D l hi SOFC St k D l t U d tDelphi SOFC Stack Development ... library/events/2009/seca...D l hi SOFC...

D l hi SOFC St k D l t U d tDelphi SOFC Stack Development Update

Rick KerrRick KerrCell and Stack Development

Pitt b h PA

SECA Conference 7/15/09

Pittsburgh, PA2009 SECA Annual Review Meeting

OutlineOut e• Summary Highlights• Manufacturing progress

F ili i– Facilities– Manufacturing System Design– Variation reduction– Use of engineering tools

• Gen 4 stack progress– Cell performance stability– Design status and performance projection– Stack development activitiesStack development activities

• Gen 3 stack performance– Syngas– Thermal cycling

• Leveraging SECA activities– Truck APU– Vibration testing– NUWC

SECA Conference 7/15/09 2

NUWC

Summary of Year’s Performance Highlights for SECA Coal Based System Stack DevelopmentBased System Stack Development

• Added 33K ft2 cell development facility• Scaled up from 105 cm2 (active area) cells to 403 cm2Scaled up from 105 cm (active area) cells to 403 cm• Fabricated > 5,000 cells and > 100 stacks of various

configurations• Completed Gen 4 stack design• Achieved stack power density of 500 mW/cm2 at 570

mA/cm2 with mean cell voltage of 0 87 VmA/cm2, with mean cell voltage of 0.87 V• Implemented low cost cell, interconnect, and balance of

stack fabrication processesp• Achieved >100 thermal cycles on 30-cell stack


Delphi SOFC Development CentersDelphi SOFC Development CentersSE Michigan

• Cell FabricationRochester, New York

• Stack Build and DevelopmentCell Fabrication• Manufacturing Development• Electrochemical Testing

Material Characterization and

Stack Build and Development• System Integration & Perform. Lab• Reformer Development Lab

Catalyst Durability Lab• Material Characterization and Analysis

• 33K ft2, added Q3 2008C

• Catalyst Durability Lab• BOP Development

• Capacity up to 8 MW in 2011


Equipment for Cell Development and Fabricationqu p e t o Ce e e op e t a d ab cat o


Delphi SOFC Manufacturing Plan- Mfg System Design (MSD)es g ( S )

• Customer Requirements• Project Requirements• Detailed Mfg Sequence ChartDetailed Mfg Sequence Chart• Define Operator Work Content• Balance Equipment Utilization• Balance Operator Utilization• Balance Operator Utilization• Balance Material Flow


Manufacturing System Developmenta u actu g Syste e e op e tTechnical Requirements:

Product Definition:

Process Flow Diagram (PFD) Process Definition:

Failure Mode Analysis:Failure Mode Analysis:PFMEA

Control Strategy:Process Flow DiagramPrioritized RPN PFMEA gy

Process Control Plan,Error proofing

Manufacturing:Process Monitoring

Process Control PlanWork InstructionsProcess Monitoring FormCell and Stack Build DatabasesStandard Cell Documentation


Process Monitoring,Standard Cell & EOP, Layered Audit

Standard Cell DocumentationGate Charts

Generation 3.2 30-Cell Stack

Data from 30-cell Gen 3.2 stack with improved interconnect and cathode Very good cell to cell voltage distribution (0.02 Volts)

MG735C676 - 30plus3 Date: 1/29/2009Fuel: 48.5%H2-48.5%N2-3%H20 Flows: 89.8(A), 151(C)

Stack Cell Voltages for Polarization Test at 60 Amps

1 00

1.10

0.70

0.80

0.90

1.00

V)

0.40

0.50

0.60

Cel

l Vol

tage

(V

0.10

0.20

0.30


0.00

C01 C02 C03 C04 C05 C06 C07 C08 C09 C10 C11 C12 C13 C14 C15 C16 C17 C18 C19 C20 C21 C22 C23 C24 C25 C26 C27 C28 C29 C30

Cell

Generation 3.2 30-Cell Stack Run Chart

Very consistent performance in stack to stack builds

Cell Voltage by Stack at 60amps

/200

8/2

008

/200

8/2

008

/200

8/2

008

/200

8/2

008

2008

/200

8/2

008

/200

8/2

008

/200

8/2

008

/200

8/2

008

/200

8/2

008

/200

8/2

008

/200

8/2

008

/200

80/

2008

5/20

088/

2008

2/20

088/

2008

0/20

08/2

008

2/20

08/2

008

/200

920

09/2

009

/200

920

09/2

009

/200

9/2

009

2009

2009

/200

9/2

009

2009

2009

/200

9/2

009

2009

2009

Build Date

AverageRange

0.80

0.85

0.60

0.70

4/10

/4/

12/

4/14

/4/

15/

4/17

/4/

24/

4/29

/5/

29/

6/6/

26/

13/

6/24

/6/

26/

7/22

/7/

23/

7/24

/7/

25/

7/31

/7/

31/

8/14

/9/

16/

9/19

/9/

24/

10/1

/10

/8/

10/1

010

/15

10/1

810

/22

10/2

810

/30

11/7

/11

/12

12/5

/1/

29/

2/4/

22/

12/

2/13

/3/

4/2

3/26

/3/

20/

3/25

/4/

1/2

4/9/

24/

17/

4/30

/5/

5/2

5/7/

25/

14/

5/15

/6/

4/2

6/6/

2

0.65

0.70

0.75

l Vol

tage

Ave

rage

0.30

0.40

0.50

ell V

olta

ge R

ange

Contamination on interconnect

Red X® verification

0 50

0.55

0.60

Cel

0 00

0.10

0.20

Ceon interconnect for contamination

on interconnect


0.50

653

657

651

658

659

660

661

665

671

672

662

666

680

681

683

684

690

691

685

700

701

702

717

703

704

705

706

707

718

719

720

733

721

676

738

746

747

754

751

756

761

757

758

770

778

762

776

699

780

766

777

Stack

0.00

Red X® Problem Solving Tools For Stack Issuesed ob e So g oo s o Stac ssues♦ Problem solving tools like Red X® are being utilized in stack manufacturing♦ Lessons learned are being implemented for Gen 4 stack design and manufacturing

♦ Example: Find & eliminate the Red X® driving high cell voltage range

Cell Voltage by Stack at 60amps AverageCell Voltage by Stack at 60amps

0 850.70

4/10

/200

8

4/12

/200

8

4/14

/200

8

4/15

/200

8

4/17

/200

8

4/24

/200

8

4/29

/200

8

5/29

/200

8

6/6/

2008

6/13

/200

8

6/24

/200

8

6/26

/200

8

7/22

/200

8

7/23

/200

8

7/24

/200

8

7/25

/200

8

7/31

/200

8

7/31

/200

8

Build Date

AverageRange

0.70

0.75

0.80

0.85

ge A

vera

ge

0.40

0.50

0.60

ge R

ange

Did not meet voltage range specs

0.55

0.60

0.65

Cel

l Vol

tag

0.10

0.20

0.30

Cel

l Vol

tag range specs

SECA Conference 7/15/09 10Red X® is a registered term of Shainin LLC

0.50

653

657

651

658

659

660

661

665

671

672

662

666

680

681

683

684

690

691

Stack

0.00

Red X® Problem Definition TreeTM

Stack Other Area

High cell voltage range

Other Performance

Characteristics

P = 0.14V M = 0.013V

Event Defect

First time quality issue: IV

Durability issue

Feature

Components Assembly Process

performance

Repeating unit to repeating unit

Other strategies

Cassette Cell Cathode Interconnect

Paste Interconnect Component

Coating Contamination

Other

•Red X® was identified as contamination in the interconnect


•Red X® was identified as contamination in the interconnect coating from a specific supplier

Problem Definition TreeTM is a registered trademark of Shainin LLC

Optimization of Cathode PerformanceOptimization of Cathode Performance

• Robust Engineering tool used for performance optimization– Engineering optimization strategy for development of new technologies in

product and process design» Application of Taguchi methods for optimized performance» Concentrates on identification of ideal function(s) and choosing best

nominal values of design parameters for optimized performance reliability at lowest cost

L18 design with a “Nominal the Best” strategy was chosen for minimizing– L18 design with a Nominal the Best strategy was chosen for minimizing cathode polarization resistance on half cells

» L18 design allows for 1 control factor at 2 different levels and 7 control factors at 3 different levels

» Experiment results in 18 different treatment combinations» Duplicates were fabricated from each treatment combination» All 18 pairs were measured at two distinct noise conditions (72 runs)


p ( )

Optimization of Cathode PerformanceOptimization of Cathode Performance

• Outcome of Robust Engineering Project– Reduced mean cathode polarization resistance by 70% compared

to the standard cathode design– Confirmed improved electrochemical performance with button cell

teststests– Full-sized cells fabricated and to be built into stacks for evaluation


Typical Gen 3.2 30-Cell Stack Performance

Power: 1.53 kW (486 mW/cm2) @ 60A (570 mA per cm2), 25.5 V (0.85V / cell avg), fuel 48.5% H2 ,3% H2O, rest N2, 750 – 800oC

MG735C676 - 30plus3 Date: 1/29/2009Fuel: 48.5%H2-48.5%N2-3%H20 Flows: 89.8(A), 151(C)

Stack Voltage and Power Density for Polarization Test35 600

25

30

e (V

) 400

500

W/c

m2)

10

15

20

Stac

k Vo

ltage

200

300

Pow

er D

ensi

ty (m

0

5

10

0

100


0 10 20 30 40 50 60 70

Current (Amps)Stack Voltage (V) Power Density (mW/cm^2) Poly. (Stack Voltage (V)) Poly. (Power Density (mW/cm^2))

Gen 3.2 5-Cell Stack – Maximum Power Data from recent 5-cell Gen 3.2 stack Produced 500 mW/cm2 @ 60A (570 mA per cm2), 0.87V / cell avg, fuel 48.5% H2 ,3% H2O,

rest N2 750 – 800oCrest N2, 750 800 CMax Power density: 765 mW/cm2 @ 100 Amps (952 mA per cm2), 0.80V / cell avg, fuel

48.5% H2, 3% H2O, rest N2, 750 – 825oCMG735C793 Date: 6/23/2009Fuel: 48.5%H2-48.5%N2-3%H20

Stack Voltage and Power Density for Polarization Test6 900

5

6

700

800

900

m2)

3

4

tack

Vol

tage

(V)

400

500

600

r Den

sity

(mW

/cm

1

2St

100

200

300

Pow

er


00 20 40 60 80 100 120

Current (Amps)

0

Button Cell Accelerated Testing (at Kettering Univ.)utto Ce cce e ated est g (at ette g U )

Cell Power Stability at 1.5 A/cm2, 825CFuel 50 H2 - 50 N2

1

1.2

Ran out of H2

0.6

0.8

Cel

l Vol

tage

, V

0.2

0.4C

00 200 400 600 800 1000 1200

Time, Hours


Cathode Stability (at Carnegie Mellon University)Cat ode Stab ty (at Ca eg e e o U e s ty)Powder

Bulk Cathode Composition via TEM

20

25

PowderFabricated Cells530 inlet ROI 1

FabricatedCell 15

omic

%

530 inlet ROI 1530 inlet ROI 2530 Outlet ROI 4

5

10Ato

0La Sr La/Sr Fe Co Fe/Co A/B


5-cell stack (#530) tested 3500+ hrs at 0.8 V/cell on 28%H2 – 30%CO – 6%H2O – 2.5 ppmv S

Gen 4 Cell Development for Coal Based SystemsGe Ce e e op e t o Coa ased Syste s


Gen 4 Stack Design for Coal Based SystemsGe Stac es g o Coa ased Syste s

♦ Delphi is developing its Generation 4 stack as the module for meeting stationary and transportation application requirements y

♦ Key stack features are:♦ Laser welded cassette repeating unit configuration♦ Stamped metallic cassette componentsp p♦ Stamped interconnects♦ Low cost coatings♦ Low cost, conventionally processed balance of stack components, y p p

About 4x active area of Gen3L t i t t / ti th G 3Lower cost interconnects/coatings than Gen3High volume production processes5 kW core building blocks

SECA Conference 7/15/09 19Gen 3 stack

Gen 4 stack

Gen 4 Stack Development Status Ge Stac e e op e t Status♦ Gen 4 stack design is complete♦ Design concepting completed in Q1, 2009♦ Gen 4 stack components are being fabricated for stack build♦ Process development and optimization ongoing to fabricate cells, cassettes,

and stacksP t t l f t i t t k b ilt d t t d i Q4 2008 t fi♦ Pre-prototype large footprint stacks built and tested in Q4, 2008 to confirm design concept


Laser welding of SOFC parts

Interconnect Development – Materials & Coatingste co ect e e op e t ate a s & Coat gs♦ Base material is ferritic stainless steel alloy ♦ Multiple coatings under evaluation

♦ Low cost conventionally processed coating demonstrated good stability in stack test♦ Low cost, conventionally processed coating demonstrated good stability in stack test♦ Graph shows typical repeating unit performance in a stack with coating tested for

durability (Fuel = 48.5 % H2, 3% H2O, rest N2, constant current of 570 mA per cm2)♦ Further development is ongoing

MG735C624

0.9

1

1.1

1.2

0 4

0.5

0.6

0.7

0.8

Volta

ge (V

olts

)

0

0.1

0.2

0.3

0.4

0 00 500 00 1000 00 1500 00 2000 00 2500 00 3000 00


0.00 500.00 1000.00 1500.00 2000.00 2500.00 3000.00

Time Hrs

Collaboration with UTRCCo abo at o t U C♦ Delphi is working with UTRC on materials and process development of base

alloys and coatings♦ Multiple alloys and coatings are being evaluated♦ Multiple alloys and coatings are being evaluated♦ Low cost, manufacturable coating application processes are being developed


Gen 3.2 Stack Durability with SyngasGe 3 Stac u ab ty t Sy gas♦ 10-cell Gen 3.2 stack on syngas (24%H2 – 25% CO – 5% CO2 – 7% H2O – Bal N2)

at 35 amps and 750C ♦ Initial voltage drop observed in the first 200 hours due to sulfur

Mi i l d d ti i th l t 1300 h♦ Minimal degradation in the last 1300 hoursMG735C727

Fuel: Reformed Diesel

1.1

1.2

0.7

0.8

0.9

1

age

(V)

0 3

0.4

0.5

0.6

Ave

rage

Vol

ta

0

0.1

0.2

0.3

0.00 200.00 400.00 600.00 800.00 1000.00 1200.00 1400.00


Time Hrs

Gen3.2 30-Cell Stack Thermal CyclingGe 3 30 Ce Stac e a Cyc g•Gen 3.2 30-cell stack being thermal cycled

•10 hour cycle from 50oC to operating temperature to 50oC•Performance evaluated at each thermal cycle•Performance evaluated at each thermal cycle•Constant current load of 30 Amps at operating temperature•Fuel of 48.5% H2, 3% H2O, rest N2

700

800

300

400

500

600

Tem

pera

ture

(Deg

C)

0

100

200

0 00 2 00 4 00 6 00 8 00 10 00 12 00


0.00 2.00 4.00 6.00 8.00 10.00 12.00

Time Hrs

Gen 3.2 30-Cell Stack – Thermal CyclingGen 3.2 30 Cell Stack Thermal Cycling♦ 95 full thermal cycles completed with minimal degradation (less than 0.5%) ♦ Test continuing to failure♦ Robustness to thermal cycling allows for coal-based system shutdowns,♦ Robustness to thermal cycling allows for coal based system shutdowns,

maintenance, unplanned outages, etc.Stack 699 Thermal Cycle Test Summary

Current: 0.33 A/cm2, Fuel: 48.5% H2, 3% H2O, bal N21

0.7

0.8

0.9

l Sta

ck

0.4

0.5

0.6

an V

olta

ge fo

r 30-

Cel

0.1

0.2

0.3Mea


00 10 20 30 40 50 60 70 80 90 100

Thermal Cycle Number

SOFC Market OpportunitiesSOFC Market Opportunities

SOFC modularity and fuel flexibility allow for early

Clean Coal Power PlantAdvanced Power Systems

flexibility allow for early market opportunities

Residential PowerStationary CHP Power UnitsCommercial Power

Stationary Power Units

Milit Recreational Vehicles


MilitaryAuxiliary & Mobile Power Units

Recreational VehiclesAuxiliary Power UnitsHeavy Duty Trucks

Auxiliary Power Units

Daimler Truck Field Demoa e uc e d e o

•System with 2 Gen 3.2 stacks provided power during period of dynamic load changes

Next Gen APUUses Gen 4 Stack


Gen 3.2 30-Cell Stack Vibration Testing• Test to Failure using Step Over Stress strategy to assess stack reliability

– Mil Std 810G Method 514.6 Annex C, “Truck and Transportation Over US Highways”

g

g y– Ambient temperature random vibration in 3 distinct axes– Performance evaluated after each axis of testing

• Successfully completed 3.75 million miles equiv. of vibration (5 APU lifetimes)• Test is continuing to an additional level of stress

Vertical

Lateral

Longitude


Stack Testing at NUWC

♦ Delphi, SECA (US Dept. of Energy), and the US Navy are collaborating for SOFC technology evaluation for undersea Naval applications

g

♦ System application represents extreme conditions for SOFC environment♦ A 30-cell Delphi Gen 3.2 stack was integrated with Steam Reformer, High

Temp Blower, and CO2 Sorbent Bed (bread board system) at NUWC p , 2 ( y )♦ S-8 Fuel and pure oxygen were the only reactants fed into system♦ Performance achieved

♦ 1 kW Power♦ 1 kW Power♦ 75% S-8 Fuel Utilization♦ 90% Oxygen Utilization♦ 50% Efficiency*♦ 50% Efficiency

* SOFC Power output / S 8 Lower Heating Value


* SOFC Power output / S-8 Lower Heating ValueAcknowledgement: Acknowledgement: Alan Burke, Ph.D.; Louis G. Carreiro, Ph.D.Alan Burke, Ph.D.; Louis G. Carreiro, Ph.D.Naval Undersea Warfare Center (NUWC), Division NewportNaval Undersea Warfare Center (NUWC), Division Newport

AcknowledgementsAcknowledgements


D l hi SOFC St k D l t U d tDelphi SOFC Stack Development ... library/events/2009/seca...D l hi SOFC...

Documents

Transcript of D l hi SOFC St k D l t U d tDelphi SOFC Stack Development ... library/events/2009/seca...D l hi SOFC...