Micro- & Nano-Technologies Enabling More Compact, Lightweight Thermoelectric Power Generation & Cooling Systems

2012 3rd Thermoelectrics Applications Workshop San Diego, CA 21 March 2012

Terry J. Hendricks1 , Shankar Krishnan2

1Battelle Memorial Institute Process & Systems Engineering

Columbus, OH

2Pacific Northwest National Laboratory Energy & Efficiency Division

MicroProducts Breakthrough Institute Corvallis, OR

We Sincerely Thank Our Sponsors: John Fairbanks, Gurpreet Singh U.S. Department of Energy

EERE - Office of Vehicle Technologies

2

Transportation Sector Energy Use Light-Duty Passenger Vehicles + Light-Duty Vans/Trucks (SUVs)1 2002: 16.27 Quads of Fuel Usage 2008: 16.4 Quads of Fuel Usage 2002: ~ 5.7 quads/yr exhausted down the tail pipe ~ 5 quads/yr rejected in coolant system

Medium & Heavy-Duty Vehicles1 2002: 5.03 Quads of Fuel Usage 2008: 5.02 Quads of Fuel Usage ~1.5 quads/yr exhausted down the tail pipe 7 to 8 Billion gallons of fuel /year used for Automotive A/C Hybrid Electric Vehicles Move Toward Electrification – Micro, Mild, and Full

Needs for On-board Power Generation Needs for Electric-Driven Cooling

Environmental Impact Reduce Global Warming Refrigerant Use in Automotive A/C Systems R-134 a Leakage - Global Warming Impact - 1,300 times that of carbon dioxide

1Transportation Energy Data Book, 2010, Edition 29, U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, Vehicles Technology Program. ORNL-6985, Oak Ridge National Laboratory, Oak Ridge, Tennessee. http://cta.ornl.gov/data/index.shtml.

Motivation - Energy & The Environment

Motivation - Energy & The Environment 7 to 8 Billion gallons of fuel /year used for Automotive A/C

~6 % of Light Duty Vehicle Fuel Use; Releases approximately 62-70 Billion kg of CO2 / year

Current Centralized A/C Systems Require 3.5 to 5 kW of Energy in Each Vehicle

Zonal or Distributed Thermoelectric Heating, Ventilation and Air Conditioning (HVAC) Requires ~ 630 Watts Cool Driver Only and ~ 2.7 kW Cool 5 Occupants

In Heating Mode, TE much more Efficient (COPheat ~ 2.3 > 1) Current Vehicular Air Conditioner (A/C) uses Compressed R134-a

Refrigerant Gas Each Vehicle Leaks ~70 g/year R134-a R134-a Has 1300 times the “Greenhouse Gas Effect” as Carbon Dioxide

(CO2) ~18.2 Million Metric Tons of CO2 equivalent/year from personal vehicles in

the US from operating air conditioners (does not include accident release) U.S. EPA Estimtates ~58 Million Metric Tons of CO2 equivalent/year from

transportation sector (primarily R-134a) http://epa.gov/climatechange/emissions/usinventoryreport.html

Thermoelectric Systems in Automobiles DOE Sees a Vision and the Potential

Significant Waste Heat Available in

Vehicle Exhaust Streams

Need for Electrically-

Driven Heating/Cooling

Greenhouse Gas & Global Warming Reduction

Advanced Thermoelectric System Design

5

Qh,out

Qc,in

Power In Power In

T h

N

T c

Compartment Cooling Flow

Hot - Side Rejection Flow

T col d

Cold Side Heat Exchanger

Hot Side Heat Exchanger

T amb , m h Ambient

•

T cabin , m c •

P

I

Power In Power In

T h

N

T c

Compartment Cooling Flow

Hot - Side Rejection Flow

T col d

Cold Side Heat Exchanger

Hot Side Heat Exchanger

T amb , m h Ambient

• T amb , m h Ambient

•

T cabin , m c • T cabin , m c •

P

I

Thermoelectric Heating/Cooling Low-Temperature Systems

Generally Do Cascading Rather Than Segmenting to Achieve Large ∆T

TE Cooling Heat Exchanger / TE Device Integration Requirements

250 300 350 400 450 500 550 6000

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Cooling Capacity [W]

Max

imum

CO

P

←1421(W/kg)

←1294(W/kg)

←1173(W/kg)

←1057(W/kg)

←Th =340(K)

←Th =337(K)

←Th =334(K)

←Th =331(K)

←Th =328(K)

←Th =325(K)

←Th =322(K)

←Th =319(K)

←Th =316(K)

Tcold=280 (K)Tcold=270 (K)Tcold=260 (K)Constant Thot

Distributed Cooling Systems

Typical COP – Cooling Capacity – Power / Mass Relationship Shown Distributed TE Cooling Systems

Create Lower Heat Flows per Unit Higher COP’s Lower Power / Mass

Generally Right Directions for Automotive Distributed Cooling

p-type NPB BixSb2-xTe3 * n-type Bi2Te3 – Bi2Se3

* Poudel, B., Hao, Q.H., Ma, Y., Lan, Y., Minnich, A. Yu, B., Yan, X., Wang, D., Muto, A., Vashaee, D., Chen, X., Liu, J., Dresselhaus, M.S., Chen, G., Ren, Z., 2008, “High- Thermoelectric Performance of Nanostructured Bismuth Antimony Telluride Bulk Alloys,” Sciencexpress, 10.1126, science.1156446.

UAc = 40 W/K Tcabin = 298 K

100 150 200 250 300 350 400 4500

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Qc Per Mass[W/kg]

Max

imum

CO

P

←T h=3

40(K

)

←T h=3

37(K

)

←T h=3

34(K

)

←T h=3

31(K

)←T h

=328

(K)

←T h=3

25(K

)←T h

=322

(K)←T h

=319

(K)←T h

=316

(K)

Tcold=280 (K)Tcold=270 (K)Tcold=260 (K)Constant Thot

p-type NPB BixSb2-xTe3 n-type Bi2Te3 – Bi2Se3

UAc = 40 W/K

Tcabin = 298 K

Preferred TE Design Regime

TE Cooling Heat Exchanger / TE Device Integration Requirements

0.4 0.5 0.6 0.7 0.8 0.9 10

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Qc Per Area [W/cm2]

Max

imum

CO

P

←T h=3

40(K

)

←T h=3

37(K

)

←T h=3

34(K

)

←T h=3

31(K

)←T h

=328

(K)

←T h=3

25(K

)←T h

=322

(K)

←T h=3

19(K

)←T h=3

16(K

)

Tcold=280 (K)Tcold=270 (K)Tcold=260 (K)Constant Thot

Preferred TE Design Regime

p-type NPB BixSb2-xTe3

n-type Bi2Te3 – Bi2Se3 UAc = 40 W/K

Tcabin = 298 K

Distributed TE Cooling Systems Generally Move Into Regions of: Higher COP’s Higher Specific Cooling Capacity (Compact, Lightweight Systems) Higher Heat Fluxes (Higher Heat Transfer Coefficients)

Generally Higher Performance Heat Exchanger Systems Required

Distributed Cooling Systems Distributed Cooling Systems

MicroTechnology in Distributed TE HVAC Systems DOE Project in Advanced TE HVAC

Systems for Automobiles Zonal Climate Control for Thermal Comfort Compact Microtechnology Heat Exchangers

Reduce Weight & Volume Low Cost Manufacturing

Coupled with Compact TE HVAC Systems Wicking Systems for Water Management

Leveraging Nano-Scale Coating Technology

Significant Microtechnology Cost Modeling Cost Sensitivities Identified Low-Cost Manufacturing Avenues Being Developed Sensitivities to Production Volumes Material and Process Cost Drivers

8

Hybrid / PHEV Vehicles

Nano-Scale Coatings

Cost Modeling Approach

PNNL Developing High-Performance Microtechnology Heat Transfer Technologies

TE Cooling / Heating Automotive Distributed HVAC Systems A Number of Microtechnology Designs Are Being

Investigated An Example of One Such Design Is Presented Here

Established geometry, heat transfer and pressure drop characteristics

Semi-empirical modeling & COMSOL Modeling

1

1.5

2

2.5

3

3.5

4

4.5

5

5.5

0 0.0005 0.001 0.0015 0.002 0.0025 0.003 0.0035

Cell Pitch (l ), m

Ther

mal

Res

ista

nce,

K/W

Hexagon

Square

Triangle

Rectangle (aspect ratio = 4)

Rmin-hex = 2.2 K/W

Rmin-sqr = 2.7 K/W

Rmin-tri = 3.9 K/W

Rmin-rect = 1.7 K/W

Process Based Cost Modeling Bottom-Up Approach to Estimating Cost of Goods Sold (COGS)

Based on Operation of Virtual Manufacturing Line – Breaks Down Cost by Unit Process

Cost of Goods Sold (COGS)

Variable Costs Fixed Costs

Capital Equipment Maintenance

Facilities/Buildings

Direct Labor Direct Materials

Indirect Materials Utilities Process-Based Cost

Model Algorithm

Model Outputs

Start with Process Flow and Associated Equipment Set

Process COGS vs. Volume and Pareto Cost Sensitivity ID Cost Drivers

Not Included Overhead & Profit

Insurance Taxes

Inventory Management Accounting Marketing

Sales

Capital Equipment Labor

Materials Energy

Facilities Maintenance

Cost Elements

Model Inputs

Process Flow

Unit Process 1 Unit Process 2 Unit Process 3

Etc.

Process to Process Comparisons Define Fabrication Toolbox

Inform R&D Agenda

Contact: Steven.Leith@pnl.gov

7

7.5

8

8.5

9

9.5

10

10.5

11

0.4 0.5 0.6 0.7 0.8 0.9 1

Effectiveness

Hea

t T

ran

sfer

Den

sity

(Q/V

olm

etal

), k

W/L

t.

Copper

Aluminum

Effect of Channel Aspect Ratio on Flux Density

4 cm

2 cm

8 cm

4 cm

2 cm

8 cm

Cost Vs Performance

0

50

100

150

200

250

300

350

400

450

500

$0.00

$2.00

$4.00

$6.00

$8.00

$10.00

$12.00

$14.00

$16.00

$18.00

0.88 0.83 0.82 0.80 0.79 0.68 0.67 0.62

Dev

ice

Mas

s (g

)

Mat

eria

l Cos

t pe

r D

evic

e ($

)

Heat Exchanger Effectiveness

Cu Folded Fin HTX Cost vs. Performance

Material Cost per Device Mass per Device

Layered Rectangular Honeycomb Designs Fine Pitch Design (#1)

Higher Fin Density Higher Performance (ε = 0.88) Somewhat Higher Cost

Coarse Pitch Design (#2) Lower Fin Density Slightly Lower Performance (ε = 0.81) Lower Cost

Manufacturability, Process and Cost Drivers Identified

0.1

0.12

0.14

0.16

0.18

0.2

0.22

0.24

0.26

0.28

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9

Dim

ensi

onle

ss G

eom

etric

Sca

le

Relative Density

Feasible

Not Feasible

Manufacturability (Folded Fin)

Air-Side Heat Transfer Experiments

Velocity probe

Porous screens to produce plug flow

Honeycomb

Thermocouple

Strip Heater Air HX

∆P

CDA

Rotameter

Q gravity

Design #1: Fine Pitch – Tested Performance & Correlation with Models

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0

Air Flow Rate, CFM

Pres

sure

Dro

p, p

si

Measured Predicted

0

2

4

6

8

10

12

14

0 1000 2000 3000 4000 5000 6000 7000 8000

Reynolds Number (ReH)

Fric

tion

Fact

or, f

Measured Predicted

Design #2: Coarse Pitch – Tested Performance & Correlations with Models

0

1

2

3

4

5

6

7

8

9

0 1000 2000 3000 4000 5000 6000 7000 8000

Reynolds Number (ReH)

Fric

tion

Fact

or, f

Measured Predicted

Cost Comparison

Metrics Design #1 Design #2

$/effectiveness $18.55 $13.66

$/W $0.055 $0.041

$/kg $47.32 $39.05

Observations & Findings Accounted for braze thickness and separator plate thickness based on variation in

heat exchanger stack height Measured thermal resistance came out to be higher than predicted thermal

resistance Friction factor & pressure drop correlated well with fluid dynamic models Model thermal predictions may be conservative (lower performance bound).

Higher performance bound will be ~ 3% lower than the predicted thermal resistance

Discrepancy between thermal model and measurements could be due to Geometric variation in the built device Delaminated layers in heat exchangers Measurement errors Modeling assumptions compared to actual fabricated devices

Summary Microtechnology Thermal Systems Required to Enable Compact, Light-

weight TE Systems TE Power Generation – Energy Recovery and Portable Power Applications TE Cooling / Heating – Distributed Automotive Applications

Microtechnology Thermal Systems Successfully Integrating into TE Systems

Process-Based Cost Modeling Has Identified High- and Low-Cost Manufacturing Pathways, Processes, and Materials High-Cost Designs Differentiated from Low-Cost Designs Performance vs. Cost Clearly Delineated

System Performance Modeling Integrated with Process-Based Cost Modeling Powerful Combination Identifies Low-Cost, Manufacturable Microtechnology

Designs Prioritizes R&D Investment Plans & Enables Business Decisions

18

Questions & Discussion

We are What We Repeatedly do. Excellence, Then, is not an Act, But a Habit. Aristotle

Thank you for your time and interest

ADDITIONAL BACKUP TOPICS

20

System Analysis Capabilities & Characteristics

21

System-Level Couples Design Analysis of: Hot Side Heat Exchanger Performance TE Device Performance Cold Side Heat Exchanger Performance

Single or Segmented TE Material Legs Accounts for Hot/Cold Thermal Resistances Accounts for Electrical Contact Resistances Optimum Heat Exchanger / TE Design Parameters Determined Simultaneously Maximum Efficiency or COP & Maximum Power or

Cooling Capacity Designs Are Possible Off-Nominal & Variable Condition Performance Analysis