoine Durieux
6
7
Approach: Nanostructured Interfaces
Thermal Resistivity (m K / W)1.00.1
Elas
tic M
odul
us (M
Pa)
Greases & Gels
Phase Change Materials
Indium/ Solders
Adhesives
0.01
Nano-gels
Lifetimethermal cyclingLifetimethermal cycling
GOAL
104
103
102
101
Our Latest CNT Data*Our Latest CNT Data*
Copper Nanowire†
Carbon Nanotube
1 μm
‡ xGNPs:25 µm; ~10 nm
Life time thermal cycling
Nan
ostr
uctu
red
Inte
rfac
es
Year 2 DOE-NSF Project
*Gao, Goodson, et al., J. Electronic Materials (2010). Won, Goodson, et al., Carbon (2012a, 2012b) †In collaboration with group of Prof. Fritz Prinz, Stanford
‡www.xgsciences.com 8
Partial Filling – Optimization of Power Factor & Thermal Conductivity
Heavy-ion Filling Yields Lower Thermal Conductivity. Low Valence Filling Facilitates
Optimization of Power Factor.
Nolas, Kaeser, Littleton, Tritt, APL 77, 1855 (2000), Nolas, JAP 79, 4002 (1996) Lamberton, G.S. Nolas, et al APL 80, 598 (2001)
•Skutterudites with partial filling using heavy, low valence “guest” atoms
George S. Nolas Department of Physics,
University of South Florida
•Half-Heusler alloys: small grain-size provides for disordered state
x= 0.3 ; y = 1.5
x= 0.05 ; y=0
x= 0.6 ; y = 2.4
x= 0.23 ; y=0
x= 0.75 ; y = 2.6
x=y=0
x= 0.9 ; y = 2.4
LaxCo4Sb12-ySny
x= 0.3 ; y = 1.5
x= 0.05 ; y=0
x= 0.6 ; y = 2.4
x= 0.23 ; y=0
x= 0.75 ; y = 2.6
x=y=0
x= 0.9 ; y = 2.4
LaxCo4Sb12-ySny
Approach: Bulk TE Materials for Vehicles
9
Predictive computations of TE materials Electronic conductivity Seebeck coefficient Thermal conductivity
Understanding of transport mechanisms on atomic level and composition trends from ab-initio
Composition screening in skutterudites
Several new compositions predicted with higher Seebeck than base-line CoSb3
Trade-offs with conductivity investigated
Collaborative work with Nolas group focuses on Yb and Eu-filled skutterudites
Approach: Materials Computation Seebeck coefficient
experiment*ab-initio
ab-initio τ(E)
Seebeck
-1000
-800
-600
-400
-200
0
200
400
600
800
1000
0.67 0.69 0.71 0.73 0.75 0.77
µ (Ry)
S [µ
V/K]
CoSb3Composition AComposition BComposition CComposition D
300 K
Fermi level
Seebeck
Research and Technology Center North America 10
courtesy N. Singh-Miller, MIT
Computed electrostatic potential
Approach: Interface Optimization Thermal characterization focuses on interface
engagement, nanotube wetting, and stability Mechanical modeling of interfaces allows screening of
compositions to improve thermo-mechanical stability • Chemical reactivity at interfaces considering phase stability • Ab-initio computations and measurements of modulus, CTE • Q1/2011: Analysis of mechanical stresses at interfaces – in-plane
stress limitations using computed and measured CTE • Q1/2011: Cross-section of leg found to be related to the critical
stress, strong implications for materials strength for cost reduction Electronic transport across contacts • Work function and barrier calculations set up and calibrated • Key numerical screening criteria identified: Fermi level and band
offsets, Schottky barrier heights
Technical Accomplishment Q1 2011
Panzer, Goodson et al., Nanoletters (2010)
11
Technical Accomplishment: Publications 1. Barako, Gao, Marconnet, Asheghi, Goodson, “Solder-Bonded Carbon Nanotube Thermal Interface
Materials,” Proceedings of ITHERM, San Diego, May, 2012. 2. Barako, Park, Marconnet, Asheghi, Goodson, “Infrared Imaging and Reliability Study of
Thermoelectric Modules under Thermal Cycling,” ITHERM, San Diego, May, 2012. 3. Park, Barako, Marconnet, Asheghi, Goodson, “Effect of Thermal Cycling on Commercial
Thermoelectric Modules,” ITHERM, San Diego, May, 2012. 4. Marconnet, Motoyama, Barako, Gao, Pozder, Fowler, Ramakrishna, Mortland, Asheghi, Goodson,
“Nanoscale Conformable Coatings for Enhanced Thermal Conduction of Carbon Nanotube Films,” ITHERM, San Diego, May, 2012.
5. Gao, Kodama, Won, Dogbe, Pan, and Goodson, “Inhomogeneous Mechanical Properties of Vertically Aligned Multi-walled Carbon Nanotube Films,” ITHERM, San Diego, May, 2012.
4. Won, Gao, Panzer, Dogbe, Pan, Kenny, Goodson, 2012, "Mechanical Characterization of Aligned Multi-Wall Carbon Nanotube Films," CARBON, Vol. 50, pp 347-355.
5. Marconnet, Yamamoto, Panzer, Wardle, Goodson, 2011, "Thermal Conduction in Aligned Carbon Nanotube-Polymer Nanocomposites with High Packing Density," ACS Nano, Vol. 5, pp. 4818-4825.
6. Marconnet, Panzer, Goodson, “Thermal Conduction Phenomena in Carbon Nanotubes and Related Nanostructured Materials,” invited and submitted, Reviews of Modern Physics.
7. Gao, Kodama, Dogbe, Pan, Goodson, “Nonhomogeneous Mechanical Properties of Vertically Aligned Multi-Walled Carbon Nanotube Films,” CARBON, accepted and in press.
8. Leblanc, Phadke, Kodama, Salleo, Goodson, “Electrothermal Phenomena in Zinc Oxide Nanowires and Contacts,” Applied Physics Letters, accepted and in press.
9. Garg, Bonini, Kozinsky, Marzari, "Role of Disorder and Anharmonicity in the Thermal Conductivity of Silicon-Germanium Alloys: A First-Principles Study,” Physical Review Letters, 106, 045901 (2011).
10. Volja, Kozinsky, Li, Wee, Marzari, Fornari, "Electronic, vibrational and transport properties of pnictogen substituted ternary skutterudites,” submitted to Physical Review B. 12
5 Full-Length Papers Accepted, after review, for ITHERM 2012
7 Archival Journal Papers Appeared or were Accepted
Infrared
80oC
50oC
20oC
ThermoElectric Module/Pellet Nanostructured Interfaces
Gao, Panzer, Goodson et al., J. Electronic Materials, 2010
1 μm
Indium Bond
Thermal Cycling
50 μm50 μm
SEM45,000 Cycles
Thermal Cycling 100 Cycles (100
C, 6min)
Clamping System
Heat Exchanger
View Port
Recirculating Chiller Input/Output
Clamping System
Heat Exchanger
View Port
Recirculating Chiller Input/OutputHeat Exchanger Interior View
Copper Adapters
Heat
Sin
k
Chilled Water3-axis, 3-angle stage3-axis, 3-angle stage
Load CellLinear
Actuator
Infrared Camera
Sample
TE Heater
TE Cooler
Barako, Gao, Marconnet, Asheghi, Goodson, “Solder-Bonded Nanotube Thermal Interface Materials,” to appear in Proceedings of ITHERM, San Diego, May, 2012.
Barako, Park, Marconnet, Asheghi, Goodson, “Infrared Imaging and Reliability Study of Thermoelectric Modules under Thermal Cycling,” to appear in Proceedings of ITHERM, San Diego, May, 2012.
Marconnet, Motoyama, Barako, Gao, Pozder, Fowler, Ramakrishna, Mortland, Asheghi, and Goodson, “Nanoscale Conformable Coatings for Enhanced Thermal Conduction of Carbon Nanotube Films,” to appear in Proceedings of ITHERM, San Diego, May, 2012.
1 μm
Nanofoil Bond
Nanoscale Conformable Coatings for Enhanced Thermal Conduction of CNT Films
Solder-Bonded Nanotube Thermal Interface Materials
2 μm
Electrodeposited Metal Nanowire TIMs
Michael Barako, Unpublished research
High Temperature IR Imaging & Characterization
Technical Accomplishments: Stanford Overview
13
Technical Accomplishment : Interface Characterization on Thermoelectric with Thermal Cycling
0
2
4
6
8
10
12
1 10 100
R" (
m2 K
MW
-1)
N Cycles
R" TotalR" CNT-SubR" CNT-Pt
R”Total R”CNT-Sub + R”CNT R”CNT-Metal
Resistances for 1.5, 2.5, and 40 micron thick CNT films varied between 0.035 and 0.055 cm2 oC/W, with evidence of decreasing engagement with increasing film thickness.
Cycles (30 to 200
C, 6min)
14
Technical Accomplishment: Mechanical Characterization of CNT Films
Maruyama Lab Samples (SWNT), 95 kg/m3
Monano Samples (MWNT), ~30 kg/m3
Wardle Lab Samples/MIT (MWNT), 45 kg/m3
Zipping/Velcro Model
Thickness (μm)
Modulus (MPa)
Density (kg/m3)
SWCNTTop 1 600 110
SWCNTMiddle 0-25 0.5 95
MWCNTTop 0.4 300 40
MWCNTMiddle 0-150 10 29
Polysilicon 5.8-8.7 155e3 2330
Crust Model
With W. Cai Group, Stanford 15
Won, Gao, Goodson, et al., “Mechanical Characterization of Aligned Multi-Wall Carbon Nanotube Films,” CARBON (2012)
Infrared
80oC
50oC
20oC
Technical Accomplishment: Infrared Thermometry Failure Analysis of TE Modules
Before Cycling After 45,000 Cycles Optical
Infrared SEM courtesy of Yuan Gao
100oC A modified Harman technique was developed to measure the TE figure of merit ZT and the electrical resistance R.
70oC
40oC
100 101 102 103 104 1050.54
0.56
0.58
0.6
0.62
0.64
0.66
0.68
Cycles
R [ Ω
]
100
101
102
103
104
1050.5
0.52
0.54
0.56
0.58
0.6
0.62
0.64
ZT
50 μm50 μm
SEM45,000 Cycles
Fracture
Optical
Thermal conductivity and Seebeckcoefficient remain relatively stable
Reduction of electrical conductivity and ZT
16
•Indium (In) foil† is obtained (25 μm thick). •Cleaned and etched using:
1. Acetone 2. Isopropyl alcohol 3. Deionized water 4. Solder flux 5R
† Indium foil by Indium Corp.
Glass
Si
CNT
In
Hotplate
•The foil is compressed between the CNT film and the glass substrate with light pressure •The stack is placed on a hot plate at 180oC for one minute. This melts and bonds the indium to the adjacent surfaces. (Tmelt = 156.6oC)
Cr/Ni/Au
1 μm1 μmIndium
Cr/Ni/Au
CNT Film
CNT
CNT
Glass
Si
1) Metallize Substrates
3) Remove growth Si
GlassCNTSi
2) Bond surfaces
Metal
Si
Barako, Gao, Marconnet, Asheghi, Goodson, “Solder-Bonded Nanotube Thermal Interface Materials,” Proceedings of ITHERM, May, 2012.
Technical Accomplishment: CNT-Indium Bonding
17
Technical Accomplishment: CNT Nanofoil Bonding
•Nanofoil‡ (NF) is a 40μm Al/Ni superlattice which ignites and exothermically alloys to adjacent surfaces •NF is placed between two gold surfaces. Pressure is applied and the NF is ignited, bonding the two surfaces •Sn-plated NF bonds Au surfaces (forming Sn-Au bonds). The resulting intermetallic is stable up to 1000oC
‡ Nanofoil® by Indium Corp.
b) NF alloys to form Sn-Au bonds to adjacent surfaces
Al/Ni Alloy
Sn Plating
CNT Film
5 μm
a) NF is placed between CNT and adjacent surface
Glass
Si
CNT
Al/Ni
Glass
Si
CNT
Al/Ni
50 μm
Sn-Au
Sn-Au
Cr/Ni/Au
Sn
Cr/Ni/Au
Barako, Gao, Marconnet, Asheghi, Goodson, “Solder-Bonded Carbon Nanotube Thermal Interface Materials,” to appear in Proceedings of ITHERM, San Diego, May, 2012.
18
Thermal Boundary Resistance
Thermal Conductivity
Using conservation of energy, Fourier’s Law, and neglecting convection/radiation, we get:
Using the temperature drop at the interface and Fourier’s Law ref
ref
refCNT
dxdTk
TqTR intint ∆
=′′
∆=′′ −
sample
ref
ref
sample
dxdT
dxdT
kk
=
Technical Accomplishment: Pressure-Dependent Infrared Thermometry of CNT Interfaces
Copper Adapters
Heat
Sin
kChilled Water3-axis, 3-angle stage3-axis, 3-angle stage
Load CellLinear
Actuator
Infrared Camera
Sample
TE Heater
TE Cooler
Compressive Measurement Apparatus
0 0.5 1 1.520
30
40
50
60
70
Tem
pera
ture
[o C]
Position [mm]
CNTSi Glass
dT/dx| CNT
dT/dx| glass
ΔTCNT-Glass
ΔTCNT-Silicon
Gro
wth
In
terf
ace
Bond
ed
Inte
rfac
e
19
0 100 200 300 400 500400
600
800
1000
1200
1400
1600
Pressure [kPa]
R' in
t [mm
2 -K/W
]
0 100 200 300 400 5000.4
0.6
0.8
1
1.2
1.4
Pressure [kPa]
k [W
/m/K
]
Intrinsic Thermal Conductivity, kCNT
Thermal Boundary Resistance, R''int Heat Sink
3. Variation in CNT Heights
CNT-CNT Contacts
1. CNT-CNT Boundary Resistance
Individual CNT Thermal Conductance
Marconnet, Wardle, Goodson, et al., ACS Nano (2011)
Heat Source
2. CNT-metal Dry Contact
2. CNT-metal Dry Contact
1. CNT-CNT Boundary Resistance
2. CNT-metal Dry Contact
Technical Accomplishment: Pressure Dependence Before Bonding
20
0 0.5 1 1.520
40
60
80
100
Distance [mm]
Tem
pera
ture
[C]
Glass
NF
CNT For constant q'', the interfacial temperature drop is reduced by an order of magnitude through solder bonding
Indium wets to CNTs and engages more CNTs in conduction, increasing the bulk thermal conductivity of the film
Thermal Boundary Resistance, R''int
Heat Sink
Nanoscale metal-CNT interface resistance (phonons) Partial nanotube engagement
0 200 400 600 800101
102
103
Axial Pressure [kPa]
R" C
NT-
In-F
S [m
m2 K
/W]
Dry ContactUnbonded (In)Bonded (In)Unbonded (NF)Bonded (NF)
21
Technical Accomplishment: Pressure Dependence After Bonding
22
100 101 102 10310-2
10-1
100
101
102
103
104
Length [µm]
The
rmal
Bou
ndar
y R
esis
tanc
e [m
m2 K
W-1
]
If the CNT-Substrate/ metalization interface resistance could be reduced, the intrinsic thermal resistance of the CNT films would outperform solders
102 1
Yang et al. (2002,2004)Tong et al. (2007)Tong et al. (2006)Pal et al. (2008)Son et al. (2008)Xu et al. (2006)Zhang et al. (2008)Xu et al. (2006a)Xu et al. (2006b)Cola et al. (2008)Hodson et al. (2011)Cola et al. (2007)Aradhya et al. (2008)Cross et al. (2010)Panzer et al. (2008)Hu et al. (2006)Gao et al. (2010)Barako et al. (2012)Marconnet et al. (2011)Marconnet et al. (2012)
RCNT-S2 RCNT RS1-CNT
S1 CNT
S2
The blue arrow shows the magnitude of R_S1-CNT and R_CNT-S2). The green data points indicate the magnitude of the intrinsic R_CNT.
23
Technical Accomplishment: Intrinsic Thermal Resistance
102 1
Yang et al. (2002,2004)Tong et al. (2007)Tong et al. (2006)Pal et al. (2008)Son et al. (2008)Xu et al. (2006)Zhang et al. (2008)Xu et al. (2006a)Xu et al. (2006b)Cola et al. (2008)Hodson et al. (2011)Cola et al. (2007)Aradhya et al. (2008)Cross et al. (2010)Panzer et al. (2008)Hu et al. (2006)Gao et al. (2010)Barako et al. (2012)Marconnet et al. (2011)Marconnet et al. (2012)
RCNT-S2 RCNT RS1-CNT
S1 CNT
S2
10%CNT
8.7%CNT
35%CNT
11.5%CNT
0.8%CNT
2%CNT
40%CNT
Length [μm]
104
101
10-2 100 101 102 103
Ther
mal
Bou
ndar
y R
esis
tanc
e (m
m2 K
W-1
)
24
Technical Accomplishment: Impact of CNT Volume Fraction on Intrinsic Thermal Conductivity
Technical Accomplishment: Electrodeposited Metal Nanowire TIMs
Cu
Cu NW Nominal geometry: •Cylindrical NWs •10 μm film thickness •200 nm diameter
a) Polycarbonate membrane is etched to create cylindrical pores
b) Catalyst Pt/Pd is deposited on one side of the membrane
c) Metal is eletrodeposited into the pores
d) Membrane is etched away, leaving freestanding nanowires
10 μm
SEM of copper NW film:
2 μm
Pt/Pd
1 10 1000.1
1
10
100
1,000
Volume Fraction [%]
k [W
m-1
K-1
Thermal Conductivity vs. Volume Fraction
CopperNickel
• Picosecond thermoreflectance technique. • Uncertainty is due to current sample preparation techniques)
In collaboration with Prinz group, Stanford
25
Where electrode was attached for electroplating
5 μm
4 μm
20 μm
No CNTs at this spot in SEM
10 μm
Uniform coating & Grain Size most of the area except where the electrode is attached
Technical Accomplishment 2011: Nanoscale Conformable Coatings for Enhanced Thermal Conduction of CNTs
Marconnet, Motoyama, Barako, Gao, Pozder, Fowler, Ramakrishna, Mortland, Asheghi, and Goodson, “Nanoscale Conformable Coatings for Enhanced Thermal Conduction of Carbon Nanotube Films,” to appear in Proceedings of ITHERM, San Diego, May, 2012.
26
A custom-fabricated vacuum enclosure with integrated heat exchanger will be built to achieve >500
C temperature gradient across a TE sample and facilitate simultaneous electrical and optical measurements.
Technical Accomplishment: High Temperature Infrared Imaging
Clamping System
Heat Exchanger
View Port
Recirculating Chiller Input/Output
Clamping System
Heat Exchanger
View Port
Recirculating Chiller Input/Output
TE Material
Hot Side Cold Side
Clamping System Top down view
IR Camera
250 W
TE Material
Hot Side Cold Side
TE Material
Hot Side Cold Side
Clamping System Top down view
IR Camera
250 W
Liquid Cooling
Hot Side
Hot Side
Cold Side
BiTe
241 °C
406 °CHot Side
Cold Side
BiTe
241 °C
406 °C
ThermalGrease
Copper
CopperTe
mpe
ratu
re (o C
)ΔT Between Top and Bottom Plates (oC)
Top Plate
Bottom Plate
500
400
300
200
0 100 200
IR
TC
IR Image
Heat Exchanger Interior View
Quantum Focus InstrumentsIR MicroscopeQuantum Focus InstrumentsIR Microscope
27
Technical Accomplishment: Bulk TE Materials for Automotive Applications
p-type partially filled and Fe-substituted Skutterudites: YbxCo4-yFeySb12
Double filled and Fe-substituted Skutterudites: BaxYbyCo4-zFezSb12
n- and p-type Half-Heusler alloys
20 40 60 80
Inte
nsity
(a.u
.)
2-Theta (deg.)
calc. PXRD pattern of CoSb3
measured PXRD pattern of Yb0.1Co3FeSb12
measured PXRD pattern of Yb0.4Co3FeSb12
Yb-filled Fe-substituted CoSb3 by solid state reaction
Thermopower of hot pressed Yb0.4Co3FeSb12 ~ 60µV/K at room temp.
20 40 60 80 In
tens
ity (a
.u.)
2-Theta (deg.)
calc. PXRD pattern of ZrNiSn measured PXRD of arc melted ZrNiSn measured PXRD of arc Melted ZrNiSn measured PXRD of arc Melted ZrNiSn
Bi2Te3-alloys for High Resolution IR Thermometry (in collaboration with Marlow Industries, Inc.)
Survey of other material systems with potential for enhanced thermoelectric properties
ZrNiSn by Arc Melting
In collaboration with GM R&D for melt-spun processing in investigating amorphous and fine-grained Half-Heusler alloys.
28
Ytterbium (Yb) partially-filled skutterudites • Partial filling optimizes lattice thermal conductivity reduction1 • Yb intermediate valence in CoSb3 maximizes filler concentration while minimizing
added carriers2
P-type partially filled skutterudites (high temp measurements at NIST & Clemson U.) Amorphous intermetallic alloys3 (in collaboration with General Motors) Bi2Te3-alloys for High Resolution Infra-Red Thermometry (in collaboration with Marlow Ind.) Survey of other material systems with potential for enhanced thermoelectric properties
Technical Accomplishment: Bulk TE Materials for Automotive Applications
Glen Slack initiated PGEC concept with skutterudites:
Fillers should be loosely bonded to the cage-forming atoms. Fillers should have large atomic displacements. Fillers act as independent oscillators (“rattlers”). Interaction of rattlers with the normal modes should lower lattice thermal conductivity. Phonon-scattering centers (“rattlers”) should not greatly affect electronic properties.
filler Co, Fe Sb, As, P
Prof. G. Nolas, Dr. Yongkwan Dong, University of South Florida
1. G.S. Nolas, et al, Phys. Rev. B 58, 164 (1998) 2. G.S. Nolas, et al, Appl. Phys. Lett. 77, 1855 (2000) 3. G.S. Nolas and H.J. Goldsmid, Phys. Stat. Sol. 194, 271 (2002)
29
Thermal Conductivity of Yb-filled Skutterudites
1. G. S. Nolas, et al, Appl. Phys. Lett. 77, 1855 (2000) 2. P. F. Qiu, et al, J. Appl. Phys. 109, 063713 (2011)
300 400 500 600 700 800 9000.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Temperature (K)
Ther
mal
Con
duct
ivity
(W/m
-K)
Yb0.03Co2.70Fe0.91Sb12Yb0.14Co2.51Fe0.61Sb12Yb0.15Co2.65Fe0.80Sb12Yb0.20Co2.58Fe0.91Sb12Yb0.19Co4Sb12 (1)Yb0.94Fe4Sb12.07 (2)
300 400 500 600 700 800 9002.0
2.5
3.0
3.5
4.0
Temperature (K)
Ther
mal
Con
duct
ivity
(W/m
-K)
Ba0.11Yb0.07Co2.80Fe0.96Sb12Ba0.13Yb0.06Co2.66Fe0.80Sb12Ba0.17Yb0.08Co2.83Fe0.88Sb12Ba0.16Yb0.16Co2.87Fe0.91Sb12Ba0.16Yb0.23Co2.1Fe2.3Sb12
300 400 500 600 700 800 9002.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5
Ther
mal
Con
duct
ivity
(W/m
-K)
Temperature (K)
Ba0.31Yb0.04Co2.71Fe1.05Sb12Ba0.26Yb0.10Co2.79Fe0.92Sb12Ba0.30Yb0.16Co2.58Fe1.09Sb12Ba0.11Yb0.03Co4Sb12.07 (1)Ba0.05Yb0.09Co4Sb12.13 (1)Ba0.3Co4Sb12 (2)
p-type Ba + Yb filled Skutterudites
3. X. Shi, et al, Appl. Phys. Lett. 92, 182101 (2008) 4. L. D. Chen, et al, J. Appl. Phys. 90, 1864 (2001)
[3] [3]
[4] [1] [2]
Increasing Yb concentration
Increasing Ba concentration
Partial filling optimizes lattice thermal conductivity reduction
Technical Accomplishment: Bulk TE Materials for Automotive Applications
30
Electronic Transport
• The effect of filling distorts the structure locally. – Soft Sb rings accommodate the distortion.
• Electrons from filler open band gap, while volume expansion closes the gap. – Band gap is also sensitive to local distortion
0.15
0.17
0.19
0.21
0.23
0.25
0.27
0.29
0.31
0.33
0.35
0 0.2 0.4 0.6 0.8 1
filling fraction x
Eg a
t Gam
ma
CaxCo4Sb12 SrxCo4Sb12 BaxCo4Sb12 Na1Co4Sb12 K1Co4Sb12
Ca
Ba
Sr
design rules
0.15
0.17
0.19
0.21
0.23
0.25
0.27
0.29
0.31
0.33
0.35
0 0.2 0.4 0.6 0.8 1
filling fraction x
Eg a
t Gam
ma
CaxCo4Sb12 SrxCo4Sb12 BaxCo4Sb12 Na1Co4Sb12 K1Co4Sb12
Ca
Ba
Sr
design rules
.but
,
CaSrBa
CaSrBa
ωωω >>
>> MMM
Ban
d ga
p (e
V)
Filler vibration is localized and strongly hybridized with Sb atoms. Effect of force constants more important than filler mass.
Technical Accomplishment 2011: Effects of Fillers in skutterudites
Research and Technology Center North America
Filling Factor x
31
Electronic Transport Technical Accomplishment: Computational Composition Engineering
Power Factor
-5.00E-04
0.00E+00
5.00E-04
1.00E-03
1.50E-03
2.00E-03
2.50E-03
3.00E-03
3.50E-03
4.00E-03
0.5 0.55 0.6 0.65 0.7 0.75 0.8 0.85 0.9
µ (Ry)
S2 σ [
W/m
K2 ]
CoSb
CoSnTe
CoSnSe
CoSnS
CoGeSe
CoGeS
Ternary-substituted skutterudites may hold more potential for n-type XCo4B12 B = Sb (Ge,Sn)/ (S,Se,Te)
Higher Seebeck coeffs than CoSb3 Filling of ternary skutterudites weakly affects the Seebeck maxima
Changes the carrier concentration significantly
CoSb3
Seebeck coeff Power factor
Research and Technology Center North America 32
Thermal transport Technical Accomplishment: Nanostructure Design for Thermal Transport
• New method developed for ab-initio thermal conductivity prediction – Grain boundary scattering term included in thermal conductivity
• Effect of scattering noticeable at 300K for 500nm grains • Grain boundary scattering much less effective at high T for skutterudites
Research and Technology Center North America
Smaller grains
d,n)v(
τ phph
q+=
−τ11
33
Accomplishment: Outreach & Engagement Industry Initiatives in Science and Math Education (IISME) – Summer 2011
Undergraduate Thermoelectrics Lab – Fall Quarter
K-12 Educational Outreach – Fall 2011-present
Mentorship of a public high school teacher for summer research experience and curriculum development using thermoelectrics
Stanford’s heat transfer course (ME131A) includes a thermoelectrics laboratory experience. Designed in conjunction with IISME teacher Lab exercise uses infrared microscopy with thermoelectric modules
Designed engineering course which is now taught at a public high school Experienced first-hand application of thermoelectric modules
We are now partnered with a public high school to provide materials and mentors for a TE design lab Introduces high school students to thermoelectric modules and their applications each semester Hands-on design lab to engage students in engineering
34
Collaboration & Coordination
Stanford • Prepares CNTs samples on TE materials • Transport property measurements of CNT-TE pellet combination, thermomechanical reliability tests on interface (300-800 K) •Process development for CNT TIM tape
Bosch • Ab-initio simulations of transport properties of TE materials and interfaces. •System-level simulation and optimization
USF • Develops high-T, high efficiency TE materials • Transport properties (ρ, S and κ) and Hall measurements (10 - 300K) • Structural, morphological and thermal (DTA/TGA) analyses
NIST •Transport properties (ρ, S and κ) and Hall measure- ments (1.8-390K) •Specific heat, Power Factor measurement at 300 K. •Custom-designed precision TE properties measurement system (300 – 1200 K)
1- Interface 2- System-level 3- Durability 4- Materials 5- Heat sink 6- Metrology
1, 3, 4, 6
4, 6
1, 2, 3, 5
Samples Information
35
• Bulk TE Materials: Develop p-type partially/double filled Fe-substituted Skutterudites, n- and p-type half heusler alloys for melt-spun processing, thermal stability tests of materials and joints.
• CNT Thermal Tape Development and Characterization: Bonding extension to 600oC, thermal stability investigation.
• Nanostructured Metal Thermal Interface Materials: Investigate thermal, mechanical, and electrical properties of metal nanowires and meshes, optimization of materials, geometries, and surface treatments for operation at 600oC.
• High-T (ZT)eff Characterization Facility Implementation: Vacuum chamber development with IR transparent window, validation using Bi2Te3-alloys
• Ab-Initio Simulations: Band gap calibration for skutterudite and half-
Heusler families, focussed computatios on phase stability, Seebeck coefficient, and transport properties.
Proposed Future Work
36
• With this award, DOE & NSF are enabling an academic-corporate team to focus on the key practical challenges facing TEG implementation in vehicles: interfaces, system-relevant metrology, and materials compatibility
• We are developing metrology for fundamental properties of nanostructured interfaces, as well as (ZT)eff metrology for half-Heusler and skutterudite thermoelectrics considering interfaces. Simulations include atomistic and ab initio results for TE materials and interfaces, and system & heat exchanger level optimization with the corporate partner.
• Key 2011 results include: (a) process development of CNT tape and several bonding options (Stanford) (b) detailed mechanical characterization of CNT films (Stanford), (c) IR characterization of TE pellets and corresponding interfaces under thermal cycling (Stanford), (d) interface modeling & optimization (Bosch) and (e) process development (arc melting, melt spun) for bulk TE materials (USF)
Summary Slide
37
Technical Backup Slides
• Bonding the CNT films to relevant substrates is a major challenge as not all materials are compatible with the CNT growth procedure.
• Recent progress utilizing a combination of metallizations allows CNT films grown on sacrificial silicon wafers to be successfully transferred to a range of substrates using thin indium foils as binding layers.
• This is a key step towards developing the free standing CNT tape for thermal interface applications.
Diffusion barrier (Ni) 100 nm Surface layer (Au) 150 nm
Adhesion layer (Cr) 100 nm
Substrate (Glass/Quartz)
Surface layer (Au) 150 nm
Solder (In) ~25μm foil
Diffusion barrier (Ni) 100 nm Adhesion layer (Cr) 100 nm
Substrate
CNT Film
Evaporated
Evaporated
Foil ribbon
Technical Accomplishment: CNT Metallization & Bonding Procedure
39
Glass
b) Clean indium foil or apply flux
CNT
1) SAMPLE PREPARATION
Glass
CNT
CNT film bonded to glass, before removal of Si substrate
Si
2) THERMAL BONDING
Hotplate Si
Au/Ni/Cr coated CNT film on Si
a) Evaporate Au/Ni/Cr on CNT and glass substrates
Au/Ni/Cr
Indium
Indium
Technical Accomplishment: CNT Bonding Procedure
SiCNT
GlassIndium
40
Technical Accomplishment: CNT Bonding Procedure
3) CNT FILM RELEASE
CNT bonded to Si
CNT bonded to glass
Original CNT
substrate, with no
CNT remaining CNT
CNT (side is covered with Au from metallization)
20 μm
Indium
Cr/Ni/Au
CNT
Glass
CNT
Glass
CNT
2 μm CNT
Indium
Cr/Ni/Au
20 μm
41
100 1010
0.2
0.4
0.6
0.8
1
Thickness of Thermoelectric [mm]
Z eff/Z
Ideal Solder and Ideal CNT
High Quality CNT
Electrically Insulating Grease
Interface Material
R’’th [W/m2/K]
R’’e [Ω m2]
Solders And
Ideal CNT ~10-7 ~10-12
High Quality CNT ~10-6 ~10-10
Lower Quality CNT ~10-5 ~10-8
Thermally & Electrically Conductive
Grease
~3x10-6 ~3x10-9
Thermal Conductive
Grease ~8x10-6 ~3x10-7
Electrically Insulating
Grease ~8x10-6 >10-5
Relevance: Effect of Interface Resistances on Thermoelectric Device Properties
Using model of Xuan, et al. International Journal of Heat and Mass Transfer 45 (2002). 42
Popular Press
“…Stanford is also working with the National Science Foundation (NSF) on a project with the Department of Energy Partnership on Thermoelectric Devices for Vehicle Applications. Here, the nanotape will facilitate the recovery of electrical power from hot exhaust gases using thermoelectric…”
http://www.eetimes.com/electronics-news/4212401/Nanotape-could-make-solder-pads-obsolete 43
Top Related