DOE/NASA/0290-02 NASA CR-175047 · DOE/NASA/0290-02 NASA CR-175047 WAESD-TR-85-0030 FINAL REPORT...

V

DOE/NASA/0290-02NASA CR-175047WAESD-TR-85-0030

FINAL REPORT FOR THE SECOND LOGICAL UNIT OF WORKCONTRACT PERIOD: MAY 1983 - MAY 1984

J. M. FeretWestinghouse Electric CorporationAdvanced Energy Systems DivisionPittsburgh, PA 15236-0864

AUGUST 1986

Prepared forNATIONAL AERONAUTICS AND SPACE ADMINISTRATIONLewis Research CenterUnder Contract DEN 3-290

For

u.s. DEPARTMENT ©F EMEROY

(HAS A-CB-1750.47,) -. GAS: COOLED fOEL CELL,SYSTEMS TECHNOLOGY. D E V E L O P M E N T .Final,Beport, Hay-1SB3:'- Hay 1984* (Hestinghouse .Electric Corp.)/: 282. p CSCL; 1QAr

M 8 6-3-19 84

. Onclas, 4 3 6 7 9

https://ntrs.nasa.gov/search.jsp?R=19860022512 2020-05-19T11:36:26+00:00Z

DISCLAIMER

This report was prepared as an account of work sponsored by an agency of theUnited States Government. Neither the United States Government nor anyagency thereof, nor any of their employees, makes any warranty, expressed orimplied, or assumes any legal liability or responsibility for the accuracy,completeness, or usefulness of any information, apparatus, product, orprocess disclosed, or represents that its use would not infringe privatelyowned rights. Reference herein to any specific commercial product, process,or service by trade name, trademark, manufacturer, or otherwise, does notnecessarily constitute or imply its endorsement, recommendation, or favoringby the United States Government or any agency thereof. The views andopinions of authors expressed herein do not necessarily state or reflectthose of the United States Government or any agency thereof.

Printed in the United States of America

Available from

National Technical Information ServiceU.S. Department of Commerce5285 Port Royal RoadSpringfield, VA 22161

NTIS price codes'

Printed copy: A04Microfiche copy: A01

'Codes are used for pricing all publications. The code is determined bythe number of pages in the publication. Information pertaining to thepricing codes can be found in the current issues of the followingpublications, which are generally available in most libraries: EnergyResearch Abstracts (ERA); Government Reports Announcements and Index (GRA andI); Scientific and Technical Abstract Reports (STAR); and publication,NTIS-PR-360 available from NTIS at the above address.

WAESD-TR-85-0030

DOE/NASA/0290-02

NASA CR-175047

GAS COOLED FUEL CELL SYSTEMS TECHNOLOGY DEVELOPMENT

Final Report for the Second Logical Unit of WorkContract Period: May 1983 - May 1984

0. M. Feret

Westinghouse Electric Corporation

Advanced Energy Systems Division

Pittsburgh, PA 15236-0864

August 1986

Prepared for

NATIONAL AERONAUTICS AND SPACE ADMINISTRATION

Lewis Research CenterUnder Contract DEN 3-290

forU.S. DEPARTMENT OF ENERGY

Morgantown Energy Technology Center

Morgantown, West Virginia 26505

Under Interagency Agreement DE-AI21-80ET17088

WAESD-TR-85-0030

ACKNOWLEDGEMENTS

The following personnel contributed to this Final Report

M. S. barrett0. E. DickeyG. G. EliaJ. M. FeretL. L. FranceN. HainesR. R. HolmanT. K. Houghtaling0. L. KellyH. 0. KulikowskiA. K. Kushk. F. LampeO.K. LanceM. T. LeB. L. Pierced. F. PierreR. RoseyE. F. SaladnaF. R. SpurrierS. VenkateshD. A. WisemanB. M. NoodleM. K. WrightJ. F. Zippay

3593E-3

ii

WAESD-TR-85-0030

TABLE OF CONTENTS

Section Title Page

Table of Contents iii

List of Figures v

List of Tables ix

List of Acronyms xiii

1.0 SUMMARY 1-1

2.0 SYSTEMS ENGINEERING 2-1

2.1 Development System Requirements 2-1

2.1.1 System Analyses 2-12.1.2 Fuel Cell, Fuel Processing, Power Conditioning,

Rotating Equipment, and Instrumentation andControl Systems Design Requirements 2-4

2.1.3 Fuel Cell Requirements 2-17

. 2.2 Fuel Cell System Requirements 2-32

2.2.1 25 kW Short Stack Fuel Cell Test Specification 2-32.2.2.2 Fuel Cell Hardware Test Specifications 2-322.2.3 Fuel Cell Manufacturing Process Specifications 2-35

2.3 Systems Integration 2-35

2.3.1 Fuel Cell System Design 2-352.3.2 Systems Interface Requirements and Control 2-44

3.0 FUEL CELL DEVELOPMENT AND TEST 3-1

3.1 Subscale Fuel Cell Development 3-1

, 3.1.1 Design and Assembly Status 3-13.1.2 Subscale Testing 3-2

3.2 Nine Cell Stack Development . 3-47

3.2.1 Design Developments 3-473.2.2 Acid Transport Test Fixture 3-603.2.3 Acid Delivery System 3-623.2.4 Nine Cell Stack Testing , 3-62

m

WAESD-TR-85-0030

TABLE OF CONTENTS (CONTINUED)

Section Title

3.3 10 kW Fuel Cell Stack Development 3-88

3.3.1 Development Status 3-883.3.2 10 kW Stack Design Description 3-90

3.4 25 kW Short Stack Development 3-94


3.5 100 kW Full Stack Development 3-94


3.6 375 kW Module Development 3-97

3.6.1 Development Status 3-973.6.2 Module Preliminary Design Description 3-1003.6.3 Acid Supply System 3-1053.6.4 Module Maintenance 3-1063.6.5 Cartridge Replacement 3-106

3.7 Fuel Cell.Materials Characterization and Testing 3-108

3.7.1 Raw Material Specifications 3-1083.7.2 Plate Materials 3-1083.7.3 Seal Materials 3-1253.7.4 Electrode Materials 3-1393.7.5 Other Stack Materials 3-165

3.8 Advanced Fuel Cell Development 3-166

3.8.1 Alternate Catalysts 3-1683.8.2 Alternate Catalyst Support 3-1713.8.3 Electrode Manufacturing 3-1783.8.4 Cell Resistance 3-1793.8.5 Bipolar Plate Corrosion 3-1833.8.6 Bipolar Plate Coatings 3-1893.8.7 Acid Management • 3-1913.8.8 Impurity Effects 3-198

4.0 ERC NINE CELL STACK TEST FACILITIES DEVELOPMENT 4-1

5.0 REFERENCES - 5-1

iv

WAESD-TR-85-0030

LIST OF FIGURES

Figure No. Title Page

1-1 Westinghouse PAFC Program Work Breakdown Structure 1-2

2-1 Rotating Equipment System Arrangement 2-20

2-2 Instrumentation and Control System FunctionalBlock Diagram 2-25

2-3 Fuel Cell System Preliminary Design 2-39

2-4 3-Dimensional ANSYS Finite Element Model 2-43

2-5 Fuel Cell System Interface Control Drawing 2-45

2-6 Fuel Processing System Interface Control Drawing 2-46

2-7 Rotating Equipment System Interface Control Drawing 2-47

2-8 Power Conditioning System Interface Control Drawing 2-48

2-9 Instrumentation and Control System Interface ControlDrawing 2-49

3-1 Fuel Cell Performance 3-21

3-2 Polarization Data for Subscale Cells DAS-011 and -012 .,3-26

3-3 Cell Voltage as a Function of Current at 185°C 3-34

.3-.4 Cell Voltage as a Function of Current at 190°C 3-35

3-5 Cell Voltage as a Function of Temperature 3-36

3-6 Tafel Plots for Oxygen Reduction on a Fuel Cell Cathodeat 1.0, 2.4, and 4.8 atm 3-38

3-7 Variation of the Measured and Predicted Cell VoltageGains with Operating Pressure 3-39

3-8 Effect of Pressure on Tafel Slopes 3-40

3-9 Voltage Gain at Various Pressures 3-41

3-10 Variation of IR-Free Cell Voltage with Time . 3-45

3-11 Variation of Internal Resistances with Time 3-46

WAESD-TR-85-0030

Figure No.

3-12

3-13

3-14

3-15

3-16

3-17

3-18

3-19

3-20

3-21

3-22

3-23

3-24

:3-25

3-26 .

3-27

3-28

3-29

3-30

3-31

LIST OF, FIGURES (CONT'D)

Title:

Nine Cell Stack Cell Edge Seal Configurations

Fuel Cell Resilient Seal Arrangement

Fuel Cell Resilient Seal Arrangement at Acid Groove

Nine Cell Stack Assembly

Phosphoric Acid Fuel Cell Five Cell Stack

Acid Transport Test Fixture

Acid Delivery System - Nine Cell System

Westinghouse Nine Cell Stack Performance - 1983

Stack W-009-10 Polarization, All Cells

Stack W-009-10 Polarization at Beginning of Phase Ic,Five Center Cells

Stack W-009-10 Polarization at End of Phase Ic, FiveCenter Cells

Stack W-009-10 Polarization at Beginning of Phase Ic,Four Outer Cells

Stack W-009-10 Polarization at End of Phase Ic, FourOuter Cells

Stack Assembly Fixture

PAFC 10 kW Stack Test Assembly

25 kW Stack Design

Fuel Cell Module

Module Flowpaths

Flexural Strength Vs. Heat Treatment Temperature

Electrical Resistivity Vs. Heat Treatment Temperature

Page

3-49

3-51

3-52

3-56

3-59

3-61

3-63

3-65

3-82

3-83

3-84

3-85

3-86

3-91

3-93

3-95

3-98

3-101

3-114

3-116

vi

WAESDrIR-85-0030

LIST OF FIGURES (CONT'D)

Figure Mo. Title Page

3-32 Corrosion Current Vs. Corrosion Potential for HeatTreated Plate Materials 3-120

3-33 Microstructures of Stackpole MF958 Plate MaterialAfter Corrosion Testing 3-122

3-34 Corrosion Tafel Plots for Heat Treated Bipolar PlateMaterials 3-124

3-35 X-Ray Diffraction Profiles of Resin 29-703 HeatTreated at Temperatures Indicated 3-127

3-36 Creep Test Configuration 3-130

3-37 Compressive Stress Vs. Seal Strain for Kalrez SealConfiguration at Operating Temperature 3-133

3-38 Compressive Stress Vs. Seal Strain for Viton/VitonCement and Kalrez/Viton Cement Comparative Seal Tests 3-135

3-39 Exploded View of a Half Cell Assembly 3-142

3-40 Tafel Plots for Oxygen Reduction on an AlternateCatalyst and Two Baseline Cathodes 3-144

3-41 Tafel Plots for Oxygen Reduction on Two AlternateCatalysts and Two Baseline Cathodes 3-146

3-42 Polarization Curves for Hydrogen Oxidation on anAlternate Catalyst and a Baseline Anode 3-146

3-43 Strain Comparison Between Standard and Stacks W009-09and -11 Cell Components 3-150

3-44 X-Ray Diffraction Profiles of Standard Stackpole vs.Kureha Papers 3-156

3-45 Comprehensive Stress-Strain Data for Kureha E715Paper at Ambient Temperature . 3-157

3-46 Kureha Paper Stress-Strain Data at 55°C 3-159

3-47 Kureha Paper Stress-Strain Data at 199°C 3-160

3-48 DSC Thermogram of TFE-6C 3-161

vil

WAESD-TR-85-0030

Figure No.

3-49

3-50

3-51

3-52

3-53

3-54

3-55

3-56

3-57

4-1

LIST OF FIGURES (CONT'D)

Title

DSC Thermogram of an Unsintered Catalyst Layer

Lifegraphs of Developmental Catalyst Cells ECPS-05thru -08

Lifegraph of Cells 3009 and 3038

Lifegraph of Stack 560

Modified Resistance Model Compared with Old Model

Corrosion Behavior of Stackpole Backing Paper

Cell Seal Design

AICM-Anode Backing for Nine Cell Stack E-009-01

Initial Performance Data for Stack E-009-01

Schematic Flow Diagram for Nine Cell Stack Pressurized

j>age

3-163

3-170

3-173

3-176

3-181

3-188

3-192

3-196

3-197

Facility (SE-1) 4-2

4-2 Current Nine Cell Pressurized Recirculation Design andProposed Single Pass Cooling Design 4-7

vi.ii

WAESD-TR-85-0030

LIST OF TABLES

Table No. Title Page

1-1 Phosphoric Acid Fuel Cell Technology Development ProgramSecond Logical Unit Tasks 1-3

2-1 Performance Summary 2-3

2-2 Fuel Cell Anode Supply Composition Requirements 2-6

2-3 FPS Top Level Design Requirements 2-8

2-4 PCS Performance Requirements 2-9

2-5 PCS Input from Fuel Cell System 2-11

2-6 PCS Output to Utility Equipment System 2-12

2-7 PCS Equipment Design Requirements 2-13

2-8 RES Operating Statepoints 2-15

2-9 RES Power Summary 2-18

2-10 RES Design Requirements 2-19

2-11 IC$ Top Level Design Requirements 2-21

2-12 ICS Input/Output Requirements 2-22

2-13 ICS Equipment Design Requirements 2-23

2-14 Key PAFC Design Parameters, Values, and Constraints 2-28

2-15 PAFC Development Goals 2-31

2-16 100 kW Stack Performance and Design Requirements 2-33

2-17 25 kW Stack Performance Requirements 2-34

2-18 Subscale Test Specifications Issued 2-36

2-19 Nine Cell Stack Test Article Data 2-37

2-20 Test Plan Sequence ' 2-38

3-1 Cells Tested Under Various Test Plans 3-3

WAESD-TR-85-0030

LIST OF TABLES (CONT'D)

Table No. Title jagg

3-2 Summary of Subscale Cells Tested 3-4

3-3 Summary of Subscale Test Data 3-5

3-4 Subscale Testing Status 3-8

3-5 Test 2x2-002 Rev. 0 Assessment of Five Electrode Variables 3-10

3-6 Test 2x2-004 Rev. 2 and Rev. 4 Pressurized ElectrodeCharacterization 3-13

3-7 Test 2x2-005 Rev. 2 & Rev. 3 Baseline Performance andFacility Comparison . 3-16

3-8 Test 2x2-006 Rev. 0 Alternate Catalyst Comparison andElectrode Evaluation 3-17

3-9 Test 2x2-007 Rev. 0 Thru Rev. 2 Alternate Backing PaperComparison and Process Evaluation 3-19

3-10 TS 2x2-008 Rev. 0 Electrode Bonding and SinteringProcess Development Tests 3-23

3-11 Test 2x2-011 Rev. 0, Rev. 1, and Rev. 2 - ManufacturingProcess and Acid Transport Variable Evaluation 3-24

3-12 Test 2x2-013 Rev. 0 Thru Rev. 5 Electrode QualityVerification Test 3-28

3-13 Test 2x2-014 Rev. 0 and Rev. 1 Pressurized ElectrodeEndurance Verification 3-31

3-14 Test 2x2-016 Non-Electrode Related Catalyst Poisons 3-32

3-15 Comparison of Measured and Predicted Cell Voltage Gainsfrom Pressurization 3-43

3-16 Stack Assembly Comparison 3-54

3-17 Nine Cell Stack Manifold Seal Test Summary 3-57

3-18 Summary of Nine Cell Stack Performance 3-64

3-19 Cell Voltage and Resistances at Rated Power (SR6 Flow) 3-69

WAESD-TR-85-0030


Table No. Title Page

3-20 Steady State Performance at Quarter Power Point 3-70

3-21 Test Times for Stack W-009-08 3-74

3-22 Summary of Stack W-009-08 Data 3-75

3-23 Stack W-009-10 Steady State Performance at Full PowerPoint with SRG 3-79

3-24 Stack W-009-10 Effect of Humidification on CellPerformance (Phase Ib) . 3-82

3-25 Stack W-009-10 Tabulation of Mapping Results 3-87

3-26 Stack W-009-11 Steady State Performance at Full PowerPoint with \\2 3-89

3-27 10 kW Metal Stack Manifold Seal Test Results 3-92

3-28 Fuel Cell Module Design Parameters 3-99

3-29 Module Physical Characteristics 3-103

3-30 Module External Interfaces 3-104

3-31 Developmental Plate Compositions 3-110

3-32 In-Plane Electrical Resistivity of Development PlatesAfter 900°C Heat Treatment 3-111

3-33 Mercury Porosimetry Data for Development of Plates HeatTreated to 900°C 3-112

3-34 Mercury Porosimetry Data for Selected DevelopmentalPlates Heat Treated to 900°C 3-113

3-35 Flexural Strength of Bipolar Plate Material 3-117

3-36 In-Plane Electrical Resistivity of Bipolar Plate Materials 3-118

3-37 Corrosion Rate of Candidate Bipolar Plate Materials 3-121

3-38 Crystallite Sizes and Lattice Spacings of Heat TreatedVARCUM 29-703 Resin. . 3-126

3-39 Summary of Load Vs. Deformation Compression Test Data 3-132

XI

WAESD-TR-85-0030


Table No, Title Page

3-40 Leaching Effect oh Several As-Received FluoroelastomerMaterials 3-138

3-41 Second Leaching Effect on Selected FluoroelastomerMaterials 3-140

3-42 Leaching Effect on Heat Treated Fluoroelastomer Materials 3-141

3-43 Helium Bubble Pressure of Various Thickness SiC Layers 3-152

3-44 Elevated Temperature Bubble Pressure of Typical Cathode 3-153

3-45 Backing Paper Comparison Summary 3-155

3-46 Melting Point Data for TFE-6C 3-162

3-47 Ion Chromatography of Various Samples of Phosphoric Acid 3-167

3-48 Endurance Test of Development Catalyst Cells ECPS-05thru - 08 3-169

3-49 Stress Testing of Developmental Catalyst Cells ST-01thru -06 3-172

3-50 Terminal Performance Stability of Stack 431, mV 3-175

3-51 Performance of Alternate Catalyst Support Cells, mV 3-177

3-52 Effect of Matrix Thickness on Cell Performance andResistance 3-182

3-53 Effect of Matrix Thickness on Performance and Resistanceof Stack E-009-04 at Pressure 3-184

' 3-54 Stack E-009-02 Open Circuit Voltages, mV 3-193

3-55 Summary of Acid Additions to Nine Cell Stacks 3-199

3-56 Total and Free Sulfur Contents of Carbon Supports 3-200

4-1 Specifications of the FMCS for 12x17 Stack PressurizedTest Facility 4-4

4-2 Verified Operating Conditions 4-5

xii

WAESD-TR-85-0030

LIST OF ACRONYMS

AICM Acid Inventory Control MemberBOU Beginning of UseCOE Cost of ElectricityCT Combustion TurbineDHE Dynamic Hydrogen ElectrodeDSC Differential Scanning CalorimetryEOU End of UseEPRI Electric Power Research InstituteERC Energy Research CorporationETU Electrolyte Take-UpPCS Fuel Cell SystemFEP Fluorinated Ethylene-PropyleneFHS Fuel Handling SystemFMEA Failure Mode Control SystemFPS Fuel Processing SystemI CD Interface Control DrawingICS Instrumentation and Control SystemLVDT Linear Variable Displacement TransducerO&M Operating and MaintenanceOCV Open Circuit VoltagePAFC Phosphoric Acid Fuel CellPCS Power Conditioning SystemPDS Purchasing Department SpecificationsPES PolyethersulfonePFA PerfluoroalkoxyethyleneRES Rotating Equipment SystemRHE Reversible Hydrogen ElectrodeS6S Steam Generation SystemSR6 Simulated Reformer GasTAG Technical Assessment GuideTBD To Be DeterminedTGA Thermogravimetric AnalysisUES Utility Equipment SystemUPP Utility Power PlantW_ WestinghouseWBS Work Breakdown StructureWDPF Westinghouse Distributed Process Family

. xiii

WAESD-TR-85-0030

1.0 SUMMARY

This report addresses the work performed during the Second Logical Unit ofWork of a multi-year program designed to develop a phosphoric acid fuel cell(PAFC) for electric utility power plant application. The Second Logical Unitof Work, which covers the period May 14, 1983 through May 13, 1984, wasfunded by the U.S. Department of Energy, Office of Fossil Energy, MorgantownEnergy Technology Center, and managed by the NASA Lewis Research Center. ThePAFC Technology Development Program efforts were performed by a teamcomprised primarily of personnel from the Westinghouse Advanced EnergySystems Division and the Energy Research Corporation (a division of St. JoeMinerals Corporation), in parallel and integrated with the Utility PowerPlant Program which was sponsored by Westinghouse, a host utility, theElectric Power Research Institute, the Empire State Electric Energy ResearchCorporation, and other participating organizations.

1.1 Scope of Work

The major elements of work to be performed in the multi-year program havebeen integrated into a Work Breakdown Structure (WBS) as shown in Figure1-1. This WBS establishes the basic framework for all contractually requiredeffort and thereby provides a uniform structure for planning, assessingprogress achieved, controlling costs, etc.

The Second Logical Unit of Work comprised twenty WBS tasks (see Table 1-1)for which specific objectives, scheduler and cost plans, and deliverables

have been established.

1.2 Technical Objectives

The primary objectives established to be achieved in the Second Logical Unitof the program were:

1-1

.

si 1

-

—

=::

ll

!!!W

lMT

IltA

NIO

US

PO

Wt

ft rtA

Nt

SV

SIIW

S

AC

CI

SS

OR

T'f

llC

TR

tCA

lS

VS

TIM

S

•r,|{i

&Hi

! j

j*1

i*ll

•*if"s

hp

sug|

?

z

e

8

8

D

K

S

!

8

i

i

1

1111111111

CO

MirnuC

'iO

MM

AM

AU

CM

tMr

J

I!

ii|S

1 r^T

lCM

NO

lOC

VIjm

oof

or C

ONCI

Pn

M

OD

ULE

S

%if5

ii

i

Hip.*.

1

tUQ

SC

ALC

/

Tc

rAC

aiiY

i —„

*I— -r

|ix^S..'?•• ••• i

. 52

f|

o

9

5e

5S

—

98

9o

9S

o

8

n

1

Jn58

:

J

n5

-n

I-

1

l

I1

1i

1

"!

INlM

inO

I

NO

I»lt

1jk

tttiin

|

•* ^

1

M

J|

•Is•**£5*

Js£

»U

ti e

cu

„ T

UT

* A

CU

ITY

1 — 1

lljji

_"*J.r~i

lijLT

3

3

33

3

88

O

O

8

SS

ni9)8

=l

J

-I

J

J

no§

J

1

1

1

i!«i

t1

|

Mil

|

AM

JA

OJ]

Jt

j

*r»

§£i

i!i —

1I

L-T.

1

M|y

r- —h;"s

U

o

sQ

3

oS

SK

8

n

co-con 1 1

n33=

J

no

z

_ "

JIM&

TA

ltA

TIO

MA

MD

CH

tCB

bU

T

|

.{!Hi

.!,

rri

f

!i|—

ill

}

s

s8

S

"

n

!.3

n9

.

M

5"

" i "» i

MIS

CiL

LA

MfO

US

fV^lf

MS

«

F~

!i

l"T

if

r

1L'T

•S -8?S= IS ! H! S fPH WAESD-TR-85-0030

ORIGINAL PAGE' ISOF POOR QUALITY

sf .

i.•*J 'ui.

ooco

<os L!1 DO

5-LJ o

2

roS.O)0

0.

u.o.

i5 c/1S 3

I *t c

!| ^

8 | I

T r — i r i r T -M •II ' ' . 1 ' 8 5 •—

! |S I! 1 j §i pi ! 1

J L. .Lj Lt.U L,L ' ^

1 S1

n iii ii pbj _;._

1-2

WAESD-TR-85-0030

TABLE 1-1

PHOSPHORIC ACID FUEL CELL TECHNOLOGY DEVELOPMENT PROGRAMSECOND LOGICAL UNIT TASKS

WBS TASKNO. DESCRIPTION

1101 Program Management

1102-01-100 System Analyses

1102-01-200 Fuel Cell, Fuel Processing, Power Conditioning,Rotating Equipment, and Instrumentation and ControlSystems Design Requirements

1102-01-300 Fuel Cell Requirements

1102-02-200 100 kW Full Stack Fuel Cell Test Specification

1102-02-300 25 kW Short Stack Fuel Cell Test Specification

1102-02-400 Fuel Cell Hardware Test Specifications

1102-02-500 Fuel Cell Manufacturing Processes Specifications

1102-03-100 Fuel Cell System Design

1102-03-300 Systems Interface Requirements and Control

1103-01 Subscale Fuel Cell Development

1103-02 9-Cell Stack Development

1103-03 10 kW Fuel Cell Stack Development

1103-04 Fuel Cell Materials Characterization and Testing

1103-05 25 kW Short Stack Development

1103-06 100 kW Full Stack Development

1103-07 375 kW Module Development

1103-08 Advanced Fuel Cell Development

1104-01 Subscale 2x2 Inch Cell and 9-Cell TestingFacilities Development

1108-01 Technology Proof-of-Concept Module Fuel Cell System

1-3

WAESD-TR-85-0030

• Establish the final system-Oevel requirements for the Fuel Cell, FuelProcessing, Power Conditioning, Rotating Equipment, and Instrumen-tation and Control Systems.

• Develop the Fuel Cell System preliminary design.

• Reduce the total fuel cell resistance to less than 0.2 mft.

• Increase the average cell beginning of use performance at pressure onthe order of 50 mV, and improve the long term performance of the fuelcell to a 2 mV loss per 1000 hours of operation.

• Develop an optimum stack startup procedure from ambient conditions tothe nominal operating conditions: 480 kPa (70 psia), 190°C (374°F),and 325 mA/cm (300 A/ft2) operating on reformate hydrogen (83percept utilization) and air (50 percent utilization), and changes

. above/below these nominal conditions.

t Develop performance repeatable and cost effective manufacturingprocesses for the electrodes, matrices, and cooling and bipolarplates.

Considerable progress was achieved relative to each of these objectives andis summarized below:

• System level design requirements were established for the Fuel Cell,Fuel Processing, Power Conditioning, Rotating Equipment, andInstrumentation and Control Systems. This was accomplished by theperformance of numerous studies and analyses which addressed theareas of system trade-offs, performance, operation, and maintenance.The specifications prepared for these systems include functional,operational, interface, and quality assurance requirements based onthe selected plant design parameters: 1) 7.5 MW gross electricaloutput (7.2 MW net), 2) a heat rate of less than 9500 kJ/kWh

WAESD-TR-85-0030

(9000 BTU/kWh), 3) full power operation with naphtha as the backupfuel, and 4) automatic control during all power operations.

A number of modifications were incorporated into the PAFC power plantdesign which resulted in improved performance, or a simplerconfiguration. These requirements, studies, and modifications arediscussed in detail in Section 2.1.

• The Fuel Cell System preliminary design was fully developed. Acomplete design description, system maintenance requirements, andstructural analysis are presented in Sections 2.2 and 2.3. Thesystem interface requirements were identified and are documented inSection 2.3.2.

• Technology in the fuel cell development program has progressedsignificantly. Subscale cell testing verified stack design andmanufacturing processes, as well as provided a base for stackscaling. The various test plans and results are reported in Section3.1. .

• The nine cell stack design has been further developed and refined.The primary areas of improvement were electrode edge sealing, acidtransport within the cells, and manifold sealing. Additionally, newmanufacturing, startup, and operating procedures resulted asubstantial 48 mV increase in cell performance, while a furtherincrease of 30 mV is projected due to thinner matrices. Completedetails and testing results are given in Section 3.2.

• Based on the achievements in the nine cell stack area, the 10 kW, 25kW, and 100 kW stacks were re-designed, incorporating a new invertedcell configuration. This configuration provides improved acidtransport. The development status and design description of each ofthese stacks are presented in Sections 3.3 through 3.5.

1-5

WAESD-TR-85-0030

• The 375 kW module design, necessary to the utility plant application,was updated to incorporate advancements achieved by the fuel celltechnology program. The major changes included addition of acontinuous acid supply system, addition of electric heaters tomaintain higher internal temperatures, and the inverted cell stackdesign. A complete design description, including maintenance andreplacement schedules, of the 375 kW module is given in Section 3.6.

f The characterization of various raw materials utilized in themanufacture of repeating cell components was addressed.Specifically, plate resistivity, porosimetry, and corrosion, sealmaterials, electrode materials, and other stack materials wereevaluated. The comprehensive results of this effort are reported inSection 3.7.

• The advanced fuel cell development effort achieved considerablesuccess by defining performance goals and identifying problem areas.A Performance Improvement Work Plan was formulated and implemented toimprove cell performance and reduce variance. The areas primarilyaddressed were alternate catalyst evaluation and support,manufacturing, cell resistance, plate coatings and corrosion, acidmanagement, and impurity effects. A complete summary of this effortis presented in Section 3.8.

• A new presurized test facility, SE-1, was constructed and checked outat ERC. This facility, which is capable of accommodating a fullsized (12x17 in.) nine cell stack, improved upon past ERC facilitiesby incorporating advanced methods of measurement and control. Thisfacility is described in Section 4.0.

The aggregate of this effort resulted in the revision of manufacturing andstack assembly procedures, stack design, and system specifications. These,along with the associated drawings, were updated and released.

1-6

WAESD-TR-85-0030

2.0 SYSTEMS ENGINEERING (WBS 1102)

The objectives of this task were to develop the final system-level designrequirements for the Fuel Cell, Fuel Processor, Power Conditioning, RotatingEquipment, and the Instrumentation and Control Systems (FCS, FPS, PCS, RES,and ICS, respectively). The following sections summarize the results ofanalyses and design studies performed.

2.1 Development System Requirements (WBS 1102-01)

This section summarizes system analysis, technology assessments, and theresulting design requirements for the FCS, FPS, PCS, RES, and ICS.

2.1.1 System Analyses (WBS 1102-01-100)

The primary objectives of the system analysis effort were to perform systemtrade-off studies, performance analyses, and operational studies to establishpreliminary system level design requirements.

These analyses identified the functional and conceptual interfacerequirements in conjunction with the Task 1102-04-100 Plant Systems Analyseseffort. System level trade-off studies that considered heat rate, powerlevel, cost of electricity (COE), operating characteristics, and developmentrisk were performed to select system level design requirements for each ofthese systems for a prototype power plant that produces 7.5 MW DC with aheat rate goal of less than 9500 kO/kwh (9000 Btu/kWh). The effect uponplant performance and operating characteristics for startup, steady-state,shutdown, and malfunction conditions associated with these systems wasdetermined by utilizing appropriate steady-state, transient, and

controllability analysis models.

A number of design modifications were incorporated into the PAFC powerplant. These modifications include:

2-1

WAESD-TR-85-0030

• Provide addition of 100 cells per module.

• Incorporate the functions of the FPS turbocompressor into the RES.

t Make plant operation at a lower pressure level at part powercompatible with RES recommended speed.

§ Provide part of the steam required for reforming in the FPS.

• Design for full power beginning of use with naphtha fuel.

The above items resulted in improved performance or a simpler configuration.

2.1.1.1 Performance and Operational Studies

Four normal power plant operating points were established in the FirstLogical Unit. These are: (1) full power beginning of use (BOU), (2) fullpower end of use (EOU), (3) part power BOU, and (4) part power (EOU). Thefull power BOU point is identified as the baseline prototype power plantdesign. The EOU condition is defined at a cell voltage degradation of 80mV. Though the remaining fuel cell operating parameters (temperature,pressure, current density, etc.) remain constant at BOU and EOU for full andpart power modes respectively, there is a difference in the operatingparameters between full and part power. The full power BOU operatingconditions for the fuel cell are 190°C (374°F), 480 kPa (70 psia), and300 mA/cm2 (280 A/ft2) producing a cell voltage of 0.690 V (0.610 V atEOU). The corresponding part power conditions are 190°C (374°F), 207 kPa(30 psia), and 75 mA/cm2 (70 A/ft2), which result in a BOU cell voltageof 0.752 V (0.672 V at EOU). The part power operating pressure represents apressure level compatible with the RES speed. Part power operation atconstant temperature maintains the pressure level of the steam generators.The performance summary for these operating points is given in Table 2-1,where the heat rates have been reduced from one to four percent below theFirst Logical Unit values.

2-2

WAESD-TR-85-0030

TABLE 2-1

PERFORMANCE SUMMARY

FULL POWER PART POWER

Cell Voltage (mV)Pressure, kPa (psia)Temperature, °C (°F)

2Current Density, mA/cm

(A/ft2)

Gross DC, kW

Gross AC, kW

Net RES, kW

Parasitic Losses, kW

BOU

690

480 (70)

190 (374)

300 (280)

7500

7200

98

202

EOU

610

480 (70)

190 (374)

300 (280)

6625

6360

103

202

BOU

752

207 (30)

190 (374)

75 (70)

2042

1878

-233

64

EOU

672

207 (30)

190 (374)

75 (70)

1825

1679

-203

66

Net AC, kW 7096 6243 1581 1410

Thermal Input, GG/h(106 Btu/h)

Plant Heat Rate,kJ/kWh (Btu/kWh)

62.6 (59.39) 62.6 (59.39) 16.1 (15.23) 16.1 (15.23)

8875 (8370) 10,028 (9510) 10,155 (9630) 11,388 (10,800)

2-3

WAESD-TR-85-0030

2.1.2 Fuel Cell, Fuel Processing, Power Conditioning, Rotating Equipment,and Instrumentation and Control Systems Design Requirements(WBS 1102-01-200)

Updated system design requirements were prepared for the FCS, FPS, PCS, RES,and ICS. These specifications include functional, operation, interface, andquality assurance requirements.

2.1.2.1 Fuel Cell System

The primary function of the FCS is to produce DC electric power and thermalenergy using air-cooled 375 kW fuel cell modules. The fuel cell modulestransform the chemical energy from the fuel and oxidant reaction,(hydrogen-richfeed gas supplied by the FPS and oxygen in the form of pressurized air suppliedby the RES) into electrical energy.

Design requirements established for the FCS include:

t Provide a gross electrical output of 7.5 MW DC at full power operatingconditions at BOU.

• Control electrical output based on current demand. The design basis(BOU full power) fuel cell module current output is 325 amps.

• Provide hydrogen utilization consistent with fuel cell performancecharacteristics and FPS performance requirements. The design hydrogenutilization is 83 percent.

t Equalize fuel and process air flows to the individual fuel cell modulesand provide for on-line checking for cross leakage between the anodeand the cathode.

• Design for stable operation at power operation conditions and forresponse to the transient operating conditions within the appropriatetime frames.

2-4

" ^ W A E S D - T R - 8 5 - 0 0 3 ( T

a Maintain fuel cell module temperatures greater than 107°C (225°F)during STANDBY.

t Maintain the fuel cell temperature between 38°C (100°F) and 54°C(130°F) during shipping, handling, maintenance, installation andnon-power operating conditions to prevent solidification of theinstalled phosphoric acid.

t Maintain the air pressure level between 205 and 480 kPa (30 - 70psia). The system design shall include the capability to isolate anyfuel cell module or modules without causing significant pressuredifferentials across individual cells.

• Maintain the fuel cells in a dry inert atmosphere having a dew point of-23°C (10°F) or less during shipping, handling, installation, andnon-power operating conditions to limit phosphoric acid water-content.

a Design for operation on the fuel cell anode supply composition defined

in Table 2-2.

• Design to allow replenishment of fuel cell phosphoric acid duringoperation and shutdown.

2.1.2.2 Fuel Processing System

The primary function of the FPS is to convert steam and hydrocarbon fuel(either natural gas or naphtha) to a hydrogen-rich gas for use in the fuelcells. Fuel is provided by the Fuel Handling System (FHS) as natural gas or asliquid naphtha. Fuel cell anode exhaust is used to provide thermal energy forthe reforming process. Steam and feedwater are supplied by the Steam Genera-tion System (SGS). Pressurized air for combustion is provided by the RES.

Secondary functions of the FPS are to reduce impurities (hydrogen sulfide,etc.) from the fuel stream which are detrimental to fuel cell performance, to

2-5

WAESD-TR-85-0030

TABLE 2-2

FUEL CELL ANODE SUPPLY COMPOSITION REQUIREMENTS

Component

HydrogenCarbon DioxideCarbon MonoxideWater-MethaneNitrogen

Composition (volume fraction)Design Basis Acceptable.Limit

.742

.182

.008

.044

.014

.010

> 0.50

diluent

< 0.01< 0.06< 0.05diluent

impurities

H2SCOSOlefinsHigher Hydrocarbons (C*)NH3Cl

Metal IonsTar/OilsPartiCulates

Composition .(ppm by volume)

< 100

< 100

< 300

< 1000

< 10

< 1< 1 (by weight)< 0.05 (by weight)< 0.05 (by weight)

2-6

—— —• — — WAESD-TR-85-0030—

remove water from the process gas stream, and to provide process heat for steamgeneration.

The FPS "top level" design requirements are summarized in Table 2-3.

2.1.2.3 Power Conditioning System

The primary functions of the PCS are to control the magnitude of the directcurrents generated by the PCS, to combine these currents, and to convert themto alternating currents suitable for application to a distribution system andto an electric utility transmission network through the Utility EquipmentSystem (UES).

Secondary functions are to control fuel cell module electrical loading, provideappropriate fault protection, harmonic reduction, power factor correctioncapability, and means of integrating system control and monitoring into theoverall power plant.

The PCS design,will be designed to meet the performance and equipmentrequirements identified in Tables 2-4 and 2-7 based on the input and outputconditions as specified in Tables 2-5 and 2-6.

2.1.2.4 Rotating Equipment System

The primary functions of the RES are to provide and circulate a supply of cleanpressurized air to the PCS and to provide clean pressurized air to the FPS.The air provides a source of oxygen for both the FPS combustion and the cathodeside of the fuel cells and also removes thermal energy produced by theexothermic reaction within the PCS. The RES uses steam generated in the S6S,cathode exhaust gas from the PCS and combustion exhaust gas from the FPS todevelop mechanical shaft power to help drive circulating and compressing

equipment, thus helping to maximize the overall power plant efficiency. Keyoperating statepoints are summarized in Table 2-8.

2-7

WAESD-TR-85-0030

TABLE 2-3

FPS TOP LEVEL DESIGN REQUIREMENTS

PERFORMANCE

H2 Production Rate, normal m3/s (SCFD)

• Gross• Net (Hydrogen Utilized)

Natural Gas Feedrate,* kg/s (Ib/h)

1.5 (4.9 x 106)1.3 (4.1 x 106)

0.34 (2,720)

OPERATIONAL

Gold Start to Standby, hoursStandby to Min. Power, hoursMin. to Max. Power, Percent Full Power per minOperating Range, Percent production rateOperation

Start/Stop Cycles (Over Lifetime)- Warm Startup/Warm Shutdown

- Cold Startup/Cold Shutdown

< 4< 1/2

7.7

23-100

Automatic Dispatch fromStandby to Full Power

10,000

150

GENERAL

Fuel CapabilityAvailability, PercentDesign Life, YearsMaximum Plot Plan Area, m2 (ft2)Maximum Height of Equipment Above Grade, m (ft)Noise Level (db)Transportability

Natural Gas/Naphtha9525

230 (2500) -

9.1 (30)

55

Common Carrier Truck (Forall subsystems exceptreformer)

* Includes raw fuel used as reformer furnace fuel

2-8

WAESD-TR-85-0030

TABLE 2-4

PCS PERFORMANCE REQUIREMENTS

Power Conversion

Rated Full Power InputRated Full Power OutputRated Part Power InputRated Part Power OutputMinimum Operating Input

7.5 MWDC

7.2 MWAC

1.875 MWDC

1.725 MWAC

.75 MWDC

Efficiency

Full PowerPart Power

96 percent92 percent

Power Factor

NormalExcess VARs Available

Unity - Part to Full Load1.4 MVAR - Full Load, Rated Voltage5.4 MVAR - Part Load, Rated Voltage+6.8 MVAR or - 3.4 MVAR - Zero Load

Fault Protection

DC PowerAC Power

Electronic circuit interruptionCircuit breaker

Response Time

Controlled Power ChangeControlled MVAR ChangeFault Clearing Trip

.5 seconds over operating range

.1 seconds over operating range

.05 seconds

Harmonic Distortion Maximum total harmonic distortion of3 percent RMS of fundamental voltageand 1 percent for any one individualcomponent when operating into asystem with 250 MVA short circuitcapacity.

2-9

WAESD-TR-85-0030

TABLE 2-4

PCS PERFORMANCE REQUIREMENTS (CONT'D)

Manual Control To bring system to STANDBY conditionduring plant startup and for systemcheckout.

Automatic Control For all normal power operation viaplant computer.

Power Dispatch In response to changes in currentsfrom fuel cell modules.

Reactive Power Dispatch In response to command signals fromplant computer independent of realpower dispatch.

Power Level Control Accuracy + 150 kW

Electromagnetic Interference Shall not result in misoperation oflocal communications equipment orany sensitive loads.

Black Start Capability None

Low Power Bleed Loading Provide 50 + 2.5 ohm load for eachgroup of 5 modules - to load modulesbefore PCS is connected.

2-10

WAESD-TR-85-0030

TABLE 2-5

PCS INPUT FROM FUEL CELL SYSTEM

Number of Fuel Cell Modules

Nominal Input Per Module

Full Power - Beginning of UseCurrent, ampsVoltage, volts DCPower, kWDC

Full Power - End of UseCurrent, ampsVoltage, volts DCPower, kWDC

Part Power - Beginning of UseCurrent, ampsVoltage, volts DCPower, kWDC

Part Power - End of UseCurrent, ampsVoltage, volts DCPower, kWDC

Module Voltage Maximum Limit

Module Current Maximum Limit

Module Minimum Controlled Current

Module Low Power Bleed Current

Module Ripple Current Limit

Module Effective Resistance

Module Effective Capacitance

20

375 kW Module

3251156375

325

1025333

' 80

1260

100

801130

90

1300 V (sustained more than 1 second)

450A (sustained more than 1 second)

50A

5 to 10A during shutdown operation

100A peak to peak, 180 or 360 Hertz

.95 ohm during power operation

Variable with magnitude of maximumcurrent reached during a load change24 MFD for 100A maximum120 MFD for 350A maximum

2-11

WAESD-TR-85-0030

TABLE 2-6

PCS OUTPUT TO UTILITY EQUIPMENT SYSTEM

Output Voltage

System Performance DegradationPerformance(Resulting from utility AC gridvoltage variations)

Utility System Unbalance

Frequency

13,800 VAC, 3 phase, 60 Hz

Bus Voltage

Variation

+5 Percent>*-5 Percent to +10 Percent(continuous)<-5 Percent to -10 Percent(continuous)<-10 Percent to -20 Percent>+10 Percent or -20 Percent

System

Impact

Rated Power95 Percent ofRated Power95 Percent ofRated Power85 Percent ofRated Power

, Unpredictable

Rated Power Output with up to 3 Percentunbalance and 2 Percent negative sequencevoltage

Rated Power Output - 57 - 61 Hertz (Trips offline outside this range due to transformercore limitations)

2-12

WAESD-TR-85-0030

Design Life

Availability

Reliability

Structural Welding

Structural Design

Footprint

Environment

TABLE 2-7

PCS EQUIPMENT DESIGN REQUIREMENTS

25 years10000 Operations (mechanical)

99 Percent

99.5 Percent

AWS Dl.l

ANSI A58.1 (Seismic Zone 4)

43.5 m2 (1500 ft2) maximum

Temperature -29°C to 49°C (-20°F to 120°F)Humidity 100 percentPressure 85-110 kPa (12.5-16 Psia)Altitude 1525 m (5000 ft)

Governing Electrical Code

Shipment

National Electrical Code (NFPA No. 70 -Latest Revision)

Truck Transportable (maximum size modules3.5 m (ll'-6") high x 2.25 m (7'-6") wide x9.5 m (3T-6") long, 18600 kg (41,000 Ibs)- shock loads of up to 3 g during shipment)

2-13

WAESD-"TR=85-0030

TABLE 2=7

PCS EQUIPMENT DESIGN REQUIREMENTS (CONT'D)

.Mai ntenancOequ.ir.eme fits Clearance for removal of major componentand assemblies

Repiaeeabie eiir filtersSelf checking diagnostic circuitsConservative thermal margins•24 hours/year scheduled maintenance

downtime

GonttQi.l ed .Mechan.i ca 1_.1 h.teffaces Equipment foundationsEquipment fencingFire protection alarms

Gon.t.ro.ll ed,El.ec.tr i ca.l „ Iriteriaces. (-H) or (-) connection to. each PCSmodule

13.8 KVAC output connections to aUtility Equipment System

480 VAC power feeder from PowerDistribution System

Ground connections from PowerDistribution System

Control interface with Westinghouse WDPFData Highway

2-14

TABLE 2-8

RES OPERATING STATEPOINTS

WAESD-TR-85-0030

Full Power

EOU

Part Power

BOU

• Air CirculatorFlow, kg/h (Ib/h)Inlet Pressure, kPa (psia)

Inlet Temperature, °C (°F)Outlet Pressure, kPa (psia)

Outlet Temperature, °C (°F)Fluid

642, 200 (1,416,000)

482 (70)146 (294)

489 (71)

147 (297)Air

102,490 (225,950)

207 (30)117 (243)

207 (30)

141 (286)Air

Centrifugal Compressor

(Two Stages)Flow, kg/h (Ib/h)

Inlet Pressure, kPa (psia)Inlet Temperature, °C (°F)

Outlet Pressure, kPa (psia)Outlet Temperature, °C (°F)Fluid

33,790 (74,490)101.3 (14.7)27 (80)

482 (70)146 (294)Air

17,919 (39,480)101.3 (14.7)

27 (80)

207 (30)

117 (243)Air

• Steam TurbineFlow, kg/h (Ib/h)Inlet Pressure, kPa (psia)Inlet Temperature, °C (°F)


8,193 (18,060)365 (53)140 (284)

21 (3)61 (141)

Steam

1,126 (2,476)365 (53)140 (284)

21 (3)61 (141)

Steam

Cathode Gas ExpanderFlow, kg/h (Ib/h)

Inlet Pressure, kPa (psia)Inlet Temperature, °C (°F)

28,230 (62,230)475 (69).

193 (379)

11,730 (25,860)200 (29)172 (341)

2-15

WAESD-TR-85-0030

TABLE 2-8

RES OPERATING STATEPOINTS (CONT'D)

Full PowerEOU

Part PowerBOU


Combustion Gas Expander

Flow, kg/h (lb/h)Inlet Pressure, kPa (psia)Inlet Temperature, °C (°F)Outlet Pressure, kPa (psia)

Outlet Temperature, °C (°F)Fluid

110 (16) 110 (16)27 (146) 114 (238)

Air (02 depleted) Air (02 depleted)

9,288 (20,480)

352 (51)

440 (824)

110 (16)

295 (563)

Combustion Products

2,457 (5,416)

138 (20)

434 (813)

110 (16)

408 (767)

2-16

WAESD-TR-85-0030

Table 2-9 gives the power summary for the RES design for full and part power atBOU and EOU conditions. The BOU part power case requires the most power fromthe electric motor, 231 kW (310 HP). Table 2-10 lists the RES designrequirements. A compact arrangement of the RES is illustrated in Figure 2-1.

2.1.2.5 Instrumentation and Control System

The primary functions of the ICS are to provide real time monitoring of allplant parameters and provide warnings of deviations from normal operatingconditions or plant configuration, to control all startup, power and shutdownoperations of plant systems, and to implement automatic corrective action inthe event of faulted or emergency conditions. Secondary functions are toprovide data recordings of plant parameters, plant configuration status, andalarm event sequences for overall plant performance evaluation use and eventdocumentation.

The ICS will be designed to meet the performance requirements identified inTable 2-11, based on the numbers of inputs and outputs specified in Table 2-12.

Table 2-13 lists equipment design requirements that will be met, and Figure 2-2shows the basic system configuration in block diagram form.

2.1.3 Fuel Cell Requirements (WBS 1102-01-300)

The fuel cell baseline technology requirements development effort needed tomeet the PAFC system design and operation goals imposed by FCS performancerequirements is discussed in this section. Current fuel cell technology andthe anticipated improvements resulting from the development effort areidentified and compared to the prototype plant requirements and goals.

2.1.3.1 Approach and Assumptions

The overall objectives of the fuel cell requirements effort were to establishthe baseline fuel cell technology requirements, define goals for design and

2-17

fABLE 2-9

RES POWER SUMMARY

WAESD-TR-85-0030

.Full PowerBOU EOU

Part PowerBOU EOU

Steam turbine (kW) 561

Combustion Gas Expander (kW) 458

Cathode Gas Expander (kW) 1175

Air Compressor (kW) -1842(Two Stages)

Air Circulator (kW) -255

Motor Power Required (kW) 97

672

458

1175

-1842

-360

103

46

22

207

-492

-14

-231

77

22

207

-492

-16

-202

2-78

TABLE 2-10

RES DESIGN REQUIREMENTS

Performance

WAESU-TR-85-0030

See Operating Statepoints (Table 2-8)

Operational

Warm Startup/Warm Shutdown CyclesCold Startup/Cold Shutdown CyclesOperating Range, Percent of Rated SpeedOperation

10,000

150

70 - 100

Automatic Dispatchfrom Standby to FullPower

General

Availability PercentDesign Life, yearsMaximum Plot Plan Area, m2 (ft2)Maximum Height of Equipment Above Grade, m (ft)Noise Level (db) at plant parameterTransportability

94

25

50 (525) ,

9.1 (30)

55

Common, Carrier Truck

2-19

WAESD-TR-85-0030

ORIGINAL 'OF POOR

tx. •-. »-C9O Z

00

Q)

O1C(Oi-

<u4->I/)

(U

<o+Jo

ICM

O)

3

WAESD-TR-85-0030

TABLE 2-11

ICS TOP LEVEL DESIGN REQUIREMENTS

System Type Westinghouse WDPF distributed control system, using local

programmable controllers, data highway, and centralcontrol room.

Manual Control Between COLD STOP and STANDBY from local control stationsof individual systems or central control room.

Automatic Control During POWER operations from local control stations of

individual systems or central control room or remoteutility dispatch room.

Alarms Lock-in type warnings of deviations from normal operating

condition or plant configuration and identification of

fault and emergency conditions and sequence of occurence.

Monitoring Real time indications of plant alignment and operating

parameters.

Redundancy As required to achieve desired reliability and accomplishsafe shutdown. ;

Uninterruptible As required to monitor plant status after loss of plantPower Supply power.

Software User friendly - no previous computer programmingexperience required.

Diagnostics Self checking and failure diagnostic programming built in.

2-21

WAESD-TR-85-0030

Type of I/O

Digital Input

. /TABLE 2-12

, ; ,,ICS INPUJ/pUTPUT .REQUIREMENTS

••.."•• •::.'''••• --•--;?• Electrica.l :.-:> ,No. Required* Characteristic Function

768 384-TTL, .. -.Level, temp., and pressure192-25VDC, switches, limit switches,192-115YAC. events, handshakes, binary

data

Digiital Output . , .. . 7^8

, An a .lag, Input.; • . Thermocouple.• High level

' '•'•••. 3 I '•."•>'. i. ?'". ..'•:.,'.:

Analog Output..;*,,Hi;gh level

• PID Controller

288

288

72

72

384-TTL,:192-115VAC,f192-Relay

Type K0-5 VDC ,

.0-5 VDC

4-20 ma

Digital contro,ls,. contac-tors, 'inhibits, enables,handshakes, binary data

Temperature sensorsfrom level, pressure,flow, voltage and currentsensors and strain. gauges

To meters, trend ;

recorders, analog devicesClosed loop controllers

TOTAL 2,256 (720 Analog, 1536 Digital)

*Current estimate .

2-22

WAESD-TR-85-0030

TABLE 2-13ICS EQUIPMENT DESIGN REQUIREMENTS

Response Times

Controller computational frequency TBDOperator display update TBDMultiplexer scan rate TBDTime to change display format TBDAlarm scan rate TBD

Alarm timing resolution TBD

Accuracy

Type of MeasurementTBD

Control Accuracy

Design Life

Availability

Reliability

Structural Welding

Structural Design

Configuration

ErrorTBD (percent of full scale)_+ 2 percent Power

25 years10,000 operations (mechanical)

99 percent

99.8 percent

AWS Dl.l

ANSI A58.1 (Seismic Zone 4)

93 m2 (1000 ft2) maximumfootprint. Skid mounted, prewiredassemblies with weather/explosionproof enclosures as required. SeeFigures 2-8 and 2-9.

2-23

WAESD-TR-85-0030

TABLE 2-13

ICS EQUIPMENT DESIGN REQUIREMENTS (CONT'D)

Environment

Governing Electrical Code

Temperature -29°C to 49°C (-20°F to 120°F)15°C to 32°C ((60°F to 90°F)inside enclosure)

Humidity 100 percentPressure 85-110 kPa (12.5-16 psia)

National Electrical Code (NFPA No. 70,latest revision)

Shipment Truck transportable (maximum size modules3.5 m (11'-6") high x 2.25 m (7'-6") widex 9.5 m (31'-6") long, 18600 kg (41,000Ibs) - shock loads of up to 3 g duringshipment)

Maintenance Requirements Clearance for removal of majorcomponents and assemblies

Conservative thermal design marginsOn-line diagnostic and verification

capability and software

Controlled Mechanical Interfaces Equipment foundationsFire protection alarms

Controlled Electrical Interfaces All wiring to other plant systems

2-24

WAESD-TR-85-0030ORIGINAL -Of, POOR QUAUTY

i

fi

v/iw

ti i«

a*M

<.>

k.n

gid

UU

MrlinO

l Mlilil U

MiM

U O

W N

OU

UM

MaiU

I «

•DT •SKT ~ tWEecu

MIWOATA

rioiTo

1 • t 11 1 K1

woe*

TH!»N9M

11

lOCAi

EICHANIC

.a. MDUTBOAW

NARO01 SC TAPE MCI OATA

«.OTTE«

| t t f t f

KWITOBINC

I

t ,

« e

f *

(*

^

MCC -^_JV

SUPl RV I SORT ^- ^ AMNUNC I ATOQ

1Nldxutr

4

.

"

1

OATA HIRWAT

tOCAL K 1 S"O VDT *ITI1

«T»OA«D

j(

r ji 4 A™"""OICUALIIPUTS

1 1

01C1TAIOU1PU1S

-4AM10C ARALOC

— I- -14 \\ ill

< \ f

LOCALf C 2

^tOCAlK 3

>} 1 ' > 'J V ' •

1 1 >

TtriCAL I/O

IITt

i»»UAi i/o ;EQUlWin '

; 1

4H

Figure 2-2. Instrumentation and Control System Functional Block Diagram

2-25

WAESD-TR-85-0030

development of the fuel cells, and determine further development effortrequired. Specific efforts during this reporting period were to:

• determine fuel cell testing performance and technology baselinerequirements to meet system development goals.

• establish and update the PAFC technology database needed for systemdevelopment requirements assessments.

• define the fuel cell design and operating requirements for thecontinuing development testing to ensure meeting system requirements.

§ provide projected fuel cell requirements for the PAFC module designspecification and update as required.

The key baseline fuel cell technology existing at the beginning of thisreporting period and from which the fuel cell development requirements effortprogressed included design and fabricating processes for:

• nominal 30.5 cm x 43 cm (12 in. x 17 in.) "Z" channel process gasrouting bipolar plates and double branched (tree) cooling-plates.

• rolled electrode catalyst layers dry bonded to Teflon wet-proofedStackpole backing paper.

• five cells between cooler plates

• a matrix utilizing .25 mm (10 mil) thickness MAT-1 and nominal .20 mm(8 mil) thickness SiC layer.

• performance development testing to be conducted with simulated reformergas (SRG) fuel and oxidized with air.

2-26

•"— - - WAESD-TR-85-0030

t Viton seals selected as the first baseline (more detail is presented inSection 3.2.1.1 Stack Elastomer Edge Seal Development).

O O

• nominal 48 cm (12.7 x 10 gal) of phosphoric acid applied at 97percent concentration for each cell.

During this development period variations in fuel cell assemblies of alternateelectrodes, processes, seal materials, seal configuration and acid loadingswere investigated. However, the above baseline configuration was the basis forcomparison of the technology development.

2.1.3.2 Technology Assessment Updates

During the First Logical Unit, the reference PAFC design technology baselinewas not fully established for assessing development efforts for meetingprototype power plant applications. Effort was directed during this earlierperiod toward upgrading technology to improve performance. This is noted inTable 2-14, where the parameters developed in the First Logical Unit effortwere updated to concur with the fuel cell prototype performance goalsdetermined during this reporting period. A review of the prototype goalsresulted in the definition of potentially more compatible fuel cell and plant(systems) operating conditions for the specified plant requirements (Refer toSection 2.1.2). PAFC development goals and test performance experienced duringthis reporting period are presented in Table 2-15.

Although most of the fuel cell operating parameters that were needed for designand concept assessments had some level of definition, the impacts of theselected operating conditions on cell life, overall stack performance, econo-mics, and risk were not definable since corresponding test data were notavailable. Work continues to identify the parameter trade-offs and sensitivityof the fuel cell to its operating requirements in plant system conditions.

2-27

WAESD-TR-85-0030

« *r-~|

6J

UJz-Aca•=cH-

•

CONS

TRAI

NTS:

11

O

«=C

*toLjj

-J

5*ii

toocLU

UJyt5£•pzQ-

•yirCDtoUJ

rain

t to

<

800

mV

fro

m life

per

form

ance

CO•pccu€o0

c•r-On•r-s-0

I 1

CO

o

co•r"COOe^.

:oCJ

^^t^~cu>0)_J

ooV

tra

int

- m

inim

um d

esig

n vo

ltag

e re

qu

ire

d

COcoor— "reco

•f—+J

uc3

Ll_

CUCoz

ratn

t -

Vol

tage

de

pend

ent

- re

qu

ire

sm

ent

da

ta t

o se

t tim

e lim

it

at

tem

pera

ture

•P Q.CO OC r^

O CU0 >

CUC XJ0•r- .oo cuo .ci- -pS- S-O 30 4-

>—*•

CM_CVJC 4-*O ^H

^C *iE -0

in r-*r**h i

1 Oo «•— *

rain

t2

- fu

nct

ion

of

tem

pera

ture

and

,

t ye

t ch

arac

teri

zed

•P O00 CcO TO0 C

reco cu^ S-CO 3O 00S- 00S- COO S-O CL

•

CM4J4-

CM -»O <Ot— CM CO

1 E CTlin u ir » ***••» c3

• pÂ E • —

str

ain

t -

low

eff

icie

ncy

an

d ce

ll p

erf

or-

tra

de

-off

s -

eff

icie

ncy

vers

us

cost

)

C 00o cuO i-

•r-CU 3O CTC COre i-P* *~~*

o cu4- US- Ccu reo. E

CM ^ÊCMO -P

•?* " ^ps ^^

OO^D f"^^- roV

str

ain

t -

elec

trod

e life

re

quire

men

tse

lectr

oly

te

loss

c04-0 0

CO -P(J (JC 'Ore a.E Ei. •>-O

t OCO S-

O_ 4-

"1 1o10CM- ^

OOOCMCM

str

ain

t -

tem

pera

ture

co

ntr

ol

req

uir

es

max

imum

lo

cal

tem

pera

ture

in

ce

llC "Ôo c

cu cuo oc cre reE •'-i. i.o re4- >

CU r-0-<

^— u_

O 00 00O CT>o fCM 1i m

o f^CT^ ^*r— *— •*

str

ain

t -

tem

pera

ture

co

ntr

ol

for

aci

dne

e va

ria

nce

& C

O p

ois

on

ing

4-^C

*r—($i.-pCOcouto•r-

co

C_J

cuucreE

O4-1_

d.

*i*~*

QCQ1—

(OE

O4-S-QJo.

•»

•uco0

str

ain

t -

tra

de

-off

st

ud

ies

of

syst

emve

rsus

pr

essu

re

leve

l ne

eded

ct_) CO

0cu uuc enre c£ ir""S- -P^3 fO

4- S-t- CUCU Q-0. 0

*—*.re•^

re too- Q."^ o00

vo «— •

tra

int

- d

esi

gn

con

ditio

n

impo

sed

on

ce

ll as

sem

bly

and

sea

ling

CO "c =°.°0 -p^— 1re s.c reo Q.•i- CU•4_> COUc co3 'U. CT)

'*O " ^ "*Q. reNX «^

COin Q;

•^^ mcî^— *

str

ain

t -

tem

pera

ture

co

ntr

ol

of

ce

ll a

cid

nd

perf

orm

ance

(c

ha

ract

eri

zatio

n TB

D)

C ^*oo c

oCO T-0 -Pc ^*re s-E -Ps. co cu

4- <->i. Ccu oa. o

*~û_

0 00 CM0 0in co^— i

1 0O °oi— CMr— •—*

str

ain

t -

life

req

uire

men

t im

pact

fro

mte

mpe

ratu

re

con

diti

on

s

co •—

(_> r-CO

CO U0C COre >E "-l_ COO CO

l+_ CU%- <~>CO X

Q. CO

-+*u.

000 00o So coCM - —

Q-CO

o

coo

CQ

CO3O3C

Oo

o3C

Oo

oCQ

CO3O3C

COo

to3O3

3O3C

OO

CO3OCO

re

<oCOc

CO3O3

OO

oCO

r-i r- Ccuo

COC7>re-p

E3

•^X

,re

CUO

COC7Ire

1E

>>-P-r-00ccuQ

COtS-

o

COcCOo•pcCOS-J-3o

CO

cuo

COS-

Xre

o

CO

3-Precoo.

xre

cos_3-PrecoQ.I

^— c•Ecus_3•Prei.cuQ.ECU

**-Xrez.

cu$_3COCOcuS-

Q_

0) .

u xc reCO ^j-*-"cu

14- CO14- CU•r- COQ re

C3cuj_ to3 COco cuCO Ucu oS. i-0. 0-

cus_3-pres.COQ. ,Ecu

1—-pcre^--r- Co -^0 Zo- —

cui.3•pret-CO0.Ecu

t—4->cre ^*

i— Xo reo s:o- —

2-28

WAESD-TR-85-0030

oo

oot—zI—I

orooooo<t

•— ooI uj

CM H>

UJ <c

CO1 >^C *»- GO

or<ca.

ooUJoou_<ca.

UJ

fnit

for

sep

ara

tio

nse

alin

g

•r- -O

r~ <ocCO >>

•r— r~~to J3CU E

•o cuto

i to10

4->c f—

'to cus- u

J->toccu£5o01 — .c•enil0

tO "Oc co f

1 *

i— Cto cuC Eo cu

•r- S-4-> fO 3c cr3 CU

U- &-

r—cucu

^-^oCOt—

to3ocuc(O

c<o

F• 1—

1

t/).M

>,[ *

•r—

4->ccu•o1— 1

^_cucue.

<0a.

,If.

<r.

Q ^X

4J 10C SrO • —

"o cu0 S-0 3

to.— COi— CUa> s-o o.

(nit

for

sep

ara

tion

sea

ling

•r- -C)r- C

<0CCD >>

to -QCU E-o 

co

• ^~*<*- X<l to•f ^rQ " —

4-> <Uc s-(0 3

4-> COO COto cuCU 5-ar o-

uire

men

t im

pact

crcuL.

cul|_

^«

1

+Jc

•r~(0s_

4->tocoocuoc10

t

oi.cuQ-

^*u.

f ^ oo OCD CT»un io <*>CM-

.E

f^Qco

—*

^M-JJ;Q

ex<D

1 — • — •c\x -r*

(J S10

-(-> 0oo -t->r— Xr~ *0CO ^^0- —

uire

men

t im

pact

crcul-coIf-•r-r—

1

4->c•r-<0S-

4->tocooCO0c

0If.S-cuQ-

<_> *~~*0 U.OoCM <£>

CO5 ^_^

CO3o3c•r-4^co0

.<*-*4—

oexEcu

1— — »c

\x «r-

0 S

4-» OOO 4->

i— X^" tOco 2:o ^

uire

men

t im

pact

crCOs-cu

14-r->

1

4->c:

•r^(0I.

4->COco0

cuuc(0E

otJI_

s~cuQ-

^^^^Q QCO CO1— 1—

c0- 33 04J X3S- 4->(O ^31 ^-oo oo

1

COcu4^toon

^ccu

• r—tO S.c cuto *S- 0

t— o_

tocutotoo

1ou•r-4->(J(0(-exs_o

14-4->c•r-(0^--

J- 4J4-> 3to exC 4->0 30 0

CU 4->O CC CUto ilE i.

0 Q.If-Or) V

Q. ^— "

CM -EC\J0 C

Qa L^S Q*0 ECM COV ^^

CO3o3c

•1—4^co0

cu0

(04^

• • totO •!-

O CO•r- CUC /*v*(0^ 4JU UCU <0S 4->

c^ oo of0

4-»GO 1

2-29

rO .r-O. co^ CX

0 0

CM 1I O

O CM

to3O3C

•f~4JCO0

•oto0

^^ ^— ,0)r—C rO

Q-4J

3 cr— i— 1

0 - —

1

and

ele

ctr

ica

l lo

sse

s

cuO)<0_*<0cur^

1cu

4-> Uc c

•r- 1010 4->i. to4J -r-(/) (/)

C CUO i.

CJ

^>O) UO fOC 4-*T3 iCe oS- Uo

(4. _prS. 0)CU-i-Q. .C

100.

"cn

•143 ^ c *^^^^ to0 CXcov^ ^—— " —

to3O3C

•r—4^Co

"CO

c••S,

^3toO-Jcoc

• CX2to

r—o

1

uire

men

ts

impa

cts

crCUi.

CU«f-•r-^>.

|

4->C

•r—tOi.

4->toco

.CJ

cuucr0

J^

OIf.S-coo.

••

Qco1—

11

^^ '

COcu4J tO(0 -Cor 4->

0)Q. CCU COcu s-1- 4J

<_) OO

1

>>4->•r-r—•r~J3>0^^u3S.4->

1

4->C

•^*

<0S-4-)toCo0

•ocutooCXE

1— 1

^_^

ll•

or™x.

E

CO•

co

11

•1—c=

cu0 's<0 T34J 000 2

^^£3 4->E j:

•r- COX -r-(0 d)2 31

1

>>4->«^-i•r*J310.U3S-4J

1

4->C•r-r0S-

4-^toCO0

-ocutooCXEH^

11

g* "

4-> 4->•o .•«- c3E ID

E CU3 r—S 3•i- -aX 0*u ys:—

i

co

ce:coOo

•"T CO

cii IUJ ef

E! ><C «H- to

o:UJ»=TUJ

Q)

O

<§

>-UJ

5-o

i.)•»•»

xi(O

03J-4J

-MC

•r—roS_

-4->toCooT3O)toOQ.E

I-H

ii

4JJC *""0>+J

•r- ••—tt> C3 =>

E 0)

JlSIs:

i

co•Jjuc3

M-

10rO

TO<UX

•f—4-

io ov•r (O

•»->•!-> r-

1°

OC •cno•r-to -MC Vt

•r™(O CI. •!-4jto toC r-^O H^O O)

Ur""rO S-C (UO -Q•r- E4-> 3U CC3 <4-U- O

<_3O

>OO•*V

io3O(UCrO

-t->Cro

•«->toC

I H

- »

•»-> ^£3 UQ. rO

+•> 4->3 COO • —

E 0)3 OlE <0M- 4JX *—rO Oz >

1

co+J• r-

•oOO

cenv>O)

T3

^vrOC

•p—Eo

c_>o>ooroV

to3O3C

•r—+->

O<_>

•4J J<3 0Q. <0

-t-> •«->3CO

O--'

is.^•5S os >

1

1n

t -

nom

inal

de

sign

co

nd

itio

nin

t -

conv

ersi

on

eff

icie

ncy

•i- rorO S-t *J

4-> toto CC OO O0

0)r^ 0rO CC rO0 E•i- 5--!-> 0OM-C 5-3 <V

U_ 0.

0O

>O0oevjA

to3.O3C•r"-!->COo

4J> *—+3 4JQ.-I-

•4-> C3 ID

O- —E 0)3 O>

-P5

li•

. j

xed

proc

ess

con

tro

ls

and

fixe

d2

ters

a

re

need

ed f

or

furt

he

r••- EM- ro

S-"T3 roO) Q.E3 OJto m«/) OJ(0 j=

cn1-c

•r™ •> CrO en

toJ- 0)(U T31 *

M- 0)<O o •

c cen QJ oc $-"-

•<- J-W) J_ T-

C o 3O (4- CTU (U

to s-.>>$-<1) O) Ol^ +J C

(U T-

<4- E ••-»O ro rO

s_ s-•+J ro QJ«/) Q. Q.

•r-. Or1- O>

o -o"O C C O COT3 «t_ -r-C J_ WO 0 Q.-0

' *

—

••o<uc

*r~

<u4->(U•o4->O)>>

->Oc1

•!->(JroO.E•r-

Co•r-(/>O

OO

<M

•

•M

E,_«

• U(UQ.Q.3

T3O)

•r-o>uS-<uQ.

>.fl+^

Crt iMi*s.s_3U

• ««-oO)N•r"s-oa.rO>«*0a.ro3:t»-0

4->

3O

^

ro•r-'

WAE^SD-tR-85-0030

2-30

WAESD-TR-85-0030

TABLE 2-15

PAFC DEVELOPMENT GOALS

Cell Performance:^)Prototype

Primary Goal Demonstration

0 Voltage:Cell Potential - avg. mV/cell

2 x 2 9 Design^2.) 705 678 ± 1212 x 17 @ Design^' 680 645 + 12

• EnduranceVoltage Loss - mV/1000 h 2 10-15Corrosion Loss - cm/yr (mil/yr) 2 (2.54 x 10'3) NA

SecondarCatalyst

Anode Loading - mg/cnr Lib/ft2). 0.25 (.5 x 10'31 0.30 (.5 x 10'31Cathode Loading - mg/cnr (Ib/ft^) 0.50 (1.0 x !0'J) 0.50 (1.0 x 10"J)

Oxidant Utilization with Air:Oxygen'3' - percent 50 50

Fuel Utilization with SRG:Hydrogen*3' - percent 83 83

Operating Temp, of Cell:Cell Avg. Temp. - °C (°F) 190 (374) 190 (374)

(1) Fixed Parameters:Natural gas feedstockERC-Westinghouse Air Cooled System ConfigurationMark II "zee"-"tree" PlatesMAT-1 matrix1982 Process Specs/Fab Controls

(2) Design Conditions (Beginning of Use):480 kPa (70 psia) Pressure Level190°C (374°F) Average Cell Temperature325 mA/cnr (300 A/ft/) Average Cell Current Density LoadingAcid Control Equilibrium80 percent H2 Utilization (SRG)50 percent Air utilization

(3) At 485 kPa (4.8 Atm)

2-31

WAESD-TR-85-0030

2.1.3.3 PAFC Module and, 100 kW Stack Fuel Cell Requirements

The performance and design requirements for the 100 kW stack and module designdefined in the First Logical Unit report were assessed and updated during thisreporting period. Revised requirements are presented in Table 2-16 for the key

PAFC operating parameters.

2.2 Fuel Cell System Requirements (WBS 1102-02)

The primary objective of this task is to develop test specifications inaccordance with the FCS requirements discussed in the previous section forsubscale (2x2 in.) fuel cells, nine and ten cell stacks, 10 kW stack and the25 kW stacks. The test specifications include requirements for steady stateand transient performance, endurance, key statepoint data, gas analyses andchemistry, and structural considerations. The following sections summarize the

results of these efforts.

2.2.1 25 kW Stack Fuel Cell Test Specification (WBS 1102-02-300)

Although a preliminary 25 kW stack specification has not been issued, systemperformance requirements and design requirements were identified and documented

during the First Logical Unit of Work. Updated 25 kW stack performancerequirements are given in Table 2-17.

2.2.2 Fuel Cell Hardware Test Specifications (WBS 1102-02-400)

A number of subscale cell and nine cell stack test specifications were issuedto perform tests to obtain the technology characterization for fuel cell andstack design and performance prediction. The 10 kW stack test and measurement

requirements and testing conditions were defined and the test specificationinitiated.

Subscale cell test specifications were issued for a group of technologycharacterization tests and quality verification tests. A summary of the

2-32

WAESD-TR-85-0030TABLE 2-16

100 kW STACK PERFORMANCE AND DESIGN REQUIREMENTS

PARAMETER

POWER

TEMPERATUREOxidant Inlet*Coolant InletFuel InletPlate Avg.

*Same as coolant outlet,

UNITS

kWdc

NOMINAL

94

182 (360)

143 (290)190 (374)

190 (374)

DESIGN RANGE

0 < 120

Amb. to 204 (400)

Amb. to 160 (320)Amb. to 204 (400)

Amb. to 204 (400)

PRESSURE

FLOW

Fuel 83% H util.Oxidant (air) 50% 02 Util,Coolant (air)

CELL VOLTAGEOpen CircuitOperating LimitOperating Design Point

CELL CURRENT DENSITYMinimum OperatingDesign Point

FUEL STACK VOLTAGEOpen CircuitOperating LimitOperating Point

FULL STACK CURRENT

kPa (psia) 485 (70) 101-107 (15-85)

g/s (Ib/hr)

mV

12.6 (100) 15 (< 120)

97.6 (775) 113 (< 900)1934 (15,350) 2268 (< 18,000)

920

800680

mA/cm2 (A/ft2)

Volts

Amps

50 (47)

325 (300)

360

315

270

350

0 - 1000

600 - 690

0 < 400 (372)250-400 (235-372)

< 385

< 335210-335

0 < 430

2-33

WAESD-TR-85-0030

. ;, TABLE 2-17

25 kW STACK PERFORMANCE REQUIREMENTS

The performance requirements for the 25 kW stack design are listed below:

PARAMETERS UNITS NOMINAL DESIGN RANGE

POWER

TEMPERATURE

Oxidant InletCoolant InletFuel InletPlate Avg.

PRESSURE

FLOW ' '

Fuel

Oxidant (air)Coolant (air)

CELL VOLTAGE

KWdc 25 < 30

kPa (Atm)

g/s (Ib/hr)

(184) 363

(136) 277

(190) 375

(190) 375

485 (4.8)

1.1 (8.75)

26.3 (209)

419 (3325)

(149-204) 300-400

(110-160) 230-320

(149-204) 300-400

(149-204) 300-400

10.1-606 (1 - 6)

1.1 (8.75)

<52.7 (

365-548 (2900-4350)

mV

Open Circuit'Operating LimitOperating Point

CELL CURRENT DENSITY

920

800

680

mA/cm2 (A/ft2) 325 (300)

< 1100

500-800

<400 (372)

'2-34

WAESD-TR-85-0030

subscale test specifications issued is given in Table 2-18. The purpose of thefirst group of tests was to: (1) verify process and fabrication reproduci-bility and the process control requirements, (2) establish comparative capabil-ities of electrodes with other vendor catalysts, (3) to determine the impact ofsome process variables and material sources, and (4) to meet the requirementsstipulated in the technology development program.

Test specifications were defined and issued for nine cell stacks W009-08, -09,-10 and -11. Test article data and the test operations are listed in Tables2-19 and 2-20. Revisions were issued for stacks W009-08, -09 and -11 toaccommodate changes in the test plan. Test results and conclusions derivedfrom these test plans from the experiments that had been conducted arepresented in Section 3.0.

2.2.3 Fuel Cell Manufacturing Process Specifications

The Nine Cell Stack Subassembly Procedure (PAFC-006) and Final AssemblyProcedure (PAFC-007) were modified for each stack fabricated during the reportperiod. The modifications to the procedures were made to accommodate thedesign changes peculiar to each stack.

Disassembly procedures were prepared and utilized for stacks W-009-04 andW-009-10.

2.3 Systems Integration (WBS 1102-03)

2.3.1 Fuel Cell System Design (WBS 1102-03-100)

FUEL CELL SYSTEM PRELIMINARY DESIGN DESCRIPTION

The FCS (Figure 2-3) consists of groups of ten fuel cell modules supported intwo rows of five together with their piping on an elevated platform. Two suchten-module groups are employed in a 7.5 MW plant. The fuel cell module designis described in Section 3.6. Each module has an overall height of approxi-mately 3.5 m (11 ft 6 in.) and diameters of 1.4 m (4 ft 6 in.) over the

2-35

WAESD-TR-85-0030

TABLE 2-18

SUBSCALE TEST SPECIFICATIONS ISSUED

Test Specification Test Objective

TS 2 X 2-001 - Reproducibility (Fab, Ass'y & Test)

TS 2 X 2-002 - Five Manufacturing Variable Effects

TS 2 X 2-004 - Initial Baseline Electrode Characterization at Pressure

TS 2 X 2-005 - Initial Baseline Electrode Characterization

TS 2 X 2-006 - Screening of Alternate Vendor Catalyst

TS 2 X 2-007 - Alternate Backing Paper/Process Evaluation

TS 2 X 2 -008 - Electrode Bonding and Sintering Assessment

TS 2 X 2-010 - 10 kW Stack Electrode Performance

TS 2 X 2-011 t - Alternate Acid Transport Evaluation

TS 2 X 2-012 - Manufacturing Variables Evaluation, Catalystand Backing Paper Wet Proofing

TS 2 X 2-013 - 10 kW Stack Electrode Quality Verification

TS 2 X 2-014 - Pressurized Endurance Demonstration

TS 2 X 2-015 - Screening of Manufacturing Variables

TS 2 X 2-016 - Non-Electrode Catalyst Poison/Contamination Assessment

2-36

WAESD-TR-85-0030

T|

toXJ >>QJ--4J tO10 •»->c ^^

•r- <_>

tO

to " toXJ >> XJ >>

 <O 4-»

•r- O -I- O£ E v><O I— <O I— O»

e JoLf) 5»

C «*O XJ4J X tO lO•i— CO O) O=> en >- _J

a.o

XJO)XJ

LO

•r- CO> en

oo4->

<o4->oo

O)

toO)

$- OO O

s-O)<uo:

enI

CVJ

CO

O

UJ_lO

(/J

C_5«c.oo

ool

(Uc

O)to(O

CO

(1)C

<utoreCO

to(Oco

to

•a<u•o

•i- CO> en

*1CM

C «*O

4-> M£- co> en

a.o

n)

a.ooo

10

oo

ro OC

 (O(O OC .

OI

Oo

Q)C

0)to(O

CO

CM

to

(U

to(Oco

(UC

a>toco

ooo

•a(UC

to01

Lf)•

UD

X toco a)en >-

E

O

i—en

•aroo

a.o

s-(O

GO

coto

O)o

CO

to

(U

X•r-S-

rOsr—

1t—

S!

to(U4JfOr^O.

S-<0r^oex

•r—co

to

"cO(Uoo

1 *

c

E3i-

toc

HH

XJ XJ•r- . 100 0

to4Jto3JH-

r^(Oco•r-4->» -XJXJ<

oo

S-orexto

XJ-

r>>o+jcos-o*t-toco

ou

ooS-

ooCO

XJ(U

u•r-

XJ•r—O(0



<o

*

2-37

WAESD-TR-35-0030

CO4->

CO Co> coro •>-

4.) OO

atCJ> OC C•r- a>a. 3a. cr<o <Ds: oo ex

o

00S-

OOin

K- •— CX COC 4-> CO rO O >O -r- •— O C -t-9 CO•r- 3 rO C O OO O1Q 004-> O C rO -r- ••>. C ' C

C r O l _ O S - 4 J 4 J . r - S _ 03 N -r- T- 3 .r- i. r- CO T-O-f-CJ-P-O-O <O O.C+J-OJ- r O c - O 4 - > > , 4 J C X•4J rO c S -LUeCOOOOO3 •— CO CO^ O CX O.OO Q. O O I I I I

ro•P<UOC3

Olil OOM $-

Q. t-t Q:LU D- O Ot— o. to o(/> «£ oo •—

S LU

C N O O>•r- -i- CO S- -O -r-Q. rO c Q-

r1- CO CXVO O rOi— 0. (2> £

o-LUI/O

<=>«VCM

LU

UJ

-colLU

.-H s-a: D:

ro

CO

CM

COooevj

(Oca -LU >- oo

(— Q (/)CO «t LU

LU CC

<o oM T- -I-

•r- •»-> U_ «/)S_ T- o Q.(C T3 LO

* o

LO

03

0)Q.O)S-

o-t->

c c•r- OrO U

+J ooQ.CO "OU CX rOCO

LO _ » LC r>. O O CM

o o oo -.0.0

ooes:oo

tvl«-4JQ;

O i--

CO

COa:00

00

uo

-"O •!—

OJ T3

3Q. O «t- Q.3 CX4J

S_ -4->ITJ J-+j ro Q. (O00 O.Z3 $-

COenrO+J

CM" "o>»

O)

(O

00c

c ^.3 U1- ro

-(->oo oo+Joo -CCO -4->

-aa)IM

00cu$_Q.

OO

CO

OJs.

GO

Q-

C£r— LU OO

Z H-a.Cl.OOLU OO LUH=OH:oo 2:1—

<

00u3:0inV

e— 3 -r-•r- O CJ3 t- OrO •'• f—•P O +»ro «i-CO C -OCX CO C0) Q. Oo; o o

oorO

^ CO3 >> S-

•— S- CTc— CO CO«C > S-

i— CM

•o(U

co

(O

s-3O

<o•oCOjrCO•MrOS-

I

(O•oXo

rO•!->00Coo

roIM ••r- ^•— O

3 O4->

CM

LO« CM

Q.4JC

3 '5o a.

M- oO ro

•i- COc >

•r- CO

ou

cCO

•r—00crOS-

00

o

Q.O

O03

cuonto

ooo

O oo.C C

0°04Jir> Q.0°+J CO

OOQ. CO

J_o-o

*»- c

O *Jc

t3 COco e3 CXC O

O COo-o

2-38

ORIGINAL PAGE- ISOF POOR QUALITY

W \i OiA

WAESD-tR-85-0030

<uo

<0c

Ol

Q_

o;

oo

01o

<u3

CO

CVJ

S-3cn

•r—U_

2-39

WAESD-TR-85-0030

pressure vessel cylinder and 1.7 m (5 ft 6 in.) over its lower flange andsupport region. The weight of each module is approximately 5500 kg (6 tons).

Each of the fuel cell modules is designed to be truck transportable by commoncarrier. The module dimensions and weight are within the envelope permissiblefor truck transportation. The structural design of the modules is such thatthey can withstand the loads imposed by this type of transportation, as well asexpected seismic loads. The module pressure vessels will be designed inaccordance with Section VIII (Division 1) of the ASME Pressure Vessel Code.

Cooling and process air is supplied through a 1.2 m (4 ft) diameter pipe whichtraverses the length of the system immediately below the platform and midwaybetween the two rows of modules. Air is extracted from this pipe and issupplied to each module through a short 40 cm (16 in.) diameter branch pipe.The cooling air, after traversing the module internals, is returned throughanother 40 cm (16 in.) line to a 1.2 m (4 ft) diameter air return line,supported immediately below the air supply line, midway between the rows ofmodules. A portion of the spent cooling air is extracted for use as processair internally to each module.

The 15 cm (6 in.) diameter fuel inlet and outlet lines are supported in tworows immediately below the outermost edges of the mounting platform. The fuellines and the 20 cm.(8 in.) diameter process air return lines, which aresimilarly supported, are connected to penetrations at the module pressurevessel lower head through lengths of small diameter piping. This piping is5 cm (2 in.) diameter in the case of the fuel supply and return, and 10 cm(4 in.) diameter in the case of the process air return.

The symmetry of the piping arrangement provides for uniform distribution offuel, cooling and process air between the modules comprising the PCS. The pipesizes are selected so that the pressure drops through the fuel cell stacksdominate the flow resistances in the system, thus promoting equal flowdistribution between modules as well as between the fuel cells within a module.

2-40

WAESD-TR-85-0030

Shutoff valves are provided in all the fuel and process air connections to themodules to enable any of the modules to be shut down independently of thesystem. Valves are not provided in the 40 cm (16 in.) cooling air connectinglines to the modules so that continued operation of the fuel cell system with amodule or modules shut down will therefore result in the continued circulationof cooling air through the shut down module or modules. The modules and theirpiping system and valves are completely encased in thermal insulation to reduceheat losses to acceptable levels.

Cable trays, located below the platform, support the electrical power leads andinstrumentation lines between the modules and the PCS and ICS.

The overall height of the PCS is 8 m (25 ft). The overall length and width ofthe ten module assembly are approximately 10 m (33 ft) and 7 m (23 ft),respectively. The module support platform is approximately 4 m (14 ft) abovefoundation level. Foundation level must be located approximately 1.5 m (5 ft)below ground to provide for module handling using a permanent overhead craneand meet the 9.1 m (30 ft) height limitation of the crane.

The physical location of the PCS ten module groups in relation to theinterfacing plant systems is suitably arranged so that the thermal expansioneffects in the large 1.2 m (4 ft) diameter cooling air pipes can beaccommodated by hinged bellows without overstressing the pipes in bending orexceeding limiting axial loads and movements at the PCS interfaces.

The smaller process gas piping is less constrained in this respect due to thesubstantially smaller diameters involved. However, some modification of theprocess gas piping geometry within the PCS is required to increase the bendingflexibility of the vertical legs to accommodate thermal expansion of thehorizontal piping.

2.3.1.1 Fuel Cell System Maintenance

Access is provided to the fuel cell modules from below for attention toelectrical connections and acid replenishment systems. Access to the piping

2-41

WAESD-TR-85-0030

and shutoff valves below the ten module group is also available. Access isalso provided from the sides of each ten module group and via an aisle betweenthe two five module rows within each ten module group to facilitate removal ofthe pressure vessel flange bolts when fuel cell cartridges must be replaced.

Detailed maintenance requirements for all of the components of the PCS have notyet been defined. However, an initial assessment of the module maintenancerequirements including acid replenishment and cartridge replacement, isincluded in Sections 3.6.4 and 3.6.5. As the PCS design matures and specificitems of equipment are selected, such as valves, pipe hangers, instrumentation,etc., additional maintenance requirements may be identified.

2.3.1.2 Structural Analysis

A structural.analysis of the entire fuel cell system piping and supportstructure was initiated using the finite element computer program ANSYS.Figure 2-4 illustrates a three-dimensional view of the entire PCS piping andsupport structure model containing over 1100 elements. The purpose of thisanalysis is to demonstrate that the stresses in the piping and supportstructure are within allowable limits or to recommend design changes to insurestructural adequacy of all components. The allowable design stresses are basedupon ANSI B31.1 (1980 edition) Power Piping Code and the ASME B&PV code,Section VIII (1983 edition), Division 1, Appendix 6.

The finite element model includes all large and small pipes, support structure,and the module vessel baseplates. The weights of the vessels are included asmass elements acting at the center of the baseplates. The work platform andhandrails were not included as structures; however, their weights wereincluded. The bolted connection between each vessel base and the support beamswas included. Specific details of this connection will be considered in thethermal stress analysis. Flexibility factors for the pipe elbows and stressintensification factors at the pipe connections are included based on the ANSIB31.1 Power Piping Code.

2-42

JJAESD-TR-85-0030

ORIGINAL 'PAGE* tSOF POOR QUALITY

Figure 2-4. 3-Dimensiona! ANSYS Finite Element Model

2-43

WAESD-TR-85-0030

The loading conditions considered in the analysis included dead weight,earthquake loads (1/2 g), thermal expansion (assuming a -29°C (-20°F) day basedon EPRI TAG guidelines), and internal piping pressure. The frame columns areassumed to be rigidly connected to the concrete base.

This analysis effort is now in progress; however, no results or conclusions areavailable at this time.

2.3.2 Systems Interface Requirements and Control (WBS 1102-03-300)

This section presents the drawings which will be used to define and control thedevelopmental systems interfaces. The Interface Control Drawings (ICDs)illustrate each system in simplified form and show the interfaces with otherplant systems. These drawings were developed as part of a complete ICD packagefor each plant system which also contains information regarding each interface,foundations and supports requirements, electrical interface requirements, andmeasurements requirements. The ICD packages provide the means to identify andmaintain control of all interfaces as needed throughout each system's designdevelopment. The developmental systems ICDs which are shown in Figures 2-5through 2-9 are listed below.

System Figure No.PCS 2-5FPS 2-6

RES 2-7

PCS 2-8

ICS 2-9

2-44

ATM ORIGINAL PAGE {§

FPS

nr

«_

iTi iFPS

r

t

-».FCS

Ld

i

SAS-

f

•

\

TB]

h--*

\

)

FCS

f1 .

>

*OUT rj.1i

OUT

KIN

KOUT

MODULE*NO. 1

AIN1

^^

FCS

t1

. 46

INS

1

-FCSu — u_

.

CIN

(

. r

1

"11

\

1

1

1

i

e t f

*

AOUT

KOUT

H* ^

*

-llFCS

OUT

KIN

MODULE*NO

AIN{ i

> 1\

FCS—

. 20

•

CIN

AKC

*TM

It

J_

|5 |

FCS— -^

Fj SILENCl

-*-RES

IDMESi

F

\\CS

[BD

PCS |

SEE t_NOTE 2 j

H1

_L

FCS

\ m

ATM

SEI

FCS •• »

~u 1.

WAESD-TR-85-0030

A - ANODEK - CATHODEC • COOLING AIR

•Typical Fuel CellModules, Modules 2through 19 not Showi

PCS-

SEE NOTE 2

PCS

RES

SEE NOTE 3

ICS

FCS EQUIPMENT FCS

FOUNDATIONS

SEENOTE 1

AELEGEND:

[xx| - Plant Interface Number

FCS, PCS, etc. - System AcronymsAE - Architect EngineerATM - Atmosphere

NOTES: 1. Equipment foundations and supports by AE2. Electrical power interfaces with PDS and PCS

3, Electrical control interfaces with ICS

Figure 2-5. Fuel Cell System Interface Control Drawing

2-45

AIR FILTER/-SILENCER

WAESD-TR-85-0030AUX. AIR COMPRESSOR

FPS —t— IMS

JCOOLINGTOWER CIRCULATING PDS

@-HF**FPs\\ ' \\ \\ \\\\ ! \v \\ \\

PUMP SEE » !NOtE 2~r~|

FPS |

FPS EQUIPMENT

'**'/'" FOUNDATIONS

SEENOTE 3

. FPS*I ICSF

S////"''

A

'SSEE

~NOT£• 1

LEGEND:

fxxl - Plant Interface Number 3. Electrical control interfacesFPS, RES, etc. - System Acronyms with ICSAE - Architect EngineerATM - AtmosphereNOTES: 1. Equipment foundations and supports by AE

2. Electrical power interfaces with PDS

Figure 2-6. Fuel Processing System Interface Control Drawing

2-46

ORIGINAL PAGE- ESOF POOR QUALITY

SGS

fFPS WAESD-TR-85-0030

RES

CIRCULATOR

INTER

M

LER

RES«

42

•- ••

)OILCOOLER

scs

All

RES JEL SCS

TBI

SAS

\\ \ \ ' \ \ \\

.RES

'////"'

SEENOTE 2

PDS |4

\ 'RES

RES EQUIPMENT

LEGEND: "' FOUNDATIONS

[xx| - Plant Interface Number

RES, SGS, etc. - System Acronyms.AE - Architect Engineer.ATM - Atmosphere.

NOTES: 1. Equipment foundations and supports by AE2. Electrical power interfaces with PDS

3. Electrical control interfaces with ICS

RES.

"U

SEENOTE 3

RESSEE NOTE 1

Figure 2-7. Rotating Equipment System Interface Control Drawing

2-47

IOuo

o

ooo>(J(OH-

Ol

I/I>>toCD

COO

O)§a.

COI

OJ

O)

Ol

2-48

PDSWAESD-TR-85-0030

SEENOTE 2

ICS

I

LOCAL SYSTEMCONTROL PANELS

CENTRALPLANTCOMPUTER

ICS

SEENOTE 3

ICS

PLANTSYSTEMS

NOTE 1i

i

i////y FOUNDATIONS //'//

AELEGEND:ICS, PDS, etc. - System AcronymsAE - Architect EngineerNOTES:1. Equipment foundations and supports by AE2. Electrical power interfaces with PDS

3. See individual plant systems ICD's for interface requirements with ICS

Figure 2-9. Instrumentation and Control System Interface Control Drawing

2-49

WAESD-TR-85-0030

3.0 FUEL CELL DEVELOPMENT AND TEST (WBS 1103)

3.1 Subscale Fuel Cell Development (WBS 1103-01)

3.1.1 Design and Assembly Status

The principal subscale design and assembly procedure utilized was thatdeveloped during the First Logical Unit of Work. Gasket requirements wererevised to accommodate cells assembled without MAT-1 matrix material as well ascells with alternate matrix material.

A metal end plate design was evaluated to provide suitable end plates forpressurized endurance testing of up to 5,000 hours. Gold plating of tantalumwas considered but since a reliable plating process was not readily available,it was concluded that gold plated nickel 200 would be evaluated. To select thegold plating process for the nickel 200 design, four simplified end plates withvarious types of gold plating were corrosion tested by immersion in 97 w/oH3P04 at 190°C (374°F), 485 kPa (4.8 atms.), with no applied potential for

over TOO hours. Following visual examination,;three types of coating weresubjected to a second pressurized, 485 kPa (4.8 atm.), corrosion test for anadditional 500 hours. The gold plating selected was a combination of 5 microns(.2 mils) of gold applied by electroplating. Five cells were assembledutilizing gold plated hardware and tested, four at pressure and oneunpressurized. The gold plating failed during cell operation.

Work was initiated on heat treating sections of subscale 900°C (1650°F) heattreated carbon end plates to 2700°C (4950°F). Based on the success of thiswork on high temperature heat treatment of carbon end plates and the relativelypoor performance of cells built with gold plated nickel 200 plates, it wasconcluded that the effort would be concentrated on procuring 2700°C (4950°F)heat treated carbon end plates for pressurized endurance evaluation of baselineand alternate catalysts.

3-1

WAESD-TR-85-0030

A design revision was completed and drawings prepared for a subscale cell usingheat treated carbon end plates that can accommodate?"a variety of matrices. Thematrix size was increased to the cathode size and is now larger than the anode,making the subscale configuration simiTar torthe stack configuration.' Anodeand cathode gasket widths were modified to provide a 0.25 cm (0.10 in.) overlapbetween the anode gasket and the matrix or between the anode gasket !and thecathode if a matrix is not used. This overlap should decrease the possibilityof cell cross leakage in this area. Teflon gaskets (of different thicknesses)are included to facilitate disassembly of the cell after testing and to providecapability to adjust for variation in electrode and gasket thickness. Thedesign provides for a nominal 12 percent electrode compression coupled with a15-20 percent elastomer seal compressive strain range.

3.1.2: "Subscale Testing

: A total of 160'cells, unpressurized and pressurized, were assembled and testedat Westinghouse. With the exception of five cells, all the testing was basedon test plans, and the status of these test plans and cells built is providedin table 3-1.

An additional sixty-nine subscale fuel cells were assembled and tested during•this'period. Testing of eleven subscale cells assembled during the first

. . . ' Tlogical 'unit1of work was also conducted. These cells were tested to addressvarious areas of work (table 3-2) as identified in the performance improvementwork plan developed by Westinghouse and ERC to achieve the power plantperformance goals. Table 3-3 summarizes the test data. Specific test'objectives and results for these cells are discussed under relevant subsectionsof Task 1103-08. : . . : - , , . ! - . :

The'overall objectives of the various test plans are:

• To"establish'the baseline technology with existing ERC/Wdesign/processes.

3-2

WAESD-TR-85-0030

TABLE 3-1

CELLS TESTED UNDER VARIOUS TEST PLANS

Test Specification

TS 2X2-002, Rev. 0

TS 2X2-002, Revs. 1 and 2

TS 2X2-004, Rev. 2 and 4

TS 2X2-004, Rev. 3


TS 2X2-006, Rev. 0

TS 2X2-006, Rev. 1


TS 2X2-008, Rev. 0

TS 2X2-010, Rev. 0

•TS 2X2-013, Revs. 0, 1, 2,3 and 4

TS 2X2-014, Rev. 0 and 1

TS 2X2-016

Cells Built and Tested

CAS-001 through -016, -001-S1,005-S1

CAS-061 through -072, CAS-112through -115

CPS-017 through -024, CPS-018-S1,-020-S1, -021-S1, -023-S1,-023-S2, CPS-045 through -047

CPS-026, CPS-028, -029, CPS-031

CAS-017, -023, -090, -091, -096-097

CAS-026 through -039

CAS-047 through -049,CAS-051 through -060,CAS-100 through -106,CAS-108

CAS-040 through -045,CAS-073 through -076

DAS-001 through -010

QAS-001, -002, -015, -016,-019 through -024, DAS-047, -048

QAS-006 through -009, QAS-009-S1,QAS-016 through -020, QAS-019-S1,QAS-019-S2, QAS-025 through -027,QAS-009 through -020

CPS-033 through -036

CAS-116 through -121

3-3

• TABLE .3-2

SUMMARY OR SUBSCAUE CEliLS TESTED

WAESD-TR-85-0030

Test Purpose Number of CellsTested

Task 1103-08ReferenceSection

Alternate Catalyst

Alternate CatalystSupport; .

El ectf ode Mariuf actur i rig

Cell Resistance

Plate Corrosion

'Quality Control

Miscellaneous

19

7

14

10

9

6

15

1.1

1.2

1.3

1.4

1.5

1.8

Total 80

3-4

WAESD-TR-85-0030

ORIGINAL PAG£- ISOF POOR QUALITY

co

UJ CO—I 00OQ ro

o>-

8

l|i

i

-sKS

SSne a

=S

SS SSSS SSSS

8= S

SSwSS 55 555 SS=S SSS^S SS RRS 53S2 =rsi= ii SS

issi iszzissi 25==

SSS5 SSS SSSz2~~ €5s 555

if II11 il

„»-s* 5 53 |i| II || if iiii ggsggss- III IIIfc. — — wt v* v> «• v Ukiw ^w* B S y*o* i i t > cSoaooBoca «59 a 32S2S »J *.^.*22& ta«u«u«w*«n •* * ^^ E*^ '^ **"**•** SS£S|6S -ZnwiggvtMtMU'* ««« St««

= = 33 mmmS" = = = „„ „" S3 So BSSS »«SS>5&»SS = = = = = 3tt «!!^y!>^^ >^-r ~— «o ».u.---. g>^K«S

SSSS SISSl 11533 S3 S3 SS §§§§ ssslii 8SSSSS8S SSS SSS

ii II II ==== IIIIII IIIIIIII III IIIII II Ii siiillil III ill

3333 3333S 33333 33 33 33 ==== 333333 33333333 333 33

?f? I??t t •-» UUUU i • i i UUUU«(J CJ I

IIs ==== Illal ===== =^ I 4-fV* UUU4JW I I ' ...... WCJW «_»«-»<_*

si=s ==-==g IIIIIIII === ===

™™» ~ ~ !ES ~«M!S« «__»_ S*~MHS •c^ ZsSiZm *«_»«_ «MSv~»SS •c<x*e ~* *~ •-*<t^*« ^^9* ^4^CC4 ^<C^^^ ^^4^^ SS ^^^^ «««««« ^«M44C^^^ 93S 399

.«... «. .««.. ««... cSSS SSSS S5SSS SSKSS SSSSS SS SSSS SSSSSS SSSSSSSS SSS SSS

Hie Illil Illil £5552 II S5S5 IIIIII isIIIIII III ill

e*>»^ *^^,o^ tfi^^ô »» **»a»r-* viwainma> ^«^ ^««0^_ • ^ IDÎ^XDW^ o^*S5Z 5353 SSSS3 3332S 2SKS3 S3 SSSS KSS-33 S3

SSS S2IC SSSSS 33SS2 SgSCE: =S RRS3 KKKSSK SS3a»»- ^^*>«rx & — 0 1 wwrMi^M v&-«n<«<« r»5f ^ ^ ooo^»r^S-. 53 ^nS SSS SSSf^rt»» o«io ^?S9I

VW1*»< 22 £?^'vC SSZ^™2 £^>S^«n ?S SSS5 SCS^v^TK SsiSSiss sis 5RSK 5~*" ==S5S iis-? Ss°ss Is ~"** s

SS5 S = = S SSZKS ZSSSS KSSCS BS SS55S SSSSSR SSSSSSSS SSK 5«-«•• «bn^« ki«4«-o «?i^»î« !S3.ov».a m^ ,««^^ «n^«^3^ ^B^«««^«« <o«^^ 3

III lili isflla IIIII Illil II sill llllll glllllll II! Hi

3-5

oo

e£

O

CO

203

OO

ORIGINAL PAGE fSOF POOR QUALITY

z

•ntsits*«si«stc/i«n to tn »n in i/»

5i||ii

IIIIII

5SSSIsss§l222sl| ilsiil Sssa x**xIsIIlIIs §lii s.Srz—Jlllll I

!!!!!!!!!!!!!! i?!!!! 111!« __ __, _

•J t» *J CJ «J CJ «_> U UUUULJU

Hi«A«au»wiut«nui cn(/»cMcn«/i«n

saasssss ssssss ssssss zsss ssss

Bss^ssss sissss ssssss sals ssss

iiiiiiii iiiiil iiiiii iiii iiii

oo^<eo--«o-»- •DCBcaoaoacD «*>-«<«-a-o ^»-o-o" î^.^^S^»»«S^™£ S2SS«0 eooeoe " •* — **~* "-o-w-w

o2£t23 5S*3SS3 to 33 3 S3SS

•» OW^^l OQ«iQ 9 3 OO^f^* ^V>-9O

?9^»?» £ESS^«« 7£?<? ?»4?^

Hill mill iiii Iiii

WAESD-TR-85-0030"

ss

1: g-s-^vfisS!"--5«/t • ^ i ÛM« V ^^

§:=!* =in • 1*1 i A;

£ = -' §

S222

IIII Illl

i> :>;>:> &»»»

iiii iiiixs^s as^^

UUULJ ^.J.J_J_

•nvivitn gS£^

??*?s ???sxsx

^•tr>»^ fÔf*.^

SSSK SSSS

SSSS

3-6

WAESD-TR-85-0030

• To provide a base for stack scaling, and

• To determine the effects of principal parameters in order to developpredictive correlations for performance scaling.

Table 3-4 shows the objectives and status of the various test plans.

3.1.2.1 Objectives and Results of Various Test Plans

TS 2X2-002. Rev. 0. 1. 2

The goal of this test plan is to determine the effect of the following

parameters on electrode/cell performance.

• Two different shipments/lots of catalyst.

• Catalyst heat treatment/nonheat treatment.

• Two different heat treatment lots.

• Catalyst loading on cathode.

• Wet and dry bonding of catalyst layer to the substrate.

• Two different SiC lots.

• Thin, < 125 microns (< 5 mils), and thick SiC layers.

• SiC material Polyox 1 and Polyox 2.

Table 3-5 shows the performance and status of the cells built in this test plan2 2with the cell voltages (terminal) reported at 200 mA/cm • (186 A/ft ),

101 kPa (1 atm), 190°C (374°F), 80 percent hydrogen utilization and 20 percentair utilization.

3-7

WAESD-TR-85-0030

OJ

to

|to

•* CD

CO H-1

UJ tO

QJS-3COCOO)

Q.

coQJ

•OCrO

QJCOcoro

O

rOO

S-

ro

COQJ CO

CU

-dOS-o.

CO4->OQJ

CU

QJ

CO

OJO

CDC $-O 4->•i- O4j CUrO •—M QJ

S- ^ CU 4->O) O -o T-4J rO o '—O 4-> I. -i-rO CO 4J O$- . O *rO 3 O) M-

O3-oO

Q.QJo;

cu4->

CO rOJ3 QJ -r-rO r- $-•r- _Q Q.S- ro OrO «r— S-

> ^ g

f *Iil 3 r

3 O »4-4-> (f- OOro Cj_ 4->4- O C

rO O QJE-r- O

4-> roQJ -i- •—> -O Q.•r- -o QJU. eC Q;

°0CO i—

•§•0$- C

-1_> rOOO) 4->

"co E

QJ OC 4-•r-

"oj coCO CrO.a c

c:o

rOM

Oro

to

O

coQJ•aos-oQ)

'QJ

CO COQJ CO (!)

-o co-o4J O T3 OCO s_ O S_

S^•r- OJ

QJ "^CO 3rO CT

J3 QJ

QJ COC tnCU QJCO 4->3

(O O

•r- >C QJH-. a:

oa> ••-•P 4JtO •!-

Q.-0^ <

CO QJ-a co -i-$- r—

4-> QJO COQJ rOI—T3J3QJ QJ

+-> 3OJ 4-> COC "ro C

£ E CUco $- coCO O 3rot-^a QJ o

ro o a>

C 0) Q.HH o: zs

o o o<O QJ CU CO

tO Q} CU CUo

°O ^ "Os- c c cO rO *O rO

C 4J 4-> -t->QJ </) CO co

^ ^^ ^^ ^

^^ n3 03 ^3

rO {^ fl3 fQC O O O

Q> S- S- S-+•» QJ QJ QJ

<O 4-> 4J 4JO O O

OM- 4- M-o o o

OIc c <= c• £ 0 0 0

Qj'.J3 4^43CO •!- T- •!-

$- "O "O -Oo-o 5-0

COCOQJOos-Q.

• OIC

"O£rO

cu^3 t^—

*^" O4-> OrO £.3 Q.

ro m> 3QJ

CCO OCO »r-QJ +J0 00 3$--oQ- 0

^ OLQJQ. rOrOQ. C

QJO> COC si

-* coorO O-Q 4->

QJ C-4-> Oro.;:C COS- cQJ CU4-* i *^ x<£ UJ

CQJECOCOQJCOCOro

Oic

»r™S-QJ

4*^C•r-CO

•ocrO

CT)C

•r—-oc0J3

QJ•oO$_4^0QJ

r—UJ

O)ocroE

O

s-QJQ.

QJ-oOS-4JOQJ

CU

\sOro4JCO

3:

0

4-> ' —rO 1CH-$- <£•QJ S

4-> 'C r- EO ro O

•f— r—4-> «* 4-rfl o QJ3 10 r-

(0 rO ^£> c- CMQJ CU —

s- •—4-> 3 •^ x.^ sj*o0. J_ S-(/) O OC 4- 4-rOS- CO CO

4-> i— i—P— p— .

•O a) QJ•r- 0 0o(O (4— 4—

o oCO

+•> c crO CD CDC •!- T-

$- 4-> +•?CO M- T-

4-> -0 "O

5 5

r_

rO

«00

*C0

+JrO3

p—<o> 'cu 01

cCO-r-CU 4~

i— OJ3 OrO S-•r- Q-S- 4->ro CU> 5

01$-C QJ•p- Q.t 033 0-4Jo wro Cll. » "

—} ^(

C 0rO «3«y ^^

oco03

co

COQJa:4JCO

CM C\J

cu

QJ

QJ

"a.EO0

•^3 ^3' OJ QJ

QJ QJQJ QJ

QJ C C

CU 4-> +Ji— 00 COa. a> QJE 4J -PO QJ QJo a: a:

(V re•o

CO Q.ro 3

O) "O

E ^ C CS_ •!- .p- T-

O E O OCj— S- CT> OiO) QJ C Cor I— o o

roCO "Oro Q-

3-o- —QJ

Jj QJ^3 r*^ ^^E •— C£ QJ ••-0 0 0

(f_ C o>QJ *^ CQ; 0 o

•O T3QJ Q)•P CT O>r—QJ C C 3

Q. O O CUE CT> O1.CO c C Qoo o to

"0 -0QJ QJ

QJ QJr~ r™a- a.E Eo o00

•oQJ

QJf—a.

o

QJpi—QJOCroO

-o -ocu Q>

OI r— r~C 3 3

•r- T3 -OO QJ QJ

c "5 oo to co

•ocu3•ocu:oto

oQJa.co

O O •— CM o CM ro O CM CO O r— CM CO Oi— O O •— CM

CUC£

ICMXCM

CO

-^ ^ -^^CU CU QJ

r— CMO OO O

CMXCM

QJ QJ QJ QJCC.CC.CC.CC.

ooCMXCM

to

CO CO QJcc.ce.cc.

inooiCMXCM

to

QJ Q) QJ QJCC.CCCC.CC

IOOCD

CMXC\l

to

CO QJCC CC

ooI

CMXCM

to

QJ

00OOI

CMXCM

CO

QJCC

QJ Q; QJCCCC.CC.

O r-

oI

CMXCM

to

OI

CMXCM

tO

-cucc.

oI

CMXCM

to

3-8

WAESD-TR-85-0030

 •*-» 3rO <O (/)

•r-s- - .CO <J i> ••- ••- ••- co

«o3

CO>

O 3 3 3 Oi_ O" CT CT+J

4->O -C -C -C 00

QJ <O (O <O QJ-O J3 -O +J

^^ro <O ro <O to4J C C C C(/) O O O o

.TJ "O "O T3

O "O -o "O

O

l/lC COo .—•£ usrO -r-

£ )5i/i >cO O»

 CO«^ COrt) c

I- OQ_ 00

a>

a;10<o

roc

rO4J

OU

Col/l

>>r—rOtJ<0U

o,-oo

QIo

<_>oocoôo

^_3

CO

4->I/ICO

1—

•oj,l

CO^_

exEoo

Olc•ôO)c0

m CTc c•r- ••"0 0Ol Ol

c§c§

•oCO

3ôCO

oo

•orô3:

-aCO

3XJCO

u00

Olc•r-001co

O •— c\j rou•^*

+->rou•^<4-•l~uCOCL00

•MtnCO

co co co co coct a: o: o: a;

oor—O

1CMXCM

00

•

COa:

.I— •

o|

CMXCM

oo

•

COa:

IT)

0iCMXCM

00

•

CO0£

<£>^—O|

CVJXCM

oo

3-9

co•=cf-Ha:<c

a§

Lul OO

HH r—u_ co

ini u.

CO O

CO • Ul

o

0300oo inoo +->oo (O

co

OJooCMX

OJ

00

ooo0V

<s>CO' E=u

i0

1

:r~

=u•r-4

gCM

*

•r*4-*23

OJr

§CO

«

4J

r—

<nco

>r_

•oco

4->

cu

tnc

. .' §°incu

1—

c'10

J3

§inin"*

i

10s:

UJ 4->a cz io

•a0

CO

B

IO

UJ •O 4->O C

t— CO

t_) •

Co

CO

toI— •|_ 10to vjH-

UJ 4-*

*~ to!s~ 01-LU i— into <o S_UJ 4J 3ce o oa. (— x

""*?

u, UJ~"

VI

O .-a»

§ U E-<

O UJ*

^~

in1- -T-3 ,—

e0 UJ—otoinL.3 r—O r-- — •

u eO —o ujCM

a)

IO «)• — •

= ««•i- .,-• — •—i eu

*5o

1tt» ?<(»•

X X

r^* ^

Q_ Q_

O O

to oo

c c

£ £

cu cu

0 o-J i

a a:

5 5

3 3

XI "O

< <

3 3

•0 -0

CO CO00 00

^ 2

in 01

*~~

0 Jo*J" in

IOi , —to

o eg•» VO

ro coCM CMto vo

co coto oo

-iO Oo o1 15 ico c3

CMCM *

XX 0

«— o

°- „

O V

to "*c

CZ «r-

? ^

a.s

O •?1 3:

«.«

< 5

3 3

T3

< <

3 3

h- 5

CO COco co0? ^

O CO

CM •—

CO fSIO ^

CO OO•— CM10 in

1* Oto to

vo tot CMto to

VO •—

to co

o o0 01 1

15 <o o

•3tr CNJ

*1£XO X

0 ^0-

•1

i- CJ

toc!c -1- JC

I—

a.

3: 5

at cc

5 5

* £"3 0

•a

<t «c

3 0

1- S

CO COCO CO

01 r-

CM O

r~

5 °in *

r-^— |

to

CM 1to

r- OCM ,—VO VO

CO

in

^3 ^30 0

1 1< <CJ v_>

CM i —tt: to

X X

3? ^0- o-

" °m tocz c

!c r

CU cu

0 0— ' — '

a: a

5 •>0 0

T3

«C «t

C3 0

•ato 3:

CO COoo co0 Oi

S i•~ •—

^ sto m

^- ^CM inVO to

co inCM in

CM r—co into 10

r>. 0110 VO

r—to1

O <O0 01 1< <V-> C_>

" — CM

X X

>> °

c? "°

" °

c c

t- t-* •>

•*•* "M

.? .?^ 3Z

< <

>» >)

0 D

•o

5 <

o a

-D

ro ro00 CO

o^ o.

CM O

OJ OJ

g; 5LO {ft

r— O>*— o

O CT>r— CO

in oi— mIO lO

oo or> co

O S5. ^JO o

1 1

o o

' CM •—

X X0 0>! >1

"o oCu cu

C_) CJ

to to

-C .ch— 1—

a. cu

3 30 0—I _l

a: a;

5 5

cu a>

-ato 3:

5 <

+-* 4-*

S'S

X)

^ tCO CO00 CO

OT r-»

(JO tor^ coOJ ?

OJ ^>o> roLO (O

o cr»O rovo vo

§ S!LO IO

r- i—o *rto vo

O» CO

CJ* VO

01 1to toVJ O

' — CM% ntX X

>, °0 0

Cu 0-

to to

o "S

1- (-» •>

CU Q.

jr ^01 01

3: x

o: K

<t «t

1 I

•o10 X

< '<

s sT3

io x

ro roCO CO

0 ^

co in

r^ co

O Sto in

CM Oro roIO vo

<— co

VO VO

«* «9-to vo

VO Ol

f- <^.

•— CM

0 01 1«t <0 0

•— CM

X X

o "oCu Cu

" "to vo

j: jzt— H-

•* •>cu cu

I S—1 _!

o: a:

5 <

i- s-o a

13

to 3:

5 <

i. i.0 0

T3

IO X

CO COCO CO

OO Ol

VO CO

CO IOCM •—

CO VOCM Olin CM

in <*CM Olvo in

CM r-CO COIO VO

in 01CO COvo vo

r~ TJ-

vo r-

CO ^*

O oi i

v/> to<t ft.

CM . —

X X0 0

0 0cu cu

•t «VJ CJ

to to

«- r-(- K-

4-> 4->0. CU

^ .COl Ol

•r- «r-3: x

o: of

< 5

1- t.o o

•oVO X

< <

.1- u.a o

XI

to x

CO COoo coCT1 CA

OJ Otn ojio in- CM

*— CMio r-m <•

in oo rovo vo

O rovo vo

r- inco rovo vo

O «3-cr» cn

in vo

0 ot i

oo oo .

<-> U

WA

•aCOX r

ininO°^ '

o4-*m

CO '

IO

to

0£

(/)

'aiut_

O. t/)

"O ui0)

10 "IV u

u- ^

1o •§HI -r-1

U iO -r-O 3o ,0CJi CU

f o;

I— ~3Z m

i

WAESD-TR-85-0030

cuu

•acu

>,1o

3-10

oI—o

coLU_JCO

Q£

>

ELEC

TRO

DE

LU>1— 1u_u_

LO O

cn i—LU LU_i s:CQ CO<t COr— LU

coCO

CM

•

LU

T3

CB

»•"•

•

>LUa:CMOO

1CMX

CM

| —coLU(—

«*OO1%v.coinit-CD .

I/)ITS

103

4->034->CO**-*

Oo .

cn

CVJ

o

ooCM

=3

i.

gCvl

Ik

CMX

OCO

eS

I/ICo

•o-co

v>Ol1-

WVI*J ' ' - --- - - . . . _ . .C

o vo *>• co voO i— O •—4-> o o o o

VI 1 1 | |» VI VI • VI VI•— W QJ W WJO i- i. t. UE 10 10 10 105> a. a. no.VI O O O O

*• o o o o o o o o ' o o o o o o o o o o o o o o o o oo oo </o oo10

4 - * c n c n c n c n c n c n c n c n c n c n c n c n c n c n c n c n

OO

UJ •

z o o o o o o o o o o o o o o o o o< CO

4-» . ' O ' O ' O ' O ' O t l ' O ' O ' O ' O ' U ' O "O T3 "O "O

O O O O O O O O O O O I / ) O O O O O O O O < / 1 0 O O O O O O O O O

4-* cn cn cn 01 cn cn cn cn cn cn cn cn cn cn 01 cn• * " < C 4 < £ « t < £ 4 « C < X < < 4 « t < < < £ • £ I VOO X

UJ • £C3 "O >i >^ >^ >» i >^ ^ >> ^ >» j >> >> >> i >»0 C J - U i - C i - t - U U l - U l - U U i . ( - U * 03 : o o o o o O Q Q o o o o o o o o o •—*— CD. O

o • o• CO

0 X Z Z I I I I I X Z X I X X X X 0

IO4-» ^3" t t «}• *»• t «et ^J- ^J" V) CD

Ul -4_ (A CO OO CO 00 CO OO CO CO OO i-

H - O H * r o c o r o ^ ^ - ^ m ^ r o o o »< JT J-1- 10CO ^- V» * ^ — J

iQL- C O O O ^ - C O O O O ^ i — *3- O> CO CO O O O O C\Jh-4J3 c o c x j c j ^ - o o c o o ^ - ^ r o ^ 1 * " * ^ • ^ • ^ • ^ f c i» / ) O O c j r - r o r o ^ f ^ j - u ^ r o f ^ m n c o « o \ o ^ > ^ o r

.UJfr-X ^ - ^ - , — r- OJ CM • — • — • — »— CM U ^

o

LU »— *— » a » r ^ . c o i f > o o i r > c \ j c \ j m ^ - r ^ i n o ^ c o ^ m ôo o»> ^ • ^ • i n ^ - c v j ^ - m r o r o ^ ^ - r o O r o c o m <uUJ O E \ 0 \ o \ & \ G \ O t & \ O \ & \ O \ o \ O \ o . *£> \O \& \O -C '

Q. UJ C

o.vtt. 0>3 C0 ^ - 4 1 • * -x r— *— » c o ^ - c o ^ - ^ r o r o c o i o t ^ . co c o ^ o i n c N j f-

qj > ^ c o ^ " i r t ^ i r t i n c o ^ " ro i to O cxj ~ in "OO (J E \ O ^ - V O i O ^ v O ^ ^ O ^ O t O 4O Ô VD *O \O ( . 01

ouj'"' < 2

"~ K-VI

3^ «o oj — » r- f— m e \ j < o r o ^ - o f - v o co o o c x j ^ o m a*X o> m r ^ . u ^ m t n t n t o r o ^ ' r o i u ) o» ro CM in X

E I D U O V D U O ' O U O V O V O ' ^ ^ ^ V O 1 / > I O ^ O \ OOUJ— 0

VO 0

VI O*l~ •3 r- 1o --* * o c o m o ' * » r ^ v o r - ' . c M o < ' O c \ j o c o ^ ^ -X <u^ ^ • t o u - » t r > u " > « * i u ' > o j ^ ' r o ^ ' u " > r o r o f O m 1—

O E Ô (O l O l O t O l O t O ' O ' O v O V O l O (O IO UO to XO •»— 'O LUCNJ

OJo

< o *o--* i n o \ « n c v j c o r o ' O * — ^ - c o o r x i o v o m r ^•*-» w i t M O i o r ^ v o ^ O v o f ^ o c o r - ^ c ^ v o i o r < » c o r - ^•«- -r-*^- « -c in

1-4 OJ

^ o j f o ^ - i r > k o r ^ c o o ^ o ^ c \ i c x j c o ^ mu o u o l O u o t ^ u o ^ O u o l ^ r ^ r * - . r > . ^ ^ ^ - , —? ? ? ? ? ? ? ? ? ? ? ? - 7 7 7 7

r— < / > b O C O t / > < ^ > l / > ^ b O t / ) b O ( / ) C O OO OO IO tO

O<— • C J O O t _ > < - > O < - J O O ( j O t J O < _ > < - > t_>

WAESD-TR-85-0030

3-11

WAESD-TR-85-0030

The results of this plan can be summarized as follows:

t The present baseline cell performance (mean of 12 cells) is 648 _+ 6 mV.

t The effect of different catalyst lots (lots 2 and 3) was within the

expected _+ 10 mV deviation.

• Catalyst heat treatment has 10 - 25 mV advantage over non-heat treatedcatalyst.

t The two furnace heat treatment runs of catalyst were within the+_ 10 mV expected deviation.

• The; effect of catalyst loading, SiC coating thickness and manufacturingprocess were interactive with + 21 to -26 mV variability over mean.

• There was no significant effect of dry bonding over wet bonding on cellvoltage (was within the -MO mV expected deviation); however, drybonding was preferred over wet bonding due to ease of manufacturing.

• The effect of new SiC lot was within the +. 10 mV expected deviation.

• Thipner SiC coating (from 250 microns (10 mil) to 125 microns (5 mil))provided 12 +_ 6 mV benefit.

TS 2X2-004, Rev. 2, 3 and 4

The objective of this test plan was to characterize pressurized 2x2 subscalecells; i.e., to quantify the map of acceptable fuel cell operation and toprovide data base for assessment of stack scaling effects. The approach was toassemble and test pressurized 2x2 cells with the following variables:catalyst heat treatment, current density, pressure, and temperature. Theresults and-the status of these tests are given in Table 3-6.

3593E-81

3-12

WAESD-TR-85-0030

UD1

CO

Ul-Jco£

oHHL_r

1— H

a:UJl_o

^C<c'0

U)o0o:h-01 1 1 ,*-— J ^t*Ul CO

-^Q r-Ûl COtvl \i—* LOex."•- ii

oo oUl </)a: CDo.«* 3

lo

a £.o

^t

CvJ

. ^^>UJa:

*oo

1CVJX

CM

1—

1 1 11 —

•4-14.O

4-1O

1—

^ ai4J U

UJ1.

<t

o 01CM O

• UJ

UceOO

^—CM ^—

X 01U

O 0 UJcooen

•r- Si4JCM i—

o "oi"-^ O

6 Us * * i

<t Oo o>CO CM U

• C4- 1 IO

VI

I/I I/I

0 Of

.

•OC

3vi »—01 011- 0

in4-1C

14->(/»01h-

c<o>^

^—

EOl

I/I

^r—

1

IO

4-1C

00

T3C0

CO

4J*

tj

c00

•OCo

CO

IOu

in01

I—. VIJ-3

£

^

^~E

-2-

>E

i

0•— <

îa4JIO0

•oOl

4->

Oli.j

4-*1001

1Ja:UJ

0 0 0 0 0 0 0 0 0 0 ' o i 0 0 0 0 0 0 0 0

< « * < < < « * < «> > > . > , > , > > > , > < > , > , > ,l - 1 - l - J - l - l - l - t t - t .o o o o a o o o o o

•a -o -o -o -a -o

< < 5 5 5 < 5 5 5 5

' t t V t . j W l . W ^ t j Wa o a o o a o o o o

-0 -0 T3 -0 -0 -0

t/i ( / > « / > ( / ) i/) t/> x be x 3:

c o c o o o c n c o c o c o c o c o c o" * ^ " " x * ^ x ' " ^ * C T r ' " * ^ o % ' x ^ " ' ^ " ' X l 1 "*" •"' *"

g ^ - o ^ r ^ o — o - a - r -C X J ^ O C O ^ J V O C O V O O O J N J U D

m * co o> 5> oo *10 10 (O vo ô ro *** 10 * ro

•»t

**r-*. c\j *omi i oo i i i i f"- iVO \O VO

# * * *

* * + *^D O* ^3 C5 '"*'• O^ ^3 vO ^J *ef

C O C O O J O C V J C O O J C n C f t ^ -

^ j 0 ^ o ^ o r ^ v o * j o r - * - i J n i n t r >

»~~ r~ f~* 00 * 00 * 00 «

* ' * ' i- ' CM *

O O O o O o O o O o1 1 1 1 10 . 0 . 0 . 0 . 0 . 0 . 0 - 0 . 0 - 0 .

oo oo oo

< < <> > > > > •i- t- l-o o o

£ fe fe

5 5 <

t iW a a o

X X 3Z

CO CO CO

ivT "****~ ^*

O r-. ^>-CM OTfr CM

CM <»•1 CM C\J

10 10

t—i I r-.

IO

Ol1 1 <T>

IO

O COI 10 ro

in r**

r-• m t^.

ro o** 10

i— CMOO 00

1 1CM CM CM0 O O1 1 1

OO OO OO0 - 0 - 0 -c J o U

oo oo oo

« < <t

> > > ! > »4- 1. t.O Q O

•oOO X X

5 5 5

>l >l >1r c c0 0 0

•0

00 X X

*»• •» TtCO CO CO

U) UO O l

CM CM CM

OJ OO CO»— f- co10 <<r 10

i i i

i i i

•CO CM IOCO •— CO

i— CO CO

o> en" i>»

in 10 r-«• •» TT0 0 0

1 1 1oo oo i/>O- O. O-t \ c i > ^

B

•aaiX

Z

ioj:uto

o4-1

CO

aiOlL.IO_1

00

•oaiia01i-i-

IO01X

uo

8eniI-X

aiu

vi01JE**M-O

01

CIO

Eo14.

01o.

4->VIai4-'

1/1

O

>I.o.co

T3OlI/In)

oi

13oeooo

CMX

o!.

CO

VI

0)

co•n

E

inCO10

*

E

CO

>E

in

OIl-01xviOl1-

ra**"01i-10

\-a

Q*(

ai

c.°4->

•oc0

•o01

4->ieu

4-»iaviaiOlIO4-*

0>t-

0CM

.

01uc

o14-uaia.Xo

o4-1

3•o•o01

IOc'g1ai

4->

vi

1u(_>41

*

I/I

*—

U

i.0)4-1I/I

VI

<0u4JcOl•o

4-*

3

Ola:

CMooii.o

oo1

•

Ol>

4->U01a.inOlu

•E10CO

cia

1/1^£

OO

4->•0

Xo•a4J3£in

co4J

T3•o

TJ

>a

£o

14-Ol£>

01UC

0

I.01o.

E

osCOo:ooo4-1

X

I14-

4->14-

VI

t.ai4J

IO

u4-1

t.1-01

4-1C013

"&

4t«

*

3-13

•oc"0

WAESD-TR-85-0030

oo• cooo

a:

oo

Q CO

a: ur>r

UJ<a' o

o~E £

•Si 51 •'•OOOJ

00 1/1 t/> 01

S- S- S- !_< a. «t <t

o £3 a o

T3 .-O T> ? T3

o> 01 o> o>I- S- I. 1.

oo o

CDOUJ tO

GO GO"OO '<UJo:Q.

CO "

LU01

OO

CMX

CM .

OH-

«=•o e

i.CM -a: o ~

o o o Eoo o ^

o ui

=3 ». •—3 i—

" O OJ"

3 r—O r-

o >O UJOJ

<uo

-

Ql- *-« a;»- a:

«M CM CM OJCN* CM CM CJ

O LO O> i—O CNJ CJ> or~ 10 vo r»

T3 .

O

U

O+J10

co

VD 10

< — cj>CJ> CO

U3 CO O> i—OJ CM - OJ roO O O OI I I I'

I/) I/I I/I (/>•Q. Q. a. a.o -u < - > < _ >

S

.§Ch

-g

OS• JT

Ol

°OJr- EO <J

o no .a «£* . *.-*

3-14

-----™ ~._ _._ ^ —_ --WAESD=TR-85-0030~—

Only preliminary results were obtained from this test plan, since there weredifficulties associated with the functioning of the facility and some prematurecell failures due to unknown causes.

Test Plan - TS 2X2-005, Rev. 2 and 3

The aim of this plan was to characterize the baseline performance of subscalecells by mutual exchange of cells between ERC and W_. This type of "RoundRobin" test also enables a comparison of the two facilities.

The approach was to fabricate and assemble eight 2x2 cells with the pertinentvariables being the assembly facility, test facility and test cells with andwithout MAT-1. The test plans are shown in Table 3-7 with some preliminary .results.

These tests are in progress both at ERC and VJ. From the preliminary data, itcan be stated that W^built and tested cells averaged 659 mV at 200 mA/cm(186 A/ft2) and 190°C (374°F) with 80 percent hydrogen utilization and 20percent air utilization, whereas the ERC cells under similar conditionsaveraged 660 mV. Further testing, in terms of rebuilding and exchange, is inprogress.

Test Plan TS 2X2-006, Rev. 0 and 1

This test plan was designed to evaluate alternate catalyst/electrodeperformance and also to perform an assessment of vendor manufacturingprocesses. The test plan for this task is shown in Table 3-8 and all the cellswere evaluated under standard atmospheric conditions at 190°C (374°F) at200 mA/cm2 (186air utilization.200 mA/cm (186 A/ft2) and 80 percent hydrogen utilization and 20 percent

It can also be observed that Prototech (DC01) could be considered as analternate supplier of catalyst. Giner fabricated cathodes (G82-5-10 and 7) hadgood initial performance which was better than the baseline, but the electrodes

3-15

WAESD-TR-85-0030

oooI—Io:o.o

0

OO

ioo

o coLc •>>.a: inLUf\ i|« :

CQ I-H<C —J|_ LU

oo

CQ

CO

0£.

og

0£.

LOop

.- x,-CVIX

ooLU

UJ J(O

ooocn"

CM

O

i0oCM

1

1 * -

£

1_

5MOCM

__:. .

'CM'^p

>«oco

•

^^B iinco

•p*jji•ocoo1 *in

I—

UJ

2£.••«t

UJOot-o

I/)

I—«t1—t/>H-l/lUJ1— lt-14.Z °UJ r-1/1 IO

cc o0. r-

£;

UJi/tV.30 r—I i—

Ol

8 °O UJ

^—

VI1. f—

o f<u31 o

O UJ.OVD

ini-3 ^-O r—

W

O UJ 'CM

<UU« s

t- 4J•M inc "v>

C£.

^™Olu

4Jc01

oJJinCD

1—•oc

r™

E=

inin^cr-

<Oz:

•ci/>

••O

OCD

.4->

O

•4->C

l/l

,•o

• co

CD

•4->10O

inOi

int-Q

3-'

^

4^^

"

£

^»

*— *

*•"*

E*--f.

^_^

i

o»-

, Ol, *•'

o•4J

0)

T3

T3. Ol

^~

*0)Ucroo

ui

4->

3

t.O

31

2:

013

>.t.O

31

X

coo

*J~ .p^*~

o

c?in

COco

ISin

CO

00

^o

1

 -M

3 3

l_ 1.0 0

3 31 1

£ =

o> o>3 3

£• £•0 0

3 31 1

3: Z

COCO

*— o

CD

*o ^>^ "~

O1 r—

1 OIO

1 CT>in

CMCM0

• or— OCM U3O

1

CM01

IO

O^ f^

oi

co »•>r— CM0 0

1 1< <0 0

fx.Q>

O

IO<t

3U.C

cr>i<

in<o

oZ

4J0)3

&.O

3

'I

Ol3

>.

0

31

X

^^r-

GO

Q

z

|

,

1

,

CM0

1

<O

<_>UJ

04->

. C• Ol

14->-

Olr-

J3

U.

3!

I/I

cn

i.0

T3

I/I

1-

>,i.O

31

r-

«*

^1 —

tn

voin-10

co

CMl£>IO

nIO

ooi

<i<_>

4->VI01

Oo:UJ

r—

J^

CM

1

co

onIO

0

co

occUJ

o4->

c.Ol

14-*

<ur-

J3IOU.

3|

o

cn

i.0

•oJl

i.<

>)<-o

31

X

«t

^1 —

CO

inroIO

CM

«»1010

o

1 —

st<o

4Jin41 "K

Q)0 t—

JD

u_3 1

0>

1.0

"O

I/I

i.1t-o

31

t—X

t — «3"

fS r—1 —

CO e~^

I §«o

CM CO

ro i~-r— ro10 vo

COO •r— O<

10cnOl

<tO

4->

Olt—

nu. •*g l

o

cn

L.O

T3

(/>

i.«t

I"

3l

1—X

«*1^ •^

^-

0inIO

CM

IOIO

or~

0»o

<t<_>

- '

•o0)X

•oou

*s>J=

IOCO

cnL.10«J

*10

T-,Ol4-*

*r~JQ<ou.OJ •i/> in

0 r^

cn uc

•<- t.+J 01HO 4J01 in3 ••-in•i•O r-Ol IO4J (J« -r-Ol 4-1t Ct- 01

+•* •*—

01 *->

O 3O J3O Olo ce.cr*

3 ^fi «?3T '*--'

3-16

WAESD-TR-85-0030

00Ioo_

CQ•a:

ooQC

UJ_lUJ

o

o •-»••-i CO

Q- CO

o in

COro

103+->0}4Jto

cc:

ooICMX<M

oocr>

<aCME

rf

ÔOr^„

co

IO

V.

sCM„

-'3

CM

*

Xooo

4-*

,_

VICo

•ooo

VI<uh-

VI

c<uE

0

4-"VIHI

t—

cIO

>,jp§tn10

i4-»•O

4-*C

OO

UJ -CD ^30 cS= 04 m

IOcj

4->C

I/I

Uf •/—\ ^H0 cX 0t— 00

o

u

on •+_ vi=3 V^ QJ

5 ov~in r- in

•a t-1- 4J 3I/) o OII 1 [ *T*

1—

UJ ^-~-

UJ o 6

VIuo •—Z r~ *"^

CU *

O — •O UJ

~VIi. r—

s sô uj*— *o

i.O r— *—•

U E

O —O UJCM

__ s"io %—

5 "SC Irt

O>-i

on

<c

w

•o

3

inOo

CO

. SI

inCMCM

CM

CM

VO

ro

^VD

*!•

r-I

VDCMO1

0

OO

*

0)3

•o

<u

ino

^

CO

CM

inroCM

rovo

•3-VO

VO

vo

.

vo

*rvo

CMo

<t

TJ

aiuc

</1

•*

cu3

TJ

3

S «T<j So a

CO^ i

vo

CM

CMro ivo

coro ivo

COro ivo

CM

VD

rovo

oo enCM CM

? o

<J 0

oo oo

<t <*

cu at3 3

•a -oOO 00

3 3

vo vo0 Oa o

CO

CM

(*** f

eo t—

S CMen

vo in

CO ^*VO COID VO

CO CMVO COVO IO

CO Or-. coVD VO

vo vol»» VO

0 •-co ro0 01 1

4 <

oo

•*

a>3

•o

01.3

oa

•~

voVOa>

en'in

COCMvo

COCMVD

inCMVD

CM

ro

CMroOi

o

"g10ui.

JO•ou.V.o•acat

l/> OO

< <;

a ai3 3

•o -aoo oo

5 I

O) <X3 Z

I-. VO0 0

0 0

CO

£1

co coO cr»O CM

en i—CM "f10 *r

CM COr^ rovo in

O CM

vo vo

CM ^l

r~ «*•VD VD

O •—oo oo

ro **ro ro0 0i i

S'<0 U

enc•

IOCO

31

•o iJa) cua <o

"£ ^JO -010 <0u. u.k I.0 0•0 T3C Ca> a>

l/t 00

< «s

AI Qj)

3 3

•O T3

I

i ir-. vo0 S0 0

CO CO-N. —CM •—

o voin f—CM ro

in coCM VO

o onen *!•ro vo

in co«*• min vo

r-. •»CM VOvo vo

CM i—

CO CO

vn voro ro0 0i i

00 00

u u

en.c

uIO

CD

31

•oataui.

JOIOu.

o•oc01^

I/) OO

1 <

HI 013 3

5 5% £

l.z <

4 OlZ 3

° g0 0

00 CO

•~ •"

O CMr^ r^»

o roro CMvo vo

S CMvo vo

O ^ÎO CMVO VO

in rop» coto vo

<r enin vo

r. coro ro0 0

S 5

cuucIO

0

cua.i.o0a.s-l

^yCU

IOc

01

00

4

013

•ooo

1.5

013

CM0

a

CO

o

en .CO

roin

i

•

,COvn

CM

m

enSoo

0 .

•o0)X

£

r—0

UOO

u4->iaCO

01on

1

00

3-17

00I

ro

ca<c

Ooa:oUJ

o%•z.oooI—4

a:D-

o

>-<.

oS

o

c/)

to3

oo

UJOL

O.OI

CM•X.

CM

C/7LU

CJooCT*

<a.OJEU

1ooCO

i•

co*->ION

f

5oCO

£sCO

DC

iEef.

Inc'0

•oco

4_>I/ICU

I/I•McuEoo4JI/Icu

1—•o10

.i"

EcuI/Iin"*

i>0

c

LU •C3 ^3O C

a. to•

IO

Jcto

LU •i— 1 ô cI Of— ooo

IO

1 4-»1/1 H- I/I3 U- CU1— Oh-5 °tO r— I/I

IO k.1- 4J 3to o oLU 1—3:

LU *— --tO Q)>LU U E

O. LU

ini.3O •—

0 SI

O LU

r?

m

^ °^C? LU^ôIO

inL,3 r-o •— •—

u 6O ~--O LUCO -

O)u

15 ra-~• " 4-* CJ

•t^ 4^-^c in

i— c CUcc

r_

32

* <

3 3

CO COO 0

=• £

< «

3 3

to inS °0 0

§ S

*f CMro ooCO CM

CO voin CMLO \o

r- LO

IO ^

co mVD vO

CT* 10

S £

in r-~in 10

i • ?u u

COEoOl

in

o•o10o

cu•ooc

<

3

^v

oCJ

<

3

1

1

COCOIO

in

i

Ol

^_oIO

in

91

S1

CJ

•oHI

cuucIO

1CO

eu

fCO

o•Q •10o

cuT3OC<

«t <

3 3

CJ OOO 0

0 0

5. <

3 3

CO IO0 So a

,

ID1 Sin

i *»•

' §

1 00IO

IO1 CO

IO

CO

' ID

s §1 1

0 CJ

<

3

(VJ

o

«t

3

§H

o

in

ooin

COr-.

£

invo

r-.

^

Jo1

Ô

5

•*

3

COoo

«t

3

oo

sCO

in

oID

CO10IO

r««IOIO

CO

^

21

(/)

CJ

•*

3

COo

<

3

0

O

COen

*

IDcn

IDIO

IDID

COIOIO

en

00

S

1

S

TJCU

'euucIOo

o oc o c o c o c o i / i i / i t o z : z c o < / >

C U C U C U C U C U C U S ^ S . ( - ( . C U3 3 3 3 3 3 0 0 0 0 3

ro co co"o "o o o ^ x* "o "o "o "o ot o t o o o t o t o t o t o t o t o o

Ol C71 Ol Ol ^ I- Ol Ol Ol Ol I-

< £ < £ < S < S < £ S

k , L - t - L . C U C U t . L , ( . c , C UO O O O 3 3 Q O O O 3

O O O o O O i O r ^ i o r ^ r Ô O O C J « - » C J < - > C J C J C J « - Jo o o o o o o o o o o

' •* 1- »t t Tt ' «»CO 00 CO 00 CO 00

1 ~^ — ^ •>. ~^ -v.

v o i o o o c o ^ o o ^ o ^ a - î c o c o ^ r ^ r i o i o ^ - e v j ^ t OC O C O C M C ^ r — O O O C M C O ' —^ - ^ ^ - ^ ^ - C M » — C J C M ^ X J

m r ^ - c o o ^ m O m r o ^ o i î o o ^ c M U ^ C M ^ c O O f — r ou n i n u n i r i v o i O i n ^ O i o i o

r - i n i n u ^ c u r o c M a * c o O *1 O> CT* ijO ^3 F~ f1^ r1* *^ VO ^~

m m m v o i o ^ v o ^ ^ i o ^ o

o o v o m r o i o c v j c o i r t m * * 'i o o ^ m ^ - r — r ^ c o f ^ v D 1 ' 1

i o i / > t n ^ c v D ^ O v o v o t o v o

c \ i c M c o r o c \ j r s . c o a » o « ^1 r— ^3 IO CO CM f**^ O^ *** f"^ ^*

i D i o m v o v o t O v o v o v o ^ o

i n r * . o ^ r o c o c o ' ^ - a « ^ o

r*« f** vo ^^ co r*^ vo f*1* r** o^

i n v o r * * o o o > o O ^ - c o c o ^ 'L O v o v o t n u ^ v o o o o o Oi i i i i i i i i i i

t / l t / > « / > ( / ) * ^ > ( / ) * ' ' l ' * / ) V > » / ) * / '^ ^ ^ ^ ^ ^ < C ^ ^ r f « tO t_> (_) CJ (J tJ CJ CJ *-> CJ CJ

t/>

i.

<

iCOoo

t

<

cu3

IDo

roCO

COCOIO

CO

IO

IO

oCOIO

or-.

ino

to5

WAESD-TR-

to to

Ol Ol

o o

TJ -O

to to

Ol Ol

1. L.0 0

i— CO

CJ 0o o

oo Sin in

COCO COCO CO

CO

CO «3-«*• in

i i

, ,

CO Olin co

COCO. •

r*% co

10 co0 0i ito i/o

c » cj

-acuX

IE

"oUto^.uIOmcuOlV.IO

to

3-18

co00UJooona.

oc>o

Q.

O

a:UJQ. ooQ_ •—

COCT> CTJ .i z: un

co i— i^ 4-

UJ O O

CQ CQ<C|_ UJ

W)as

•a: 3z: 4->o: n)UJ 4->h- oo

I ^^^^«cCM

UJ

OO

CMX

CM

ooo*—C2*

CNJEu<^0oCM

'• t

Co

•eM

r—

^

X0CM

•4-1

CNJ

SCO

«

E

^r-

(AcO

^f*uco0

VI

. h-•oc

ftEOlinin*

iIO

cIO

UJ •

o 1?z o4 CO

•4J

CJ

JC

Ul •o -a0 cX 01- CO

V->

*j^j

10 if. in3D u_ Ol

51—10 i— in

ID i_(— 4J 310 0 0UJ 1— X

1—

Ul r-—~to o»UJ o Ece. —O. UJ

int.3O r—

X i — • — •0 SIo ^Ô UJf

ins- .—

. 3 .—O OI-~ *

U Eo uj^-ôVO

int.

O i— ^-

U E

O ~—O UJCM

Ol<J

•"" C

"•*" 4-* CS4-> VI E

c ince.

i—OlO

0— •

VIOluo0.

•o4J

•UJ

1-fe4roCO

f:

infûl<oOl1.3

IO

<t

0>3

TJ

IO

*

^jOl3

•o

CO

•—

§~

CMr-in

00in

f;in

rocoin

OCO

o0itot_>

£uoI.o.01

in<o0.

•f.UJu.1-%4COCO

A

ID

fÛJ

<o.cOJ1_

^

<o>

T3

<4J

T3

to

IttoCO

^- 'COo

COin

in00in

coin

CM

in

CM

CM

*~

__

o1

o

'inVI01uoi.a.

•o

«UJ

l—^4CM

^^r,

inpÛJ

tojraii.3

5

T3

l/>

•*

4-*ai3

•oto

1•—

CO

oCM

inroin

r-in

inin

§in

r-

^

CM

O1

to<_>

-in-inOl

oa.01

in<oCu

«»UJL^1-

XCM

f.inf*ûlto01

3

l/»

5

Ol3

T3

*

4-»Ol

T3

10

00

ro

COCNJ

•

ininin

COvoin

ooin

ovo

.-<-

roStoo

VI •ai u0 °>- Ova.

01

•a in4-> IO(/> O.

n •Ul UJU. U.

*— h-*< *<

CO CO

r. •

VO VD0 0CNI CNJ0 0a. a.tv 01o oa. a.0 (J

to 10

to .to

5 5

01 Ol3 3

•o -aCO to

<; «t

4J 4->Ol Ol3 3

•O T3

to to

oo SCNI CM

t CMro ro

" "

uo ro•**»• nm to

co roin 10

CM COO roVO IO

CO •—•— roIO (O

*— ino^ r^

^r mS §1 1to uo0 0

o

1-<£

a

•o

Ol

• • i3^• t.a

X

Sin

8VO

"invo

invo

CMvavo

invovo

covo

roOitoo

0z

s-<

£*a

•oto

Ol4

^

L^

X

sin

o0vofmm

roinvo

invo

r-.r«.vo

inr«*VO

ON

in

.rÔ

toU

in- inaiuoi.o.01

in10

cuco

u3•oOL.O.

CaiOluoto\s

to

oz

t.«t

£

•oto

Ol

*^

X

s-v.in

8vo

"".

VO

CMCMVD

5VO

5VO

r-.

•^

inr**0

1tol_>

inVIOluo1.a.tvVI

o.co

u3

i.a.caitvu10

*s

IO

O

Oli.•X.

^*o

•ouo

Ols.

J^t-Q

1X

coin

8VD•

rovo

rovo

invovo

^0<tovo

r»vo

VO

i —o1IO

o

WAESD-TR-85-0030

aiX

S

2.0JZuto

u4->

CO

01Olt-10

,

to

: •

•oOl

4-110014-t-

. 4J(OaiX

oo0o,

1—X

3-19

- ; , WAESD-TR-85-0030

had high decay rates. The performance of this cell is shown in Fig. 3-1.Vendor cathodes DC05 and 10 provided acceptable initial performance and DC04also appears to be good but these electrodes need further endurance testing.Alternate catalyst materials DC-11 and DC-12 initially appear to beunacceptable. A number of these tests are being continued for up to 5000 hoursand a detailed analysis of their performance will be delayed until the testsare completed. However, based .on the available data, it is apparent thatreliable predictions regarding the performance degradation of certain catalystscannot be made based on test duration of less than 3000 hours.

Test Plan TS 2X2-007, Rev. 0 and 1

r

The objective of this plan was to assess the performance of alternate backingpaper, alternate wet proofing methods and effect of changing Teflon levels inthe backing paper on 2x2 subscale cells. The performance histories for variouscells in this test plan are shown in Table 3-9. '-.'. .

One series of cells, CAS-040 through -043, was fabricated from electrodessupported by a possible alternate support layer, namely, Kureha E715 carbonpaper. The backing or support layer was wet proofed to two levels of Teflonusing two processes, the standard dipping process and a paste spreadingprocess. The terminal cell voltage of these cells was 600 mV or lower and thecell resistance was high, 11 mfl or greater (since the backing paper

resistance was 50 percent higher than Stackpole paper). The poor cellperformance can tentatively be attributed to the higher resistance of the cellscompared to baseline cells. Additionally, the rapid degradation of the cells,approximately 12 mV in 800 hours, indicates possible flooding of the backinglayer at both Teflon levels. This data indicates that the E715 paper is not anacceptable alternative or substitute for the material presently being used forelectrode layer backing.

The use of the alternate "paste" wetproofing process, or "silkscreen" process,was found to be comparable to the double dip process, and average cell voltagesfrom the two processes were similar. This indicates that the silkscreenwetproofing process is an acceptable alternate process for further study.

3-20

WAESD-TR-85-0030

ru

QZ<r

co

03O)

az

(40

U)GG

I

'J,-Co -

SSG)in

G)G)in

G)G)G)

in

G)

G)

G)G)inoj

G)G)

in

G)G)G)

G)G)in

C/)

G)G)CD

S.inrs.

(9G)inu>

G)to

G)inin

G)G)in

- 3DU110A 113D

u

OJQ.

O)

Ico

o>

3-21

WAESD-TR-85-0030

t

test Man 2X2-008, Rev. 0

A total of ten cells were built under this tes,t plan to evaluate the effects;bf(1) dry and wet bonding, (2) alternate sintering environment* viz., steam, airand argon. ' This plan also was aimed at providing an acceptance screening ofSchold Mix MAT-1 compared to Ross"Mix, MAT-1. The performance of the variouscells is shown in Table 3-10.

The results from wet/dry lamination comparisons conclusively show that drybonding produces an acceptable result, which was also shown earlier in TS2x2-002. Sintering of electrodes iii argon is also an acceptable-process (alsoconfirmed by ERC test plan TSE 2x2-006 results). There was no significantdifference in subscale cell voltages whether Ross Mixed MAT-1 or Schold MixedMAT-1 is used,; and this clearly shows that Schold Mixed MAT-1 matrix isacceptable.

r.i: . • '

Test Plan 2X2- 010, Rev. 0

The main aim of this test plan was to verify the quality of electrodes for a10 kW stacks Seven cells were assembled and tested to assess the variation inthe performance of electrodes within a batch and batch-to-batch. During theinitial phase of this plan, the baseline processing procedure was revised dueto the change of electrode sintering gas to argon; hence, this test plan wascancelled.

TS 2X2-011, Rev..O, 1 :and 2

The purpose of this test plan was to study the effect of alternate acidtransport medium, especially zirconium pyrdphbsphate insulation material,Teflon content in MAT-1 matrix, and the thickness of SiC coating. A series oftests Was designed and cells were tested as described in Table 3-11.

3-22

CO

co

o,o

UJocoCOUJoso.Cfl

2 ^Qi 00UJ *

5 5CO «J">

CO.

CO Z

011)

t— I COo =»

LJo§

COooI

CMX

OJ

CO

o2

CMEU

1OoCNJ

1

•»

4->3

&-S

SCM

•»-

CMT

O

B4 .^_

gi

Co4J

!cO

i/iCU

UJ

\

UJoo3:K-

CJ

!*•It-

S 04 fl)r- 4-»I/I O

r-r-l/>UJ .K-

UJ •—

Ul O

O. UJ

VI

3O r-

'euO 0

O UJ

—I/I<- r-3 p-O CUx u

• C5 i .1

toi/it-3 r-O • —x cu

O UJCM

CUU

f— C

t- 4->

ji .12C l/>

•-* CUae

o

VI

c

p

o*•*

IA0)r-

C

,_*n

i<IO

4-1C

00

TJcO

CO

54JC

no

•ocs4-»•o

+Jtn

VI 'uo

^^

3.

^^

i

^

^^^

1

<3.5.

a•—•

u10•oca

i/>

o

^j• *

T*.*.o3•a£a.cu

ae

ce

L.

4

*

•otn

4

4JCU

•aI/i

ro00

CM

sinp~

COCOto

a>roto

toto

roto

Or .

o0

I/Is

roCM

•*

—4uVI10

X f•P- OU 4->*J 10a co

0)•0 TJC 0<D V.

01 U-o cuO r-1. UJ

8 IPr- IOCU t/1

in in

U 4J/ CU4 3

s I13

5i x

• CU4 3

4J >,0) S-3 0

S hi4-» r—OO X

CO COCO CO

CM i—

to inCO COin tor— . r—

r— CMco oin to

r— Ult- CMto to

co roCM CVJto to

to roCM CMto 10

CT to

to to

CM CO

S O01 1

t/1 t/14 <ta o

roCM

«*

!-

4U

VI

_cuIOCO

ai•ao».uCU

f—UJ

CUE<o

« Ul

4J 4J0) O)31 3

0 0

3 2X i/i

0) CU3 3*

>> >i

fe £

1 T>

CO COco cor- CM

in COro r—us r-.r— r~

to roto to

to co£ S

CM «rto to

«* oCM *»•to to

OO C7I

in to

^ tn§ o

01 1

0 Q

I/I OO

C7I 4->V. W4 3

a 3

T> T»

l/> OO

t. d)4 3

>» 4->

fe 5

•o -oI/I 00

co ooCM CJ

•* toan oor- inro »—

«t tnf— roIO IO

in *rto to

o toto «crto to

in «»•to to

P- w—

10 r*

to r-

8 8t/i t/i4 4O O

OO

C7I

4

3

•oOO

01

4

4J

3

•D

S

^

CMtotoCM

oCOto

COinto

ininto

into

«»•to

cooo

4O

00

en

4

<=

•ooo

1.4

>,o

31

X

CO

V

totoCM^f

toto

intoto

Sto

toto

CT>

to

cr*0ooo4o

ae

enw4

0

TJ .

OO

U*4

>,

0

3

X

CO

CJ

o

CM

CM

to

COto10

IOto

toto

CO

to

0

o1

t/14a

TJCU^f.I/IV)oce

u

COr_

I1

Of,

TJW

! au

u.CUI/IaoCJ>c

1/1CU3

TJCU

IO<uh-

4-1

CUX

31

I—X

is£TJ

*O

u

cu10

CO

CUeni.IO

100

WAESD-TR-85-0030

3-23

.WAESD-TR-:85-0030

o»— 11•^^

••

J

LU

LU

CO

t H

o;«=c

a:oa.V}SZ

1Ot— I

feeCD

<cCOCOLU0On'O_

•— CD

1 | I

CO O£

iii *'~ *LU . I—:^j tJCQ isCc3* LL;1— =3

1

CM

m

LUa:0'_^p

•=cl_

•

LUc£-

0

•

1 1 1

a:'

i—oo

1,CMX

CM

COLU

"5"op

coint j6*.±ftJ

</»3

-[ %

<aCO.

ooOO!

IS.

CM

g

i0o<M

1

•

5

i.

w0CM

it

-

4_l

CM

X000

6

^

• •

c0

TJCO

°in

K?

C 0CU "O - , . -i- O6 —

. • TJ o '. . £ oV) i — (O . T? "Ocu •— cj : o o ' • •t 1 U l l I . I 4 - > 4 J I I

TJ *- C . Vr- V- ',. t^ n) « *» <->

CM "D T3 ,. > ) C l C C C C U C U C Ci — O O O O O 4 - > 4 J O O^ »— o m a. ' — r— ' — c j i O i o i — ' —

§ < t - ' » - f — * * - ^ l 4 - ' f - O O »»-**-a ) o o r ^ c u o c u o o c j o c u c u

</) H* LU r— f— O ^ -f- h- h - ^ ~

i - i - c u o c o - r - o - a . -~~^

S o o z ^ o o - o o ^ o o t — " o o z r s j r > i ^ r ^

c- I T . O O O O O O O

• ' 'LU 13

^ «* «I <t <t «t

.10 < D O ) L . l . S - S . S - S _ U S - l - L . S_ 1-<_> 3 H O O O O O Q O O Q O Q Q

^' " - o - D - 0 - o - o - o - o - o - o x . - o - o - o - o

: oj*, XLU • . -i—0 T3 ECO C ' • CT1 O) CD C7) C71 Ol O) C7> O1 C7) O^ O1 C7) CD

<c o(_> - . .C

o ' O Q O O O O O O O O O O a 1=1 co10

4-> COi*- >n •. .

oi^ i i i i is- t i jE i is : a: 3= . ' 8>

Z3 1O

« C l O l > C O C O C O C O C O C O

00 ,O O CM CM; CO ro LO LO. -r" •/%^ 1 - 3 C , 0 0

OOL U O « - « * T f r « 3 O C O C O C O C O - C M C X JH~ O Ol if m ^^ ^ WO IO C3 <*^ CO CO

f~ ^* r™ CO CO CM CM f ^ lm~ r™

LU r— *— * -OO CU >LU O Ect: . * — • OCM c o i r > i n a > c o * t c T » o o o * *O-lij ^ t T » I I C M C M C S I ^ - C M C M » — I O C M C M

VO LO LO 1O LD LO LO LO LO li*> LO ^OVI1-3O i—

0 *" ?

O O J ^ L O I l S r ^ S L O L O L O ^ f C - 0. - - V O L O L O L O L O L O L O l O L O l O ^ J

VI CDt- •— V.3 r— t—O OJ ^^

o LU — - « j - co *— m c o o c O f — * * c o cu

oVI • Ot- o3 .- 0O r— •*"* O>3: cu >0 ° -S- ' LO to o< ov ' in r. ro ro c o « r 'O U A • • c o i r > i - i L O L O C o v o L n L O « — C M r s j C M v —

cuo

•— c ___••- *S a"*» «i e J - '•r— -r-. *•«-* CO IO • ^~ ^~ ^~ LO CT* r-" ^T LO CM ttl

»— i cu - CO r^» LO LO- co r-*' f^ r^ ^~ f— CO cooc. •— •—

r~ CM CO f in LO C7^ '"C3 " CM CO f r 00

O O O O O o O o O O O o O O•— .1 1 1 1 1 1 .1 :). 1. 1 1 1 1 1 I

O •— O Q Q O O O O^ O Q Q O Q Q O

3-24

WAESD-TR-85-0030

One of the important conclusions from this test plan is that the insulationcoating from zirconium pyrophosphate is not acceptable since the cells hadlower cell voltages and also 60 percent (approximately) higher resistance thansilicon carbide. It can be seen once again (previously, shown in TS 2X2-002)

s "

that thin SiC coating is highly desirable (also confirmed by ERC Test Plan TSE2X2-003 results) and the measured resistance of SiC coating is about.02-.03 mft/micron (0.5-0.7 mfymil).

The effect of MAT-1 carbon layer on cell internal resistance was assessed usingsubscale cells DAS-011 and DAS-012. MAT-1 was assembled in DAS-011, butomitted in DAS-012. After continuous testing at 200 mA/cm2 (186 A/ft2) for200 hours, the polarization data and internal resistances were measured at190°C (374°F) and ambient pressure. At the current densities of this study, nomajor differences were observed in the internal resistances of the two cells.Since the presence of MAT-1 increases the interelectrode distance, one wouldexpect a higher resistance in DAS-011 than that measured in DAS-012. Theadditional resistance was not detected, presumably due to the electronicconductivity of the MAT-1.

Figure 3-2 shows the IR-free cell voltage vs. current density relationships forthe two cells. At very low current densities, both cells exhibited similarvoltages. With increased operating current density, however, the voltagedifference between these cells became significant. If one plots the voltagedifference as a function of current density, a linear curve is obtained,revealing that the MAT-1 in DAS-011 contributes approximately an additional3.1 rrfi to the overall cell internal resistance. At 190°C (374°F), themeasured electrical conductivity of 100 w/o H3PO. is 0.5/ohm-cm.Assuming that the tortuosity, factor for a loose random pore structure is /3,the calculated electrical resistance through a typical MAT-1 of thickness0.229 mm (MJ.009 in.) is 3.09 mft. This estimated value is in goodagreement with the additional internal resistance of 3.1 mfl obtainedexperimentally. In conclusion, the inclusion of MAT-1 increases the cellinternal resistance by ^3 mfl and the inferred resistance is about,012-.02 mfl/micron (0.3-0.5 mSVmil). As a result, the terminal voltage of

3-25

WAESD-TR-85-0030

0.95

0:90

0.85

§ 6.80;O

0)

ow

C

6.75

6:70

6.65

6-.6010

Subscale Cell Tests190°C, 1 atm., 100 w/o H3P04

Air Flowrate: 860 cm3/min.H2 Flowrate: 80 crrrVmin.

DAS-012-(w/o Mat-1)

DAS-011(w Mat-1)

I

20 30 40 60 80 100 200

Current Density. mA/cm2

300 400 600

Figure 3-2. Polarization Data for SubscaleCells DAS-011 and -012.

3-26

_ ......... ....__.. . _ . . . . . _ . . ._. WAESD-TR-85-0030

a subscale cell without MAT-1 at 200 mA/cm2 (186 A/ft2) should be about 16mV higher than the cell with MAT-1. However, additional data are required tocompletely define the resistance contribution of MAT-1.

TS 2X2-013. Rev. 0. 1, 2. 3 and 4

The major objective of this test plan is to verify (using subscale cells) thequality of the electrodes and cell components that are incorporated in thebaseline Stacks W-009-10, W-009-11 and W-009-12. A series of 27 cells werebuilt and testing is in progress as shown in Table 3-12.

From the tables, it can be observed that the QAS-006-009 (SI) cell voltagesranged from 635 - 652 mV with an average of 645 mV (based on 200 hour data) atthe specified test conditions. The same electrodes in cells QPS-009 through-012 performed at an average voltage of 640 mV (range 627-647 mV) with SRG at485 kPa (4.8 atm.), 80 percent H2 (SRG), 20 percent air utilization,325 mA/cm2 (300 A/ft2) at 190°C (374°F). These are the electrodes thatwere used in assembling in Stack W-009-10.

The electrodes that were obtained from heat treatment run No. 4 wereincorporated in Stacks W-009-11 and W-044-1 and these were tested in subscalecells QAS-016 through -019, -019-S1 and -019-S2 under atmospheric pressure andin QPS-017 through -020 under pressurized conditions. The atmospheric testcells showed an average of 648 mV with a range from 624 mV to 680 mV (based on200-hour data). The same electrodes in pressurized cell conditions with SRGexhibited 659 mV (range 647-669 for 3 cells) indicating that the effect of SRGwas about -25 mV as expected with the range from -15 mV to -30 mV. Theseelectrodes appear to be acceptable for stacks (with a minimum performance >630 mV) and will presumably aid in understanding the data from the stackdevelopment program.

TS 2X2-014. Rev. 0 and 1 .

This test plan was designed to evaluate alternate catalysts in terms ofendurance under pressurized conditions. During this reporting period, tests

3-27

WAESD-TR-85-0030

Iro

ZDcrLUooQi

o

03 IT)

CfL

=3

LT)

IV)rO

rd4->CO

CO

5IX

CXIX

co

t ^

oOen

<3>CME*«

•=OoCM

«Co

N

+J;3

(-•r-•X.

OCM

•t•

4JO

CM3:

oCO

E«^

f—

inco•r-+J

•oCO0

inCDr-

incCD

g

OC-J

4JIftCUH-

TJCIO

.aE<1>

in"*

i(O

4-*

OO

UJ •o -oZ 0a co

*0

+f(r.

00

UJ •

3: o1— CO

o4J

O

oo i*_ in=3 <£ <u<oo r- in

1- JJ 301 o 01— *""1-Z f—UJ .— • — •OO QJ >UJ u Eof —*0. UJ

inj-O r—

oS?o *~ *O UJ

*

l/t

^ ~O a)-— -z o >

E

8"-VO

1-O f— —

o Eo ^—O UJCM

0)U

•— c^

c "wioe

•—0— •

|

rO, l*-

W

n)•o

J?_ • —ro r—c n)£ CE -r-o Ec o

ccen 5>- o

o o o o oZ z Z Z Z

en en en en en

«««t«

>> >i >> >> >>=> o o oo

o-o-o-o-ouo oo oo oo oo

en en en en en

>i >> >> >i >>3 O O O O

M CM CM CM CM

E 3:3: 3:3:

o co oo cocoCM CM CM p— ro

ct vo in vo oPO in o cooo o o CM o

n oo o o CM*!• ro o r— rovo vo vo vo vo

ro CM 0 CM•* «» O i roVO VO VO VO

«3- r— CO CO**• in ro i ro

co CM vo en m«• in •* o <">10 vo vo vo vo

co •» o oo ooco co en r— co

^~

oo10 r- co en o>

3 O O o Oi i i i ito t/> oo oo</)cro-o-cyo-

^(O rC 10•r- C

C Oc

en -g•r- O3; _j

o o o o o o oz z z z z z z

Ol CT> O1 CT> O1 O1 CT1

<<t «t< <t«t <

>,>,>,>,>,>) >,a o o o a o o

-o -o -o -o -o ~o -o

en en en en en en en

>i >> >> >i >> >> >io o o o a o o

,3. . - . - j.r— 1— t— 1— (— 1— t—3; a: 3: 3: 3: x 3:

CO CO CO 00 CO

in in in in

CM CM CM O O CT> ro

ro ro ro »— CM vo ro

co o p- co r~ CM m*a- CM m en en CM fvo vo in in in vo vo

•— ro T»-in ro r^ i ivo vo in

CM ro in rovo *»• CM I i CMvo vo vo vo

p^ o r— oo r-. ^J~ vovo co «j- en o CM ^~vo vo vo in vo vo vo

'CM CM r*. co ro co i —r p*. r< r « en oo en

t— CM0000

vo i» oo en en en o

it i i i i i

cr cr cr cr cr cr cr

^_t— IOIO CC 1-

E Oo cc

JC

0 •?_J 3:

o o o

CTI en en

>> >i >^o oo

•O -OT3

oo oo oo

en en en

><><>>o o o

«r *f «*•3: 3:3:

<r »>• i-o o oro ro ro

r*. i — ror— «*• invo vo vo

co ocnCM in «*•vo vo so

in co CMo coco

in vo ro ooi i i

OO l/)00

cyo-o-

•<uX

IET3

'o<J

oo .

u

CO

QJ

ro

1

in

•o i—ai ai4-* IJm0 C<j> <j)

"O1 -r-

^ i-1

3XI0)

CvJ

1

to

»—t/)

1*i"-'

3-28

WAESD-TR-85-0030

CNJ

1coUJ

co«*I —

o1—0

*" *

1—UJh-

0

\—

OHH

1— 1cnUJ>

pI— 1_J— io-UJooanI—UJ_iUJ

«

in

ÛJOL

CC.

O

UJQ-*

CO

O1

C\JX

CNJ

1 —

UJh—

1co^^^LO

*f

o

to

•

^J*Oi.

oo

0oO*

<&

CM

SiinCMCO

iu'<>«0CM

Oaioo

CMX

Xg„

E400

^

I •l/>co4-»

•oco

•M(X*011-

Ul

8

UJooi—«c

•S~

oo1—UJ1

ooUJOfa.

t0

0oo•—

£o

o0

VIJ-0X

ooCM

«o1 ••_

in4JcOl

U

V)Ol

t—XJc03

"agin

•oc

oo

.+1

<->

_jjcoo

T3Co

CO

•

o

^tit S>— inra l-*J 30 0(-X,_

0) >

w^1.1

,_

U EUJ

,-"QJ^-X

E

^_r— "~«Ol>

0ÛJ

0)Uc

I/I c

sac

^

O«— '

<O i—C 09»- C

i 1c

CT ^£ 5

O O O O

CTI CT* CT> tyt

««X >>>>>>a o a o

D T3 T3 T3

HO OO OO 00

en 01 o> 01< < < <t

0000

NJ CM CM CM

r x x x

S * «>• 1-OO CO CO

ro co co co

^ r^* r**> r*vn m m in00 COCO 00

CMxoin

oo- o o

i i i i

M CMx^x

10 10 o o

in CM r r*.IO |O IO IO

n cr> 10 inin «r »j «»•

o> o *— CNIo oo oi i i i

OO OO 00 OO1. Q. Ck o.

inC 01

•2 **4J aiie u

••-> c:vi aii 'd

IIr- Ol

•oTO

t. ••-

vi nin u

0.—

^CNlCOfl.

oooo oooo

o> en en en

<« <

>,>,>,>,oooo

T5TJT3 T?

ooioioio

en 01 01 en

oo o a

CM CM CM CM

XXX X

co coco coin in in uo

^^ ^f «^f 1* p* r

CM CM CNJ CMX^X X^x

S mo> CMcoco o

10 10 in 10

1 1 1 1

CM CM CNJ CMX X X X

O CO CNI CMIO VOU3 10

^™**— *^-^^**CM CM CM CMx x x x

^C3o"co"4- VO CM

IO IO Ô IO

r~ oo ocncoco o

"

ro - in 10o oo oi i i i

o. o_ a. Q.O'O'O'O'

a p—c af- C

i ic oc

en 3•»- o

oo oo oo oo

en en en 01

<««t

>,>,>,>,oooo

T3 -O T3 TJ

0700 oo oo

en en en en

S- S_ i- 1.o o o o

Tt Tt «»• «s-x x x x

*O \O ID \OID ur> un tnCM CNJ CM CSJ

* * * *CVJ CM C\J CJn: ai i zO» CO O^ ^r** r . f"* 10ID \o \£> in

i i i i

CT> CO r*. CNJU^rO <J3 <NJ*o *o "^> m

t— OÔ» \OID vo in o

".f~ oo cr> ooooo

1 1 1 1OO OO OO OOCi. o. CX o.o-o-o-o1

co

(dN•»-

2CM

X

iin<>

3o

*

•aaiK

•E

,2ou00

u4-1m

CO

aieni.n

oo

xc3acai

c3U.

*»0.

T>ai<aCDi.

4_><Oai •xuoooeni

i—x

3-29

WAESD-TR-85-003Q

were initiated with an alternate catalyst, DC-06, in two cells, along with twoWestinghouse baseline cells (CPS-033 through -036) as shown in Table 3-13.Further tests are planned in this area with various types of catalysts.

The preliminary tests in this test plan were performed with gold-platedhardware (plated bipolar plates). The initial results showed a poorperformance with an average of 580 mV under pressurized conditions whichindicates that gold plated hardware may not be acceptable.

Test Plan TS 2X2-016 ",

The objective of this test plan was to evaluate the effects of variousnpnelectrode related poisons such as by-product materials dissolution atvarious temperatures. Two sealing materials, viz., Viton and Teflon wereassessed at 190° and 200°C (374° and 392°F) temperatures with and without:MATrl. An attempt was also made to evaluate the effect of alternate endplates. Testing of the cells built in this plan is in progress and preliminaryresults are shown in Table 3-14.

Although the initial performance data on cells CAS-116 through -121 exhibit ascatter, it is, safe to conclude that Teflon is an acceptable sealant. Furtherre-building and re-testing is required to obtain any endurance data and/orother conclusions.

EFFECT OF VARIATIONS IN KEY OPERATIONAL PARAMETERS ON UNPRESSURIZED CELL

PERFORMANCE

To normalize the experimental data for unpressurized subscale tests, theeffects of minor variations in test temperature and current density on cellperformance were investigated using cells DAS-019 and -021. Both cells wereassembled using dry bonded argon sintered cathodes made using Lot No. 3 heattreated catalyst, dry bonded argon sintered anodes made using as-receivedcatalyst and carbon layer (MAT-1). Tests were performed at 80 percent hydrogenutilization, 25 percent air utilization at 200 mA/cm2 (186 A/ft2) (5.13 amps).

3-30

WAESD,-Tft-85-qq30

oI—I

<:»—IU_

I—401 C

OO

co03

CO

OO

I— <=rO 00

LU OO

QLU

CO t-t O

CO

C£.roooLUo:

oc.O

CO<oCO3

-MCO

CJoc

4-1IOo

CVJx

oo

u E

E O O E4-1 O —'<£ OUJ

OtU-

OID

U E

O —O UJ

•r- HI3 t-

L 01IO4-1

"so.•—•o1- O

OO !/>(/),*/>

1-1-1-

1-1- t- i.o o o o

U} too 0-0,-uC-J O 4-1 4-1O O OO OO

poo o

<o CM o rouo vo in uo

- ro ro oo roo oo oI I I I

OO l/)OO OOO. 0-0. 0.O CJCJ O

•ao

jro

i00

oo

3-31

C/J

oCOt—t0o.1—co

. —1

1—oQUJ1—

i— UJ

nUJ

UJ O—I OCO OH

h- CJUJ

UJ1

0

^ô

0

CVJX

CM

1 —coUJ*~~

CO

roin4—o(/>(O

' 3 •^_>

fC

to

o0§<a

CM

o

iooCSJ

£

ii.

5

gCM

m

CM3:

VI

%r—t .

Z3

»

e$

co

Ji•Qco

1 •

011-

<n+jcgoo

VIOJ

•aai

s01VIVI

14J(O

4J

1-

Ul •0 T3O Cz o< oo

J<3

+i£

UJ •o -oo c3: o(— CO

CJ) »

(/> 1 VI=1 I*. 01t— o>—t—</l r— VI

10 1-t- *J 3«/> O Ouj 1—3:

t—

UJ r— —OO Ol>

O- UJ

VI

30 r-3: r-^^

O O E0 Ô UJ

*"VIt- ^~3 .—O OJ*—*

^ o:eouj—VO

VIt.3 .—0 r-^*

§ °—UlCVJ

010 .

*— c

5 2^-"c 1/ii-i 01

cc

r—

T

010

VIVI QJO> 4J+> 10

a.•o

-o cC 01ai• i **; ^

•* .r- r- 10 to r— to' *0 10 ' Ol Ol *O 0)Ol Ol VI VI Ol VIVI V) VI

c c cC C 0 0 C 0O O •— r- O r—

••- t- Ol Ol - - Ol•>• •» 1- »- > 1-

0 0oo z z; i/i oo oo

01 01 01 01 01 01

< < < < £ < £ « £

> , > , > , > , > , > ,

O O O Q O O

•a -o TJ -o -a TJOO OO OO OO OO OO

Ol Ol Ol Ol Ol Ol

« £ < « « < <

_

o o o o o a

z i z 3: a: 3:

CVJ ID CM CO iO VOo> ro o* CM ro roin r* uo ro r** r-»

r CM vb m 01 i-*VO 1C U> <c «O VO

ro oo o<10 iO (J3

«* rf r~ en >O >~

IO (O 4O t/) Ô 1O

r» ir>' o >o oo tovo o f^ ^* to r^

r~ f~

to r» oo' en O ^~- i- •— CM CJ

i i i I i IOO OO OO (/> OO OO

00 0 O 0 0

WAESD-TR-85-0030

•aX

'£•aoo

j=

IOCO

OlOli. • • •

1

00 '

"OOl

m01k.

i *• 10

Ol^c

o§o>

.3: • .

3-32

WAESD-TR-85-0030

Figures 3-3 and 3-4 show the effect of varying cell current between 4.5 and

5.5 amperes on cell voltage at 185° and 190°C (365° and 374°F). Both cellsvaried approximately 2.9 mV per 0.1 amps at 190°C (374°F) and between 2.7 and

3.0 mV per 0.1 amp at 185°C (365°F).

Cell voltage as a function of temperature is plotted in Figure 3-5. Over therange of temperatures examined, the cells vary 1.2 mV/°C (.7 mV/°F) (DAS-019)

and 1.1 mV/°C (,6raV/°F) (DAS-021).

Similar data was previously determined for wet bonded, non-heat treated

catalyst electrodes. The agreement between the sets of data indicates very

little effect of process changes. The new data was utilized in the data

acquisition programming to perform the normalization of actual test values.

3.1.2.2 Performance Improvement From Pressurization

The effect of operating pressure on performance of subscale fuel cells

continued to be studied. This work was initiated in the First Logical Unit.Test cells were wet assembled using 93 w/o H3P04 and measurements made at

190°C (374°F) over the pressure range of 101-687 kPa (1 - 6.8 atms). The cells

were operated for over 300 hours at pressure to stabilize the electrodeperformance and improve the reproducibility of the results prior to taking the

measurements for pressurization effects. Polarization data was obtained ateach pressure and the cells were equilibrated at each current level forapproximately ten minutes before data was taken. Cell internal resistance was

measured using an in-situ technique with an AC milliohmeter (Hewlett-Packard

Model 4328A).

Typical cathode potential-current density relationships at pressures of 101,

240, and 480 kPa (1.0, 2.4, and 4.8 atms) are given in Figure 3-6. Linear2

Tafel regions were observed at current densities below 150 mA/cm (1402

A/ft ), the measured slopes ranging from 106 to 109 mV/decade and independent

of operating pressure. Increased pressure did not affect the mechanism of the

oxygen reduction reaction. Based on the assumption that pressure effects on

3-33

WAESD-TR-85-0030

CO

o

LJ

O

uin

s "5 <c

*f inS «M

£ £

to

in•in

a.Ea:

in

in

S00CD

COIS.(O

S(OCO

Sin(0

C9»(O

Sento

(9tvjto

oto

oooo

cQJ

J-

O

Uc3li-

IO

(/)(O

a;CT

0)o

coiCO

3'a>

3-34

WAESD-TR-85-0030

at

|

{' oa01

cn

o

u

u

OI

03m

I

m•

CM

S s03 tVI

o>

T

ni

ni

CO

in•

in

in

COQ-ecr

i

in

8O(O CO

COCO

eginCO

eg*•10egcnCO

sCMCO

egto

ooO

(O

O)

i-

CJ

<o

OJcnre4J

ro0>

3-35

WAESD-TR-85-0030

en —

CO

UlQL

LJCL

LJ

o>

ut»•

(9(VI

(SO)

QJ"O

(S00

(D

Stoto

CDinU)

S^»(D

SCOCO

8CM(O

A» - I I.W3

O)i-

ret-

Co

u

<o00re

re4->

"o

IT)I

3O)

3-36

WAESD-IR-85-0030

the activation and concentration overpotentials of the hydrogen electrode aresmall and can be neglected, the mathematical equation for predicting voltageincrease resulting from pressurization was re-examined. At low currentdensities, where the concentration overpotentials of air electrodes are small,the expected voltage gain is given by:

AEp = [2.3 RT (2 + 4/ac)/4F)]-log P [1]

where P is the operating pressure in atmospheres, a denotes the apparent

transfer coefficient for oxygen reduction, and F is Faraday's constant. In

practice, a

expression:

practice, a is obtained from the measured Tafel slope b using thec ' c

a = 2.3 RT/(b F) [2]dr* t*

Using an average Tafel slope of 108 mV/decade, the cell voltage gains atvarious pressures were calculated using equation [1] and the solid line inFigure 3-7 shows this gain. The experimental voltage gains at currentdensities of 100, 200 and 325 mA/cm2 are presented in Figure 3-8. At both100 and 200 mA/cm (93 and 186 A/ft ), the observed voltage gains are inexcellent agreement with those predicted by equation [1]. Thus indicatingthat, in the operating current density range where cell voltage losses aremainly due to the activation overpotentials at the cathode, Eq. [1] predicts

the gains from pressurization.

p 2As shown in Figure 3-8, the voltage gains at 325 mA/cm (300 A/ft ) areslightly higher than predicted by Eq. [1]. Pressurization results in increasedoxygen solubility which, in turn, reduces the concentration overpotential.Thus, the additional voltage gains at higher current densities are presumablyattributed to the reduction of concentration overpotentials at the cathode.

The influence of operating pressure on cell internal resistance is given inFigure 3-9. Increasing pressure from 101 kPa to 240 kPa (1 atm to 2.4 atms)

3-37

WAESD-TR-85-0030

LU-Xoc

ao

0.95

0.90

0.85

< 0.80*-LU

I0.75

0.70

0.65

0.60

i r T T T

I I

SUBSCALE CELL CPS417190t, UNHEAT TREATED PtCATHODE: 0.51 mg/cm'ANODE: 0.29 mgAan2

AIR FLOW: 966 std. cm 7min.H2 FLOW: 73 rtd. cmVmin.

Tafel Slope(V/dacade)

I I 1 I10 20 30 40 50 60 80 100 200 300 400 500

CURRENT DENSITY, m A/cm2

Figure 3-6. Tafel Plots for Oxygen Reduction on a Fuel• - . • • : , . . - Cell Cathode at 1.0, 2.4, and 4.8 atm

3-38

WAESD-TR-85-0030

140

120

100

> 80

<

UlO 60

o

40

20

Subtcate Call CPS-O17190°C. Pt/carbon catalyitCathode. 0.51 mg/cm2

Anode: 0.29 mg/cmz

Air Flow: 966 Std. ctr»3/minH2Flow: 73 Std. cm3/min

I I I Q . I I

A 200 mA/cm2

D 32S mA/cm2

0.110 V/decade

I I I I

3 4 5 6 8

OPERATION PRESSURE, atm.

10 15

Figure 3-7. Variation of the Measured and Predicted CellVoltage Gains with Operating Pressure

3-39

WAESD-TR-85-0030

0.95

0.90

0.85

oc

-•' 0.80



0.70

0.65

0.60

SUBSCALE CELL CPS017190°C. Pt/CARBON CATALYST

CATHODE: 0.51 m|/cmZ

ANODE: 0.29 mg/cm2

AIR FLOW: 966 >ld. cm]/min.H2FLOW: /3 ltd. em3/min.

PRESSURE TAFELSLOPEUtm) (V/d.c«Jt)

t" Btlll

2.4 O.tOI

4.1 0.109

I I I I

10 20 30 40 SO 60 80 100

CURRENT DENSITY, mA/em2

200 300 400 500

Figure 3-8. Effect of Pressure on Tafel Slopes

3-40

WAESD-TR-85-0030

140

120

100

80

60

40

20

SUISCALE CELL CPS 017

Ult, Pt/CARBON CATALYST

CATHODE: LSI mi/cm2

ANODE: l.» *|/cm2

AIR FLOW: IM ttd. cm3/min.H FLOW: 73 ad. cmj/min.

O 100 mA/cm2

A 200 raA/cm2

325 mA/on2

13 RT 4SLOPE-—Jp—(2+ J-)

ac • .2 3 RT ; ke - 101 mV/dtadt

1 I I I I3 4 K B

OPERATION PRESSURE, aim10 IS

Figure 3-9. Voltage Gain at Various Pressures

3-41

WAESD-TR-85-0030

results in a relatively significant decrease in cell resistance. At2 ?100 mA/cm (93 A/ft ), for example, the resistance drops from about 10.5

m|i to 9.0 mft, while.above 303 kPa (3 atms), the measured resistance is

relatively independent of pressure. .This.behavior is related to the acidconcentration decreasing with increased pressure, essentially due to increased

partial pressure of water vapor at higher pressures. As has been^welldocumented, the electrical conductivity of H^PO^ increases with decreasingconcentration in the range of 86-102 w/o. Under pressurization, the decreasedcell internal resistance is primarily associated with decreased acid

concentration.

At higher current densities from 200 - 400 mA/cm2 (186-372 A/ft2) and

pressures above 240 kPa (2.4 atm), an empirical equation for total cell voltagegain from pressurization was derived by taking into consideration the combinedeffect of pressure on the Nernst potential, cell internal resistance, and both

activation and concentration overpotentials at the cathode, namely: ^

AE_p11 = (2.3 RT/4FH2 + 4 a.) log P + 1.22 x 10"4 i

• ••'' • <~ rgn+ 4.8 x 10"3 P - 1.93 x 10"5 iP - 1.95 x 10"2

2where i is.the current density in mA/cm and P is the operating pressure inatms. -

t -L

For the apparent average transfer coefficient of 0.85, Table.3-15 compares the

measured voltage gains with the cell voltages predicted by both/Eqs. [1] and[3]. At 485 kPa (4.8 atm), the observed voltage gains are higher thanpredicted by Eq. [1] which accounts for voltage gains in reversible potentialand activation overpotentials of the cathode. As shown in Table 3-16, thetotal cell voltage gains given by Eq. [3] are in excellent agreement with the

measured values, the gains predicted by Eq. [3] being within + 2 percent ofthe measured gains.

3-42

VJAESD-TR-85-0030

LO

1

CO

1 - IU— 1— iCO=£r~

*^~^*

,!3-z.ot— 4

<Crvjt—4

C£IDOOooUJO£Q_

OC£LL.

OO2Ti—*

05

UJCD<C

I0>

I1

UJo

CDUJh~C_3t <

' OUJctrQ-

Q'Z.«cQUJo;IDOO=£UJ5£

LL.O

<£^

OOO1 <

O£<=cQ.

OC_)

co

4-> --~rO ^?

>QJQ

C

ro

QJ >01 Ero - —

-(->i — i — iO CO> 1 —

-a •QJ CT+0 UJo

•r- >>•O JDQJS-

Q_

CM

"~ - ^4-* >0 Ero^

•»Ol I

C •-<] l —

+ 0"UJ

S-UJ >i< .0

>•— E

QJ0 C

-O roQJ OlS-3 QJ00 OlrO rOQJ 4->s; •—

o>

>5

4^

00C - "rjxjQ E

O+-> •*«.c <QJ ESs_3O

QJ^— ^~**

3 E(/) 4->l/) fOfl> « J*>\JJ> "^^

S-Q.

o cr> i — o o co LOo o i — o CM o i —

i i i + +

-acro

rO

-t-> CC -f-QJ rO-> O)OQ. QJ

i — ^ O O f ^ i £ > C O c T > Oli — < — i — L Q o i C M c O +•* <Oi — i — i — i — i — 00 -l->

C i —S- OQJ >z

-oC QJ•i- S-

300 00+J rOC QJQJ E

QJ --.> *"~**

o ci. •<-CL rOE 01

•T—

' QJL O L O i o c O r ~ . r ~ - c o L O E 0 1O O O L O O O ' — C M O O o r Or- •— r- •— r— 'S- -(->

O i|_ •—o

II 0) >c

o .,- -oS 00 QJ

•r- S_•*-> S- 3C rO 00QJ • ro

•r- 00 QJ QJO c "O E•r- .^ 04- (0 J= 1ii __. i %* 01 •*—^

i — L O O f - C O l — r ~ ~ Q J r O ci — i — C M ^ O l C M p O O O J O - r -i — i — i — i — i — Ooi ro

ro QJ O>S- +J J=QJ i— -!-> OJ

•+- O O)00 > +J rOC fO 4JrO i — i—i- i— 00 o

+-> QJ f— >o <o

4_> .|— -r-j^^ ' \^C QJ 4-> QJ

J QJ J= C +J' S- 40 QJ 0

rO 4-> -r-Q. II O "O

O O L O O O O O O . Q . Q JO O C M O o O o <" — i- i-r— CM CO CM <\J CM CM •»-» QJ Q.

QJ O> • —JZ roo4-> » II

OC•• CT o C

0 <3 T- 00 — 4-> •!-O (O ^O> + > rO

c o c o c o ^ - r ^ c o c o + J L I J O Q J

«d" •* «* CM CO LO V£>

i — CM CO

3-43

WAESD-TR-85-0030

3.1.2.3 Correlation of Performance Decay with Cell Voltage

Electrochemical investigations were completed on the correlation between

performance deterioration and cell voltage at ambient pressure, 190°C (374°F),20 percent air utilization and 80 percent hydrogen utilization. After repro-

ducibility testing at 200 mA/cm (186 A/ft ) for approximately 1600 hours,three subscale cells SC-037, SC-038 and SC-039 were operated at 80, 100, and

2 2200 mA/cm (75, 93, and 186 A/ft ), respectively, so that their initialIR-free cell voltages were in the range of 700-800 mV, above 800 mV and below700 mV. In the course of testing these cells, 100 w/o H3PO, was added

twice to the cell acid reservoirs. Electrode flooding was found in SC-039after M850 hours of continuous operation. The IR-free cell voltages andinternal resistances are depicted as a function of time (up to 400 hours)

in Figures 3-10 and 3-11, respectively.

With an initial IR-free cell voltage above 800 mV, subscale cell SC-037exhibited a performance deterioration rate as high as 6.0 mV/1000 hours (seeFigure 3-10). At 80 mA/cm2 (75 A/ft2), the IR-free cell voltage of SC-038

decreased with time at 2.1 mV/1000 hour. Prior to electrode flooding,o

however, the performance deterioration rate of SC-039 operating at 200 mA/cnr

(186 A/ft2) was only M.6 mV/1000 hour.

As shown in Figure 3-11, the in-test addition of phosphoric acid resulted in nobeneficial effect on the internal resistances of SC-037 and SC-038. After

refilling the acid reservoirs with H^PO^, the internal resistance of SC-039decreased MJ.5 mft, indicating that electrolyte dryout may have occurred in

this cell. As seen in Figure 3-11, however, the. measured internal resistance

of SC-069 continued to increase with time.<v

3.1.2.4 Dependence of Cell Performance on Hydrogen Utilization

The effect of hydrogen utilization on cell performance was re-examined insubscale cells ERC-3036 and CPS-023. ['Note that cell ERC-3036 was previouslytested at Energy Research Corporation for about 700 hours and was transferred

3-44

>E

UOuUUlccu.I(X

f

1 1 1 1 1 1 1 1 1 1

rTTO^J-D-Q-°--/fc D-0— r^ J^800- f T- S"— -0 0—0

^-SC-038, 10 mA/cm2 . j

7 R O - lAdd !Add780 Ucid lAcid

S S1

760 - -

740- i-SC-037, 80 mA/cm2

720 - A

700 -

680 -

6'°^J^-

640 -

0 200

T cUNPRESSURIZED SUBSCALE CELLS

190°C. 1 a tm. , 100 w/o H^POt,Cathode : 0.43 mg-Pt/cm2 |Anode : 0.29 mg-Pt/cm2 iAir Flow Rate : 5 Stoich •

, Hydrogen Utilization : 80*SC-039, 200 «A/cm^ * 1

1 rTllectrode1 d Flooding

^ A A A A A ! * J rt,tl U U n 'l^v^^

1 1 1 1 1 1 1 1 1 1400 600 800 1000 1200 1400 1600 1800 2000 2200 24(

TIME, HOURS

Figure 3-10. Variation of IR-Free Cell Voltage With Time

3-45

WAESD-TR-85-0030

aE

lil(J

<DC

16

15

13

12

11

10

Add Acid

idd Acid

UNPRESSURIZED SUBSCALE CELLS

190°C, 1 ato., 100 w/o H3POi,Cathode : 0.43 mg-Pt/cm2

Anode : 0.29 mg-Pt/cm2

Air Flo« Rate : 5 StoichHydrogen Utilization : 80%

-SC-039i 200 mA/cni

I I I I I200 400 600 800 1000 1200 1400

TIME, HOURS

1600 1800 2000 2200 2400

Figure 3-11. Variation of Internal Resistances With Time

3-46

WAESD-TR-85-0030

pto WAESD for testing.] At a constant current density of 200 mA/cm (186

A/ft2) and air flow rate of 32.8 cm3/min (2 in3/min) (25 percentutilization), experiments were performed using hydrogen and SRG (simulated

reformer gas containing 74 percent hydrogen, 25 percent CO- and 1 percent

CO). With hydrogen as fuel, the cell voltage remained constant withincreasing utilization up to 90 percent, but it started to drop significantlyfor a utilization of 92 percent. In the case of SRG fuel, the cell voltagedecreased linearly at approximately 1 mV per each 10 percent increase inhydrogen utilization over the range 20 - 70 percent utilization. Above 70percent utilization, cell voltage decreased more rapidly with increasing

utilization, presumably due to the CO blocking and/or poisoning effects on the

platinum catalyst.

3.2 Nine Cell Stack Development (WBS 1103-02)

3.2.1 Design Developments

In accordance with the performance goals established for the prototypepowerplant, the primary objectives of the nine cell stack developmental effortwere to:

• establish startup and operating procedures for pressurized stack

testing

• establish baseline performance, reproducibility and endurance at rated

power operating conditions

• increase stack average cell performance to 680 mV at operatingconditions of 480 kPa (70 psia), 190°C (374°F), 325 mA/cm2

(300 A/ft2), 83 percent fuel (SRG) and 50 percent air utilizationswith a maximum degradation rate of 2 mV/1000 hours of operation.

The nine cell stack design, developed during the First Logical Unit of Work,

was refined and developed further. The refinements were primarily in the

3-47

WAESD-TR-85-0030

areas of electrode edge seal improvements, acid transport improvements withinthe cells, arid manifold sealing. Additionally, some minor changes were.inoorporated into the design to address performance problems primarilyaffecting end cells of numerous stacks.

3.2.1M Nine Cell Stack Elastomer Edge Seal Development

The initial nine cell stack design incorporated elastomer cell edge seals of asingle thickness at each edge of the electrode pair. The seal thicknessexceeded the thickness of the electrode pair to provide a compressed edgeseal* The seals provided edge sealing only and did not prevent internalcrbssieaks within the seal boundary at the edge of the electrode pair. Edgecrossleaks were to be prevented by application of a sealant material along theedges of each electrode* Problems with obtaining an effective edge sealantapplication caused this design to be abandoned after testing only one stackduring the First Logical Unit of Work.

The elastomer cell edge seal design was revised to address the problem ofinternal crossleak at the edge of the electrode pair. The single thicknessedge seal was replaced by a double layer seal. The thickness of each layerWas sized to the thickness of the corresponding cell component (electrode orelectrode and MAT^-1). One seal was made wider than the other for the seal tooverlap the edge of the electrode to block crossleak at the edge of the'electrodCi Figure 3-12C illustrates a typical section through a cell with theoverlapped double Viton seal as compared to Figure 3-12A, which illustratesthe standard Teflon and MAT-1 seal arrangement, and Figure 3-12B, whichillustrates the single thickness Viton seal.

A 10 cm by 10 cm (4 in. by 4 in.) single cell was built to assess the'scalability of the overlap design against crossleak before the nine cell stackdesign was revised. For this test, the proposed overlap condition wasduplicated on all four sides of the electrode pair, which was assembled withacid-wetted MAT-1. Pressurized helium gas was then introduced to one side of

3-48

TEFLONSEAL

VITONSEAL

VITONSEAL

VITONSEAL

WAESD-TR-85-0030

.CATHODE

n \^^^^^^^^^

ANODE /"XVsA. TEFLON DESIGN - STACKS W009-01, 02, & 04 THRU 07.

CAULK ^ ^CATHODE

MAT-1CAULK

ANODE

B. VITON DESIGN - STACK W009-03.

CATHODE

C. VITON DESIGN - STACKS W009-08.

-VITON SEALr r- ANODE

\"

MAT-'

•' --Kyf.-ss 3 t

CATHODE 'X\\D. VITON DESIGN - STACKS W009-10 AND 11.

Figure 3-12. Nine Cell Stack Cell Edge Seal Configurations

TEFLONSEAL

VITONSEAL

VITONSEAL

VITONSEAL

3-49

WAESD-TR-85-0030

the cell with the other side connected to a manometer, the helium source andmanometer were then interchanged and the test repeated. For both tests, theCell and seal withstood a maximum 6.9 kPa (10 psi) differential with noobservable leakage* Tests were performed at room temperature only.

Following the success of the single cell test, a five cell engineering stackwas bliilt to develop the manufacturing techniques and to test a multi-cellstack which had acid applied with the standard syringe technique. Thistechnique was considered to provide less uniform acid wetting than that usedfor the 10 cm by 10 cm (4 in. by 4 in.) test assembly in which the MAT-1 wasfloated on acid for uniform wetting. Helium leak testing was performed on thefive dell stack using the same methods as applied to all previous nine-cellstacks* At helium pressures up to 356 mm (14 in.) of water, crossleaks of

o "3less than 20 cnr/min (72 ft /h) were measured. This was considered quitesatisfactory in comparison to crossleak results obtained with previous stacksbuilt with the standard Teflon and MAT-1 edge seal design. The nine cellsta'ck drawings were revised to reflect the double layer Viton seal. The firstapplication of the design to an operating nine cell stack was W009-08.

A 'second major change to the basic cell design is illustrated in Figures 3-12,3-13, and 3-14. The cell assembly (anode, MAT-1, cathode) between the bipolarplates was inverted and the cathode was extended below the acid channel inthe upper plate. Thus, the acid supplied through the acid channel directlycontacts the silicon carbide layer of the cathode. The silicon carbide layer-serves as the primary acid transport member for the cell and addition of acidto an assembled cell is significantly improved by this change. As shown inFigure 3-12, all previous cell designs suffered from poor lateral acidtransport because the acid had to first move laterally through a length ofMAT-1 lintil contact with the silicon carbide layer was made. Independent

tests to assess lateral acid transport rates associated with both MAT-1 and•silicon carbide 'have shown that the MAT-1 has a slower rate. As shown inFigure :3-14, the feature of MAT-1 extending beneath the seal inboard of theacid channel has been retained in the revised design. The MAT-1 serves as the

3-50

-WAE'SD-TR-85=0030"

Piu

C<u•gCDc

O)

CO(1)

0)

cr>

3-51

WAESD-TR-85-0030

O)>oo

CD

ia;enc(OJ_

J-

CBat

a>

a>

QJ3

CO

Q)i.3

ctc<8UJ8-ec

3-52

^ -- — - ^ WAESD-TR-85-0030

main barrier to anode/cathode crossleak at the joint between anode and seal,

and it also cushions and prevents damage to the silicon carbide layer of thecathode caused by seal edge shearing effects.

The inverted cell design has been used in three stacks built to date; StacksW009-09, -10, and -11. However, only Stacks W009-10 and -11 employed a MAT-1

layer as shown in Figure 3-14. Stack W009-09 was assembled without MAT-1, butwith a thicker silicon carbide layer, to test the feasibility of stackoperation without MAT-1. Acid additions to these stacks after assembly havebeen successfully performed within much shorter time periods, whereas for

prior stacks, acid additions were difficult to perform in a reasonable timeperiod. The acid addition data for individual stacks is provided in Table3-16.

3.2.1.2 Nine Cell Stack Manifold Seal

Limited process gas manifold seal tests were performed in the dry room on

Stack W009-09 after initial performance tests. The purpose of these tests wasto begin building seal leak test experience on a relatively inexpensive testbed and to identify seal configurations for further development on the 10 kW"Metal Stack" (Section 3.3.1). Seals of Viton elastomer (both 0-rings and/flat gaskets) and expanded Teflon ("Gore-Tex" joint sealant) were tested. Totest 0-rings, a.special gasket/frame (Figure 3-15) was made with standard0-ring grooves on both sides to seal between the stack to frame and frame tomanifold. Prior to starting the tests, the frame to flat aluminum manifoldjoint was proven to be leak tight. The "Gore-Tex" and flat gaskets wereapplied directly to the manifold. Tests were performed at 150°C. to insurethat the standard gaskets (uncured Viton sheets) on the end not being testedwould remain leak tight. The test results are summarized on Table 3-17.These tests show that flat, cured Viton 0.32 cm (.125 in.) thick by 0.64 cm(.25 in.) wide produces a good seal, two layers of 0.48 cm (.189 in.)"Gore-Tex" shows some promise, and 75 durometer 0-rings with standardcompression (25 percent) is not an acceptable seal. These test results were

further verified by the "Metal Stack" tests (Section 3.3.1).

3-53

WAESD-TR-85-0030

•>r- OCSi ^~1 O O m O

8

on.!=CQ.

u> s!— OI <_)

co"

CQCQ

COCO

oSC

CO

§

O. O. £

•—.— COC\J CNJ M

> <_> f» PO

1O — JD O

O 3 CO C2<4-» 0.0O- -^O & C •—

UJ 0> L. 0>

• O3 1 I

t:

OCNJ

CO

Oo o i

c a»— oo -—^0^0o 3 mro

4-» O. OQ- ^ O (BUS- *—

QJ U CNJ

S c fo. •«-» t. ^*o ** LU ••- cr*

—

4-» t—r-

;V".B2f

cn+j•o .,-0at »

CNJ CNJ «/>—

QJ QJ O»- £.£

<DQ.QC <o

ai 6 a>a. at w» t.h- at at»££ 2SC 4j U •^•-i-

l- ^ J= L> CO

. *-a> 3 *>•»- t-4-> C « E E* <o a» <£.ZIX =

.E <0o fc. §

i / ) i . u _•-*-» c a; en o

-

r- 4^ a> at ~o (/i< O ( j U i — *— 4J O) UJ

T^^cc lc l / * 5QJ 3 +J . - - O. Ot _|•M C * E E «S U CL.«. (O at 5 5 u at

t oe * joVI t- U.

^ QJ at w••-» c Q) a> o

"u -MS oca PNI uj

^§r&

at i -o at *jat w *o <ot/i »« « o z

l^ilsi"io tS*"1 "S** «/»-^ i ic*O QJ -C OE O I— CO

o o-

jc -o a> oe.*J O fc.OJ £ QJ (/>

z *•£ £

•^Q-.C 01 —^ -^ 4_» (Oo 4-» E c c(O CJ (O •.- i-

CQ 3C —I 1/1 U.

3-54

WAESD-TR-85-0030

ro

ca

—cn

|

O

cn

S3

' t I- 41 '" fl*£> ro •—

CM O CL ro Or*. CM - - CO J(J CJ O O tX d. O ro o

QJ 10

O •— _Q OCnOi

cn

§

O 3 ro C2><^* CL oa. -^ Q C2* c ^

3

cn

^~^ ^0 J

0 ro •—

o o ro en»«o *oo_ >»cr^^^.Oc^* Juj l - l . '— r v . l i t

3 «— cn LO u_ O ^ ro

3Ol

SOro

A3 •

r-»°cn

§3

O VO_ O

1 — <iZ o

oO 3

^— '"" in

0^ 1« o*O o00 §*— io:eC gQ. ?

O §CJ 3

^_->^ n

1 0

1 §LU 3C/)l/><: o^ g0 o•-r ^

I CM

ro

COCM

S

0

COCM

r_

COCM

roro

Oro

» -— — — - |VO CM

CM o a. ror^ CM -^ ^tJ <J O OX Q. CO

t QJh— c cu •— ots\ O •— & O 101/J »— O = ro ro

01

3

*-> do

, A* t

\C

k^

JX

2-

•ootu

*"

jT

1XuJ

tt.(•((U2

QJ

CO

S C!OO.+JI.^^D w>o ^* LU *u •»" cn ••-v. o

LOr-

Z3

IS> (/)

CJ O (_> •*• —n »— a P C

QJL.3 Q.

4-> V» EC (/* Q)QJ QJ r— QJE L. cn4J -O O_ 00 C CU

.»•••t-

o

cn

O

o

^_ -

0

o

,

) cn c

trn o%- ro

CM

O CM— ro

O IT— ro

in

en

LTt COO «• cn

*

O CM cn

CM CM0-C- CTv

m r-^in • •

• VJS CM— <T <7>

ro

ro uo o r-t— ro CM in o»

2£

O CM

ro ro

min •>— v cn

LO O

o

mm CM

in10

ro :n

00

ro *— *r- O

LO

,_

ro ^ ^*~

LO

LO-

LO

ro

j_QJ

CM ro *o J h

QJ

CO r-. Jr- ro CM ^ icn*

ro in•— ro

- or^CM m cr>

5«S«*-CU

toCX

c

oi *~

m *r r- *r ••- "^- oF~ CM CM - cn u- Lf> <c

*/>^^ rE'*E<*E

Cc oo -*-

4J (Ccu re uC71 L» -r-

u re r— a

^^4^ **£

C r- QJO C — 1-

-p» O QJ O

<O 4J C QJu re c to co

.c -o QJ as c: ex or c ex a.**- u o or ~+J O l_ re >» O CX< P— -P- t- -OQ J f Q J t / l Q f h- I/I f~ <T. CXf— 4-> >» QJ

+ 1- QJ CX OJ Wl I

CX c O -^ QJ •»

4-> «/> 4^ *— CX CX (O 4-» C?£ c GT"*^ i «^"S

: : <o ••- c E (-> c w u o CD P~ LO <CX, t^ O +J *P* j{ 3 CO *^- y CX"— ' y

*j o •*- ^ ^ (j •*-* cu f •—> ex *••*i/) cn o +•* h~ -p- <J r— o ^ t >| C 1. O 1- £ « C 4 "OXUJ •— f- CX. CI CJ ""• I— *^ O ^~ O CU O "*~o <o ^^ *^ *j re 3 ^— re i T^ _c L.O •*-* tJ 4-> E C C <_) r— f^ re <o QJ 3 ^- •*- ^- i tt— t_> CO2_ILOU.LO 1— 3

s i= «4~ C r> 4-» •**o QJ •«- o c <o *

CJ

§ p— o ex 3 eccu c 0.3:

t» CJ O 3 «0 O>0 ooo E C

P- QC -o •£

o

o oro co ••— r— in

o10

o mCM 0 •p- CM m

CM

O Oro r— ••— CM *D

CO

0 ^ro o rop— CM •—

oo,_

O *^ro 10 ro^~ *"" *~"

OOCM r*.IO CO "r- r- in

CO "CM r- ro•— CM •—

m cn jP— T—* *

O COCO r—

oin co

ls§

u. t-o _c **

— ' cni— -0 C

"*oi *2or0 1> b.Z 3 QJ >* Z

S lo0-^ 2O fc- Cn-O Ul.-• O» C •*- LOr— CXf- E LU

s i§5 £§ -=E §0 ^-tt 0

^ a c X ^c_> *» o i- o<C U) (-} O c£

£ fc

OJO • CM

*— ro ro •^- •— •— r— cn

0°-CMO ro o •«• .-.— P-CO

. 0°^O fO if IDro CM *— CM •— •

CM o m min ro •— *o m c

o

re

m^

• inin

m

0m

inin

O

. O VO Okro . • 4->'•— CO •** QJ

o

f.co co m i*-F— r-- 10 ro

E(J

Oin

r^ CM m reCM • < Xr— CO VO

C•p-re

* 4-»

S en co </» t-. • ex

»— cn \o "o0 Cin re

.S'Sco ro re

^ ' §3*£<

i? •in QJ ro >»*r co • =-oS

*"" "" re m cu6 -^ io1o ex k.

lO in 4J O 4-»O r-. . 1 o **> *»-in'ro - w re Q *

= ^1, , .^ Epo00 U. « C3 c O Eex. o E E Q* *- t>

t, "O m— c•0 • 3CM-0

QJ 4J CT QJ*r- trt |/l • CU "f"'P— 0) f- fc. "^ F~O.F- (A VI Q)CX (J QJ C CX CX<t >» ce o x

§ *"* i- ^ £ •—c 3 o re co

t- O *• « O O O</)•*- (9 4J -r- 4-» 1

1_ Q; CX-O (J r- Oft. E CU CU U 7

CX CU Vt CX >»e t— MI LO o .*

o o cu t>t_i ^ t- D ** *O

cn • >d E LO — 00> O 4-» O< Z 00 O --»— •—•

^ CM ro

3-55

WAESD-TR-85-0030

ORIGINAL 'OF POOR QUAOTY

PROCESS AIRSUPPLY MANIFOLD

COMPRESSION PADTIE BAR

ACID INLET

FUEL RETURN MANIFOLD,

TERMINAL/BUS BARFUEL OUTLET

COOLING AIRRETURN MANIFOLD

INSULATOR/HEATERPLATE

COOLING AIROUTLET

GASKET/FRAME

COOLING AIR1UPPLY MANIFOLD

TIE ROD

COMPRESSION PLATE

FUEL INLET

FUEL SUPPLY MANIFOLD

GASKET/FRAME

ROCESSAIRRETURN MANIFOLD

CID COLLECTION BOTTLE

LEVELING SUPPORT

Figure 3-15. Nine Cell Stack Assembly

3-56

— — WAESD-TR-85-0030

TABLE 3-17

NINE CELL STACK MANIFOLD SEAL TEST SUMMARY

External Leakage

1) GORE-TEX 0.32 cm (1/8 in.) ^< 400 cm3/min

o Final Manifold to Stack Gap ^0.013 cm (.005 in)

o Average Seal Width ^0.51 cm (.20 in)

2) 0-RINGS (75 Durometer, 0.53 cm [3/16 in]) ^60 cm3/min

o Standard Compression ^ 28 percent

3) GORE-TEX (2 Layers of 0.48 cm (3/16")) % 9 cm3/min

o Final Manifold to Stack Gap ^-0.11 cm (.045 in)

o Average Seal Width ^ 0.64 cm (.25 in)

4) FLAT CURED VITON ^Ocm3/min(2 Layers at 0.16 cm (1/16 in.)Thick by 0.64 cm (1/4 in.), ^ 50 Durometer

o Compressed ^ 40 percent

NOTE: 1) Extruded edge seal gaskets cut flush with razor blade for all tests.2) All data at 150°C (300°F).

3-57

WAESD-TR-85-0030

3.2.1.3 Plate Cracking

Other refinements were made to the nine cell stack design to addressperformance problems primarily affecting end cells. Cracks were observed inboth end plates and cooler plates on a majority of the stacks that had been

disassembled following testing. In most cases the cracks were located at theprocess ends of the plates in the nonfunctional grooves that are there toreduce material thickness for heat treatment purposes. The design of thesegrooves was modified to reduce the depth and gussets added. Both of thesechanges are designed to improve the bending capability of the plates near theedges, which are subject to bending moments resulting from gasket loads. Therevised design (Figure 3-16) was incorporated into the cooler dies and moldedplates will be tested in Stack W009-12 (Third Logical Unit of Work).

In parallel with the above plate design modification, a thicker insulator platewas incorporated in the stack design between each end of the stack and thestack heaters. The thickness was increased as much as practical within theheight limits, which are controlled by existing manifolds. The insulationplates were made 3/8 in. thicker at each end. The purpose of this change wasto provide more uniform load distribution to the repeating stack components byproviding a stiffer member between the stack and the flexible heaters, which donot react well to loading. Also, a single layer of carbon paper was placedbetween the copper current collector plate and the carbon end plate at each endof the stack. The compressibility of the carbon paper compensates forvariations associated with thickness matchup between terminal post foot and thepocket stamped into the collector plate.

3.2.1.4 Stack Assembly Status

During this period of time, five additional nine cell stacks were assembled andtested. Table 3-16 provides a summary of the characteristics for each newstack and also provides a summary for previous stacks to allow comparison.

3-58

WAESD-TR-85-0030

ORIGINALOF POOR QUAUTtf oc tn

< Ij

ore

GO

01o

a)

O)3

•ou<co

o.too

voi

CO

a)

3-59

WAESD-TR-85-0030

3.2.2 Acid Transport Test Fixture

The Acid Transport Test Fixture was developed to provide a technique formeasuring wicking rates and wicking capabilities of various candidate fuel cellmatrices. The initial design of the Acid Transport Test Fixture is shown inFigure 3-17.

the test fixture consists, basically, of a single cell combined with the ninecell assembly non-repeating hardware. The anode and cathode end plates weremodified to include nine resistance measuring contacts recessed in each of the"A" and "B" faces of the ends plates. The Viton insulations located in the

recesses of the plates provide a small spring force to maintain contact between

the resistance measuring contacts and the electrodes and insulate the contactsfrom the end plate. The resistance measuring terminal wires are cemented

within the end plate grooves and routed out to a rotary switch. The rotary

switch is used to connect the individual pairs of contacts to a milliohmmeter.

An acid level indicator tube is provided at the anode end plate near the acidinlet to monitor the inlet acid head. Three acid level indicator tubes are

provided at the cathode end plate in a location opposite to the acid groove tomonitor the acid that moves across the cell from the acid groove. The stack

assembly heaters are included in the fixture so the wicking process can beobserved with the cell at the desired temperature (presently 54°C (130°F)).

The first test was conducted to determine the sensitivity of the technique usedto measure the wicking rate and to establish improved techniques for futuretests-i The test results indicated that the wicking rate measurement techniquecould be improved by insulating the contacts in the anode end plate from the•anode backing paper. This can be achieved by providing an anode with nineIi3 cm (0.50 in.) diameter cut-outs at the anode resistance measuring contactlocations. The cut-outs could be replaced with blotter paper or silicon

carbide. Since the cathode backing paper and cathode end plate are both good

conductors and make contact with each other, the only resistance measuring

connection needed on the cathode side of the fuel cell is to the cathode endplate.

3-60

WAESn-TR-85-0030

ORIGINAL PAGE !SOF POOR QUALITY

t-

CJ

oo.

ICO

0)5_

cn

3-61

WAESD-TR-85-0030>•

The test results indicated that in a single cell with no MAT-1 the cellresistance,after adding acid reduced from 85 mfl to 10 mft in 24 hours. Thetest results also indicated that with an acid inlet head up to 10 cm (4 in.)arid the, acid outlet closed off no acid could be forced horizontally out of theopposite edge of the cathode in a 24 hour period. The post test examinationafter disassembly of the fixture revealed that the acid wicking in the siliconcarbide layer of the cathode progressed approximately 2/3 of the distanceacross the cell.

3.2.3 Acid Delivery System

An Acid Delivery System (Figure 3-18) was designed for bench-top testing inorder to obtain design and operational information. The bench-top design was

, modified to include in-vessel testing in conjunction with a phosphoric acid. fuel cell stack and provide startup and operational information for the final,„" design of the acid pumping system.

;, The pump selected for the bench-top testing is a pneumatically drivenJ commercial metering pump made of PVC material. Although the pump has an ,'

adjustable stroke, the volume of fluid delivered (with the shortest stroke... possible) is much greater than required for addition to an operating stack. An

overflow line from the acid reservoir on top of the stack returns the excessacid to the pump sump outside of the pressure vessel. A metering valve was

: added between the acid reservoir and the inlet connection on the stack to\, restrict the flow of acid and allow sufficient time for the excess to flow:.T; through the overflow line.

3.2.4 Nine Cell Stack Testing . , ,;

The overall performance results in the nine cell stack testing activity atWestinghouse are summarized in Table 3-18 and Figure 3-19. Note that Table3-18 is cumulative from the start of the stack testing program. Detailsregarding the performance of each nine cell stack tested are given in thesections which follow.

3-62

WAESD-TR-85-0030

ACIDPRESSURE EQUALIZATION

OVERflLOU//)CIO DELIVERY

ACID DRAIN

SUPPLY

Pt/MP SOCT/ON

/ICJO PUMP

PNEUMATIC.LINE'S

SOLENOID

SUPPLY

SUPPLY

CONTROLPAH EL

Figure 3-18. Acid Delivery System - Nine Cell System

3-63

0) 1- I00,

a4->inIflr-O>

.OlCJ

WAES§-tR-g5-6&30T «i ID <t CM CO <t

• '— • •— • • "•-»o z o z o o z

•— cn , CM r- o'. CM i— <t CM «i i— CM «

ID roro o co«=c i— CM «4 rocM CM

* in rocn ^ ^- . ps. coin <t i— ,— >4 ro,— CM

o z: o z o o ozoozooo

csE

l/li—Olr-

o: 3

<X. CO

z ro

in *ro «t o <C co CM in T

: CM ro z ro CM ro

r-evjin. cn

1/1Ol

c3

1.

5.CD

1/1

r> r» O * roi~»'-Tr-.i— CNjro

<0 i. i.•*> 01 3O CL O

incvi

co o«s- r—

inOOO OOCO r- t-».

CO

ICO

DCOu_a:LUQ-

.o

to«-co CM *.

—* inc\J ro

O .* * r- * «

O o enCO r- CO

* ro *ro in ro inID i— i—,

r- co O

oo.cnc•p-i_IO

<£,

cn

10O)

CQ

O

>-

oo

• a> Ecn*J 4-"

<> c 10< aio coin Vr

. O Si ~-<*- C3s. a> a:fl) OlQ. c

(O

to ^~-M- ai CMi- ova:

, ai c —•'o. re

a:

a>cn10

o>

CM

— - *^•l —roco'

* #Oi — r^CMCOCMI * *"

*CM

vo

—' *CM r— ro CM m

ro m ^- ro in m ^o

co

U01

0 .1 — ^CM•com oco ^— * *o *-TO ro m *—^0*3-

<u —.I— ro

Co V W Tj- cno o r- co cnr ^ c i— i

*m

vo 10 ID VD in C M C O I O

•oai•o

UXOl

ai

<— CM ro <s- ino o o o oI I II I

cn cn cn cn cn8.O.O OO

o o o o

IOU

i.01

inOlc3

O

— .«t

•O OlE ^ oi "U cn

•*-* oi ~o cu cIO 4J 3 "O IO

IO r- 31-r- O CD (J r-

•r- d C (J "IC2» -0 .1/1 -r- CM c .—

cn^ c *J *" "aic o o c +j o•r- Ol COO*> in u> . , c J-i/i -F- cvro »/> oioi i E 3c cxr-~ *J+* <-<^i- 5

> l— r— OOl O Olo CM o cn <_> m

r- £ Oi— 4_> o

•— CM ro «• in I

* **

3-64

WAESU-TR-85-0030

LLJ

Qill

ccLUCOCOo

ozLUCJLU

COzO

QZOo

cc.LUO.O

<

i!z 2

SP I0< N

E ocN —

."I <

en

I"3 coo

§2

O)oo,

9

Iss

IDO

g

coo

inod>

IsS CO r-

§OO

OJ9o>

LOLOCO

.

col- co

oo

I 1133 U33X3

8 S E S1133 id33X3

8 « £ S1133 id30X3

£'£'181133 ld33X3

E 'I S1133 ld33X3

I 1133 ld33X3

Z 1133 id33X3

oLOCO

OoCO

oLOLO

«00LU

CO00

oLU

LUV)

»

LUzD

CO00

Q.<

sOQLU

ro000>

IO)ocroE-

O

O(T3

1/1

d)O

0)1/13O

_E0>c



'30V11OA 1130

3-65

WAESD-TR-85-0030

Additional stack testing at ERC established an initial performance baselineusing pure hydrogen fuel in Stack E-009-02. The performance level of Cells 1thru 5 was 624 mV at the following operating conditions: 480 kPa (70 psia),190°C (374°F), 325 mA/cm2 (300 A/ft2), 80-83 percent fuel utilization,40-50 percent oxidant utilization. A new performance baseline was establishedin Cells 1 thru 5 of Stack E-009-04. This new baseline is 672 mV at the sameoperating conditions mentioned previously. The 48 mV increase is attributedmainly to the increased inlet gas temperatures of the new test facility, newstartup procedures and new manufacturing procedures may have contributed tothis increase. A quantitative separation of the individual contributionscannot be made from the existing data. Factors which will provide additionalimprovements in performance were demonstrated by comparing the resistance ofcells with 175 microns (7 mil) and 75 microns (3 mil) silicon carbide layers(.12 mft and .09 mfl, respectively), and then comparing performance of cellswith the standard 250 microns (10 mil) carbon plus 175 microns (7 mil) siliconcarbide with the thin, 150 microns (6 mil) carbon plus 75 microns (3 mil)silicon carbide (624 mV and 654 mV, respectively). These thinner matrices areprojected to result in a 30 mV improvement over the present 672 mV baseline andthus meet the goal of 700 mV. Future stacks which incorporate both matriximprovements and operating procedure improvements are required to confirm thisexpectation. Reprbducibility will also be demonstrated in future stacks.

The goal of 500 hours of continuous operation at 480 kPa (70 psia) was achievedin Stack E-009-04, but there was significant performance decay due to aresistance increase. The longest stable performance was achieved in StackE-009-02. Over a period of 359 hours (between the peak and final performance)there was an average loss of 8 mV/cell. A number of factors have limited theachievement of long term endurance. These factors include facility upsets inAp, power failure to controls, cell assembly defects, cross leaks and acidmanagement problems. Procedures have been identified to improve each of thesefactors in future stacks.

Though the factors mentioned above were'present, some limited data was obtainedusing SR6 containing 75 percent hydrogen, 24 percent C02, and 1 percent CO

3-66

WAESD-TR-85-0030

from cylinders of premixed gases. The expected performance loss on SRG is 20mV. Stack E-009-04, which was tested after revising the operating procedures,showed a 20-30 mV SRG loss. Higher losses were observed in those cells whichwere diagnosed to have cross leaks and poor temperature distribution. Higherthan expected SRG losses are suspected to be caused by the following factors:

• cross leaks• fuel maldistribution• inaccurate flow measurement• poor cell temperature distribution caused by low coolant and process

gas inlet temperatures.

Steps have been identified to improve test facility and cell design to correctthe problems mentioned above.

Stack N-009-06

Test and disassembly data were developed for a pressurized nine cell stack

(W-009-06) tested at Westinghouse and ERC. The stack was operated for over 500

hours, of which 325 were at 480 kPa (70 psia). Average stack performance for? 2the center five cell group at 140 mA/cm (130 A/ft ) was approximately 718

mV/cell (190°C (374°F), 480 kPa (70 psia), 70 percent hydrogen utilization, 40percent air utilization).

The following is a summary of important observations/conclusions:

• The final average stack performance of the center five cell group at

480 kPa (70 psia) at Westinghouse was within approximately 6 mV/cell of

the initial average stack performance at 480 kPa (70 psia) at ERC.

• Initial OCV testing at ERC indicated possible problems with Cells 1 and9. Throughout the testing, Cells 1 and 9 were substantially below the

stack average. Eliminating these two cells from the average showed

3-67

WAESD-TR-85-0030

this stack performance approximately equaled Stack E-009-012performance, 725 mV/cell and 721 mV/cell respectively at 140 mA/cm

2(130 A/ft ). These two cells (1 and 9) were found to be corroded

upon disassembly of the stack.

• Cathode plate and backing paper corrosion was found in Cells 1, 3, 6and 9.

- Cell 1 oxidant inlet area and acid channel area- Cell 3 fuel outlet area of the cathode plate

- Cell 6 fuel outlet area of the cathode plate- Cell 9 oxidant inlet area

• Low performance of Cells 1 and 9 made it necessary to test the stack ata reduced current.

Stack U-009-07 ,

Stack W-009-07 underwent testing in two phases of operation which totaled778 hours.

Major results during each phase of testing are given in Tables 3-19 and 3-20and noted below:

Phase I

Total Phase I test time was 385 hours which includes 262 hours at atmospheric

conditions and 123 hours at pressure. Testing included OCV tests, polarizationtests at 143kPa (20.7 psia) and 480 kPa (70 psia), and process gas utilizationtests for the anode and cathode. The desired full power hold was not achieveddue to problems with the cathode air flow.

3-68

COWAESO-TR-85-0030

</>Q-

in CM >—Si - r-- Oo *f> en coi— CO r— "

cse*

<ua:

C\JCO

0

0CM

o

inCOCM

O

in00

0

o*!•

0

O•a-o

tO

o

IT)CnCM

0

OCO

0

o •—0)CT

CM

ininLO

CO

9>

CO

o tor— CO

If)

i r > c r > O v o c r > r > . ^ D c o L n co

oo•co

00co

fO

O Q-S'oco f^-

(V01«0

OCO O O•si- r-v m

(O

COoo

in

oO-

eni

co

oCM

encccon

CO

• *

CO•k

Is*^~. ol-~

CM,_!inCO

O

^—COco^_^

CO

co

r—

•r-tocxop^»

• —0CO"d"

to

i ^ c T > c n c M c y i t o r ~ « i — e nCM i— CM in CM CO '— C O C M

o o o o o o o o o

1 i— CM CM i—vo o •* co '—in to in «* in cr> o co

in tn

envoCM

CO ^co •in •*

OOUJa:o

CMoooIf)

co

in•oinco

COco

in

ina.

oco

incr»

cooo

oinO)CT(O

. o o o o o o o o o

0)to<o

inCMin

en «3-oo m»d- co "*

o o"

co oen POCM CM

in«*in

in•

CO

in• •r*s»

0)c•r-Ccen0)co

CM• •

0

A

co

— .t~»

I v•

r—inCO

u_ocoPO• ^

dO•

<y\COr—

fO•IS)Q.

0r**-»**- *0COJ"

CO00

 o

01CO

i.

10Q_

O)

toto

toa>a:

i n i n i n o o m o o0 0 ^ - c o c o c o r v . m ^ - c nC M C M C O ^ J - C M C O C O C M C O

O O O O O O O O O

10 cn 1-7

CM CMn o

o

co co en co m

CD

UD•

co

in

o<oo

3-69

1—-^ .HH

OO-

c*UJ3OO.

C£.UJ

o:<

o o-CM ji.

CO •*

1.1 LUUJ *«

°3 5

^ 1Ou_O£.UJQ.

UJ

CO

>-<UJ

CO

38 essu

riza

tio

no

f po

wer

)

•• $_ inLO ex inI™" (11 ' ^~ a* o•• -a •—coCM U o

CU +J•>•»->

r l|_ CU

co^-a

coLO• •

Q• •

coCM

•*

CO

cnCM• •r-»^—• •

*cnCM*^r-.

•3-• •

o4j!H"™

0%

co00

cnCM•r-^

cuc^•r-1—

*CU

IOo

vof n •a: coco cn

LO

CM CO:c cn

CO

a: CMco cn

co•CM <f

3: cn

<*

4->ccu

"cu J-3 3

1 1 g_>

^^U-Q

LO

^^co"— '

r-™a>r—

U.Q

VOf*^co

cn^—

U-

co^—i*ocn•—

^»u.O«^>

co'C3cnr^

OO

*a.E

1—

ro

inCX

f^

^>^^

LOr*** r * CMCM 00 f

.*°inexo

r*» vo CMCM CO «*

10•r—

o^^^

LOvo •r^ o CMCM cn «*

,_^

•2in

O

^' CO

vo iI-*- o oCM I1-- •*

ex.

O)s_inin CM CMCU 3C Ot_

C . o^ 9^

C5E

A

(/)

(UQ^

WAESD-TR-85-0030

c?^j- r^ LO o co r>. -i— vo oLO LO LO cn LO r**. vo ^ r »o o o o o o o o o

>E

•«

cucn(O

•«->"o>

eA

t/)cu

fV^

LO vo vo p%. cn CM * LO cnCM f) CO LO 00 VO CO CVJ COi ^ r ^ v o L O v o L O V o r > . L o

L O L O L O C O L O r - L O ^ - V O

o o o o o o o o o

r>

Acu01(0

*o>

00 00 LO f"} r~" ^s, V0 00 ^1"CM co oo LO cn vo oo CM cnr * . f * i 0 v 0 v o v o v o r ^ i 0

Ca

•sinOJo:

c n c M O i o c n c o c n o o« ^ - i o v o r ^ ' * c o l f > « ^ - r - >o o o o o o o o o

>E

cuOlIO

r—O>

C O r i - r * > C M ^ - C M C n c O ' —r— O LO LO Cn LO P1** r— CO^^ r^* vo vo vo vo vo f*** vo

•

<g

- #t

I/)

CUC£

LO oo r * cn LO co LO oo vo« a - « 3 - L O L O ' * r - - L O c o v oo o o o o o o o o

>E»

CU01

•2r—O>

co LO cn *^ co cn vo co cnCM ^— vo f*»» co r^» co co cnr > . r ~ L O v o r * . L O v o r > . v o

=tfef^

"cu0

^"~ CM co ^ LO vo r^» co cn

*\ ^rt

VO

V0

•s»LO^J-

r— r—cn •LO vo

>CO

" '"• cn • .

LD VO

icoVO

^>LO r-COLO VO

10

VO

•s*o

CO CMcnLO VO

UJCD^f ^^rv* OUJ <

«f CO .

WAESD-TR-85-0030

COCO

o«

VO CD

oo oo

U_ c3O •!-

<* to

cn .• cn voco oo r^ co

co •— CM r>>

&E

O)CC

!<

LO LO

0 0 0

vo co O• • •

o o •—

vo00oVO

CO CM |v>CM Is* r>-vo vo vo

en*3-VO

LOcn

•CO

>r-^voco

oo

10

oCVJ

ovo

a(/ID.

co

ooco• • •

0 0 0

ro CM covo CM co o

• •O i—

CM CM VOvo co eno o o

e«*

•CO

oO-

oa.a:

VO

oooen

VOr-^ coCM cn

r ^ v o v o v o v o v o v o i o v o vo

vocnLO

.Oevi

o-

<UJ,o

OCM

CO

cno«=»•co

tocoQ.

O O)a:

CO VO i— r—LO• • •

O O O

CMLOcn n> co vo. • . • •

o o o o o

o1

•ocn•VO

a:ou.o:UJQ-

UJ

CM CD O O*•••. rv c\J Ch m VO CMoo oo cn r— ft co «*

V O v Oi — cni — vo

CJ> O ^ i —covo

LT)O

vo

o<cUJ

oo

coCOCM

cn•"•

•«CM^^CO

CO£

•i—(—

«

0)+J(Oo

VO•

CM CO3: CT>

<««

•4->c0)

"cu J-3 3u. o

U-o

1^^p>.co

ocn1—

oo

«ex£

QJ1—

I/)exor«.^"*oCO •*«d- CO

0}Q_

^^•k

CDs-3COCO Ovu 2:J-0. i«

cua:

CM

CMO

o

CD

LO VO• •

o o

VO COCO LO

VO CO ,—r^ LO . LO

VOVO

O O O O O O O

c o c o c n c x j c n L O c n c o i —CM o co vo co LO' oo CM cn co

envo

UJCD

UJ

vo•vo

LOVO

VO

3-7100

WAESD-TR-85-0030

oo

oQ-

CM

O

CM

ftco

O

U- toO -r-CO (/)

in —vo *

• en oCM en co co o

n co •— «d- m

<ucr<o

CMun

OD-

LUo h-CM C£.

CO Z3cr

LU—I t-CQ <C

I— LUO

a:o

CMO

in* •O

91

enoo

O •!-

o

en in

CM o en r-» *!•: en i— CM co

inO)

C£.

CTa)

in CO CM o CM en oO co m enCM

o o o r— i—

C O r — ^ •O c o 1 — ^

O r— r—

CM10

G

COt

o

en10in

•in

O.

LU

LUoo

ino

co• •o

ften CD

co to

r^ oin co

en o vo oin m co <o

inoo

O O O i — r- i— Oi— •—

co

in

cn enco r-

O

CM

01cr(O

in

en

men

m

 s-c 3(U * (/)

r~ . CL (/>Q) i- E tt>3 3 Q> i_

LU O h— O_

CM CMO

CJ

•— CM CO in 10 co en o; <->LU *t

3-72

- — — —-_- ^ — WAESD-TR-85-0030

Phase II

Acid additions in the dry room between testing phases recovered cell voltagedegradation. Average cell voltage increased 119 mV from the end of Phase I tothe start of Phase II. Total test time was 393 hours with 99 hours underatmospheric conditions and 294 hours at pressure. The stack was operated 125hours at the full power condition: 325 mA/cm2 (300 A/ft2), 190°C (374°F),50 percent cathode air utilization, 70 percent hydrogen utilization.

Stack W-009-08

Stack W-009-08 was tested a total of 567 hours which included 103 hours underatmospheric conditions and 464 hours at pressurized conditions. Duringpressurized operation, hydrogen fuel testing at full power conditions ran for173 hours while SRG fuel testing lasted 177 hours.

Testing was accomplished in five different phases of operation. Open circuitvoltage tests were conducted. Hydrogen utilization tests occurred duringPhase V. Table 3-21 lists the various conditions and test times for the fivephases of testing. Performance of this stack is indicated in Table 3-22 foreach phase. These data indicated no voltage penalty from power or pressurecycling for cells with stable performance. Acid addition helped recoverperformance and enhance performance of several cells.

Stack U-009-09

Test operations lasted a total of 157 hours, which included 60 hours atatmospheric conditions and 97 hours at pressure. This stack was the first ofthe inverted design with anode on top and cathode on the bottom. No MAT-1matrix was used and the silicon carbide layer on the cathode was increasednominally by 50 microns (2 mils) to improve acid transport. Dry laminatedelectrodes were also used for the first time in a full size stack.

3-73

WAESD-TR-85-0030

CM

OO

00o

OOI

OO

2

VOLU

tDa:oo

S--.C-O) 4_>S-r-0 30-

c•— o'S 4JU_ (O

s-O)Q.O

CM• 3:

JZS- +J0) .r-TC ^§ so. c

o'"— •»->3 n3

Lu S-OJQ.O

XJCU CN 0•i- -i-5^ •!->3 «0V) tl/l O)flj r^.w tiJ- 0Q-

0 .•r* Cs- o<V •!-C" _Ll^u .(_>O. (OW> J_O O)1°

0)E

£n- 'en«3 C•l->-r-0 +J

1— 10&-•fllVL*Q.O

O)•(/)(O

JC ''•o.

tos-

JZ

o •— oo oo oLf) CMr""

t/1s-x:OO OO CM O Lf>IO 00 CM

</)5-.cCO * OO O CTlr~ CM r . CM vo

^^

</>S-^zCM O 00 O OOOO OO i — CM

(/Is_^: !

O <• r^ O CMi — in oo f CTIi— ^- r—

1— 1H-t . >

H-» t— 1 >-l t— 1 >

r^r^r~~

oor~r—

«*UDpf-^J

oo0

r»~<-Din

• •

en

104->O

t—

3-74

00

to —-CU CUOj.li.S ot-<0 V|_ O

•!-» CO U-to Q3•r- OCO C 4JCU o

Q£ O "Oa.* s:o^ f cCO o CU•— <4-

c toC2; »^ c

1/1O)•oo

oCU

o o o o o o o o o

vo ooco covo vo

VO CM

O r—VO vo

G <3

CMCM

1CO

UJ1

H-

et1—

0

OO0

1cr>oo^c

^T^^fL_

00

u.0

C£

s:

CM

Q.

ca°cu •O CMc z

u oo m

Cj_s- •CO CM0. 3C

su_0

EU_o0 0

c 5 S 5 . B S S t S S S * "O O O O o O o O O V O

inr^ co r— vo r— m CT> oo co ^"in CM j . oo r . r— vo r~ o •in m ^^ co co co co *^ co co

cui— •— vo

CO

cu

o o o o o o o o o i nII

co

in0)

co

<u•oo

UO)

vo i — ini i i

"— CT>O

ini i

Oi

O

< _ ) C _ > c _ > o o o c _ >

CT»ino

tO

c_>

WAESD-TR-85-0030

cuo 3-75

cu+Jo

voin

cuc<0

_1 \

+*

CUs_•t

COEU

oin

c

. ~u0)

o•^•»_&

>-,JOECUCOCO»oOl

•^S-3

"°to

^_3

0-i->

COE

0inCOCOO

i,_r»

CU

4ucuQ.CO

" »"D;«r—

T3C(0

*%co• ^

TJ

U

HQCU

-o

•\xfOcu

COCOooo•

1

c'i

co^EU

ococ>

^x<0CU

r—

r^ •(OCS-cu•MX

UJ

><00

* — .•o•r-o

coE0

0en*—c

u3

•

~|

CSL

VO

T3•r-U<0

IE00o>CO

c•r-

•oCUT3•o(0

CO(0seno3

ootor^r-~CUo

*co-p10i_cuQ. •

ECM O

CU OCO OJ(O i—f—

Q.TDCU

CU CS- -i-O *0

CU CUco s-

•cui-3COCOcui.Q.

(0

toS-_cVOp *

COS-:

0or^

n*— *cuI/)(0-CCU

It-CD

's-3

O.c+JCOcu4->

r^J0o

i—

r— CM CO

enCM. CM

r— 31

«i Ô CMO 0. O.. U-

^ <s> oCO

cu — O «10 m c CM(O CM <O 3Z

-C^. EfL r~* t. £

00•>vi_ ino s-o co.. Q.

co WAESD-TR-85-0030

T3 0C CO

UJ CM

01C

C r—C ^^.r- CMCTCO

CO

co

—o±A

az "OQ CU

s.••- co^"

M- ECO CO ~— J

u c(0 J- \-+> t— <_)

• — > CO U_Q .r- CO- co S-\— CO O

O COO CO

CMf^

CM

ct COt— •eC COO CO

•aCM CO OCM 0 J-

CO CTV OO CO

UJ O r-—13 UJ

1 —0CM^£

CO1

^—oCO CM«C i£ O1— 0

^^ ^"

oo co

O *^» ^_ ' J*"1*"*^ ^». *^*'i— « f— O*»

or a> *^£ i/) 1Q ^5

S -C •• C"T> C^ CM1 LU OÔO CM t-O

J — *,,

voi— < r—

i— CMCO CM(/» •> O-

J= CO Q- ^«Q. . . u. CO

CO CO— -ca

A

CO CM^— O 31CO CCM <O S«

•-1 . E O"— ' i — i- in

r- 0 l_ COtO O CO C£J= co Q- cyi0- •• ^x

CO 3

***^

^ o

o_ co0V0

CT; c

^B ^^c •*• • CT

CO. CO

O

r—

CO

. f—,

^s.

inV0

COCM10

^>10

0

i_

1f^ÔCVI

*01oo

in

0

i—CM

«5

O

CMp*10

COenin

CM

O*0*0

inCO<0

0f*»

o

*0

r*^ôô

*o

CMin10

CO0V0

co

^1010

inin

O

i —o

CM

O

O<0<0

CM

0

P-.

10

f*în<0

<±

^V010

O10<0

!

O

*

O

ps.in(0

co

0

inin10

CM

^^(0

in

inin10

inin<0

^«r»>0

CM

O

inin<0

r—

O

oinV0

o-in

<0

CM

^10

OinV0

r>*100

*o

10r— •

10

co

o

COr—10

CO010

r^

inCOvo

CM

<0

OCO

O

<0

O

10enin

co

0

«ÔÎT)

O

«^

oo

>,

r*» en10 •*

^>

r>-O COin .10 in

^- o

o en

co co

0 ô

O i—in

.vo m

^— »»

i** CM

0 S.

5,inco inVO CM•

m

^>

oi — CMCT> .in m

en

CO •

iollU COCU Q.js

r—c~

Sr-O r-

13 CU•*-> 03<"" L)_CO O

.

i.J=

CMCO

II

COCOCOi.

4^fO

cso;i/>^s .5

COS-f™

inr™ ~

couCOt-3COCO

s-o.+>

f,I/IS-

covoco

T3CO II-o-o -or

' +-o•r- COO

CM

<UCO(OJZa.l+m

O

COE'•^4_)

4Jt/)CO

+•>

"(O4Jo

1—CO

o •—

•CO

Eu

COCO

1

-ocla+JCOi.•o•r—OfO

'•ao

• •!co- euoin

• r*COS- 1

-Qen co«0 -a

•a

"v> -oCO ••—cu ua.4J '(0(0 -U

oCO

oen -a

cII <0

-;,. ,

CU coCO' i —<0 i —-C (UCu O

c^* "OO CU

oCU fOE •—•r- Q.-t-> CU

s-4-*t/) *

CO C4-> 2

Or— T3

4-> 3O JC

i— to

CVJ CO

3-76 . . * ~ "

•—-••- — WAESD-TR-85-0030

Testing was conducted in two phases. During Phase I, the stack ran for onehour at atmospheric conditions when an emergency shutdown occurred due toextremely high temperatures. Phase II testing lasted 156 hours. Atmosphericpressure tests ran for 59 hours. The remaining test time included 36 hours atfull power conditions with hydrogen fuel and nine hours at full power usingSRG. OCV and utilization tests were conducted.

The stack exhibited an improved ability to wick acid due to the design changewhich provided partial surface exposure of the acid transport member (SiC) tothe acid groove. Stack resistance was higher than noted in previous stacks,partly due to the thicker silicon carbide layer. However, Cell 1 was thesingle largest contributor to the increased resistance.

After the high temperature excursion in Phase I, the stack was returned to thedry room and examined. No visible damage was apparent but cross leaks weremuch higher than before testing. Additional tests showed that the stackretained acceptable bubble pressure.

Significant accomplishments achieved with this stack included:

t Recovery of stack performance after an over-temperature excursion ofnearly 56°C (100°F) was experienced.

• Completion of a full startup on SRG.

t Demonstration of an emergency shutdown from rated full power.

• Mapping of 100°C (212°F) shutdown from two atmospheres with SRG.

• Performance of numerous startup and pressurizing cycles that verifiedthe results obtained from stack W-009-08 (indicating no performancepenalty resulting from pressure cycling).

3-77

WAESD-TR-85-0030

t Acid addition in the test loop to recover cell performance afterapparent electrolyte loss.

This stack provided a number of diagnostic data including effects of:

• Oxygen gain of 80-85 mV at atmospheric test conditions.

• Negative voltage condition startup.

t No apparent enhancement from the reversal of process gases toelectrodes (exchange of anode for cathode, and vice versa).

• Current step and flow interruption cell voltage and responsecharacteristics.

Stack W009-10

The stack accumulated 1007 hours of test operation with 117 hours atatmospheric conditions and 890 hours at pressure. During pressure testing thestack ran 290 hours on SRG with 254 hours at rated conditions.

The stack design is similar to Stack W009-09 except that MAT-1 was included inthe matrix. Minor design changes involved increased thickness of the insulatorplates between the end heaters and current collectors to more uniformlyredistribute the load transferred through the flexible heater elements to theend plates. A sheet of carbon paper was also placed between the currentcollector and end plate at each end of the stack to cushion any high loadpoints caused by the interface of the terminal point base in the collectorplate and the end plate.

Test Results

Table 3-23 summarizes steady state performance at full power condition withSRG. At peak performance, which occurred after 45 hours in the SRG endurancerun, the center four cell average was 630 mV.

3-78

WAESD-TR-85-0030

TABLE 3-23

STACK W-009-10 STEADY STATE PERFORMANCE AT FULL POWER POINT WITH SRG

Date, Time

Time on Test(at pressure)Current

Temperature

Pressure% H2 Util.

% 02 Util.

Cel

Cel

1

1Cell

CelCelCel

111

Cell

CelCelCel

111

3/19, 9:24

428 hours

322A

192°C

70 psia70%

31%

Voltage

1,2,

3,4,

5,6,7,8,9,

mVmV

mVmV

mV

mV

mVmV

mV

(Res,(Res,(Res,(Res,(Res,(Res,(Res,(Res,(Res,

mft)mft)mft)mfl)mft)mfl)mft)rrtfi)

mft)

547442

486611

635

591

589349620

(0(0

(0(0

(0

(0

(0(0

(0

.23)

.19)

.24)

.21)

.16)

.20)

.21)

.15)

.16)

Stack Voltage, V

Phase Ib1st FP data pt

3/19, 9:24

428 hours

322A

192°C

70 psia70%

31%

1st FP Data

4/4, 3:00597 hours

338A

190°C

70 psia

70%

40%

Phase IcPeak

4/6, 10:00652 hours

336A

193°C

70 psia

70%

33%

Last FP Data

4/6, 16:00890 hours

341 A192°C

70 psia

70%33%

4.704

(Stack Resistance, mfi) (2.60)

616 (0.17)611 (0.16-0.18)

515 (0.23)620 (0.19)

647 (0.16)594 (0.18)

608 (0.15)544 (0.15)

623 (0.15)

5.158

(2.50)

618 (0.18)620 (0.17)

508 (0.26)

628 (0.19)

655 (0.16)

608 (0.18)630 (0.17)

403 (0.165)667 (0.15)

5.124

(2.50)

611 (0.19)572 (0.08-0.12)

447 (0.33)623 (0.21)

647 (0.18)576 (0.21)596 (0.22)

550 (0.20)620 (0.16)

4.922

(2.90)

3-79

WAESD-TR-85-0030

Effects of humidification were studied during Phase Ib operation. The designlevels of moisture were chosen for the anode and cathode stream, which were4.4 percent and 1 percent by mole, respectively, for the anode and cathode.Table 3-24 presents the results of the tests. For the anode fuel, moisture wasadded to hydrogen first and then to hydrogen with COo later. About 5 mV gainper cell was observed in both cases with humidified anodes. For humidifiedcathodes, an average of 5. mV increase tn cell voltage was also observed.However, an endurance run with the humidified hydrogen with C02 fuel wasconducted but cell 1 dropped 47 mV from 611 mV to 564 mV. The stack was thenreturned to dry fuel.

Polarization test data are shown in Figures 3-20 thru 3-24. The first figureis for the polarization test with hydrogen and the following four with SRG.The SRG tests were conducted at the beginning and end of Phase Ic. Thehydrogen polarization curve shows an average 20 mV higher than that of thefirst SRG curve.

Before the stack was depressurized, 13 mapping points were conducted. Table3*25 .shows the results and cases used in the tabulation. In summary, the goodcells of stack 10 showed higher voltage gain, due to increase in pressure thanthe theoretical value. The temperature gain, effect 1.5 mV/l°C (.8 mV/l°F),was considered to be high at 166A and 325 kPa (47 psia). The current effectvaried with pressure operating level. A high pressure, 690 kPa (100 psia), thecurrent effect rate was -0.57 mV per ampere increase.

Stack W-009-11

S.tack W-009-11 was of the same cell arrangement as stack W009-10. At assembly,the stack was lightly loaded at an average stack pressure of 55 kPa (8 psi) andan average cell strain of 6 percent based on 190°C (374°F). The objectives ofth,e: test were: (!) to determine the minimum initial cell compressive load andstrain for acceptable cell performance, (2) to determine stack and cell opera-ting, characteristics over limited operating; map including startup and shutdown

WAESD-TR-85-0030

ooCO

• oCO <C o

^ ^ enCO CO i—

rO'r-COQ.

Or>.

OCVJ

co

CM& « & « O c M c o c o r - » o c o c o i n« d - c o c o « 3 - c o c o i n c v i ' — O C Mo o c o v o v o v o v o v o v o v o v o v o

in«*•in

.a

UJ

Q.

oo

co co

rO

o 'LoSn tt

CO O

>1co o

i « f c « c o o o c o i n o r — co in^ • c o c M ^ - c o c v j m c M ' — c n c v joo co vo v o v o v o v o v o v 0 i n v o

co«3-•LO

Oz.

0£OU_a:UJQ-

oo• *10

" Oco ct oi— en CM"»«» co enco co i—

• r -ocoQ. +

CM

Q)

O

.0. co oCM fc«

•* $- cr>CO O COCO CM CMvo vo vo

co r— r«. CMf— 1— *fr •*vo vo in vo

coco

o

* §CM ZJ, °CO t—t

tu fe

Oo• *

CMr~

9i

'COr~

*^^co

<encoco

oOCMenr—

10tf*.

COo.or»-

CM0o•oc<o

CMm

o

If)8 c o c o c o o m c o c o i n

O r x i r — c M « * o e n < -v 0 v 0 v o v o v o v o i n m

oco

•IT)

oUJ

oC3

0CM

*CM

^^CO

«o•r-

O CO<f. 0 Q.0 0^- en oCO •— 1

cuo

CM >)n: jzoa *e

<*CM •

n: «*

>»r—cos-

•r-<£.

i « a * o o o c o o c o v o c o o, — O C M ^ - C O C M i n i — O O " *m ^ v o v o v o v o v o v o v o v o v o

co«*•

Lf)

oo

1en0oi2

C_3

<00

r-

COCo+J•r--acooenc

•r—4->rOs_CUQ.0

.<JrO

4->01

^CM

""co

cueh^

•k

CU-PrOO

cCcnCOco

<««

-4-JCeus_3O

Oooen

oo

•*0.ECUh-

(O•^

COO-

or^

rO•r-COQ.

«t0)s_3COCOO)s_O.

>,

"coCM

1C

4->Ccu+Jcoo0>•oocct

o

if)g i n i n O c o c M c O f — v o m

CM CO • fO CM m r-r O O f*5

^ • v o v o v o v o v o i o v o v o v o

o>-pcoocuTJo

rOO

E e E•— CM CO in

s eco en

CM CM 'a: o cu

oCD a>o o

o o a>oa>o

CM

5•

in

CO4->

'o>.*

<uCDrO+J

"o

orO

oo

3-81

WAESD-TR-85-0030

O O Q.f- O O »«CM O O CO—. o CT» r-> ooCO CM f—

0) V O. in

« I § £Oh- H- O.

CM CMr o

ni nt ni

cnLJ •o<r ->- ni

n 2

IU Ilk

>tf t. m»- « ni

V •O) X

U> • Ml» » r> : 5 :n ni n»

' ^ » - « O W — C3~

o•r-4->(O

OO.

O

I

Oo

.Vu(O

•!->OO

OCNJ

CO

OJ

3

3-82

a*</>IB

ag sin c\j

WAESU-TR-85-0030

O> V O. i/>*•» E E 0>•O •»- 01 V.O I- I- Q.

CM CMz o« »«

1

I I

OOO

UJQ

Olos-OJ

cOl

O)

01ISf

01co

co

IM

s_

'oo.oI

oo

Oo

ooCD

OO

OoID

OOin

oo oomg(M

g

<NJi

ro

O)

3Ol

•A" '30V1TOA 1133

3-83

2 >i-r«—<j—v>«2 •- *t o Q.2 - •• o _ g

-WAESD-TR-85-0030-

o.i

CO O> O O 10r- »- r . u> CM

It N

MCNJ CMC Oi-> E E i

IB *r- C 1O I- I- CL »« »«

I

E

CO

UJQ

cu

01-»->c01oa>

u

(Uto

n.*4-

M

S-<ooQ.

ICTiOO

8OO

8CM

C\JI

ood)s_aCD

•AM 30VI10A TT33

3-84,

.r 2 «*r «• •*-a, oo •• <_> «*

o s g«r t- •- r*. in cvjn « ii u n n

WAESD-TR-85-0030

U)

f. ** 6 Ej§ 3 £ £

I I

(U

3o

ao

u

8\L/

!S. x:s :

X o

-i s

8(O

8 8CM 8S

cCD

OO

<oNI

OQ-

Or—t

Oo

(O

coCVJ

Ico

(Us_3CD

•A» •33V110A 1133

3-85

t/»

-Io>u>5o.

.

r*. o «/»«• o a.. i .- o 5? »*co o> o o if>f— ,— r* in csj

WAESD-TR-85-0030

n

O)

H ii

O.

H

CVJ

h- t- a.

J_ _L 1

ooo

oo

O)os-O)

i.=3O

>: 1^^ . i.i

(O

H- N2 .?UJ s-

oa.

cr>oo

o

CVJI

ro(U

a>

OO00

g §CD

§ gin

8cn

Oa(M

oo

•A«" •39V110A 1133

3-86

WAESD-TR-85-0030

fOo

Ocu

co CO

cu

(O e EVO -3-c\J ir>

o * o *4-> <_> 4.) O

o oCVJ i— CVJ r—

fO

ni-OCVJ

X

CM

*

0)3

tftozI-H

O-a-

co oi—iLU I—

^ 3

(O•U

o.X

(1)

(O

(O

O

a>enc

OO

^ 1

cr> co LO• • • coo o o co

OO

EIT)

,0 *•r- (/)

< < </> OLc£ O O D-O •*• CVJ OIO CO « - ^~ Or— J I ^ r—I O O I Io *o <o o oco •— •— co ^>

o°o

ooOen

cr>ooi

rO•r-«/)a.

4Jo-01

0)

JO16•r*s_

10Q.Oco

3

(O

0.

o §

<c <LO in

VOVO

a>(Sj S-

C& C2* ^0) CO +J

4J +J l_ j_ (OC C 3 3 i-d) a) i/> </> <u$- s_ i m o.i- s_ Q.cu

a>

cu</>n>O

<tj

00

r- ITSr- *f t^

T3C

OJ CO

fOin

•ac<a

co

•o

CM

•octo

CO

3-87

WAESD-TR-85-0030

procedures, and (3) to continue verifying stack and cell performance withcomponents to be used in the 10 kW stack and with the inverted cell stackconfiguration.

Only the first of the three testing phases (Phase la) totalling 159 hours wascompleted by the end of this report period. Table 3-26 summarizes the results.

3.3 10 kH Fuel Cell Stack Development (WBS 1103-03)

3.3.1 Development Status

Design modifications were made to account for the larger shrinkage (5.2percent) incurred with the A4421 regrind plate material. Molded, heat treatedplates were inspected and statistically analyzed to establish the appropriatedimensions. All drawings affected by plate shrinkage were revised andreleased. The instrumentation drawing was also completed and released. Basedon nine cell work, the electrode configuration was re-evaluated andre-designed in an inverted configuration with the cathode on the bottom toprovide better acid transport. Assembly, process gas piping, andinstrumentation drawings were revised as necessary to incorporate the"inverted" cell configuration.

Plate/electrode subassembly and final stack assembly procedures were alsorevised for the inverted cell design. Fabrication of all non-repeatinghardware was completed and plate subassembly initiated.

Samples of polyethersulfone (PES) process gas manifold material were tested inhot acid for compatibility. Some surface attack was observed (primarily inthe outer glass fibers) but does not warrant a special surface coating forshort term testing. This subject will be reconsidered following completion ofthis test. Flatness was difficult to maintain when machining the manifoldsfrom extruded slab material. This was overcome by making light cuts andreclamping frequently without bending against the machine table.

WAESD-TR-85-0030

TABLE 3-26

STACK W-009-11 STEADY STATE PERFORMANCE AT FULL POWER POINT WITH H,

Operating Conditions

Date, Time

Time on Test, hrs

Current, A

Temperature, °C

Pressure, psiaALoad, psi

%\\2 Utilization

%02 UtilizationCell 1, mV (mft)Cell 2, mV (ntt)

Cell 3, mV (mft)Cell 4, mV (mft)

Cell 5, mV (mfl)Cell 6, mV (nfl)

Cell 7, mV (mfl)Cell 8, mV (n£l)

Cell 9, mV (mft)

Beginning

5/14, 12:00

78 hrs

345 A

190°C

70 psia070%

33%

169 (0.25)594 (0.25)

554 (0.25)

591 (0.19)

574 (0.23)

586 (0.19)542 (0.24)562 (0.20)

581 (0.22)

Phase laPeak

5/14, 18:0084 hrs

345 A

188°C

70 psia0

72%

33%

181 (0.28)601 (0.29)

557 (0.25)

601 (0.20)

586 (0.23)594 (0.18)550 (0.24)574 (0.20)591 (0.22)

End

5/17, 10:00

148 hrs

336 A

193°C

70 psia070%

33%100 (0.65)594 (0.26)

557 (0.25)

596 (0.20)

576 (0.24)589 (0.20)550 (0.25)569 (0.22)586 (0.24)

Stack Volt V (nfl) 4.53 (3.0) 4.62 (3.0) 4.49 (3.4)

3-89

WAESD-TR-85-0030

A prototype metal stack was designed and assembled according to test proceduresusing aluminum plates in place of graphite plates and prototypic edge sealgaskets. The primary purpose of this prototype was to develop a manifold sealsystem. It a.lso provided experience in building a larger stack and anopportunity to evaluate the stacking assembly fixture as shown in Figure 3-25.Based on manifold seal tests on a nine cell stack, five different sealconfigurations were tested on this metal stack with plates aligned as well aspossible. Since the aluminum plates had no flow channels, only one manifoldwas required and only one end of the stack was involved in the leak test. Twodifferent manifolds were used for these tests. A PES manifold with 0-ringgrooves was used to test 0-ring seals 0.48 cm (0.19 in.) dia. of 50 durometerand 75 durometer Viton. Based on nine cell stack tests, the grooves werereduced in depth to allow greater squeeze on the 0-ring. An aluminum manifoldwith a flat seal surface was used to test flat cured Viton 0.32 cm (0.13 in.)by 0.64 cm (0.25 in.) of 50 durometer and "Gore-Tex" joint sealant 0.48 cm(0.19 in.) size. Table 3-27 summarizes these tests. The leakage data isreported for both chambers (anode and cathode) in the .manifold and for twodifferential pressures 1.7 kPa and 34.5 kPa (0.25 psi and 5.0 psi). Thesetests show that the flat 0.32 cm Viton gasket is the best, followed by onelayer of "Gore-Tex" on flat Viton, then.the 50 durometer 0-ring or two layersof "Gore-Tex", and finally the 75 durometer 0-ring. Future tests will evaluatethe best of these on a misaligned stack.

3.3.2 10 kW Stack Design Description

The 10 kW stack mounted in the test loop pressure vessel is shown in Figure3-26. The stack contains 44 cells in eight groups of five with two cells ateach end. Each group of five cells is separated by cooling plates. Theremaining cell components in the five cell group are separated by bipolarplates. The nine cell stack discussion relative to the "zee" and "tree"pattern bipolar and cooling platesi respectively, is applicable for this stackand is, therefore, not repeated herein. Additional design details werepreviously described in the "Final Report for the First Logical Unit of Work."

3-90

ORIGINAL 'PAGE ISOF POOR QUALITY

WAESD-TR-85-0030

X•1—u_>>.O

§

CO

O(O

oo

CMi

cocuS-=3

3-91

WAESD-TR-85-0030

TABLE 3-27

10 kW METAL STACK MANIFOLD SEAL TEST RESULTS

(ALIGNED STACK)

Aluminum Manifold Tests^)

• Flat Cured Viton 0.32 cm (1/8 in.)Thick by 0.64 cm (1/4 in.) Wide

• Gore-Tex 2 Layers of 0.48 cm (3/16 in.)

• Gore-Tex 0.48 cm (3/16 in.) on Top ofFlat Viton 0.32 cm (1/8 in.)

Polyethersulfone (PES) Manifold Test(3)

• Viton 0-Rings (0.53 cm (0.210 in.),75 Durometer)

• Viton 0-Rings (0.48 cm (0.187 in.),50 Durometer

Compression(Percent)

32

(0.045 gap)

(0.10 gap)

43

41

Leakagecm /mi n

0.172 kPa 3.45 kPa(0.25 psi) (5 psi)

0/0

0/5

0/0

0/0

19/53

6/12

19/24 High

0/5 11/82

(1) Flat face sealing flange.(2) Leakage data given for each half of one manifold.(3) Reduced depth 0-ring groove 0.25 cm (0.100 in.) vs. 0.38 cm (0.150 in.).(4) All tests at room temperature.

3-92

WAESD-TR-85-0030

ORIGINAL -PAGE; ISOF POOR QUALITY

Acid Supply

Process GasTwo CavityManifold

Test VesselLoop No. 1

ElectricalConnectors

Cooling AirReturn Plenum

> 44 Cell Stack

Stack Support& Cooling AirReturn Pipe

Process AirSupply Heated Air

Return

Figure 3-26. PAFC 10 kW Stack Test Assembly(Non-Inverted Arrangement Shown)

3-93

. WAESD-TR-85-0030

3.4 25 kW Short Stack Development (UBS 1103-05)


Consistent with the 10 kW stack, this stack was re-designed for an invertedcell with the cathode on the bottom and the anode on the top. This providesbetter acid transport since the acid must now run over the cathode in the acidgroove area which insures good contact with the acid supply. Detail andassembly drawings were revised as required. .

3.4.2 25 kti Stack Design Description

The 25 kW stack mounted in the test facility pressure vessel is illustrated inFigure 3-27, which has not been revised for the inverted cell design. Themajor difference would be to move the process gas connections from left toright and vice versa. The stack assembly is supported from a carbon steellower support plate which is mounted on the cooling air outlet pipe and twosupport columns. Many of the 10 kW stack design features described in Section3.'3.2;r;are applicable to the 25 kW stack and are not repeated here.

1 •* ' '

The 25 kW stack contains 104 cells in 20 groups of five with two cells at eachend of the stack. Each group of five cells is separated by cooling channelswith the design identical to that used for 10 kW stacks.

3.5 100 kW Full Stack Development (WBS 1103-06)


The 100 kW Stack Preliminary Design was updated to include the most currentstack design features. The major revisions included in the update were asfollows:

3-94

ORIGINAL PAGE-ISOF POOR QUALITY

-KAESD*TR-85-0030

Cooling AirReturn Plenum \

EtoctHca!Connectors

Acid In

Process ASr Supply

Heated Tie Rod(Heated Air)

Insulator Tube

ProcessAir Return Fuel Supply

Figure 3-27. 25 kW Stack Design (Non-Inverted)

3-95

WAESD-TR-85-0030

• Revision of the stack and process gas supply lines to reflect theinverted cell arrangement, i.e., the inversion of the anode andcathode. Gas supply line connections to the vessel and stack manifoldswere repositioned.

• Stack related provisions for a continuous acid pumping system wereadded. This included an acid distribution box at the top of thecooling air plenum and designated acid and vent line routings withappropriate vessel penetrations.

• The process gas manifolds were modified from a two piece design to amolded four piece design. The shorter sections should be more suitablefor injection molding than the half-height components.

• Other minor changes were incorporated. In addition, the 100 kW StackPreliminary Design Description was revised to reflect the changes madeto the 100 kW Stack Design.

i

3.5.2 100 kW Stack Design Description

The 100 kW Stack Preliminary Design (inverted) was developed in detail for thestack mounted in the Westinghouse-provided test facility Loop 1 pressure vesseland the stack in the prototypical 375 kW module pressure vessel. Except fortest assembly differences, the basic stack design is essentially identical forthese two test configurations.

The facility vessel test assembly is very similar in concept to the 25 kW stackdesign and employs identical support structure and piping components wherepossible. The main differences are the longer stack, cooling air outlet plenumbox, electrical power take-off wires, acid lines, and pressure vessel. Inaddition, process air in the 100 kW stack is piped directly from the coolingair outlet plenum box (as in the 375 kW module design) instead of via aseparate vessel penetration provided in the 25 kW stack.

3-96

WAESD-TR-85-0030

The module vessel test assembly incorporates maximum component commonality withthe Loop 1 test assembly configuration and the 375 kW module design (seeSection 3.6). In this arrangement, the 100 kW stack and its cooling air outletplenum box interface with the module designed internal support structure.Process air is supplied from the cooling air outlet plenum box. A departurewas made from the prototypical module piping manifold arrangement on the vessellower head, to avoid the need for blanking off and, possibly, bleeding the

piping legs for the three missing stacks.

The 100 kW stack preliminary design contains 419 cells in 83 groups of fivewith two cells at each end of the stack. Each group of five cells is separated

by cooling channels and is of a design configuration identical to thatdescribed for the 10 kW stack". Many of the 10 and 25 kW stack design features

as described in Sections 3.3 and 3.4 are applicable to the 100 kW stack, andthus are not reported herein. Additional design details were previouslydescribed in the "Final Report for the First Logical Unit of Work."

3.6 375 kW Module Development (WBS 1103-07)


The fuel cell module preliminary design was completed and is illustrated in

Figure 3-28 and described in Section 3.6.2. The nominal design parameters ofthe module preliminary design are presented in Table 3-28 at beginning of use

of the plant. The major changes made were:

1. Inversion of cell components (anode and cathode) and relocation ofprocess piping to accommodate the anode-over-cathode configuration.

2. Addition of electric resistance heaters into the cooling air outletplenum to maintain module internals temperature above 38°C (100°F) to •

prevent crystallization of the phosphoric acid.

3. Addition of a continuous acid supply system.

3-97

ORIGINAL PAGE ISOF POOR

WAESD-TR-85-0030

MANIFOLD-...

FUEL SUPPLY

PROCESSAIR RETURN

FUEL SUPPLY

PROCESSAIR RETURN

ACID PUMPACID TANK

COOLINGAIR

HEATEDAIR

PROCESSAIR SUPPLY

FUELRETURN

FUEL CELLSTACKS

FUEL RETURN

Figure 3-28. Fuel Cell Module

3-98

WAESD-TR-85-0030

TABLE 3-28

FUEL CELL MODULE DESIGN PARAMETERS

PARAMETER UNITS NOMINAL

POWER

TEMPERATURE

Oxidant Inlet*Coolant InletFuel InletPlate Avg.

*Same as coolant outlet.

PRESSURE

FLOW

Fuel

Oxidant (air)Coolant.(air)

CELL VOLTAGEOpen CircuitOperating LimitOperating Point

CELL CURRENT DENSITY

MODULE VOLTAGE

Open Circuit

Operating Limit

Operating Point

MODULE CURRENT

kWDC

kPa (psia)

kg/h (Ib/h)

mV

mA/cm2 (A/ft2)

Volts

Amps

375

186 (366)

147 (297)

191 (376)

191 (376)

483 (70)

181 (398)

1400 (3086)

27,842 (61,380)

920

800

690

300 (280)

1540

1340

1156

324

3-99

- WAESD-TR-85-0030

A stress analysis of the moduli preliminary design was-completed anddocumented. The analysis indicates' that -the'"modiH'e structural design satisfiesASME Code requirements and is adequate to withstand the loads expected duringtransportation, operation, and maintenance.

A preliminary maintenance assessment of the module installed in the Fuel CellSystem was performed. The most important maintenance considerations weredetermined to be the monitoring and replenishment of the phosphoric acid andthe periodic replacement of the fuel cell cartridge at approximately 5 yearintervals during the 30 year life of the plant. (The fuel cell "cartridge" isthe assembly of four 100 kW stacks and associated hardware that fit inside thepressure dome.) These maintenance considerations are discussed in more detailin Sections 3.6.3 and 3.6.4.

3.6.2 Module Preliminary Design Description

The fuel cell module is illustrated pictorially in Figure 3-28. The moduleconsists of four stacks of fuel cells supported inside a containment vesselwith their cooling air outlet passages discharging into the space formedbetween the stacks. Cooling air is admitted to the pressure vessel cavity,surrounding the stack assembly, and flows through the stack cooling airpassages to the space between the stacks. Corner seals, provided between theadjacent stack inner corners, minimize the leakage of coolant air into.theexhaust plenum. The cooling air enters and leaves the vessel through the lowerhead, allowing the vessel cylinder and upper head to be easily removed forrapid replacement of the fuel cell stacks without disturbing the externalpiping. Process air is piped to the stacks from the cooling air outlet cavitybetween the stacks through holes in the top plate which separates the inlet andoutlet cooling air cavities. Fuel is piped to the stacks from a stainless :

steel piping assembly in the lower region of the pressure vessel which is fed,in turn, from a penetration in the lower vessel head. The fuel and process airreturn lines from the stacks are led to similar piping assemblies andpenetrations at the 1'ower vessel head. .Figure 3-29 illustrates all of theabove flow paths in a schematic illustration of the module.

3-100

STACK

PROCESS GASMANIFOLD

.STACK

COOLING AIRFLOWPATH INCOOLING PASSAGES

PROCESS GASMANIFOLD

PLENUM END PLATE

HEATED AIROUTLET PLENUMBETWEEN STACKS

COOLING PASSAGESBETWEENEVERY 5 CELLS

STACK

PRESSUREVESSEL

STACK SUPPORTSTROCTURE &HEATED AIROUTLET

ORIGINAL -PAGE fspOOR QUALITYT WAESD-TR-85-0030

PROCESS AIRFLOWPATH INCELL PLATES(FUEL SHOWNDOTTED)

PROCESS AIRTO CELLS -TAPPED FROMHEATED AIROUTLET PLENUM

COOLING HEATEDAIR AIRINLET OUTLET FUEL

PROCESS AIROUTLET

Figure 3-29. Module Flowpaths

3-101

WAESD-TR-85-0030

Electric resistance heaters are installed inside the cartridge cooling airoutlet plenum for use in preventing crystallization of the phosphoric acidduring transportation/storage, and plant shutdown. The heaters are energizedfrom an external source of power and are controlled to maintain a minimum,temperature of 38°C (100°F) at the fuel cells. Explosion-proof temperatureswitches mounted outside the module pressure vessel, (or cartridge shippingcontainer) monitor the cell temperature via a sensor bulb in the cartridgesupport plate and capillary tubing. The electrical leads from the switches tothe heaters and the capillary tubing penetrate the lower vessel head viasealing bushings.

. The |,lectrical connections at the stack consist of copper connectors clamped tothe copper collector plates with bolts and supported from the insulatorplates. 4/0 nickel-coated TFE insulated cable is swaged to the copperconnectors and led to the appropriate connectors on the adjacent stacks and asealing bushing at the pressure vessel penetration. The cables are supportedat intervals along their lengths to prevent whipping under short circuit faultconditions.

Two 6 in. diameter penetrations are provided in the vessel lower head toaccommodate instrumentation. Each penetration has the capability ofaccommodating several multiple lead compression fittings. It is expected thatthe need for instrumentation will progressively reduce with time as the fullstack and module test program proceeds.

Support for the fuel cell stack cartridge against seismic loads in thehorizontal direction is provided by four adjustable support screws on a 90°spacing around the upper circumference of the pressure vessel. The supportsare individually adjustable to suit the particular dimensions of initial orreplacement cartridges following installation of the pressure vessel. Fourguide bars supported between the stack compression plates protect the cartridge

from damage during installation of the vessel.

The major physical characteristics of the module are listed in Table 3-29. Themodule external interfaces are listed in Table 3-30.

3-102

WAESD-TR-85-0030

TABLE 3-29

MODULE PHYSICAL CHARACTERISTICS

Vessel Height

Vessel Diameter

Base Plate Diameter

Weight

Module Contains 4 Stacks of 419 Cells

Cell Plate Dimensions (machined)

Stack Height

Stack Weight

Vessel Cylinder Weight

Automatic Disconnection of ProcessFluid lines during cartridge replacement

Pressure Vessel - Carbon Steel to ASME Section VIIIDivision 1 for 586 kPa (85 psia) and 204°C (400°F) capability

3.5 m (11 ft, 6 in)

1.37 m (4 ft, 6 in)

1.68 m (5 ft, 6 in)

5400 kg (6 tons)

83 5-cell groups2 2-cell groups

0.419 m x 0.3 m(16.5 x 11.75 in)

2.44 m(8 ft) (approximately)

621 kg (1370 Ib)

1088 kg (2400 Ib)

3-103

Q)4->fOct:

oUL.

(OS-OJo.E<

LU

LU_l

QO

S-.C

CT-i£

^_^s_.C

_O

^^*

*— 1c_>|o 1

^Hu_0—-I

.«ol

Q_l

^11

-•CO

•f—COa.

'%_^

(UN

oo0)ex

Q_

CM

«*00

r^CM

oooCO

9*

r—VO

l~^

«*

^^

t-~cnCM

co00^-

*

0r^

•

J3f—

Otnr-~^^^

•aO)cn

. c(O

u.

—VO

c1—4

1S-

•r—eC

cnc

*^m

"oo

CM•sfr•d-

VOCM

^~cn,CM

*iOOLf)

VOCO

.

VOVOco

vor^« r**~*

cnvo

^—^_a

oLT)

~ _-

•o(Ucnc(O

u_

—VO

+J3O

1s_«^

<cnc»^

oo

or~

«*,-T

0

^~*co

cocnv_^

cnr^CO

VO1 .«3-

cnvo

**™ .

JD

0LT)^~.

•a<ucn(O

r™~U_

s<d-

1 \+^3O

1

S-•r—

<

totom\UUOS-

—CO

00cnco

,_cn^_^

vor~»co

coCO^f'

or^

^— •»jo

Oin*^*

•oaicn(0^^

u_—CM

C»— t

1

r—O)3

cnvo•—

COr«.CO

vocn

—e£

Z.

LOCOco

VOr^>*" <f.

z.

cnvo

^~-*J3

OLf)r^^^

-aOJcn(0

l»

u_Q

= CQCM 1—

•)->3 C0 •-.

1 1

r— T3»h-

a?Ms

C•o01wtstvl

1O^ — '

cnc•^x:to

' 3-Q

cn

r—<cmy^to

(Us-•^

s6s•»->

:cn3Os-:

-4->T3O)(U

4_

^—

O)to(/><D>

O

toc0

4->0<ucc0o0

•r—

S-1 \+^uO)

r—

TO(U

•<-*

*3toC•r-

LUU-1—

O4->O3

TOCooS-O)o.a.oo

TOO)+jToou1

QJ

(J•i—C

•oO)

T3CfOs--Pto

cn -»-»> '£<o E '->

•r- +j0 -I -r-^ '-^ 3•3- cn cn o

C t = s -•k .f— •!— .f—

O) +J -l-> ON (tj <O•r- $_ 5- Cto O) O) (L)

o. Q. a.OJ O O O

• 3 VO O O

LO «3- «a-r- CO if)

O r- r— i—

(Ucn<o

-l->

o5>

(U

3-ao

WAESO-TR-85-0030

:3-j£• ^

tO 13a. i—E 3 CQn3 J*t

0«d- LO '«3-CM r coco co r-^

O)4->

4-> roc or

. i- S <03 0 Q- a;QJ <D a>^ 3 's•o "o -ao o o

3-104

WAESD-TR-85-0030

3.6.3 Acid Supply System

A continuous acid supply system has been incorporated. Phosphoric acid is

supplied to the stacks through Teflon tubes clamped to connectors in thestack. The supply lines to the stacks receive acid from a distribution box

attached to the cover plate of the cooling air outlet plenum.

The acid pump consists of a piston, cylinder, and check valve assembly. Thepiston rod of the pump extends through a sealing gland into a double-acting aircylinder. All of the pump parts exposed to acid are made of Teflon. In

operation, compressed air is admitted to the piston rod side of the aircylinder causing the acid pump to withdraw acid from the reservoir. The

compressed air supply is then valved to the opposite side of the air piston andthe piston rod side is allowed to exhaust, causing the acid pump to dischargethe swept volume of acid into the supply line to the acid distribution box atthe upper end of the fuel cell cartridge.

The acid pump is mounted from the acid tank, which is suspended below themodule vessel lower head, clear of the vessel thermal insulation. Under normal

operating conditions the acid tank and pump are maintained at a temperaturebelow 65°C (150°F). Heat conducted into the tank from the module pressure

vessel is removed from the tank by free convection. Under low temperatureambient air conditions heater coils around the tank and pump are energized to

maintain a minimum temperature of 38°C (100°F). Explosion-proof temperatureswitches mounted on the acid tank monitor the tank temperature and control the

electrical supply to the heaters. Provision is made for monitoring the acidlevel in the tank and for manually replenishing the acid supply when required.Monitoring may be visual, by means of a translucent Teflon sight gauge affixedto the tank. Final details of the selected approach will be verified in

testing prior to being applied to the module. Provision for filtering the acid

supply from the pump and for minimizing the recirculation of corrosion products

and debris may also be incorporated if testing experience shows such provisionsto be necessary.

3-105

WAESD-TR-85-0030

3.6.4 Module Maintenance

The only consumable, apart from the fuel and air supplies, requiring periodicattention is the phosphoric acid supply. Nominal module acid consumption is

•5expected to be in the region of 1.2 cm0 per hour (approximately 0.25 gallonper month). However, consumption conservatively could be four times thisvalue, or 4.8 cm per hour (1 gallon per month).

The acid reservoir suspended below the module pressure vessel has a capacity ofapproximately 15 liters (4 gallons) with, perhaps, 11 liters (3 gallons)effectively available to the pump. This volume is sufficient for one year ofoperation under expected nominal consumption conditions or, conservatively,under maximum conceivable consumption conditions, three months. Therefore,acid replenishment is not expected to be necessary more frequently than everythree months. Acid level monitoring and logging of acid tank levels (possiblyonce a month) will be required in the early operational phases. These may berelaxed later as consumption rates are more accurately characterized. It isalso conceivable, depending on the rate of contamination of the acidexperienced in operation, that periodic removal of the tank for completedraining, cleaning, and replacement of the acid may be required. This would beperformed during a scheduled shutdown period during the module life. Possibly,operational experience may show this to be unnecessary during the five yearlife of the module, in which case tank removal and cleaning would be performedonly during the cartridge replacement shutdown period.

3.6.5 Cartridge Replacement

Cartridge replacement is expected to be required at five year intervals. Thecartridge and pressure vessel dome must be removed together to satisfy liftingheight limitations and protect the cartridge from the elements. The cartridgeremoval and replacement procedure outlined below satisfies these requirementsand avoids disturbing the external piping connections to the module lowerpressure vessel head. After removal, cartridges will be transported to anassembly area for packaging in a shipping container prior to being returned tothe factory for refurbishment.

3-106

WAESD-TR-85-0030

Replacement cartridges will be shipped from the factory in special shippingcontainers equipped with heaters and designed to maintain the requisite fuelcell environment. Assembly of a replacement cartridge into the power plantwill follow the reverse of the procedure described below.

Replacement of a fuel cell module cartridge requires, first of all, that thepower plant be shutdown, depressurized, and cooled. Following a sufficientperiod of cooling, the insulation segments at the pressure vessel boltingflange and the bolts can be removed. The pressure vessel dome can then belifted, using a crane, until its lower flange is level with the cartridge lowersupport plate, a distance of 44.5 cm (17.5 inches).

While continuing to support the vessel dome from the crane, temporary bracingis deployed to ensure safe working conditions while lifting fixtures attachingthe vessel lower flange to the cartridge lower support plate are bolted intoposition. After the vessel dome is securely bolted to the cartridge, thetemporary bracing can be removed and work started to separate the cartridgefrom its connections to the module lower pressure vessel head. .This requiresthe separation of electrical power takeoff leads, acid feed, drain and ventlines, heater power supply leads, temperature monitors, and cartridge holddownbolts. Any diagnostic or control instrumentation provided in the module mustalso be disconnected. The cartridge and vessel dome assembly can then beremoved and transferred to an assembly/disassembly room for preparation forshipment to the factory. During transfer from the plant to the assembly/disassembly room, cartridge heating can be maintained, if required, by the useof a transportable power source with temporary heater power and temperaturemonitor connections.

In the assembly/disassembly room, preparation for shipment to the factory willnormally involve removing the pressure vessel dome from the cartridge andtransferring it, together with its lifting fixtures, to the replacementcartridge. The spent cartridge is then placed in a shipping container fortransportation by truck. (The replacement cartridge will have been delivered tothe plant site in such a shipping container.) Transfer of the replacement

3-107

WAESD-TR-85-0030

cartridge and pressure vessel dome from the assembly/disassembly room to theplant follows the reverse of the procedure described above.

The shipping container envisaged for use in shipping cartridges to and from thepower plant has not yet been designed. The design of the shipping containermust allow the. cartridge to be heated and maintained in a suitably "inert"environment during shipment. In addition, the shipping container must protectthe cartridge from handling damage during shipment and must be designed tominimize the volume and weight of the shipment.

3.7 Fuel Cell Materials and Component Characterization and Testing (UBS 1103-04)

This section addresses the characterization of the various raw materialsutilized in the manufacture of repeating cell components, as well as thecharacterization of repeating cell components, including chemical, physical,mechanical, electrical and corrosion behavior.

3.7.1 .Raw Material Specifications

The Purchasing Department Specifications (PDS) previously prepared wereutilized for the procurement of raw materials for repeating componentmanufacture, Additional data were accumulated and will be utilized in futurerevisions of the PDSs as required.

3.7.2 Plate Materials

This section summarizes the characterization effort devoted to bipolar andcooling plates and the raw materials used in their formulation.

3.7.2,1 Plate Evaluation

Emphasis was placed on the evaluation of "Zee-Zee" bipolar plates producedprimarily from graphite or carbon powders exhibiting approximately equiaxedparticle shapes, namely Asbury Regrind A99, Asbury Regrind 4421 and Stackpole

3-108

WAESD-TR-85-0030

MF958 powder. These three materials produce plates which are generally flat to

within less than 0.8 mm (0.030 inch).

Plates were heat treated to 900°C (1650°F) and those evaluated are identifiedin Table 3-31.

The in-plane electrical resistivity and mercury porosimetry for a number of

these plates are given in Tables 3-32 and 3-33.

Based on the corrosion data obtained by ERC, it was concluded that additionalheat treatment to higher temperatures would be required for adequate

pressurized cell performance. Sections of larger plates previously heat

treated to 900°C (1650°F) were heat treated to both 1500°C and 2700°C (2732°F

and 4900°F) for evaluation of selected mechanical and physical properties and

for autoclave corrosion studies. Typical mercury porosimetry data for the two

regrind type materials and Stackpole MF958 powder plate sections are given inTable 3-34. It is to be noted that the porosity of the three materials

increased from 3.6 to 5.0 percent after 900°C (1650°F) heat treatment tobetween 11.0 and 14.7 percent after 2700°C (4900°F) heat treatment.

Weight loss data for ZFA64, TZA27 and TZA30 resulting from the second heattreatment cycle to 2700°C (4900°F) were determined. The data indicate a weight

loss ranging from 1.2 to 3.2 w/o.

Flexural strengths at room temperature were determined for the three materials

after heat treatment at 900, 1500, and 2700°C (1650, 2732, and 4900°F). The

average data are plotted in Figure 3-30, together with reference data on A99(33 w/o resin) material. After the 900°C (1650°F) heat treatment, the regrindA99 material had the lowest flexural strength while the Stackpole MF958 (33 w/o

resin, dry mix) plate material had the highest. Heat treatment at 1500°C

(2732°F) resulted in a small strength increase in the three materials while

2700°C (4900°F) heat treatment resulted in a strength decrease, particularlyfor the MF958 powder plate material.

3-109

WAESD-TR-85-0030

TABLE 3-31

DEVELOPMENTAL PLATE COMPOSITIONS

PlateNo.

ZZ29311299

21300

11302

11330

11332

ZZ341

ZZ346

11367

ZZ404

ZZ405ZZ411

ZZ414

ZZ419

ZZ464

ZZ465ZFA64

TZA27TZA30

ZZ53911597

11608

11612

ZZ642ZZ643

ZZ650

ZZ660

ZZ660

Type Graphite

Poco PXB-5Q-325Airco Speer TexacoWilmingtonStackpole MF958

A99 Regrind (-40)

A99 Regrind

A99

Stackpole MF958

A99/MF958

A99 Regrind (-40)

Stackpole MF1001Desulco 9026

A99

A99

Asbury 4421

Asbury 4421 Regrind

Asbury 4421 Regrind

Asbury 4421 RegrindStackpole MF958 (Lot 1)A99 Regrind

Stackpole MF958

Stackpole MF958

Stackpole MF958

Stackpole MF958

Stackpole MF958Stackpole MF958

Regrind A99

Regrind A99

Regrind A99

Mix

DryDry

Dry

Dry

Dry

Dry

Dry

Dry

Dry

Dry

DryDry

Dry

Dry

Dry

Dry

DryDry

DryDryDry

Dry

Dry

WetWet

Dry

Dry

Dry

Comments

Regrind prepared from coldpressed material .

50 w/o of each powder.

5.7 day heat treatment.

10.7 day heat treatment.

First powder lot.

Second powder lot.

Second powder lot.

Second powder lot.

Second powder lot.

Second powder lot.

Second powder lot.

-40 mesh.

-40 mesh.

-40 mesh.

3-110

WAESD-TR-85-0030

TABLE 3-32

IN-PLANE ELECTRICAL RESISTIVITY OF DEVELOPMENT

PLATES AFTER 900°C HEAT TREATMENT

Plate No.(])

ZZ293

ZZ299

ZZ300

ZZ302

ZZ330

ZZ332

ZZ341

ZZ346

ZZ367

ZZ404

ZZ405

Resistivity Resistivity^2)p Plane p Thickness(mfl-cm) (mfl-cm)

2.98

2.01

2.12

3.42

3.07

1.53

2.00

1.78

3.26

1.69

2.37

Ratio(2)Pp/Pt

-

-

-

-

-

-

-

-

•

-'_

Table 3-31 for compositions, etc.I2)uata to be collected.

3-111

WAESD-TR-85-0030

TABLE 3-33

MERCURY POROSIMETRY DATA FOR DEVELOPMENT

PLATES HEAT TREATED TO 900°C

Plate.No.uj

ZZ293

ZZ299

ZZ300

ZZ302

ZZ330

ZZ332

ZZ341

ZZ346

ZZ367

ZZ404

ZZ405

ZZ411

ZZ414

ZZ419

ZZ464

ZZ465

SkeletalDensity(g/cirT5)

2,152.182.18

2.102.08

2.24

2.23

2.23

2.08

2.252.10

1.951.97

1.97

1.83

1.83

BulkDensity(g/cmj)

2.112.11

2.10

1.941.91

2.112.13

2.11

1.98

2.061.78

1.871.86

1.911.76

1.77

SpecificPore Volume(cm-Vg)

0.010

0.014

0.016

0.039

0.042

0.027

0.021

0.026

p. 025

0.040

0.084

0.024

0.030

0.017

0.022

0.020

SpecificPore Surface

(mz/g)

4.77.1

8.9

17.621.6

14.4

10.9

14.1

, ' , . 12.7 \v

20.623.511.414.3

7.8

11.9

10.8

Porosity(Percent)

2.1

2.93.4

7.58.0

5.74.4

5.5

5.0

8.315.0

4.55.6

3.2

4.1

3.6

U)See Table 3-31 for compositions, etc.

3-112

WAESD-TR-85-0030

TABLE 3-34

MERCURY POROSIMETRY DATA FOR SELECTEDDEVELOPMENTAL PLATES HEAT TREATED TO 2700°C

Skeletal Bulk Specific SpecificNo.v 'Percent

ZZ341

ZZ367

ZZ465

Subscale

Density(g/cm-3)

2.10

2.09

2.07

2.14

Density(g/cm-3)

1.87

1.83

1.76

1.73

Pore Volume(cnT/g)

0.059

0.069

0.083

0.109

Pore Surface(nf/g)

25.3

24.4

33.3

27.3

Porosity

11.0

12.6

14.7

18.9End Plate(A99 Graphite+ 33% Resin)

Table 3-31 for composition, etc.

3-113

!__ -^

ii' Z CO COk_ — uj LU

co tr: c:_^_ ^fc III^^ ^-^ Uj

CM «H Q£ -J -I< <

. - M K- »-u_ u. r^v o oUJ LU t*\ 1— 1—

t— ce: o:i «•" <— + &* cr<! CM CM

Z ZZ CO ;? ;$• ~ ~ ITi .* 1,I co co CD1 LU UJ U_ rHu. ae or s: cr> CMi bsP ft«P LU «-" -*•

22 ?5 i *^ "^O O — 1co K> o- a Q

a. z z+ i XX

+ JC l_4 1— 1

' u a: a:t~* CT> CD < O O

CD CT> \— ui UJ<t <C CO LSi Ci

II I' II II II

•-« • o x D

r

Li s/' ^i- <'i \L V! \1 \J_ \ii' •1

it i*i i*i i i l it i it it

0 0O) CD

l l l|!!l!!!in!IIMUII>]JIIIUMIJIMim!J

*••

**—

A \trn «

• ,TX frl

/ /I/ // J -

/ // ^

/ 'J'/ /S*/ ,yi

/ Ss *"s sS // s/ 1/ // // // 1

f S / /s / /

s ' •< / 1\< i• • \\ • \\\ \\ \\\ \«— « fe\ \

WAESD-TR-85-0030

O

S "

<n

o

•«n>

ooen

oo5>. 0CM ~

t3UI

o Qa"* «•<M §,

UJJ"~

ao t-

•*+ -?cj D«j»2.

a 2S <*CD I—^^ ••-«

1-

a $a»-•

oQra*-«

o3UJ •

QO

O Q O O 0*°Is- to in - en

O)t.3^J(0S-o>D.

niUJI —

•4->COJ

•»->w<yS-

1—4->re£•+jenCO)

•»->oor—ro$_3Xd)

Ll_

«OCO

1

oo

Ol

301

LL_

H13N3H1S

3-114

-WAE-SD-TR-85-0030

The in-plane electrical resistivity of the three materials was determined and

is provided in Figure 3-31. As would be expected, the resistivity decreased

with increased heat treatment temperature.

Shrinkage data were obtained on small samples of Stackpole MF958 platematerial heat treated to 2700°C (4900°F) and ranged from 0.5 to 0.8 percent.The accuracy of these measurements is limited and more reliable data will beobtained from full size 2700°C (4900°F) heat treated plates.

Helium permeability testing performed on small (4.5 cm (1.8 in.) diameter)samples of Zee-Zee plates heat treated to both 900°C and 2700°C (1650°F and

o o4900°F) indicated a very low permeability of 8 x 10 cm /sec (0.48 x-8 310 in. /sec) or less for Stackpole MF958 material, while the regrind A99

material gave values of about 6 x 10 cm /sec (0.36 x 10 in. /sec).

Based on the screening tests performed to date, the most promising graphitepowder for production of large flat as-pressed plates is Stackpole MF958.Additional samples of this material were heat treated to 900°C and 2700°C(1650°F and 4900°F) for flexural and electrical resistivity measurements. Thedata obtained are given in Tables 3-35 and 3-36 and are in relatively goodagreement with the screening data presented earlier.

The backup to this material will most likely be the Asbury Regrind A99 platematerial. Several additional plate samples were evaluated after 900°C (1650°F)heat treatment (ZZ650, 660 and 666). The average flexural strength was53.2 MPa (7,700 psi) and ranged from 48.5 to 55.6 MPa (7,000 to 8,100 psi),while the in-plane electrical resistance for the same samples ranged from 2.99to 3.17 mfi; the average was 3.08 mfi.

3.7.2.2 Plate Corrosion

Autoclave corrosion testing of the three principal candidate plate materials,Stackpole MF958 (33 w/o resin), Asbury Regrind A99 (M9 w/o resin) andAsbury Regrind 4421 (M9 w/o resin) was performed in 97 w/o H~PO at

3-115

WAESD-TR-85-0030

*B

_ —* Q>

tn t- OOi t/).p-_

O£ C

ro i— to— +•> <O 0)

O •»-> QC4-> O

O CO *~»•>» O CO X

CSJ— *J-C\JCO >>

— *S- lf> L.•— *CT» O

•o u_c T3 s: i•r* Ci. -f- Q. O

o: j* 2:•— oCM cr> re +->^s- cr> -4-> o

. <•<(/) _j

i i i~ X O •

„„

-

""* /

«•

—

p p *1

'/ / ' -/ / ' -/ / '9 f m

/ / 1

/ / 1

/ / 1I I 1II •

II I• I */ / 1 mm

/ ' /

/ / /

s / /s / // / f< t /. 1 . , , . . . , , , 1

en CM cc

im-iin in-ni 'i i TA T i crn\j °

Ooen

o0o

g(VI

oo(M

0ocsl

oa0)^*

0oin^^

oo(M•

aoO)

ooI)*0

)1)

-

oo••

a.UJ

i-UJ

K-

UJ

1-

1—<UJ^~

<D

<o

a.1

ci-1 —a>

3-4-J(Os_OJex

1—4->cO)E

(O

i.1—4->ns0)

.Z

</)>

>>*J

>

+J1/1

djQi

(Ooi_

4->Ocu

UJ

•CO

1OO

OJ

3

•r—U.

3-116

WAESD-TR-85-0030

l»- e— -CO HJ 4->

i. crCD 3 COl X CDC £ S-"3 r- -l->

DC U_ OO

oin<=!•

«»^~

.,_! ^~»/>| 0Q.I *J

ooo

#t^-*f^

CT»

ODr^

£| .5sl co

inr-»

_i<t— H

o:LiJI— . 0<2:UJ1—< I— J=_i co 10 *->a. cn j_ o

«3 3 C0£. S- X CU

in < £*£7 o <££00 0-00 I— (LU CQ

cn n-§ 0

I- zU_

•r- O(Sl\ CM0-1 "

CMfO 1 *a. r~-21 r^

p»

OzLUC£.t—OO

1

C£0 +->X CLU CU_J *J Eu_ to -t->

CO <Oa: cu

s-1—

-J ob] oo I cn~ j

0mcr>•>^~o

-t->oono*in

CO

<*in0+Jr>

U3PO

o0cn

A

IO

in^»•*

o"CM

"O-» CU•U X0 -r-

•— 2

s- >»cu £?•a Q3O •" — •>

*~\ Q. c r-^• «^» U1

in "£• i/> mo cu cor— z c a: •

to — cn•<- cu o coi- 4-> co ^^mcu (a m 34J i — cn r>jro Q- U- co r lSi—4 S CO- —

OO

ACMi —

O+->oo

*o

«*COCO

o-I-)k£>

cn^0

ooCO

A

or—

«>r•1-

r~

oocn

00o

A

soo4->

oir>

«>»

—r>LO

O+->no

cn1-

OokO

««^

*CMLn

o.

CM

-a- 0)-I-) X

°'£s- >»cu s-•0 Q3O •"-*Q. C CM

3 </»Scu cuc a: •>

- — co.oo

CO - 10m 5cn hJU_ IO fvlS co- —

OoCO*o

o4->

oor^*

CO

•*'d-1

o+Jcnenin

ooo*or—

cnCO10

0ocn

0inin*

r«.

04->

0oCO

Ain

oCMino+jcncnro

Oinin

•»10

—in•*

o^CM

~a^~. CU-(J X

^sS- -!->cu cu-0 33O •.•"-*Q. C CO

•r* ^^'3 l« VDCU CUc Q; •

• CMO ^~

CO -^VOm 3cn MU_ CO MS co> —

3-117

WAESD-TR-85-0030

i/> e2 °

«(*•

o:UJ

<=£

UJ

<c_JQ_

o:

oD-

CQ



UJ

UJ

Q-I

10

m

10o

(O

J-cu

OO

oo

ooCO

cu•ooQ. C

cuc

00^S 5u_ roS CO

ir>•oO) .X CTi

M- CO

LT>

s-Q-

o•

CVJ

oo

ro

oo

S-QJ

oQ. CVJ

2 W kDO) O) TDc o: cu •

*-" X 00O •!- O

00 ' S VOin 5cr> >>Mu_ *i> s- tvis oo Q>_^

oo

oo

0o

cu•oo •Q. C ro

2 M . voQ> O) T3c o: cu »

—- x cviO •<- *f

OO ^S v£>m "S-en •»-> (viu. co cu r>j•S. CO 3-

3-118

WAESD-TR-85-0030

190°C (374°F) and 485 kPa (4.8 atms) after heat treatment at 900, 1500, and2700°C (1650, 2732, and 4900°F). The dynamic hydrogen electrode (DHE), usedas a reference electrode, was polarized at 0.1 mA/cm2 (0.09 A/ft ) andexhibited a stable potential of approximately 22 mV below the reversiblehydrogen electrode (RHE). After each corrosion test, the density of the acidwas determined and showed a change of less than 0.5 w/o.

The corrosion potential-current density relationships (using one side ofspecimen geometric surface area) for the 900°C and 2700°C (1650°F and 4900°F)heat treated samples are shown in Figure 3-32. Relatively linear Tafel sloperegions over two to three decades were observed for the samples tested. The2700°C (4900°F) heat treated samples exhibited slopes of from 57 to 85mV/decade which were somewhat lower than those of the 900°C (1650°F) materialwhich gave slopes of from 116 to 145 mV/decade.

Table 3-37 summarizes the corrosion data at 0.8 and 0.9 volts versus RHE. Atsimilar test conditions, the 900°C (1650°F) heat treated plate material madefrom Stackpole MF958 powder (33 w/o resin) gave a slightly lower corrosioncurrent density than the two regrind materials (M2 w/o resin), presumablydue to the lower resin content of the former. At 0.8 V versus RHE, the 2700°C(4900°F) heat treated materials exhibited corrosion currents of 0.0017mA/cm2 (0.0015 A/ft2) or lower while for the 900°C (1650°F) material,

2 2corrosion currents ranged from 0.17 to 0.77 mA/cm (0.7 A/ft-). Increasingthe heat treatment temperature from 900°C to 2700°C (1650°F to 4900°F) thus'reduced the corrosion current density at 0.8 V versus RHE by at least twoorders of magnitude while the 1500°C (2732°F) heat treated material showedonly about a factor of two reduction.

After corrosion testing, samples were gently washed in hot water to removeacid residue, vacuum dried, mounted and polished for optical microscopicexamination. Microstructures of the corrosion surface cross section are shownin Figure 3-33 for the Stackpole MF958 plate material. It is noted that thecarbon phase resulting from pyrolysis of the resin binder was selectively andseverely corroded in the 900°C (1650°F) heat treated plate resulting in a

3-119

WAESD-TR-85-0030

UJ

1.2

1.1

_eoc 1-0o"oQL

o 0.9(0 .Qo0 0.8

0.7

0.6

0.510-3

Bipolar Plate Materials190°C. 4.8 atm, 97 w/o H3PO4

O ZZ341 (Stackpole MF 958). 2700°CA 2Z367 (A-99 reground), 2700<CO ZZ465 (Asbury 4421 reground). 2700°C• ZZ341R (Stackpole MF 958). 900°CA ZZ367R (A-99 reground), 900°C• ZZ465R (Asbury 4421 reground), 900°C

I I I I II III I I I I I Mil I I I I Mi l l

10,-2 10

Current Density, mA/cm2

Figure 3-32. Corrosion Current Vs. Corrosion Potentialfor Heat Treated Plate Materials

3-120

WAESD-TR-85-0030

00_letH- )

Q;LUI—«£2

LUI—<_1a.a:<o

r^ ^

7 S00 LU

3 <

S 2^ Q^ z

<ou_oLUI—<o;zoHH

oo0C£C£.OO

^™*sj-OQ.COZ

O

3

r^en

»E4->(O

00•

• -«t

f_3O0eni —

• •

</>co

•1—-(->T5CO0

1— •<o

4->CCUE

•r—S-cuQ.XUJ

•!->CCMCU ES- 0i. 3 <t0 E

c •>o >•i/i <no •S- 0S-O -4->o 10

4-1CCM0) ES- (Ji_ ^3 «a;o Ec »o >I/J 00o •s- oi_O 4->O <O

CJ+J 0cCU ••E cu

-t-> I.(0 3CU -t->S- fO

t— s-cu

4-> CL•0 ECU CUI I—

Oi — i

CUr—O.Ero

00

</)-«ro

•r—s_CU

4->(0S

2(Oa:

r~» *d- OVO LO ID VD VO CVI

« £ > O " O > d - l O O C V J C V J O• • • • • • • • •C O , — O r — O O C V J r - O

* *o r^ LOi™~ f— l~~

r ^ ^ j - o r ^ c v j o O o or ^ - c o O r — ' — o r ^ c o o

• • • • • • • • •o o o o o o O Q O

o o o o o o o o oO o O O O O O o Oc r i L O 1 * * - c n L O r ~ . c n i o f * *

i — CVJ i — CVJ i — CVJ

a: o a: r*- oz «*t ^ - r o ' > ~ i — c v i i — m . v o mIO | ID « - 1 «^ VO | UDc o c i c o c o ' i c o «* < ; ^ *M l«sj M M ^ r v l M Q . I V I1^1 l_ i>g r-vi I— r-g M M N

-a -a -oc c c•r- «^ *r-S- S- S-

CO CO CO CT) en CT>LO LO in cu cu <uen en en e c . - c e . c z

•o -o "a u_ u_ u_C c C 2 Z S r - r ^ - ^ -•i- .,- -i- CVJ CVJ CVJS - S - 5 - C U C U C U « * « ^ - ' 5 l -c n c n c n i — i — i — « * ^ 1 *cu cu cu o o o

C£. c£. O£. Q. Q. Q. > > > > > >^ -^ ^ J_ 1_ 4-

c n c n c n o o o 3 3 3c n c n c n ra <o re JD ja J3I I I 4 -> 4 -> 4 -> I / I ( / ) ( / >

< = C < < : o o o o o o < « C <

<4-oco

•r—4->10r~Oa.tos-

4->Xcu>>JO

T3CUC•r-<O4->J3O

CU3

(O•>

•>

OO•

O

4-><0

T3CU>S-cu10J3O

10<0f

Z>CU COS- •S- 0

0 04->

cO (O•r- +Jin roo -oi-s- cuO CT0 ro

-l->O r—•r- o•a >oc s-(O CU

-Co enZ -r-* J=

3-121

ORIGINALOF POOR QUALITY

BIPOLAR PLATE MATERIALS AFTER CORROSION TESTING

Stackpole MF958. Graphite, 9006GHeat Treatment

200 X

Li*' y *j ii-i"'Lju' <y . ". . ''Kjflplim HI"11 ...• J j... "jp^r:-- :"«=»rrX.^?:,\:--'^;.X%v-vj'i/if^r*^î-:.*.'- 'iV4.T'.v-:V'. v7Cv'--» 5C';~i^??i5\-'-*?»-VJ>

l-.f*n;-.-*.i. -i«. ( ^ : , : ^ ^ ^-i^:.::v-:;> ';;-; vv.;r. ;:-. »i; .:.*^fr yVii>(frii>fcffVt,.-i-i- -£if'i riin fciT*-Ty:ri ii;.i v-frff y^ytÂiâ^^ .^Â- -Ad

Stackpole MF958Graphite, 2700°CHeat Treatment

200X

Figure 3-33. Microstructures of Stackpole MF958 PlateMaterial after Corrosion Testing.

3-122

WAESD-TR-85-0030

surface layer of relatively large loose particles similar to the StackpoleMF958 graphite powder used in preparation of these plates. The 2700°C(4900°F) heat treated sample shows relatively uniform surface attack withlittle or no selective attack of the carbon binder phase apparent.Metallography of the Asbury Regrind A99 and Asbury Regrind 4421 samplesrevealed similar corrosion behavior. The appearance of the 900°C (1650°F)samples agrees with the observations made previously on samples of platecorrosion from pressure-tested stacks and subscale tests - namely, that theresin carbon phase of the 900°C (1650°F) heat treated bipolar plates seems tobe selectively corroded while the graphite powder particle shows very littleattack.

Limited autoclave corrosion data were obtained on bipolar plate materials in93 w/o phosphoric acid at 190°C (374°F) and 690 kPa (100 psia). Plate sampleswere heat treated for times up to 100 hours at 900, 1100, and 1200°C (1650,2000, and 2200°F). These tests indicate that no improvement in corrosionbehavior was obtained from longer heat treatment times at 900°C (1650°F)(4 versus 100 hours) and that only limited improvement is achieved attemperatures of 1100°C and 1200°C (2000°F and 2200°F) for 100 hours (seeFigure 3-34). These tests confirm the conclusion that heat treatmenttemperature has the major impact on corrosion behavior, while longer times attemperature result in relatively little or no improvement at the lowertemperatures examined.

3.7.2.3 Plate Raw Materials

Particle size distribution for a number of graphite powders utilized in plateproduction, including as-received samples of A99, 4421, PXB-5Q and MF958 (twolots) as well as reground A99 (-40 mesh) and reground 4421, was determined.

Samples of Varcum 29-703 were heat treated to 900, 1500, and 2700°C (1650,2732, and 4900°F) to provide samples of the binder resin carbon for X-raydiffraction analysis. X-ray diffraction profiles show a contraction in the Cdirection (002) from 3.885 to 3.472 A coupled with an increase of about

3-123

WAESD-TR-85-0030

1.0

0.9

: 0.8*o

I o.

2 0.6

1C£.

0.!

0.4

BIPOLAR PLATE SAMPLES190°C, 6.8 ATM, 93 W/0 H3P0

EQUILIBRATION TIME: 24 HRS

HEAT TREATMENTTEMPERATURE i TIME

— O— 900°C, 4 hrs

— D— 900°C, 100 hrs

—A—1100°C, 100 hrs

— V —1200°C. 100 hrs

10 10,-2 ID'1

CURRENT DENSITY, mA/cm210

Figure 3-34. Corrosion Tafel Plots for Heat TreatedBipolar Plate Materials. .

3-124

WAESD-TR-85-0030

2 times in the L002 crystallite size as a result of the 2700°C (4900°F) heattreatment, see Table 3-38. Typical diffraction profiles at 900, 1500, and2700°C (1650, 2237, and 4900°F) are shown in Figure 3-35. Selected sampleswere used for determination of resin carbon weight loss resulting fromadditional heat treatments to temperatures above 900°C (1650°F); at 1600°C(2900°C) the loss was 1.3 w/o while at 2700°C (4900°F) it Was 1.6 w/o.

3.7.3 Seal Materials

The stack design relies on effective edge cell-to-cell seals as well asmanifold-to-stack seals. Seal tests were conducted to simulate the sealconfiguration planned for the 10 and 25 kW stacks. Selection of the materialstested was based on several specific parameters such as gas scalability,structural stiffness, acid corrosion resistance, compression set, andstability at 200°C (392°F). The commercial availability of stock sheet formsis an important consideration; therefore, the thickness of samples tested arenot necessarily the thickness for the stack application. Generally, gasscalability, structural stiffness, short-term compression set, and operatingtemperature stability parameters are exhibited via a sealing test at ambienttemperature and 200°C (392°F) . The general seal test configuration is shownon Figure 3-36. A flowmeter capable of detecting flow rates as low as 0.003cnr/min (0.011 in.3/h) air is placed in series with the gas inlet toprovide a more quantitative measure of seal leakage. The testing was done ona stack of five or six picture frame seals attached to heat treated 10 x 10 cm(4x4 in.) graphite/resin plates on one surface with Viton cement. A linearvariable displacement transducer (LVDT) is used to define the seal displace-ment as a function of time. The load and displacement values are recordedcontinuously on a strip recorder to evaluate the structural characteristics.Dry nitrogen gas was used to internally pressurize the seal stack to either108 kPa (1 psig) or 135.5 kPa (5 psig). Usually, two stacks were used toevaluate the ambient temperature and operating temperature, 200°C (392°F),conditions.

The perfluoroelastomer Kalrez was tested at both ambient and operatingtemperatures. This material was tested in the standard dry seal configuration

3-125

WAESD-TR-85-0030

TABLE 3-38

CRYSTALLITE SIZES AND LATTICE

SPACINGS OF HEAT TREATED VARCUM 29-703 RESIN

H.T.Temperature d0o2/A <hoo/A L002/A L100/A

°C

>900 3.885 2.06 9 18

1500 3.667 2.08 11 27

1800 3.645 2.08 13 28

2200 3.535 2.08 15 34

2700 3.472 2.10 20 48

3-126

WAESD-TR-85-0030

oo> o00

o(O

!N30U3doin o

COoCM

o o

(3 Oa T2 >«

a.LU

V)UJcc

I ~Ol yj

^Sor:

CM nz^:e/9

U-Q

tn

0<

2O

o.

UJO

£

COinCM'

oqCO

s

oCO

(Ocu

COor^I

CT>CM

ooCM

CO

IIo^!S- LJ>

LOooI

CO

S-3O)

COr-.'CO

cqCOCO

00O)CM

(O<N CM

0O) CM in

COpO

(SdO) A1ISN31NI

3-127

WAESD-TR-85-0030

§ ooo(O

oCM

I-csS<su:CO-

ua

gfc

_CMCO

o<a.wai

CM.

!

oOJM

= -0

<u '

oS-Q-

S-34JfO

O !

S-

OJS-

ooi

OO

O)S-3ai

(SdD) A1ISN31NI

3-128

WAESD-TR-85-0030

!N3DU3do00

oco

oin o*t oCO

o :co :

in

ooar»CM

ccX ,

OZO

CLtoLUO

3

(OintN

CM I

z;co

00CO

incoIb.

+JfOa)

oCO

a;

a > '

s_Q.

(O

II

inco

tos_

Lf>OOI

CO

O)

ino>T—(N

inr««'en

<Pin COinr>-

co

COr«.'oo

coin(O

O)CO

enCM

po

(SdO) A1ISN31NI

3-129

WAESO-TR-85-0030

DISK

VITOI^ >EALAl_ r

\ \7\rv-^.-i i/.,..

W*rf'S{S~ j 1}V

•\ »fc>i-: ;l fl*

^^E^ j JuJ B# *

/

1.v - l k ,-vJ

^ ,•_•! tt*^. tjQ

•*»-_4 ^S$fy*fli

1

1

~l

A

H.TS7 4^fl

Figure 3-36. Creep Test Configuration

3-T3Q;

WAESD-TR-85-0030

using no acid. The Kalrez tested, Compound 1-58, was 60 durometer Shorehardness. The Kalrez seal stack was cycled four times at ambient temperatureto compressive strains of 10.3 to 28.9 percent producing tangent elasticmoduli values of 9.85 MPa (1428.6 psi) to 16.01 MPa (2321.4 psi). During thefirst cycle, the seal stack was leak tested at 10.5 percent strain. Theresult was corner leakage at a pressure differential of 108 kPa (1.0 psig) ofnitrogen gas. On succeeding cycles, the seal stack was compressed toprogressively higher strains and the corner leak became nonexistent even at aninternal pressurization of 446 kPa (50 psig) of nitrogen gas.

Ambient temperature stress-vs-strain cycle data indicate a good strain range,0-22 percent, with a maximum seal compression pressure of 1.67 MPa (240 psi)with relatively linear behavior beyond the 3 percent strain value. Themodulus of elasticity values for Kalrez is similar to Viton with tangentvalues equivalent to 10.4 MPa (1500 psi). A summary of these modulus ofelasticity values is shown on Table 3-39 for the various loading cycles. Thistable also provides a comparison between the Kalrez and the Viton values forsimilar conditions.

The operating temperature test cycles are shown on Figure 3-37. These cyclesindicate a consistent tangent modulus of elasticity of 6.894 MPa (1000 psi)over the seal strain range of 4 to 20 percent. This also indicates that thematerial becomes softer at temperature. Again the modulus of elasticityvalues are linear and repeatable.

No evidence of seal leakage could be detected for an internal nitrogenpressurization of 135.5 kPa (5.0 psig) at both test temperatures. The minimumseal contact pressure during these leak tests was 1.554 MPa (225.4 psi) atambient temperature and 815.6 kPa (118.3 psi) at operating temperature.Kalrez seals also demonstrated the ability to withstand internalpressurizations of up to 446 kPa (50 psig) at both ambient and operatingtemperature with no leakage.

3-131

WAESD-TR-85-0030

<co

CO

COcoLU

OO

cr»co 'z.

I OCO I— I

CO C£.•=£ O

o•

co

o«=co

O

>-

CO

.X. i- O

V> U »"~

w -D 3 £ £ —

U £ O »— O.UJ U £ UJ —

•*- >>

W 4-> V> ^ 2

"> CU.— 4-» QL — 'C7> 3 V» E "-

M C -O «J V)<C O •— Q.

ui r— E UJ *-

• VI

E H— O U.

H-^ -t.O

ti 2 i- "E CUJ I— l/l - — • •»-

CU .C

c*c "01 E *c<O •»— C U *^

^ J= CU — •

_| CU t+_ I- *—•

V)C 4-»cu vt cu

^ o •»-CU I- 4-»CX CU 1-

i- ^ CL

oo>

E*S C

'O Q. to f •—E D.OO;-

' *re -^

11 ° 3.•z. '

o o<O to4-» •*-c *-

Sea

l S

tack

T

est

Co

Ma

teri

al

No.

D

esc

rip

tio

n S

u

rx. r—

cos

in

1

in mro Cvl

K°.

O ^CM CM

CMrx ^-

• cr>ro *-

cn inen rx

m r^J-CM

to

cu

o

s

U3 CO

^ in

~~CM rx,

CTt

sCM

CM O

O ~CNJ CM

CO ID

85CO CM

CM O

gsto in

^•CM

<o

CM

0)

r^rt

gco

CM—*. CM

CNJ rx

CM

VD O

O ^CM CM

rx, CMi— in

*r •—ss

ro O

co en

in

%

ro

CU

o

at CM

en in

CM —• CM

CM rx

ro

CM

CM mcoO

CM CM

—CM •—

CMO

CM O

33

CU

O

CM CO

ro rx

tno

^~en ro

voIT)

O

o «•CM CM

ID

^cnCM CMen •—

o oooCM CM

!?CM

4-»

CM ^

sS

CM

°s

co en

*°s

O rxen co

vo

in

O rx

S3

O ^CM CM

cn enrx, CM'i— IO

comCO Om vo

m «•

«o

CM CM

(J *~to u

CM

: '. .

o •—

co-—rx r—

°fO

VD

oCM

CO

CM O

O **CNJ CM

ro rx

in rxCO *"o •—

CO LO

g CMin

10 CM

IO

CM CO

.* CU

S-o

CM

rx. rof- CO

oorx •—

O tx

8OCM

COCM enm^rCM O

o *»CM CMVO •

55CO CM

vo m

SI

ss;

s

CM ^p

^ a>S'o

OO <-)

CM

rx rof- ro

00

en ro

osCM

roCM encCo

o p

CM CM

vo cn

vo»—

o-oCM O

m ro

SCM

s

CM init eu

OO O

CM

GO-—

s§ro »—

~c7pjr^

§OCNJ

enCM CO

3S

O r

CM CMID •

CM cnr*. «—§5

rx o

ro O

3S

4->

CNJ VO

a cuu*—to u*-» >>oo o

CM

CM cnmO

CM CM

CO r-

• CMCM rx.

in

CM

So

o

i— CO

SCM

ino

ss

—§§

CM-~* cu

CNJ

co

4-*

enoCO O

co o

• in

• oCM

gl'

oo

00°010r^ Oro C\J

0>0

COO

u> *mo

,

ro» eu

CM •—

ro

-

ro

CU<J

"SSC 3•*- f

^O (rtcu

o «~°*0

"? cu —

v> mV) *•» CO

4-> cr» eu cu

• 0 - 0 — 3 0 3••- S i- TO • 1oex 3 o cu o eucx-o u O.CM a.*o c *» E P

«O CDO >>r- 4J ^ 4J .

•— •— o c « cCU Vt >> CU O CUi. 3 o .t- vt •*-

t- Q) «" -CV> 1. 3 4-> I *•»ex ex o re in «

ro E > CU > CUCM O CU r- ->. i_ U U '— O

CM U- D. >« CU >,

*™*-o c y "u• cu cu c >> c

^ c cu o u o

Lf> +J yl T) Vfl

^o^k lc l£ vi ro g £ g** «o o * ^ O

X UID (0 1 1 1

CU nj vt - CM rol_ -a3 c eu cu cu»O -.- *J t) O UCU.C— ^2*0*

*— CNJ fO

3-132

OF POORUAESD-TR-85-0030

CMC

CM

O

O

OCM<£>

O

II

00OO

CMCM •

• CMO v_-

CME ECJ O

OO LDO *d-uo •O r—

II II

I— <C «tQ UJ |—

•-, o; O

•a: «c <cLU UJ UJco oo oo

i— CM OO

O

GD

COM

PRES

SST

RES

S

(/)a.(OQ.-^

o0CM

o00ro

oo

oa:

re<yooNai

(O ^^ii O

os- oO CT>

to i.s- 3

•»-> +JOO ns

t.i— QJ<O Q.a» E

OO QJ

t/> O1> C

«^-l/l 4->I/) ro01 S.I. OJ

•4J Q.OO O

oo c

^ s.t« 3d> en

E co oo o

i—COI

CO

a)i_3

OCTt10

3-133

WAESD-TR-85-0030

Another key aspect of the mechanical testing is that of permanent set at bothambient and operating temperatures. The results of this study indicated afterone cycle at ambient temperature Kalrez exhibited a 2.75 percent permanentcompression set, and after four cycles permanent compression set of 4.8percent occurred. Kalrez demonstrated larger compression sets of 5.8 percentand 7.8 percent after one and four cycles at operating temperature. Thesevalues are consistent with other elastomers which have been tested.

As part of this study, a comparison of material characteristics was madebetween Kalrez and Viton. A comparison of modulus of elasticity values atambient temperature, operating temperature, and finally at ambient temperatureafter being at 193°C (380°F) forva period of time is shown on Figure 3-38. Itwas found that the Kalrez seals were slightly stiffer than Viton seals at bothambient and operating temperatures where the elastic moduli of Viton are 8270kPa (1200 psi) and 4830 kPa (700 psi), respectively. Kalrez seals incurred asmaller permanent'compression set after four cycles at ambient temperaturethan did Viton seals. Kalrez demonstrated a permanent compression set of only16.5 percent of the maximum compressive strain developed during testing; Vitonexhibited a permanent set of 51.3 percent of the maximum compressive straindeveloped during its testing. Both materials demonstrated good sealingcapabilities under compressive loads normally encountered and with internalpressurization of 135.5 kPa (5.0 psig)..

A short compression/seal leakage test stack made of Armalon seal material wasassembled and tested. The Teflon coated fiber material, which indicatedacceptable acid resistance, was tested at ambient temperature conditions. Thedry stack was compressed and leak tested at three compression levels of 10,15, and 20 percent strain. At all three loading conditions, the seals showedleakage at 135.5 kPa (5 psig) internal nitrogen pressurization. This leakagedecreased at the higher compression strains. The exact source of this leakagecould not be located due to a limitation of the size of the test unit and thetest assembly. The material exhibited large permanent strains and verynonlinear behavior at ambient temperature conditions. Even though the tangentmodulus values are on the order of 13.79 MPa (2000 psi), the large permanent

3-134

WAESU-TR-85-0030

Figure 3-38. Compressive Stress Vs. Seal Strain forViton/Viton Cement and-Kalrez/Viton CementSeal Comparative Tests

3-135

WAESD-TR-85-0030

strains (95 percent of maximum strain obtained) indicate a lack of compliance.This lack of compliance of the material has precluded consideration of Armalonas a viable candidate seal material.

Mechanical testing of candidate seal material Kel-F was completed for bothambient temperature and operating temperature of 198°C (390°F). This sealmaterial has similar elastomer characteristics to the Viton compound. Ambienttemperature tests were conducted using four compression cycles for seal strainvalues ranging from 15 to 30 percent. The Kel-F exhibited tangent modulus ofelasticity values on the order of 8.67 MPa to 10.45 MPa (1258 psi to 1515psi). The original hardness of this seal was measured as 61 Shore durometer.The seal material exhibited a relatively small amount of permanent strain(3 percent) after two cycles to a maximum of 20 percent strain. The test unitindicated good sealing characteristics over the range of 108.6 kPa to 135.5kPa (1.1 psig to 5 psig). A 10 percent strain working range is available forsealing once the seal has been loaded to 30 percent strain. At ambienttemperature, no evidence of seal leakage could be detected for tompressivestresses greater than 853 kPa (109 psi) under an internal pressure of135.5 kPa (5 psig). The Kel-F seal material also demonstrated the ability towithstand internal pressures of up to 445 kPa (50 psig) under a normalcompressive stress of 2.5 MPa (352 psi). Since the Kel-F material exhibitedexcellent seal characteristics at ambient temperatures, no leak tests wereperformed at operating temperatures.

A second Kel-F mechanical test performed at 198°C (390°F) was completed andthe material exhibited relatively linear characteristics over the range of 4percent to 20 percent strain. The Kel-F material did not soften during the198°C (390°F) test phase. The increase in modulus of elasticity values forfirst cycle loading was on the order of 33 percent. The modulus of elasticityvalue changed from 8.674 MPa (1258 psi) to 11.69 MPa (1696 psi) for theambient to operating temperature conditions. It should be noted for both

ambient and operating temperature conditions, the amount of permanent straindeveloped was on the order of 4 percent after the seals had been compressed to15 percent and 20 percent strain values. .This is considerably less permanentstrain than any Viton compound previously tested.

3-136

wAESD-TR-85-0030

The seal stiffening effect after exposure to high temperature was alsoevaluated during this test phase. The evaluation indicated the Kel-F materialincreased in stiffness from 11.610 MPa (1684 psi) at 198°C (390°F) to 19.305MPa (2800 psi) at ambient temperature after exposure at 198°C (390°F) forMO hours. This stiffness increase is consistent with Viton type elastomers.

Compression creep tests of the Viton edge seal material were initiated. Fourunits are currently being tested at the following conditions:

TestTest UnitCondition No. 1

Temperature 149°C/300°F 198°C/390°F 198°C/390°F 54°C/130°F

CompressivePressure (using 1.72 MPa/ 0.689 MPa/ 1.72 MPa/ 2.76 MPa/Constant Load)' 250 psi 100 psi 250 psi 400 psi

Test Time 400 hours 1000 hours 1600 hours 400 hoursas of 6/1/84

Each test unit consists of five layers of Viton seal material separated bygraphite disks, Figure 3-36. The specimens were loaded, and deflection versustime data are presently being recorded. Each specimen will be tested until asustained creep limit has been reached. Documentation of the test resultswill be presented upon completion of the entire test.

Assessment of acid exposure of various fluoroelastomer material beingconsidered for cell to cell and manifold to stack seals was initiated.Samples, as received, were leached in 94 w/o acid for 150 hours at 185°C(365°F), and ion chromatography (Oionex Model 14 Ion Chromatograph) wasperformed on the resultant acid after filtering through 8000 A openingMillipore filter and dilution to 47 w/o H^PO.. An HgPO. sample was run in the

Teflon beaker used for leaching to establish the blank. In addition, weightgain/loss tests were performed on the various samples. Test results, Table

3-40, indicate rather substantial removal of F~ and. SO." from the samples as aresult of leaching. Selected samples were re-leached and relatively minor

3-137

WAESD-TR-85-0030

o«=J-oo

LU_1co<Cu-

oo_i<orLU1—e£

C£ill111

soi—oo«£iLUOC£.O— i

LU

LU>t— ILUC_>LUOf.

1oo•i

«cOSLU>LU00

z0

1—t_>UJLuu_LU

CJ3z1— 13TO<LU

1

•to

•r~to^^>>

frt*w

C<tf

>_o

t*^*^*

TJ<Dto3

>>

CLro»wuv*oO ^

••4-> O»g,=o J=S- 0>

jC-r-0 O>

3cO O)

r-4 S-o

«(-• <o

to r"i£_3 OO — «_J

-C OrO

0 >in•— c^_

(UTS•M CLJtL ._»V^» *^™

fTt t_'U W•ooo -am c00 ro

+-> raro S-

4J

i \ rn<— J UJO- Cooa: ose^0s*

•* -acr> +-»Ca;£O

<_>

1OJ «d-0

H- 003:01— 1UJ3 1 00

0>- Zco

Q-Q. 1

r—* 0

oot— ioo>-_jeiZ 1< LU

a>01SrOJT —»0 »«

**— •*

•

•*->13

C0

•r-+Ja.S-oto(Uo

(!)

"o.S<o

oo

.0 'rC

r~*J3

(U+J>>

'oS.•poa)

r—UJ

OJ

OJ

CO

in

0a>00

r—•

to

+

COi-£

oon

in oOJOr- •*o o

c-oo <u.^3>o

,__10<0z>

_*orOr^JD

(U+J>^

'oS--(-)oa>r^LU

OJ

in

in

o0en

•3-•

+

toS-j=OO

»l<T> o•* ooSr™ ^—

S "°O at£ S-> 0

OJ1o«io2p>

3 i-rO QJS- J3

•*-" Eto rC

O) O)

• - L>> >>

o -o os- a> s_+J i. 4J00 0QJ r— 0)

•— o r—LU O UJ

O OJO r-r— • r—

^- oo

to oo

OJ OJ

ro Lfj

000 c±

• •

0 0

+ +

to mo min to "oO 00 0)0 1 >i£ tO '•-

o i>N 0-C T3Ol -i-•r- o-1 <

OJ OJin r—OJ /—

in in

oo oto

o ojto ro•* •—

oo in» •ro o

+ i

UaiJ3JD3o:ccno ro•r- tO

s- toojO -r- •*E 0

r— O) •0) S_ 0a. a. z

r— r—1 |

o; ce:< << SQ- a.z» =»

^n£O(O

r—J3

CU-(->>i

OS-

4->Oa>r^LU

toO

to

O

OJinro

r—•

OO00

+

LUr-»-H

ooO2

r_

1o:<os=»

\x<J<0

J3

(U+->>>

oS-+JoO)

LU

oo

in

to

tooo

OJ•<*

+

<Tiinooi>i

o_M

S-os"aiQ-

ZO1 —1— •!

^

>-

<o:CD

_^oro

r-»J3

(U4->>»

*Oi.

-t->O(Ur-*LU

•*«!tr—

00

-

oo<•

in•oo

+

*_0)-QA3a:c too m•1- (V)

to <y»•r™ (—O

CJ<_l03

0)CT)CrOjro$_o"oooz

•sJ-

in

oo

oo

o<4-0)I —c•r-

cfD-oo

i£•z.fit_Jco

3-138

~ -— — = —~WAESD-TR-85-0030

changes in the anionic species levels were noted, Table 3-41. A third groupof samples heated at 190°C (374°F) in air for 100 hours prior to leaching showa large increase for several samples in the F~ due to prior heat treatment,Table 3-42. These data indicate that all of the fluoroelastomer materialsexamined as possible seal materials have the potential of being a source ofF~ and SOA~ contamination or poison to the cell.

3.7.4 Electrode Materials

Characterization of finished electrodes by half cell measurements wasinitiated and other evaluations were continued.

3.7.4.1 Design and Optimization of Half Cell Fixture

The design and optimization of a half cell assembly for electrochemicalevaluation of individual electrodes (both cathodes and anodes) wasaccomplished. Several generations of cell designs were evaluated in theoptimization process. As shown in Figure 3-39, the final half cell fixturemachined from.Teflon utilizes standard existing subscale test hardware(compression plate, carbon end plate, current collector, heating element andinsulation). A reference DHE was fabricated and calibrated in the half cellover the temperature range of 18 - 193°C (64 - 380°F). At 0.1 mA/cm2

2(0.09 A/ft ), the potential of this reference electrode was measured withrespect to a RHE in the same electrolyte. Preliminary tests were run in thiscell to identify optimum testing conditions, establish electrode pre-treatmentconditions, and to select the electrochemical method for polarizationmeasurements.

3.7.4.2 Effect of Sintering Temperature on Cathode. Performances: Half CellStudies

The baseline manufacturing practice requires the fuel cell electrodes besintered to enhance the Teflon binding strength. The sintering temperature isthought to be one of the key parameters affecting the wetting characteristics

3-139

WAESD-TR-85-0030

oo_ii— iC£.LUH-<

â:LU

O1—00<

LU0C£O— ->~J

_lLU

5. Q

00 £0

LU LU— 1 _JCO LU< oo

•z.o1rÔLULULULU

CDZi— io:o<LU_-l

QZOoLUoo

^_^

•COi.oJ=oLO

J_o4-

f ôLOCO

-u>top*

C5Q-oo:rs«^~o>^~W-

4->•i—s-ocuc~U(Ocu'cus_CO^~to

'1—i.cu

•t-><oe •1 >T^

cu^£COtoenT3cu4JUcu

^™O)oo-M*

^0-l->cu

o0

^J-oI— oo. nCDi— «LU3

>- oo00 i2(~HLJ0cu

««00 i—^^ CJOO>-'<«c

J_

cuenc<o.c • >

<_>*«•

4->

cQ

•r—•(->Q.

• r»S-uCOcuo

cur"~

Q.

OO

S-0).a

^s-(0•acu4-*>»

r—o+jocu

LU

ooooo

V

LO

^

oooOO

o•oo

+

iOoLO

•oo cuo >•*£. '•<-cuN 0cu cui. J-

^™fO t/)^<

1/>f

1

<

•»-> +->-= :en en

•r— »r—r— i—

^ 2?CU 'CU> >cu cu1 \ , k

^ ^2 2

-(-> S- +J L.U CU U CUO) -Q O) -Q,^ f= =' CZ n^ £3LU tO LU fO.

O COO CO^-

V V

Lrt i rt

V ^

O OOoo rôo

LO i —• .

l£> 10CO

+ +

LOLOl£> -0CO CU

1 >VO T-

o cuU S-

Lu CU o' s- e_l • r-=LU CO (U^ < Cu

T Ta; ce:< <1 1 *fuu ^—1 CUi<£ >

CS0s_^2

J^s-(O•ocu-p>>"os_+Jocu

LU

•3-oooo

V

1 f\

V?

oCOr—

r .•

^»

+

s_cuJ3J33

C£

c eno f-

•r- VOCO OO

•r- •-»•

Ocu •s- o

CU 2:

1a:<Q.5»

3-140

WAESO-TR-85-0030

CM

•*

CO

LU_l03<Ch-

*oo_i

. e£i— tOC.LUI—«Cs:a:LU2O>—00<=£

LUOon0

_lU.

QLU(—=cLUC£t—

1—sCLU3;z0

i—» sN»J

LUU.U.LU

CDZt— 1:E0<LU_l

•

t/>S- O3 3O 0J= <O

>OLO Cr— «r~

S- "O0 CU

S_T-

t-c_) -o0o-o<r> c• — ra

4J S-<O CU

4->ui <o

i— 3rO•r- -0S- CUCU N

•!-> -i-tO CE O

S- OJQJ-0

>>4->i— O0 .Ca.

c:rOS-r—CU <O> 5.CU 4->10 3

CUc co o

t^ 1 j

c?4^Q- T3COCU31 -E ^

to .&? <o a»5: 2 =31 -I-

CU _cc s. a>•r- CU .,-

2 CU•0 5CU to

SI CU CU0>— s_fO Q. Ocu E ii-

— 1 <tj CU• OO £3

(/>4->ccu

^0C_5

1CM «3-

1- 031 00CDt— • (LU3 •

1>- COCO O

zza_a.

* 1OO i—•— ' c_>oo>-_1<cZ 1< Li-

CUa>c(0

<-> >e•

i *

co4->Q.•r-i_Ol/tcuQ

a>Q.E(d •00

S-(!)JDEfi

a*4->>>os_

4->Ocu

LU

CMLO

LOy

LOV

oco

CT>•

t

1

S.cuo3o;C C7>or^.. u>w CM•r- «*ocu •s- o

CL. Z

f—1

1 —

a;CL.5»

S-CU

X3E(O

+->j:o>

•r™r—

CU4J>1

*oi-4->Ocu

LU

IOLO^~

LOV

LOV

oID

r~

•i—CM

1

s-o

^E"a)CL.

^—1H-31

<Q_5*

Ca)

os-.0a>^~a.ECOoo

o00CO

LOV

LOV

ooLO

•t

cr>CM

0•

r^

i

co

-t-> CTt•>- «d-5» i —

OTJ r-CUS- •3 O0 Z

CM1

h-31

(_><QS5>

O*JC

• ™

•acuS-3

4->U(Os-

*4—

CU (/>r— +Ja. cE <u•a Eoo en

<os.

<*-

^~f—

iO >OLO SCM 01

LOV

LOV

•*COCM

•»LO

1

Co+-> LO•t- CM> •—

O•O i—cus- •3 O0 Z

, —1

H-I

0<.Q2>

O4->C•^"

T3CUJ-3

4->UrOi-ii*r-

CU^™

o.E(O

OO

0 •*CO CO

r^

LO LOV V

LO LOV V

o o0 0CO CO

0 •>

r~^ ^~

O. ir .

+

LOLO10CO

1LO0

LU1— LL.!-, 100 -1O LUS ^

r-~ r^

i ii— i—E :co u.Z. -J=. ^

</>4->CCUE0)COS-

(4-

r—~r—.COEto

TOcoa.eo

<_3

S_(Ocur~

0

(U4-»_>>"oi.

4->Ocul~LU

CMr«-^~

LOV

LOV

0CO

CM•

r—

+

i—oooocu^ -o

cN 3CU OS- Q.

r— E<a o^ 0

r-I

1—a:i

_i•=c^

cuzt— «3:

^LU_l

O1—o:oi— iC£.a.ct:H-<

<

• f

HH

OOOC7>r—

1—<f.

00o:oo3T

0'0r^

Q£OU.

QLU1

<LU•a:i—i—<LU01

*

3-141

WAESD-TR-85-0030

>>!o

to:n<o

O)

T3O)-aO^~a.X

LU

CT>

ICO

01

3O1

3-142

~"~" ~" WAESD-TR-85-0030

of the Teflon-bonded catalyst layer. Electrochemical behaviors of three

"oversintered" and two baseline cathode samples were studied in the half cellat 190°C (374°F) and 101 kPa (1 atm). The "oversintered" 'cathodes were heatedan additional 6°C and 12°C (11°F and 22°F).

Throughout the current density region of this study, the polarization datawere highly reproducible on the two baseline cathodes. The increasedsintering temperature produced a substantial reduction in the achievableelectrode potential. At 200 mA/cm2 (186 A/ft2), for example, the hightemperature sintered cathode exhibited a potential of only 650 mV versus RHE,about 60 mV less than the baseline samples. Furthermore, the measured Tafelslopes also increased slightly with increased sintering temperature.

2 oAfter polarization at 200 mA/cm (186 A/ft ) for 40 hours or longer, onlyslight improvements (less than 5 mV) were observed on the two baselinecathodes. The performance of "oversintered" cathodes was significantlyimproved, but was still 20-30 mV below the baseline level. This result

suggests that increasing the sintering temperature improves the hydrophobicity

of the catalyst layer and thus retards the penetration of electrolyte into theporous structure. It also raises a question as to whether reducing thesintering temperature below the 360°C (680°F) baseline level would result inimproved cathode performance.

3.7.4.3 Half Cell Evaluation on Alternate Electrocatalysts

The electrochemical behaviors of several alternate catalysts were investigatedin the half cell apparatus at 190°C (374°F) and atmospheric pressure. Shownin Figure 3-40 is the Tafel plot for oxygen reduction on DC-06 catalyst ascompared to two present baseline cathode samples, C234-4 and C249-8. To

2optimize the catalyst utilization, each electrode was pretested at 200 mA/cm

2(186 A/ft ) for approximately 90 hours. The polarization curve of DC-06

catalyst was nearly parallel to those on the baseline cathodes. However, theobserved electrode potentials on DC-06 were 30 mV higher than the baseline

2 ?cathodes in the current density range of 5-600 mA/cm (4.7-560 A/ft ). At

3-143

WAESD-TR-85-0030

O0.

eao

HRLF CELL STUDIES958

988

69> 850

. see

750

700

650

600

190 C, 1 atm.. S9.E u/o H3P04RIR FLOH RHTE: 370 ml/mm.PRETESTING TIME: 90 hrs.

GRLVRNOSTRTIC METHOD--+— Cl?3-«<DC-006), 99 mV/dec.~o~ C23«-4(H.T. Pt), 95 mV/dec--«-- C249-8CH.T. Pt), 95 mV/dec

IB iea leee

CURRENT DENSITY, mfl/cm*2

Figure 3-40. Tafel Plots for Oxygen Reduction on an AlternateCatalyst and Two Baseline Cathodes

3-144

WAESD-TR=85-0030

2 2200 mA/cm (186 A/ft ), for instance, the measured cathode potential on

DC-06 was V40 mV versus RHE, as compared to the baseline performance of

mV.

Tafel plots for oxygen reduction on alternate catalyst DC-10 and DC-05 areillustrated in Figure 3-41. Electrochemical behavior of DC-10 was quitesimilar to the two baseline cathodes, showing a Tafel slope of ^86 mV/decade.However, DC-05 exhibited slightly better performance than the baselinecathodes. The measured electrode potential was 726 mV versus RHE at 200mA/cm2 (186 A/ft2).

Figure 3-42 shows the polarization curves for hydrogen oxidation on a baselineanode, A212-2, and an alternate anode catalyst, DC-03. The anodic overpoten-tials increased significantly with increasing current density; consequently,

no linear Tafel region was obtainable on either electrode. The alternateanode, DC-03, exhibited lower overpotentials than the baseline anode. At

2 2200 mA/cm (186 A/ft ), the observed hydrogen overpotential was M5 mVat the anode A212-2, ^50 percent higher than the literature referencevalue. This suggests that the platinum catalyst at this baseline anode may

have been contaminated by impurities from the Vulcan XC-72 carbon supportand/or the wetproofed backing paper.

3.7.4.4 Effect of Electrolyte Contamination by Viton Gasket CorrosionProducts on the Anode Performance

Half cell tests were performed to investigate the impact of electrolytecontamination from Viton gasket-acid reactions on the anode performance forhydrogen oxidation. The contaminated electrolyte, prepared by immersing andsoaking Viton gaskets in 97 w/o H3P04 at 190°C (374°F) and 485 kPa

- 2-(4.8 atm) for N170 hours, contained 320 ppm F and 68 ppm SO. as determinedby ion chromatography. This contaminated solution was then mixed with variousproportions of fresh electrolyte to provide a series of test electrolytes

containing a range of contaminant concentrations.

3-145

WAESD-TR-85-0030

OQ_

UJC3O

ccO

900

B50

800

750

700

650

600

190 C, 1 atm., 99.6 ujXo H3P04RIR FLOW RRTE: 370 mIxmin.PRETESTING TIME:. 90 hrs.

GRLVRNOSTRTIC MRTHOD—•— DC-010,—*-- DC-005,—•»•— C234-4(H.T. Pt); 95—o— C249-6CH.T. Pt). 95 mVxdec.

96 mVxdec.103 mVxdec.

10 100 1000

CURRENT DENSITY, mfl/cm*2

Figure 3-41. Tafel Plots for Oxygen Reduction on TwoAlternate Catalysts and Two Baseline Cathodes

50

30

oQ_

O

CJi—i1=)O

10

e

190 C, 1 atm., 99.3 u/o H3P04HZ FLOW RflTE: 31 ml/min.PRETESTING TIME: 35 hrs.

GRLVHNOSTRTIC METHOD—p~ R212-2 (,0.36 mg-Pt/cm-2)—•— DC-003 (0.50 mg-Pt/cm»2)

0 50 100 150 200 250 300 350 400 450 500

CURRENT DENSITY,

Figure 3-42. Polarization Curves for Hydrogen Oxidation onan Alternate Catalyst and a Baseline Anode

3-146

WAESD-TR-85-0030

Using pure hydrogen, the equilibrium potential of a typical WAESD-producedanode, A212-2, decreased with increasing concentration of impurities in the

electrolyte. At the highest impurity level used (i.e., 320 ppm F" and 68 ppmp

SO;; ), the equilibrium potential decreased by 20 mV. In the practical

current density region of 200-400 mA/cm2 (186-372 A/ft2), however, thecontaminated electrolyte resulted in about 5 mV higher polarization potentialthan that measured for the uncontaminated electrolyte.

3.7.4.5 Baseline Catalyst Characterization

Characterization of Lot No. 3 catalyst was performed in both the as-receivedcondition and after heat treatment at 900°C (1650°F). The data is in

relatively good agreement with both the vendor data and the PDS requirements.In the case of Lot No. 3, the as-received metal surface area is somewhat

larger, 16.3 nfVg. (79.6 x 104 ft2/lb), than that of Lot No. 2, 13.6 m2/g(66.4 x 104 ft2/lb), or the PDS value (11 + 2 m2/g). Heat treatment of

the catalyst to 900°C (1650°F) results primarily in changes in the platinum.

The metal surface area decreases from about 190 to 40 - 45 m2/g (9.28 x 105

to 1.95 x 105'^ 2.20 x 105 ft2/lb) while the surface area of the carbon

support remains unchanged. The particle size as determined by both trans-

mission electron microscopy and X-ray line broadening increases from 15 to 20

A to 39 to 44 A as a result of heat treatment. Increases in Fe, Ni and Cr

have been noted in the heat treated catalyst; this may be related to either

non-uniform distribution of foreign particles in the as-received catalyst orto possible contaminant pickup from the metal trays used to contain the

catalyst during heat treatment.

3.7.4.6 Compression Behavior of Electrode Materials

Compressibility tests were performed on Stack W-009-09 anode-cathodecomponents and Stack W-009-11 anode-matrix-cathode components. Both sets of

components were fabricated using the following procedure: Cathode (900°C(1650°F) heat treated catalyst, dry laminated, argon sintered) and anode

3-147

WAESD-TR-85-0030

(as-received catalyst, dry laminated, argon sintered). However, the Stack 09

components incorporated a thicker SiC layer and higher platinum loadings than

the Stack 11 components.

Three assemblies were tested. The first assembly consisted of five layers of3 3Stack 09 anode-cathode components wetted with 1.1 cm (0.07 in ) of

98 percent w/o phosphoric acid (H-PO.) per cell layer. This specimen wassubjected to five cycles at 56°C (133°F). The second unit consisted of five

layers of dry Stack 09 anode-cathode components and was subjected to fourcycles at 56°C (133°F) and two additional cycles at 198°C (390°F). The thirdspecimen consisted of five layers of dry Stack 11 components and was subjected'to three cycles at 55°C (130°F), two cycles at 191°C (375°F) and then one

cycle at 55°C (130°F). A typical compression cycle was as follows:

• load to 1000 N (225 pounds)• 30 minute load relaxation• unload to 8.89 N (2 pounds)

Although each compression cycle followed this general pattern, there was somevariation in maximum load and relaxation time. A LVDT and a load cell were

used to record displacement and load as a function of time. The data werereduced and evaluated to determine stress-strain relationships, permanent setcharacteristics and load relaxation effects for use in the stack analyses.

At the operating temperature of 56°C (133°F), the wet test unit with Stack 09anode-cathode components exhibited a very nonlinear stress-strain curve. Inthe 0 to 15 percent strain range, the material was extremely soft as indicatedby a low modulus of elasticity. The 15 to 19 percent strain range is a transi-tion region which was strongly nonlinear; From 19 percent to the maximumstrain range, it is characterized by relatively stiff, linear behavior. After

the third compression cycle, the stress-strain curves demonstrated good

repeatability. In the range from 19 to 22 percent strain, the average modulusof elasticity (based on nominal area) for cycles 3, 4 and 5 was 5.49 MPa

(796 psi). After five cycles, the wet specimen incurred a 10 percent permanent

3-148

WAESD-TR-85-0030

set. This permanent set represents 43 percent of the maximum strain developedduring the compression loading. During the 30 minute relaxation periods, thisspecimen showed a load relaxation of 10 to 20 percent of the initial load.Although this relaxation did not reach a steady state value, the amount ofadditional relaxation for longer time periods would be relatively small.

The structural properties of the dry W-009-09 anode-cathode components at 56°C(133°F) exhibited a strong correlation to the wet structural properties.Although the permanent set values were slightly lower, the modulus ofelasticity and load relaxation effects were nearly the same for both testassemblies. Overall, the phosphoric acid had a negligible effect on thestructural properties of the W-009-09 anode-cathode components.

At 190°C (374°F) the Stack 09 components appeared significantly softer than at56°C (133°F). Additional tests are being considered to fully determine the*structural properties of the Stack 09 components at this elevated temperature.

At the operating temperature of 55°C (130°F), the Stack W-009-11 anode-matrix-

cathode components also exhibited a very nonlinear stress-strain curve. Forexample, the cycle 2 test data indicate a modulus of elasticity (based onnominal area) of 117 kPa (17.0 psi) in the 0 to 10.6 percent strain range anda modulus of elasticity of 4.09 MPa (593 psi) in the 15.0 to 18.0 percentstrain range. At 191°C (376°F) the Stack 11 components exhibited a fairlylinear and repeatable stress-strain curve. The average tangent modulus of

elasticity at this temperature was 6.45 MPa (936 psi). After three cycles,

Stack 11 specimen accumulated a constant permanent set of 11.5 percentstrain. No additional permanent set is expected for a higher number of cycles.

Figure 3-43 compares the stress-strain behavior of the Stack 09 components,the Stack 11 components and the standard anode-matrix-cathode components.(Standard components refer to the electrodes used in Stacks W-009-01 throughW-Q09-08). Note the Stack 09 components and Stack 11 components weresignificantly softer than the standard components. This difference wasprimarily due to the soft behavior of the Stack 09'and Stack 11 components in

3-149

WAESD-TR-85-0030

(VdN) SS3U1S TYNINON

«» •« e» *<M CM -< ~

in o in a

|in

4Jc0)coQ.

OCO

OO

0fO

C•o.

<O•ac

oo

ca>ai

•)->cucoco

oo

co

ro01s-cn

(IScD SS3H1S 1VNINON

3-150

WAESD-TR-85-0030

the low strain range. The increased stiffness of the Stack 11 components ascompared to the Stack 09 electrodes is the result of the thinner SiC layer andreduced loadings on the Stack 11 cell components.

The primary design impact of the softer Stack 09 and Stack 11 electrodesoccurs with the stress distribution in the cell. Due to the soft electrodematerial, the stresses will tend to redistribute around the outside sealregions. This redistribution is being investigated in the nine cell stackfinite element models.

3.7.4.7 Silicon Carbide Layers and Powder

Helium bubble pressures were determined for various thickness layers ofsilicon carbide (all acid-filled for 15 minutes with 98.4 w/o H3P04) from

Cathode C204-4. The ambient temperature data obtained are tabulated in Table3-43. There is no apparent relationship of bubble pressure with coatingthickness. The bubble pressure varies from 89.7 kPa (13 psi) to 165.6 kPa(24 psi). These data are in agreement with previously reported bubble

pressures for acid filled silicon carbide layers.

An all-Teflon Gelman in-line filter holder was modified to perform elevatedtemperature bubble pressures up to 230°C (446°F). Data were obtained over therange from 190°C to 228°C (374°F to 442°F) and are given in Table 3-44. Thebubble pressure is somewhat lower than the ambient temperature average data,but not below 65 kPa (9.5 psi).

A second lot of SiC used for cathode insulation layer was evaluated.Comparison of the particle size distribution of this lot of material with thatof the previous lot indicates the two lots of powder have relatively similarparticle distribution.

Bubble pressure was determined at ambient temperature for cathode C214-2fabricated with this new lot of SiC. The values ranged from 82.8 to 117.2 kPa(12 to 17 psi), which are similar to values obtained with the previous lot ofSiC.

3-151

WAESD-TR-85-0030

TABLE;3-43

HELIUM BUBBLE PRESSURE OF VARIOUS THICKNESS

SiC LAYERS (CATHODE C204-4)

Sample S1C Thickness Bubble PressureI dent. mm in .kPa psi

1 0.10 0.004 165.6 24

2 0.10-0.14 0.004-0.0055 144.9 21

3 0.14 0.0055 103.5 15

4 0.11-0.14 0.0045-0.0055 103.5 15

5 0.11 0.0045 124.2 18

6 0.22 0.0085 96.6 , 14

7 0.22-0.27 0.0085-0.0105 124.2 18

8 0.27 0.0105 96.6 14

9 0.18-0.27 0.007-0.0105 110.4 16

10 0.18 0.007 89.7 13

3-152

WAESD-TR-85-0030

TABLE 3-44

ELEVATED TEMPERATURE BUBBLE PRESSURE

OF TYPICAL CATHODE (C204-4)

Test Temperature BUBBLE PRESSURE

Sample No. (°C) kPa psi

1 190 65.46 9.5

2 193 78.55 11.4

3 208 77.17 11.2

4 228 65.46 9.5

3-153

WAESD-TR-85-0030

Initial evaluation of zirconium pyrophosphate applied as an alternate to SiCwas performed on cathode C221-5. Bubble pressure for a 0.13 mm (0.005 in.)layer at ambient temperature ranged from 131 to 172 kPa (19 to 25 psi), andwas slightly higher and more uniform than data on SiC layers.

3.7.4.8 Backing Paper

A limited evaluation. of Kureha E715 carbon paper was performed. This paperwas evaluated as a possible alternative to the Stackpole PC-206 papercurrently in use. Comparison data are given in Table 3-45. The thicknessvariation of the E715 sample measured was larger than that of the PC-206 andthe average in-plane electrical resistivity of the E715 was about double thatof the PC-206. Based on these data and subscale cell resistances, thedecision was made that E715 was not an acceptable alternative to the PC-206.

X-ray diffraction of as-received samples of PC-206 and E715 are shown inFigure 3-44. These data indicate that the PC-206 is somewhat more "graphitic"

than the E715, having both a lower d002 and higher

Evaluation of the structural properties of the Kureha paper E715 wasinitiated. The Kureha paper is an alternative replacement candidate for theelectrode layer support as well as a possible acid transport member.Compression tests of E715 at both ambient and operating temperatures werecompleted. Two test units, each 20 pieces thick, were compressed in a dry (noacid) flat loading configuration. A typical compression cycle consists ofloading to a specified strain and then unloading to a 22 N (5 Ib) referenceload. Figure 3-45 indicates the compressive stress versus strain curve forthree cycles at ambient temperature. According to Figure 3-45, a compressivestress of 480 kPa (70 psi) would produce a 12 percent strain in the Kurehapaper. A similar test of MAT- 1 (carbon layer) indicates the same 480 kPa(70 psi) would induce only a 4.1 percent strain in the MAT-1. This verifiesthe Kureha paper (E715) is much more compliant than the MAT-1 (carbon layer)at ambient temperature.

3-154

03

Io*T

•

O

•— CO

LO °0CVJ «—

O10If)

oo

co•

o

CsJCOO

•

O

cvjO

OCOOOI

LO00

Ia;

ooo

IS)

zoIS)

LO t—I«a- D£I <C

ro a.

ooo

a.«=ca.CD

VOgoQ-

CU

IO

00

LT)

CO

O•

00

o•

VO

oroLT)r*»co

CO

oo

coo

oo

coO

CMvo

en•o

(U

'o.1

•a:CO VO

I/)

ioroCU

<4-O

in4-><D •CU

enoo^_^^_<Oco

•r—

^

iW

intoCUcj*o

•r—.C1—

•

C

•1—E

E

A

</)

toCU

.o•r~.C(—

•X

t•I

10inCUc^U•^.cH-

cu01rOi.cu

oU

CT>

A>>

4->•i—IOCCUo•oCU

+-)

UU****01

«»

>.+•»•^incCUo•o*» t

CM

S enO mo Q>

* £cu Q-In o.

CVJ

§

eU

(D CU

I I §•> Q. CL>> i? •*->

(O CU •!-> flj CU•— I- '•- 3 33 3 ino w> o,-i ro j.10 CO Oo s: Q-

to0>a:

in"-in

S-3

oo

CTI*o

LO

inO

Oo

O •—

NJeU

<i•A

LU3=Q£

»

t/)

>

S^

-»

*

r^

T3CU4->ocin10cu

r— CUC 33 0-

•^r- CLO fCO O^T CUr +J

• >»O i.

4->• CU

Q- E

cu toc o•1- i->— ocu a.torO C7Ico 3:

00o10

•0 £• «o

Q.

^—T3C •>(0 *r

0*-* o_in coV DC

•—•aa.cu4->E-r-CrOl(-CUin -i- u

•(-> S-co c cu— * CUCL

•o^-••-oLO OCO .4->i—^3- Or-s. c •>

o• «o

ooo• «T>

• cvj co *nr>

3-155

WAESD-TR-85-0030

S-01D.rOa.

a;3

O)

OQ.

U10

o

<L>

oS-a.co

OrOS.

toS.I

X

ICO

CDS-

O)

J.N3D>J3d 'AilSN31NI

3-156

WAESD-TR-85-0030

CL

U

LJ

m<r

a:a:LJa.a:a.

ua:

£ } f f ( 8 j

c See

gill— ew «*»

•v

ScuSCD

oro:i—CO

UJD_

o:

BCD

(ISd)

LCt

r^UJ

3

S-O

<5 ^o 3^ -»->

CO ^

s^h S

toCO

a>a.Q-

•5J-I

<DS-3en

3-157

WAESD-TR-85-0030

The second Kureha paper stack was tested at operating temperatures. Figures3-46 and 3-47 present the stress-strain curves at temperature. Note thatcycle 1, cycle 2 (loading only) and cycle 5 are at 55°C (130°F), whereas,cycle 2 (unloading), cycle 3 and cycle 4 are at 198°C (390°F). In addition,the cycle 4 data (Figure 3-47) indicate, a yield stress at approximately500 kPa (80 psi). The plastic strain associated with this yielding accountsfor the offset of cycle 5.

3.7.4.9 Differential Scanning Calorimetry

Characterization of Teflon 6C powder as-received, in a non-sintered electrodelayer (C209-1) and in a sintered electrode layer (C233-8) was performed usingdifferential scanning calorimetry (DSC). The thermogram of the Teflon 6C isshown in Figure 3-48, data from a computer-controlled DuPont 1090 analyzer.The glass transitions around 20 and 30°C (68 and 86°F) are in agreement withthe values reported for polytetrafluoroethylene (PTFE).^ ' The virginmaterial has a melting point peaking at.339°C (642°F). Upon codling, only aportion of the virgin crystalline phase resolidifies at about 310°C (590°F).The recrystallized portion of Teflon melts at a lower temperature than thevirgin material, at about 326°C (619°F), as reported.(2) No furtherchanges, both in the temperature and heat of melting, are observed uponcontinued cycling between 200 and 400°C (392 and 752°F). The thermal data aresummarized in Table 3-46.

Figure 3-49 is a thermogram of a catalyst layer (C209-1) that had only beenheated to 93°C (200°F) to drive off the NH4HC03- On first heating, a

broad exothermic transition peaking occurs at about 180°C (356°F). Thistransition was not observed in as-received TFE 6C. It was also absent on asecond heating scan. This is attributed, to the exothermic effect with thestress relaxation of the cold-worked, fibrillated Teflon. The enthalpy of thesolid-state reaction was found to be different in different samples, 25.2,27.7, and 8.6 J/g (10.8, 11.9, and 3.7 Btu/lb).^*^ These data may be an

* 'Based on the portion of Teflon in the catalyst layer (40 w/o).

3-158

WAESD-TR-85-0030

a.LJ

CD

LJCLCECL

LJn:

^ s5=5

c« - ^ S2S = s. s 2

• Ql CD ^^ ••p 275 SS MD ^^• •—• C/1 •-<| tC^CI R i s= 2! V S

in

«M

SCVJ

C9O)

en

UJD_

(ISd) 3AISS3ddNOD

oOco

oun

oe

•r~toi-

oo

Ol

oo

<uCL(Oa.

3

Irocus_3Ol

3-159

WAESD-TR-85-OQ3Q

CM

- o>

CD

Q.cnD_

ccLUOf

(ISd) SS3S1S 3AlSS3ildH03

oOcnn

ooCTl

(O

(Oo

(Os-

1000OJ

+->ooS-O)o.(O

o_(TJ1

O)

CO

O)S-

cn

3-160

iGiNAL WAESD-TR-85-0030

OF POOR

-—1-—-f—-4

T

T

T

*

<ou>oS-(U

oI/Oo

CO

CO

<u301u.

*-- --«-•g

4--

I T

3-161

WAESD-TR-85-0030

Transition

TABLE 3-46

MELTING POINT DATA FOR TFE 6C

OnsetTemperature

PeakTemperature

EnthalpyChange

J/9

1 Melting

Freezing

2 Melting

329

314

320

340

311

325

+ 40.5

- 23.2

+ 23.1

3-162

WAESO-TR-85-0030

f

I•t

t

i

t

-i 1 j-8

T

t

T

T

iii

i

4t

*T

H 1- H H

•t?oe\j

s-O)

00

T3O)S-O)

OlO

S-OJ

ooO

Iro01S-3

3-163

WAESD-TR-85-0030

important indicator of the state of fibrillation. Further work in this areais suggested, including thermomechanical and dilatometric studies on catalystlayers.

Unsintered catalyst layers (including the one shown in Figure 3-49) have amelting transition peaking at around 335°C (635°F) which is reasonably closeto the peak melting point of virgin TFE 6C; however, the heat of melting isonly about half that of as-received Teflon C&Q J/g).

After melting, no further transition is observed, either on cooling orreheating. This suggests that the Teflon, once melted in the Pt/C matrix, istotally amorphous. It is thought that Teflon penetrates into the carbon poresand forms thin films around carbon/platinum aggregates, thereby disallowingany opportunity for crystallinity. For the same reason, no success has beenexperienced in obtaining electron diffraction patterns of Teflon in electrodes.

Catalyst layers that have been sintered at 360°C (680°F) on Stackpole supportsshow no transition at all.

3.7.4.10 Ammonium Bicarbonate

Emphasis was placed on determining a satisfactory technique to determine theparticle size distribution of ground ammonium bicarbonate. Utilization of the

Microtrac particle size analyzer with either isopropyl alcohol, toluene orShell Sol as the fluid proved satisfactory. A number of wet grinding resultswere analyzed to assess the effect of the ball jar size on particledistribution. This work was in support -of manufacturing and had as its goalthe preparation of layer grinding batches of bicarbonate. Initial drygrinding experiments were analyzed for particle size distribution as well.

3-164

WAESO-TR-85-0030

3.7.5 Other Stack Materials

3.7.5.1 Manifold Material

Evaluation of polyethersulfone (PES) was initiated for possible stack manifold

application. Coefficient of thermal expansion tests for this materialconfirmed that it is anisotropic. The material expands differently in allthree directions and some shrinkage occurs between cycles. The criticalcoefficient of thermal expansion in the "X" direction of measurement was found

to be 2.02 x 10"5/°C (1.12 x 10"5/°F). The coefficients of thermalexpansion in the "Y" and "Z" directions were 2.55 x 10'5/°C (1.41 x 10'5/°F)and 2.17 x 10"5/°C (1.20 x 10"5/°F), respectively. These values are basedon an average of two tests per direction. Compressive creep tests wereinitiated for the Victrex (PES) manifold material. Two specimens are being

tested for 1000 hours at 199°C (390°F) and 20.75 MPa (3010 psi). The resultsof the test will help predict orthotropic creep rates for the Victrex material.

3.7.5.2 Stack End Insulation Material

The thermal expansion of the insulating block material (H-18408) was evaluatedin three directions up to a maximum temperature of 232.2°C (450°F).Preliminary results indicate that the material has nonlinear expansioncharacteristics and is anisotropic. The largest variation occurs in the "Z"direction. The "X" and "Y" directions are relatively consistent with valuesof 1.99 x 10"5/°C (1.11 x 10'5/°F) and 1.95 x 10'5/°C (1.08 x 10'5/°F),respectively. The "Z" direction expansion is greater by a factor of 3.74, fora value of 13.2 x 10"5/°C (7.37 x 10"5/°F). This value affects the amountof compressive force developed in the stack during thermal excursions. The

second cycles indicate that some shrinkage occurs during the first cycle.Therefore, it may be necessary to thermally cycle the finished parts made from

this material at least twice to minimize the shrinkage effect.

3-165

. WAESU-TK-85-0030

3.7.5.3 Phosphoric Acid

Ion chromatography was performed on three samples .of HoPiL, Table 3-47.The data indicate that all three are essentially the same with regard to F",

Cl", N03" and S042".

3*8 Advanced.Fuel Cell Development

To achieve the plant performance goals, the fuel cell technology needs to bedeveloped to obtain an average start of life performance of 680 mV/cell at

325 rtiA/cm2 (300 A/ft2), 190°C (374°F), 480 kPa (70 psia), 83 percent fuel

utilization and 50 percent oxidant utilization. A performance degradation

rate of 2 mV/1000 hours was considered desirable. Although significant

progress has been achieved in the development effort to obtain the above

goals, further improvements are required. Problem areas and potential

solutions were identified and a Performance Improvement Work Plan was'developed to achieve these goals.

Performance improvements are being sought by:

t improving catalysts.9 reducing the cell resistance.

• better understanding of operational parameters.

Efforts are also underway to reduce cell performance variance by:

• reducing the matrix thickness variations,i improving quality control techniques.

•"• improving fabrication and assembly procedures.

Performance degradation rates are expected to be lowered by the use of a more

corrosion resistant catalyst support, improved corrosion resistance of theplates, and acid management improvements. The advanced development effort is

described in the following areas:

3-166

WAESD-TR-85-0030

TABLE 3-47

ION CHROMATOGRAPHY OF VARIOUS SAMPLES OF PHOSPHORIC ACID

Analysis, ppm by weight

Acid F" Cl" N03" S042'

Baker Analyzed(0260-3)

Fisher Certified 3 4 2.5 13ACS (A242)

MCB, Reagent Grade 3 5 4 11

3-167

WAESD-TR-85-0030

• alternate catalyst evaluation.• alternate catalyst .support.• electrode manufacturing. -• cell resistance.• acid management. • • \• plate corrosion..• plate coatings.• impurity effects.

3.8.1 Alternative Catalysts

Heat treated platinum is being used as the cathode catalyst in the presentelectrodes. Improved catalyst activity with alloy catalysts has been reportedin the literature by several investigators. Evaluation of developmentalcatalysts DC-05 and DC-06 was, therefore, initiated according to testspecification TSE 2x2-008A. A heat treated platinum cathode was alsoevaluated to provide a baseline for comparison.

Initial performance of the developmental catalysts was 25 to 30 mV better thanthe heat treated platinum/Vulcan. Catalyst DC-06 showed the bestperformance. A total of 1,760 hours of pressurized, 480 kPa (4.76 atm),testing showed that the IR-free performance degradation rate of thedevelopmental catalyst is 15 to 25 mV/1000 hours. This is much higher thanobserved for the baseline heat treated platinum/Vulcan catalyst. Cellperformance data are shown in Table 3-48 and Figure 3-50.

The performance level of the alloy catalysts decreased to that of the heattreated platinum/Vulcan in 1,760 hours of testing. Similar results werepreviously obtained with alloy catalysts tested at ERC. An increase in cellresistance during testing was noted in all cells. This is presently thoughtto be due largely to the corrosion of the 900°C (1650°F) heat treated platesused in these cells. Cells containing 2700°C (4900°F) heat treated plateswill be tested to verify this hypothesis. Testing of all cells will continuein order to obtain long term degradation" rates.

3-168

WAESD-TR-85-0030

ooo

'-Ooi

ooQ-oUJ

OO

OO

OO•*CO

UJ_Jco

OOooI

CMXCM

UJOO

s £I "»uj tos eoen

ooUJ

Q

UJ

•

10co

•r~4->T)N•r-

•r—-M3

*•— »S_•^<o

•*-

1 _>+•

c .rd-O•r™XO

a«oLO

-ocnj

^_CM

:H^**

^—CU3

v»-

fc«CO00

*

O

—<E

LOCMCO

A

O0oCT>r—

«t<o•^10Q.

Op«

10co•^•pTJcooenc

• r—-l-><os-<ua.o

, — 3»CU E0

ft!— CU

fo enc <«

••- •!->6r:s~ o0»h*

^^«a--o ooo>^^J LO10 CMCU ^h-LO

C0<4-S- 0o LO£<

X^^ "^r- S-CO -!->O "3s

-pto

cu >>O 18C -P< <0

0

4J

cu to•o >»0-—

JC <«4J -P<o <o00

t—^^ •

cu o0 2

O-P 1 —

-PC r^to CO • —o oS CO CM

o; LOvo

SLO

^ r^to — •cu

Q. COCT>VO

•K

-— O<O «d-•i- r .+j - —c coI-H en

vo

o%LOr .

r—

I—

5gr

co0

C_)o

LOooQ

LOO

100Q.OLU

PCOp>.^^

oLOVO

•VOr~--_-*

r>.enLO

,-*•»^—•XIr^• —iCrtVO

enLOt .r--

r^1

1-<2

coooo

vOo

1oQ

VO0

1OOQ.OUJ

00co(•* *

(•>.LOvO

LOVOr--.

voenvo

-•LOVOr^

voenvo

enLOr »

r^i

I—eS.

^

co0

1oQ

VOO

10Q

1 —O1

ooQ.OUJ

•*coi*— 0LOVO

•COr^^ -*

envovo

^^<T>CMt^

- —r^vovo

CT>LOr>-i—

r-|

[-<s:

•PQ-

-PQ-

1—o: •

000

1OO •Q-0UJ

toO)O)

"34->r~O>

CUcus_

(+_

at.i— tcus_•ato0)tocu_c4->ccu<0a.c

•r—

to4-O)JOE3Z•K

3-169

WAESD-TR-85-0030

1

— . «o co0 0 O2 co co= o. a.• P 0o in ui

0 <3

w

"3*; co co

• 3 ? ?Q) 0 O• S a o

_. 0

CO

75a 10 coP O O

1 |

^ 0 Po P Q.£

«0

CM

O

<

Eto >oc CM5 <0 c*• • o=5 P 5C 0 « JO O ±fP a> E

^» **09 ^ 3

= «* ,?•CO to *

A ^ rfQ. O COO iv co

— 1qT-

1 1

IV CO0 01 1

CO COa. a.P P01 Ul

0 0

CO0

P *.^^ c^«

1 1

—

D •"o

P® • _

k- b^^

w .

w s: S ^o "• i- ^^oi i_; c^0 I J

.

£_

j

<00

;<too—

> <jmrvp«KP vihiO

«•^^ ^

c ^= S 0003S 2

5 . • •^ <^ • *<-oIO

1 1a» co6 o

: <ICPD9

<3 OD

<3 <3DL <D o —

1 1

ooq^CO*

o0

c\T

ooT*

O

^ *RC) 0

(0w£

ai2I-

00o

IDO

C/OQ.

0)O

4->I/I>)

'(O-!-»<O

C_5

O.O

O)o

O

CO

a.<os-O)

OLOI

o>'S_3cn

A 3QV11OA 1VNIINH31 1130

3-17Q

Six additional cells (ST-01 thru ST-06) were tested to compare the performancestability of developmental catalyst cathodes with the 900°C (1650°F) heattreated platinum/Vulcan cathodes under atmospheric stress test conditions(870 mV terminal voltage) and nominal atmospheric operating conditions, perTest Specification TSE 2x2-008. The cell performance at 200 mA/cnr(186 A/ft2) is presented in Table 3-49. The IR-free performance ofdevelopmental catalyst cells degraded faster than that of the cells with heattreated platinum catalyst, both under stress conditions and at nominalatmospheric operating conditions. Of the developmental catalysts, performanceof cathodes with DC-06 catalyst degraded much slower (36 mV/1000 hours) thancathodes with DC-05 catalyst (77 mV/1000 hours). Performance of the cellswith the heat treated platinum cathode increased during the test period. Thismay be due to continued wetting of the catalyst layer.

Based on the data available, it appears that the performance of cathodescontaining developmental catalysts DC-05 and DC-06 is less stable under stresstest conditions than under nominal atmospheric operating conditions, anddegrades faster than the heat treated platinum catalyst.

Cell 3009 containing a 900°C (1650°F) heat treated platinum/Vulcan cathode hasbeen tested at 101 kPa (1 atm)-and 200 mA/cm2 (186 A/ft2) for 11,000hours. A lifegraph of Cell 3009 is shown in Figure 3-51. During this periodits terminal voltage dropped from 643 mV (peak) to 619 mV. This represents anoverall degradation rate of 2 mV/1000 hours. The performance degradationappears mainly to be related to a decrease in catalyst activity. The cellresistance appears to be quite stable.

3.8.2 Alternate Catalyst Support

Analysis of stack and subscale cell data obtained under pressurized testconditions during the First Logical Unit of Work indicated that theperformance degradation rates of cells containing heat treated platinum/Vulcanelectrodes could be as high as 25 mV per 1,000 hours. Currently, performance

3-171

WAESD-TR-85-0030

vo0

1

±3at:_l_

H-

1—CO

CO^J_lUJC.J

H-CO^ -

t §I ( *

CO <->

I.I i-J

y •*m £< gH- £

a.0_iUJ>UJ0

u_oCDz1— 1J—coUJ

I/OtoLUC£| —to

^_,COoo

1CMXCM

LUCOH-

OLUa.cot

coLU1—""

*~~•r-.4J3

i.•r~(O

""*

-(-»

fO"O•»••

XoasinCM

•bcno

fVIWXJ

GC

^— *r '™cu

' 3v+-

*SCOCO

•t

CMEU

"*""'*»

EOO*""

«COoocr»

•«03

•r—toQ.

in^~"

• •toco•r~4->

TDCOoO>C

+J<OL- •cuQ.O

CH-1 CU »

CO O)CU S- 0)O>LU "3 >C 1 4-> EIB o; r--C 1-4 Of-3 >

to> tO tO CT>E s- s- cu c

CU -C 1- "-A I * i % »J•» 4^ +J ^^

LU 4- CO to v/>CD < O O)«C O VI- -4->1^ •— O

10

^- — .LU COLU CO CO CDor s- s_ cu cLu CU J= S- •<-

1 -4-* +J *ô; 'i- vo co to•-i < co cu• — • CO t)_ +J

oeC

i— «

cu o>LU CU CO C(— &-(/)••-

O rtl J_)cu *^—1 «4- J_ tO_l CU 4-> CULU CQ tO -*JC_)

Xr-" *r™i— S-n\ ' *CU •*••O "3

-(->CO

CU >>"o rnO ">

5S< «

^—f"^ •cu o0 2=

CU:>

•r—-•-> •»->to ocu cu to•o >»o <—jz. «S4J 4JA3 A30 0

p^rî

r—

COvO

0OVO

«±r~<o

«!•^fvO

00Or^

COVO

oto

+Ja.

r*.o

1h-to

COtoCOS-4Jco

tnoioo

or~|

,CTlVO

^J-invo

00CT>vO

coVOvo

or~-

1 —vo

ot/>

1 %+ft

Q-

CM0

1t—to

"oS-

-4->

coo

ino|

o0

VOCOi

cr>r>.vo

0CMVO

r«.0r»-

cr>invo

ini

CT>r^vo

ooo

i %+•a.

co0

i—to

toCO(U

4Jto

vO0

Oo

0

1

vOOsJ

1 —

r>.cr>vo

cr»CMr^

r>.envo

voCMr>.

i .envo

oto

»_i^â.

•*oi

I—co

oS-

4->Coo

VOO

1to0

CM

+

r_

r~.vO

00OVO

00COvO

1-ooVO

cnVOVO

00CMvO

oto

4â.

moi

i—co

COtocus_4Jto

4->ex.

h-

co^~+ •

r_

00VO

incovo

a>r*.vo

vocoVO

COVOVO

coCMVO

CO

CO

*J^â.

voo

11—LO

'os.

4->co0

4->Q-

1—a:

•(/)C0

•r—->(T3N•r-r—•^• j. ^^3

^5

^— *

-Pc<o

TD

XO

>soînCM

•oITJ

^~s

§•r_cu3

v»-

a<coCOVA^

•«fVI

°EU

~~v<E

0Or—

f«

000cr>r*~

•«(O•r-toQ.

in

• •

"os_+Jco

co

r^

oS-

•0 +JC to crO -r- O

_ 0to asC CM CUO _ p—• -c•r- in 4->4->(13 4- CNn *r~tj *^

•1—i— tO tO•r- C 104J O3 -r- C

4-> O*"*+ fO «r~

CMlM +JO -r- <0^ ^^ *^-^* ^ .

•i- +J•(-> 4-> CC 3 CUtf* f \*O ^J

-o — » c•r- CMOX O Uo - —

cu>S 4-> •(->CM C >». »O r—

Ln !•! O••- s-

T3 X •!->C O O<rt C1)<w y^

^"H • niCM a;

31 OT O)^ irt e;• • «v> c4-> rO

i— r— COCU O3 > CU4- J=

^L «J

>S (0co c cuCO •- S-

E 3•> t- to

0 CU Co -»-> a»C^^ ^^ ^Jr^ r™" I *

cuM f 1 TH•* ^J ^J

to - <u•1- •£» +> •to e o coa. a) i—

O i— r—m i-». cu a»i— CO to O

• •

</)C'o•r-

4->

•oC00

CT>C

•ptoO)

U^^^

totocuS-•Mto

3-172

™

•oceo

— _i

_

—

— M

0

C0OO)

— _c

COw

aO

—

cCO<J

>^

X0)ooCO

• =0

0

CM

^^<E

oCM*o

o

»

EaT-

1 1

_

a«b_c

^£CO

a! ""COCOoCO

"oO _

00

<] —

o

<J<J "

o

QiQ

c O

** u^Cp CO ^T^O N ^^

"* 5 o *"~ ^ ^)5 'J5^^ ^^ ^4^

CM *. <3

5 a O_ -"a

u. 0 1

CO «O ^CO CM <j)

<01 ^f 1

oo

w

ooo%

T-

oo

oT^

oo

o"

oooCO

oooaT

ooo

oooCM"

K CO IOo d d

WAESD-TR-85-0030

(0

UJ

COroOro-ac

ooco

a>o

i.a>a;

LDi

CO

0)i.3a>

A ' aovnoA nao

3-173

WAESD-TR-85-0030

degradation rates of similar cells, based on short-term test results, is 8 to12 mV per 1,000 hours. It is important to note that the 2 mV/1,000 hoursdegradation rate was demonstrated in Stack 431 operated at atmosphericpressure for more than 16,000 hours (Table 3-50) and Stack 560 (Figure 3-52).Recently, this rate was confirmed in subscale Cell 3009 which has been runningfor over 11,000 hours (Figure 3-51).

During the Second Logical Unit of Work, Gulf Corporation's ShawiniganAcetylene Black and 2700°C (4900°F) heat treated Vulcan XC-72 were selected asalternative catalyst support materials. A suitable catalyst support materialshould have a low corrosion rate, high surface area, and be electrochemicallyclean. Shawinigan Black carbon has a l.ow sulfur content (0.01 percent).As-received Shawinigan Black has a surface area, 90 nr/g (4.40 x 105

ft2/lb), similar to that of 2700°C (4900°F) heat treated Vulcan, 80 m2/g(3.90 x ICr ft/lb). The corrosion rate of Shawinigan carbon at 0.8 volt isat least an order of magnitude lower than that of Vulcan XC-72. Thesedesirable properties formed the basis for its selection.

Platinum/Shawinigan electrodes with 25 percent and 30 percentpolytetrafluoroethylene were fabricated. Fabrication of Shawinigan electrodeswas more difficult than Vulcan electrodes. Platinum/Shawinigan electrodeswere tested for over 7,500 hours at 101 kPa (1 atm). Performance data for thecells tested are summarized in Table 3-51. The electrode performance is ^20 mV lower than that of heat treated Platinum/Vulcan at 101 kPa (1 atm). Theperformance decay rate (2 mV/1,000 hours), however, is similar to that of heattreated platinum/Vulcan electrodes.

Performance of 2500°C (4550°F) heat treated platinum/heat treated Vulcanelectrodes was very low. Platinum/heat treated Vulcan electrodes tested inother ERC programs showed better results than those obtained here. Theperformance in this program was still 20 to 30 mV lower than that of heattreated platinum/Vulcan electrodes. The poor performance of heat treatedplatinum/heat treated Vulcan electrodes, may be due to a large platinumcrystallite size. This hypothesis, however, needs to be investigated further.

3-174

WAESD-TR-85-0030

TABLE 3-50

TERMINAL PERFORMANCE STABILITY OF STACK 431, mV

CELL NO.

1

2

3

4

5

AVG. CELL

INITIALPERFORMANCE

670

660

670

670

650

660

PERFORMANCEAT 16,382 HRS.

630

630

620

630

630

630

PERFORMANCEDEGRADATION RATE,mV/ 1,000 HOURS

2.4

1.8

3.1

2.4

1.2

1.8

OPERATING CONDITIONS:

PRESSURE:

TEMPERATURE:

REACTANTS:

CURRENT DENSITY:

CATALYST:

PLATES:

MATRIX:

CONTRACT:

1 atm180-185°C

H2/AIR

100 mA/cm2

Pt/Vulcan900°C HEAT TREATED MK-I

MAT-1

DEN3-205

3-175

WAESD-TR-85-0030

1 1 1 1

"

. •.

• 1

i'•

*m *

• *\

r• * * V™ •

•••

• • ~}

),

•j

. }f>

« "fI «' -s 1• > {

» *• " <•

a •;

5 I' .' S '«

o. .• i.. •. ° '.** »• •* «

^ o* * J)

• • IT".-_• CM O

.' *"» <

\ = E -r' 5 o• • u>

• •' o „- . : £ < -i*

« • aA" '. O• or -

sr j• *

1 \™

*j0> CO N. <o *

o(O00

o(0

CO

oCOto ••

o0CM

OCMK09

O

CMto

O(O^» •1O

oCOCMIO

o§*0CMCO

0

COCO

0<0oco

oCOCOCM

o3

CM

OCMy>

of^o(OCO

0s

oo. 6 d 6

o30zzo

ocUl0.o

o.oCM

10

S zsO OC

« oo(OCO

o<-oLO

003

JCCL03&.cn

evjIT)I

CDS-3cn

'aovnoA inao -OAV '30NV1SIS3U1VNU31NI

3-176

WAESD-TR-85-0030

-

E

oo1

o1 —Qi

0 — -»a. CNJ0 OOO O

1I— CNJ00 X

r- >- CNJLO _l

1 «=C UJoo | — oo

«=c \~LU O

CQ LU CO•=£ \— UJI— <C 0_

Z OO

LU 1 —1— 00

| LU<=C h-

LUO

LUO•z.<ca:oC£

D.

UJCDe£

o

, >

LULUa:LU

ce.

_jz— 'ccLU1—

OOa;~^o• ~_1_

'

1 1 1oo3;1 —sf

1—t— zOO ujo oS LU

O£

NX

^fLU

-J

1— (

1— t

^^

~

3 LU 00LU Ot*~ LOOO OO CM

t— LO

1

_l .LU O0 2=

wI~T f« —

^ ^y ji-. »_^ ?T

,3: o ^~™ ^C ^^< o eo —i

t—oo

_J

t—

0

Q£UJ

1Z -J 0

0 u_ oo

^ LU0

zO LU

1— 0=C 00z oot— * LUS CC=1 CU

u-

O UJ

J. ZLU 0— C_3

1—aia.CU

oo

CO IT) r~- i — LD OO O '-OsO ^C i CO ^*5 " '-O ^O- - ^.^ - " ^^f ^ - x^ ~ » . .

r^ ^ ^- i CO ro ip CO^ * — oo r~* co CM P~» oo

O cr> o O i rvj - 30•y\ in co co i r 10O vO * o vc 1 %O '-^—^ *— X x. x ^_^ ^_^ ~~^

«s- oo r-- co i r— oo•Jl OO OO «* 1 - •*

r- oo vo oo co en LO <s-<~O lO VO lO OO «* ' X > v O^, . . . ^ _ ^^ ^_,

tQ LO ^J f f*^ LO ^^ COOO C?v CM •* O - CO CM O

O O C O O O C M O O O « T0 1 0 0 " — O O O O O U T )

CM CM r— CM 1

CM LO tO OO CT> CMO O O O O i — P ^ - C O

1 1 1 1 1 1 OO OOOO OO OO OO OO ! /1 OO< X « f e t < : ^ < ^ O O O O

to to to LO to to to LO• • • • • • • •

o o o o o o o o

C U Q _ C u C u C U : C U - l - > + - »cu cu

1 — 1 — 1 — 1 — 1— • H-z: 32- DC - 3: a: 3:

CO-o c -a -a -a -a . -a -a+-> O -t-> -4-> + - > - ! - > -t-> 4->o o z o o o o o o o o o o o o

T3 -0 T3 T3

oo oo oo oo^1 " "O "O

X X X +-> X 4 - > - l - > -UOO OO OO OO

oo oo oo oor— r— p— ^-»

• • • •

o o o o

L O O O O L O O O OC M O O O O O O C M O O O O O O

•0 -0 T3•> « . OJ « » CU • CU

Ct— CH- c> Ct— 1— 1— C> C>IT3 31 <I3 "r" ( -r— ro "T* 33 ~T" {& «r— (O «^01 en c n c u c n en <D en cu•^ O -r- O T- O •!-<-) «C_) .0 -r- O -1- OCO CO C<1) C° CO CO C to in-C. — .C. — -CT3 JZ' — 3CM 3 CM JTOJ JZ13oo oo c>o oo > ^ oo y)

CMEo

^ cE O -f— £4JO <ITS o ^>. o

I/) O"> CMOC i— i— 3: CM0

•r- >,"O -4-*c ••—O •• l/l0 <1) c

S_ .. COCT •• 3 01 Oc a> +* -t->••- 1- "5 c 4-»4-> 3 S- IT3 CITS (/> co -i coS_ </) Q. o S-CU OJ S «3 S_O. S. CO CO 3Ol cu I— o; cj

3-177

WAESD-TR-85-0030

3.8.3. Electrode Manufacturing

Some of the performance variations observed in subscale cells and stacks maybe attributed to the variations in electrode manufacturing processes.Occasionally, partial burning of the electrodes has been observed during the

air sintering process. Subscale cell and out-of-cell tests were, therefore,conducted to study the effect on cell performance of the following variables:

• sintering gas environment.• electrode lamination pressure.• amount of filler ammonium bicarbonate used in the electrode

fabrication process.

Cathodes sintered in various gas environments were tested in subscale CellsMF-01 thru -04 at atmospheric conditions. Although no significant differencein cell performance was observed in the limited number of tests conducted, it

was decided to use an argon gas environment to prevent the possibility ofelectrode burning.

Thermogravimetric analysis (TGA) of the unsintered electrode layers in air hadshown that burning occurred around 160°C (320°F). It is known, however, thatheat treatment of the catalyst powder is carried out at 900°C (1650°F) in anitrogen environment without any noticeable burning. Experiments indicated

that burning can be limited by reducing the amount of air available at the

catalyst surface.

Effects of electrode lamination pressure and the amount of filler used inelectrode fabrication on cell performance was studied in Cells MF-05 thruMF-08 at atmospheric pressure. No difference in the cell performance

associated with these manufacturing variables was observed. However, very lowlamination pressure sometimes resulted in poor lamination of the electrode. A

more detailed study is necessary to optimize the required laminationpressure. Elimination of the filler from the manufacturing process can result

3-178

; WAESD-TR-85-0030

in a simplified process with significant cost savings. Further testing instacks is required to confirm the above results.

3.8.4 Cell Resistance

Subscale cell tests indicated that a 31 mV performance improvement could beachieved by reducing the present matrix thickness, 0.043 cm (0.017 in.), by50 percent, 0.023 cm (0.009 in.). These results were further verified in ninecell stacks. It was recognized that manufacturing processes would need to berefined to fabricate thin and defect-free matrices. A brief account of thecell resistance model and experimental results are presented in this section.

Previous effort under DOE/NASA contract DEN3-67 identified the potentialresistance contribution of various individual cell components. Significantreduction in cell resistance was achieved in that program by heat treating thegraphite-resin plates. Initial estimates indicated that the electrolytematrix accounted for a large part of the measured cell resistance.

The present MAT-1 matrix consists of a carbon layer, 0.025 cm (0.010 in.), anda SiC layer, .018 cm (0.007 in.) thick. It has become apparent that theresistance of the carbon layer is not directly measured by the AC millioh-meter normally used to measure resistance. Two subscale cells were tested toobtain a preliminary estimate of the resistance of the carbon layer. Althoughthe thickness of the matrix in Cell 3028 was 60 percent greater than that inCell 3029, the measured resistances of both cells were the same. The peakIR-free performance of Cell 3029 was 20 mV higher than that of Cell 3028.This suggested that the resistance of the carbon layer is not being measuredby the present resistance measurement technique. A similar gain in perfor-.mance for cells without carbon layers in matrices had also been observed inStack E-009-02.

Understanding of the effective cell resistance network was, therefore,desired. The previous resistance model did not include the interaction ofelectronic and ionic resistance of the electrode catalyst layers and the

3-179

WAESD-TR-85-0030

carbon layer in the cell matrix. The model was modified to include this

interaction. Figure 3-53 presents the revised model. When the cellresistance is measured using a mi 11iohmmeter, current from the meter passesthrough all the resistances. The path of the current generated in the cell isdifferent than in the milliohmmeter case. Thus, the actual cell resistance isdifferent that the measured value. Resistance equations for the new modelwere derived and individual component resistances estimated using theseequations. The estimates also confirmed that the matrix accounts for most ofthe cell resistance.

Resistance of the cell matrix can be reduced by reducing the thickness of theSiC and carbon layers in it. However, these thicknesses may only be reducedto the minimum required for these layers to perform their functions. The mainfunction of the SiC layer is to serve as an electrical insulator between anodeand cathode, while that of the carbon layer is to provide a bubble pressurebarrier to prevent gas crossover between the electrodes. Very thin layers ofSiC and carbon are sufficient to perform these functions.

As a first attempt, the thickness of the SiC layer was reduced from 0.018 to0.008 cm (0.007 to 0.003 in.) and that of the carbon layer from 0.025 to 0.015cm (0.010 to 0.006 in.). These thin matrix layers were evaluated in subscalecells (TH-01 thru TH-06) at atmospheric pressure. Performance and resistancedata are presented in Table 3-52. Individual resistances of the SiC layer,carbon layer, and hardware (electrodes, plates, and contact) in standard cells(2x2 in. and 12 x 17 in.) were estimated using the test data. It is again

apparent from these estimates that the matrix layers offer most of theresistance. A resistance reduction of 2 mft resulted from 0.01 cm (0.004

in.) reduction in the SiC layer thickness, which represents a performance gain

of 10 mV at 200 mA/cm2 (186 A/ft2). A carbon layer thickness reduction .of 0.01 cm (0.004 in.) resulted in a performance gain of ^ 9 mV at 200mA/cm2 (186 A/ft2). The total observed gain at 200 mA/cm2 (186 A/ft2)

2 2was 19 mV. At power plant operating conditions, 325 mA/cm (300 A/ft ), aperformance gain of 31 mV can be expected. This was verified in nine cellStack E-009-01 and is a significant improvement in cell performance.

3-180

WAESD-TR-85-0030

O)T3O

•oO

•o(U

COQ.

IO

a;•oo

a;o<a4->t/1

O)o:

I-u

•oo

ouo£o<aO

SB.oo

in

a.

utooco111

1Cu'CD

ozo

X

E

oo<rt-uui

azZuo

oooz

ooo

111>-

a.

<uU)

0)

0) eno tnC (II(0 ^4J CU) D

•H 4JW O O(!) -HK C O

O -HJ-l U-l

.. JJ .-it: o oE^ (D OO •—! Oj2 OJ W

roinro

(U

3-181

WAESn-TR-35-0030

• ' ''

•1—•4-> ..3

. . " . — .S-

•r—to

•* ,r

t

4->C

n i to

y ^• • < X"• ": ^ o

c/lXJ J*

• • • < / > £Uj CM

• ' <* -o

1 §<c ^_^UJ ^C-3 ^™z • —ef

cl ^•£ £• ' 2 s

CM a- S, LO .•-• i H"i CO r-J CM

o • Ef • -laJ ° 0T". 'CO • Z • • >•< ° ^V ;hr </> - „oo °

UJ LJ^ CM«tf«

. . i<;: -y-r- O

' ? 0^ CT>x •—•— i

. . o: _-I— fc5 -^s "*u. *~oi — •*0 ^1 1 1 -J**^Uj »•i, Ot E^ t

. 0•z.oCJ

<J3•Zl—t1—«cQC.UJ

• Q.O

G£»

CU(J/rtfO4->t/)

•r—I/)CUa:

!CUo

^_^

aiCUs_

u_1

o;i— i

"(Oc

CUI—.

- ,_

"cuo

X

S-4-><o2

p«—r^

CU0

Afll\w

crto

•»->"o5>

•r—e

•«to</>CUc-*u•1—jrI—

-!->'in*o

,_j

^ 0)<O CTICU tO

O_ • ^l_l_ ^— '

±J-."°< >

' —fO•r"•4-*•r-CI-H

-l->

COtO

—1

_!«£*oCU0

,<o

«M

c»— 1

(/) (/)

S- -(->3 WO CU:n i—

r— •

CU Oo z

S-cu>1tO

— 1

co_as_fOO

i-cu>,«_lo•r™OO

r-'- LO O ;'' i — CO r—

5t CO - LO " LO- J

OO LO O r— IS. O

CO CO •* LO ^1" LO

LO -!j- co o r«< '. r-CO CO CO LO ^ LO

«* LO LO CM LO CT»CT> CT> O r— CT> COLO LO f** P>. LO LO

ro r^ <* i— O cor^ r^ oo co t LOLO LO LO LO LO LO

^3- LO LO CM CO Oo CT> o i — o> cr>r~- LO f** r>. LD vo

r»» i LO r— - 'cr>CO t~» 00 CO l*» LOLO LO LD LO LO LO

1** LO «3" i— CM COCT> CO CT> O CT1 t -LO LO LO r . LO vo

LT> CO LO O 00 CMr^* LO ^ r^ **^ ^LO LD LO LD LO LO

o o o o o o' * < * ' * « * ' * • *r^. r^ r~. r^ r^ i —

LO t— CM LO fO «d"o o o o o o

1 1 I I 1 1•n: 3: 3= a: re a:-I— h- 1— I— H- H-

o LO o o LO of™ ' r^

co ro oo r^ r*^ r^

•-QCU-a3

UC•r-

-|_>

OcS-a>^^(O

c0nS-*ouat_c-t->s_o

Lt-

co

•r—4->(JCUS-S-ooa:^^^•K

3-182

. ... WAESD-TR-85-0030

The 0.008 cm (0.003 in.) thick SiC layer did, however, cause an electricalshort resulting in low OCV and cell performance. The manufacturing of thinSiC layer should be improved to yield a stronger, uniformly thin layer.

The 0.008 cm (0.003 in.) thick SiC layer was also evaluated in a nine cellstack (E-009-04) at pressure. Performance and average cell resistance dataare presented in Table 3-53. A performance gain of 14 mV was expected based

oon the measured resistance reduction. The observed gain at 325 mA/cm(300 A/ft2) was 18 mV.

3.8.5 Bipolar Plate Corrosion

Corrosion of the cathode side of bipolar plates has been observed in subscalecells and stacks after pressurized operation. The following factors aresuspected to play a significant role in plate corrosion:

• acid volume expansion.• high water partial pressure in oxidant exhaust stream.• defective wetproofing of backing paper.• fuel starvation in a cell driven by other cells.• oxidant starvation in a cell driven by other cells.

The objectives of this task were to identify the cause(s) and to gain anunderstanding of the mechanism of bipolar plate corrosion. Conditionssuspected of causing bipolar plate corrosion were investigated by introducingthese conditions into subscale cells and evaluating their effects on platecorrosion. The ultimate goal is to find a means to prevent or protect againstbipolar plate corrosion.

Subscale Cells PC-01 thru PC-08 were tested for 600 hours at atmosphericpressure to evaluate the suspected contributors to corrosion. Acid expansionand higher water partial pressure in the cathode exhaust stream were obtainedby running these cells with pure oxygen at greater than 60 percent utilization

3-183

WAESD-TR-85-0030

TABLE 3-53.

EFFECT OF MATRIX THICKNESS ON PERFORMANCE

AND RESISTANCE OF STACK E-009-04 AT PRESSURE

(1) DECREASE IN MEASURED RESISTANCE

Matrix Thickness, mil Average Cell (12 x 17 in.)Resistance, mfl

SiC Layer Carbon Layer At 325 mA/cm2, 70 psia, 190°C

7 10 0.12

3 10 0.08

Difference 4 0 0.04 mft

Average Resistance of SiC layer = 0.01 mft/mil

(2) MEASURED VOLTAGE GAIN

Cell No. Matrix Thickness, mil Cell Terminal Voltage, mVSiC Layer Carbon Layer At 70 psia, 325 mA/cm2, 190°C

1 7

2' 7

3 . 7

4 75 76 37 38 3

9 3

10

10

10

10

10101010

10

662

654

676 Avg.

673684641*678 Avg.708

680

669.8

676.5

*Cells 6, 7, 8, 9 were diagnosed to have.electronic shorts

(3) CONCLUSIONS

Based on resistance reduction:

predicted gain . = 14 mVobserved gain, Cells 6-9 =7 mVobserved gain, Cells 7, 8 and 9 = 18 mV

3-184

— - - - - - - WAESD-TR-85-0030

and then running them on air. This operation is referred to as an oxygencycle. Fuel and oxidant starvation due to gas maldistribution were simulatedby restricting the gas flow and fixing the current density via an externalpower supply.

Maximum acid expansion was expected in Cell PC-01. No change in performance,however, was observed after two oxygen cycles. Cell PC-02 could not beoperated at 84 percent oxygen utilization, (partial pressure of watercorresponding to plant conditions) hence it was operated at 78 percent oxygenutilization. Performance of the cell decayed by 25 mV after two oxygencycles. Upon disassembly of Cell PC-02, a platinum deposit was found on theSiC matrix and on the anode graphite plate in the acid reservoir area. Therewas no apparent reason for this rather strange occurence. No significantdecay was observed in the other two cell (PC-03 and PC-04). During oxygencycles, acid was seen to have expanded in the acid fill tubes for all fourcells. Upon disassembly, no visible softness was found on any of the cathodeplates.

Cell PC-05 was operated in an oxidant starved mode and driven by an externalpower supply. A decay of 31 mV was observed after two oxygen cycles. Nosoftness of the cathode plate was evident upon disassembly.

Cell PC-06 was run in a fuel starved mode during the oxygen cycle. This cellrequired nearly 2V from the external power supply. After one hour ofoperation in this mode, a significant increase in cell resistance was noted(from ^ 4 mK to 60 mft). Testing was terminated when acid in the filltube turned black. Significant corrosion was observed in the acid channelarea on the anode plate.

The results of these experiments may be summarized as follows:

• Up to 12 percent acid expansion and corresponding high water partialpressure did not cause any apparent softness of the cathode plate or

3-185

WAESD-TR-85-0030

the backing paper. This may be due to the lateral transport ofelectrolyte in the 2 x 2 cells allowing the acid to escape from theedges. The above experiments also suggest that similar acid expansionexperiments in 12 x 17 stacks would be helpful in isolating theeffects of acid expansion from any other effects of high pressureoperation.

• Oxidant and fuel starvation caused significant performance loss infuel cells in a driven mode. Fuel starvation is more detrimental thanoxidant starvation although fuel starvation causes anode platecorrosion; no apparent cathode plate corrosion was found in either ofthese cells.

Cell PC-07 was tested to evaluate defective wetproofing of the backing paperon plate corrosion. The cathode backing paper in this cell had ahon-wetproofed area (^ 1.3 cm (0.5 in.) diameter) in the center. This cellwas first conditioned and tested at 200 mA/cm2 (186 A/ft2), 101 kPa(1 atm) pressure, 25 percent oxidant (air) and 80 percent fuel (hydrogen)utilizations. It was then tested at 325 mA/cm2 (300 A/ft2), 101 kPa(1 atm) pressure, 84 percent oxidant (oxygen) utilization and 80 percent fuel(hydrogen) utilization to simulate acid expansion and high water partialpressure in the oxidant stream. After two cycles of acid expansion, nosignificant decay in cell performance was observed. After 600 hours oftesting, the cell was disassembled, and no corrosion was visually apparentunderneath the non-wetproofed area of the backing paper. A small area of thecathode plate near the acid reservoir was found to be corroded.

Cell PC-08 was assembled with a specially fabricated cathode. The backingpaper of this cathode had a 1.3 cm (1/2 in.) wide SiC strip in the middle inplace of standard backing paper. This provided an acid path from the catalystlayer to the cathode plate. Cell PC-08 was first conditioned and tested at200 mA/cm2 (186 A/ft2), 190°C (374°F), 101 kPa (1 atm), 25 percent oxidant(air) utilization, and 80 percent fuel Chydrogen) utilization. It was then

3-186

WAESD-TR-85-0030

tested at 325 mA/cm2 (300 A/ft2), 190°C (374°F), 101 kPa (1 atm), 84percent oxidant (oxygen) utilization and 80 percent fuel (hydrogen) utiliza-tion to provide a 8 percent acid expansion and high water partial pressure inthe oxidant stream. After two cycles of acid expansion, a steady decline incell performance was observed. Testing of this cell was terminated after580 hours.

Careful examination of the cell components during post-test disassemblyrevealed no apparent corrosion of the plates. The anode backing paper and thelower half of the anode plate were found to be wet and may have been the cause

of the poor cell performance.

Potentiostatic corrosion testing of as-received and wetproofed Stackpolebacking paper was also conducted. Test results are presented in Figure 3-54.Corrosion behavior of this material was very similar to that of pyroliticgraphite. The corrosion current at 1.05 V for the as-received backing paperwas V x 10"3 mA/cm2 (6.5 x 10~3 A/ft2) (in comparison to 4 mA/cm2

(3.7 A/ft2) for the 900°C (1650°F) heat treated A-99 graphite/resin platematerial). The corrosion current for a wetproofed backing paper at 1.05V was^ 2.5 x 10"2 mA/cm2 (2.3 x 10"2 A/ft2). This is higher than the valueobtained for the plain backing paper. This may be due to some residualdispersing agent from the flourinated ethylene-propylene (FEP) emulsion.These results suggest that the backing paper is graphitic and any further heattreatment of the backing paper may not be helpful.

Corrosion testing of Asbury A-99, Asbury 4221 and Superior 9026 molded plateswas completed. The Asbury 4421 plates showed the best corrosion resistance in

these tests. At 0.85V, wet-mixed Asbury 4421 plates showed a three-fold

reduction in corrosion current as compared with standard Asbury A-99 dry-mixed

plates. The dry-mixed Superior 9026 plate showed no improvement over thestandard Asbury A-99 dry-mixed plates. The improved corrosion resistance ofthe Asbury 4421 plate may be attributed to the higher purity of the graphiteand to the wet mixing process.

3-187

WAESD-TR-85-0030

QE POOR

g•40.

H

W

^j;:-:.: : ;- V { - \'-^^-(~- '.._:.: _.:.-L.l i^:::.:'tf:_i. ! : | ::'•_ : . . - iZ

rrr~.— . — .1 . . .... — . ; —: "••~-~T"">|-""-:'V: T^-.—•{- '• :-:-.;•=.---;• :-r~~:. •••••----:• -1;—I_t—i _^

s.O)a.<oa.0)

orecaai

r—Oa.

i." "rf).!Z

5-O

(O

(1)ca

ot-o

CJ

oo

3cn

**i._ — o u}.^_.. o -""* — - o inc?> • -- c\ • -— CD co---- r* rx-— — • -\£ ,

- o o — o — o

"rvix;:n.T.o.i

3-188

-•— WAESD-TR-85-0030

3.8.6 Bipolar Plate Coatings

A possible solution to prevent or minimize plate corrosion is to apply aTeflon (e.g. FEP) coating on the cathode side of plate (except the rib area)and thereby reduce the plate surface area in contact with phosphoric acid.Another solution may be to heat treat the plates to a high temperature (e.g.2700°C (4900°F)) to form a structure which is more corrosion resistant.Because Teflon (FEP, PFA) is a poor conductor of electrons, the bulk of thecoating should be sanded off the rib area to minimize the electricalresistance added by the coating. Another option which minimizes theelectrical resistance of Teflon coating is to use carbon or graphite filling.

A test plan was written to evaluate plate coatings which includes thefollowing items:

• Evaluate the corrosion resistance of FEP, FEP/Graphite A-99,PFA/Vulcan XC-72 coated plates, and 2700°C (4900°F) heat treatedplates.

• Determine the stability of each coating.

t Determine the effect of coating thickness on measured cell resistance,coating stability, and corrosion resistance.

• Demonstrate subscale cell performance.

The objective of plate coating is to form a protective layer against acidpenetration into the graphite/resin plates and thus prevent plate corrosion.The development of a protective coating with a minimum increase in fuel cellelectrical resistance is being pursued through vendors and at ERC. Thecurrent status of the development is presented below.

3-189

WAESD-TR-85-0030

Two subscale cathode plates were coated by a vendor with FEP to yield0.0013 cm and 0.0025 cm (0,0005 in. and 0.001 in.) thick coatings.Preliminary tests indicated that these coatings were electrical insulators.By. shaving most of the coating off the ribs, the increase in electricalresistance due to coating was reduced; however, the cell resistance was stillhigh.

To lower the electrical resistance of Teflon coatings, development of aconductive graphite/Teflon composite coating was initiated. Materials andplans for a composite coating were submitted to a vendor for formulation andsupplying coated subscale plates for future testing.

DuPont's "Topcoat Finish Black" (dispersion of Teflon and graphite) wasreceived and evaluated. The electrical, resistance of a coating less than0.0013 cm (0.0005 in.) thick was not acceptable. Discussion with DuPont alsorevealed that an unacceptable anionic wetting agent was used in "TopcoatFinish Black". Further work with DuPont is planned to establish materialssuitable for fuel cell use.

Presently, bipolar plates are heat treated to 900°C (1650°F). Higher heattreatment temperatures are expected to increase the corrosion resistance ofthe bipolar plates. Ten sets of Westinghouse subscale cell graphite/resinplates were successfully heat treated to 1150°C (2100°F). Five sets of theseplates were heat treated to 1500°C (2732°F) by a vendor. Heat treatmentcycles for these trials have been developed by ERC under in-house programs.The other five sets of subscale plates were successfully heat treated to2700°C (4900°F) by a vendor using ERC procedures. Visual inspection of theseplates indicated no delamination or blistering but the plates were bowed.Careful sanding of the plate reduced bowing satisfactorily. Dimensionalchanges due to heat treatment above 9006C (1650°F) were less than 1 percentand current MK-II stack hardware should accommodate these changes. Asexpected, the porosity measured by mercury porosimetry (pore diameters between50 microns (2 mils) to .006 microns (2.4 x 10 mils)) increased formaterials heat treated at 2700°C (4900°F).

3-190

WAESD-TR-85-0030_

3.8.7 Acid Management

A comprehensive acid management system study was submitted during the FirstLogical Unit of Work. Some experimental effort on different acid

management-related issues was also initiated during that period. The effortduring the Second Logical Unit of Work focused on four different areas asidentified in the Performance Improvement Work Plan document:

• Improve lateral transfer of electrolyte during non-steady stateoperation and acid addition by improving matrix and transport and analternate seal design.

• Test acid volume control methods.

• Estimate acid requirements based on individual component

characteristics.

§ Establish criteria and procedures for acid replenishment during stackoperation.

Three nine cell stacks were tested in support of this activity. StackE-009-02 was assembled with a seal design which extended the silicon carbidecoated cathode and the carbon layer over the acid replenishment channel. A

diagram depicting this design is shown in Figure 3-55. The design improvesthe wicking of acid into cells. The effectiveness of this design was

demonstrated by acid addition to Stack E-009-02 after its first pressurecycle. The cell OCV increased significantly as shown in Table 3-54. Thisstack had five standard cells and four cells which contained no carbon layerin the matrix. Cells without a carbon layer had very low OCV before the acidaddition, but increased to the level of the standard cells after acid addition.Apparently, acid was effectively transported to the cells. This data supportsthe recommendation that this seal design be used in future cells.

3-191

WAESD-TR-85-0030

ANODE PLATE CATHODE PLATE

ANODE

A '

cn cao

CATHODE WITH SIC

TFE SHIM'

SECTION A-A

f-I TLfLruTJirLrLr

CATHODE SHIM

ANODE SHIM

•) STANDARD SEAL DESIGN

CATHODE PLATE

CATHODE

SIC

CARBON LAYER

ANODE

ANODE PLATE

ANODE BIPOLAR PLATE CATHODE BIPOLAR PLATE

ANODE

B

CATHODE WITH SIC

TFE SHIM7

SECTION B-B

CATHODE SHIM

ANODE SHIM

j\j\i\nj\nj\r

CATHODE PLATE

. CATHODE

SIC

• CARBON LAYER (OMITTED IN E-009-02. CELLS 6-9

ANODE

ANODE PLATE

b) ALTERNATE SEAL DESIGN FOR MPROVED ACD TRANSPORT

Figure 3-55. Cell Seal Design

3-192

WAESD-TR-85-0030

TABLE 3-54

STACK E-009-02 OPEN CIRCUIT VOLTAGES, mV

Operating Conditions: 1 atm, 125°C, H^/Air

BEFORE AFTERCELL NO. MAT-1 ACID ADDITION ACID ADDITION

12

3

4

567

8

9

Avg.

Avg.

YesYesYesYesYesNoNoNo

No

MAT-1 Matrix Cells

SiC Matrix Cells

888878

923

913

886

719

46

137

898

66

896895

937

943

988772

1003895.

903

932

893

Total Acid Addition ^ 65 cc 97%

3-193

WAESD-TR-85-0030

Stack E-009-03 was assembled witfh CePlPH - 5 containing Kureha 604 paperinstead of Vulcan carBda'niay'eri- 'Thi'S rtiaterfal was tested to see if itimproved the lateral transport of acid during operational changes that resultin acid volume expansion. The stack was -tested for : a total of ^ 160 hours,including ^ 90 hours at 480 kPa (70 psia). An average performance of653 my/cell in Cells 1-5 containing Kureha paper in cell matrices and661 mV/cell in-Cells 6-9 containing MAT-1 matrices was observed at 480 kPa(70 psia),< 190°C<(374°F), 325 mA/em2 (300 A/ft2), 83 percent fuel(hydrogen) and 40 percent oxidant (air) utilizations. A reversal of polarityin Cell 3 caused stack testing to be terminated. Kureha paper in the cellmatrices resulted in a slightly lower average cell performance (^ 10mV/cell) as-compared to cells containing MAT-1 matrices. Several otherdifferences: were observed between Cells 1-5 containing the Kureha papermatrices and Cells 6-9 containing MAT-1 matrices:

.' * -"%. . -.' '

• Open- circuit voltages of Cells 1-5 were lower ( 50 mV/cell) than:-,,„,,.. those. ,in Cells 6-9. This may, have been due to internal: short, circuits

or slight gas cross leaks.

• Dtiie'-to the more porous structure (and thus higher electronic t •resistance) of Kureha paper compared to the standard carbon layer, themeasured resistances of Cells 1-5 were ^ 50 percent higher thanthose of Cells 6-9. This higher electronic resistance was notexpected to adversely affect cell performance since ionic resistanceof the matrix has a much more significant effect on performance thandoes electronic.

• SRG loss in Cells 1-5 was 82 mV/cell at the design point compared to aloss of 34 mV/cell for the cells containing MAT-1. Overheating alsooccurred in several cells containing Kureha paper in the matriceswhich was attributed to reactant crossover.

Post-test disassembly showed degradation of the Cell 1 cathode plate near theoxidant inlet and both the Cell 3 anode plate and anode backing paper near thefuel inlet.

3rV94

- . - -.- ' WAESD-TR-85-0030

Based on the limited experience thus far with Kureha paper in cell matrices,Kureha paper is not satisfactory as a barrier to crossleaks. Post-testdisassembly also showed that it does not prevent plate wetting by transportingthe electrolyte laterally.

During upset conditions it is possible to get relatively large changes in acidvolume. There are two methods for dealing with this potential problem:

• allow the acid to expand laterally and collect in the replenishmentchannel (modified matrix tested in Stack E-009-03) .

• allow acid expansion perpendicular to the electrode (tested in StackE-009-01).

An acid inventory control member (AICM) was initially developed and tested onContract DEN3-205. The best component design from that effort wasincorporated into the anode of Cells 6-9 in Stack E-009-01. The basic designof this member is shown in Figure 3-56.

The stack was operated for 530 hours, of which 100 hours were at 480 kPa(70 psia). As shown in Figure 3-57, Cell 6-9 containing AICM anodes hadperformance equivalent to the other cells. Disassembly inspection showed thatone cell with the AICM and one cell without an AICM had cathode platecorrosion. Apparently, the AICM did not prevent this type of corrosion.

The amount of acid required to fill standard or modified components isobtained from analysis of ERC electrolyte take-up (ETU) data and fromWestinghouse calculations. Since components can vary in thickness anddensity, the amount of acid to completely fill them during the assemblyprocedure cannot be ascertained with any significant accuracy. Thisvariability, however, can be accommodated by applying a constant amount duringthe initial assembly and then adding acid through the acid replenishmentchannel in each cell after assembly. The optimum conditions for post-assemblyacid addition remain to be determined.

3-195

WAESD-TR-85-0030

ccQJ

co

o2

*O<ffl

Ooi

UJoc

..uQJ£

S-o

oS 2u. Oo ocg H

ft *111

(J.(OCQ<u13OC<c

O UJUJ •-

UJ Ow o

IT)I.

OO .

QJS_3

-3-196

OF POOR

WAESD-TR-85-0030

wQQ

Wo

oo^oo

o10

ooS-o

fOQ

<UO

o<4-S-<uQ-

(O

toi

CO

O)5-3CD

'TVIlNHlOd 1VNIWH31

3-197

WAESD-TR-85-0030

Analysis of acid replenishment requirements during stack operation has beenlimited by the relatively short periods of operation for each stack and thevariations of the cell components built into the stacks. Each stack was foundto accept acid when it was added. A summary of the acid additions ispresented in Table 3-55. Apparently the cells had not received enough acidbefore operation. Improved procedures for acid addition prior to operationare required for the future.

3.8.8 Impurity Effects

An effort to catalog and understand the level of impurities in variousstarting materials was initiated during the First Logical Unit of Work.Spectrographic analyses showed that as-received Vulcan XC-72 (standardcatalyst support) contained ^ 1 percent sulfur. Because high levels ofsulfur may cause catalyst poisoning and create problems during plant startup,an effort to remove these impurities was undertaken. Various pretreatmentsincluding leaching in phosphoric acid and heat treatments at 900°C, 1800°C,and 2500°C (1650°F, 3300°F, and 4550°F) were examined as shown in Table 3-56.

i

Both free sulfur and total sulfur were measured. Acid leaching removed only asmall amount of sulfur. Similarly, heat treatment to 900°C (1650°F)("standard" heat treatment temperature after catalyzation) did not change thesulfur content. Thus, as expected, sulfur appears to be strongly bound withinthe carbon structure. Further heat treatment to 1800°C (3300°F) and 2500°C(4550°F) did remove ^ 95 percent of the sulfur. It is possible that anintermediate temperature between 900°C (1650°F) and 1800°C (3300°F) may alsoprovide a similar reduction in sulfur levels. However, such heat treatmentsdo change the chemical and physical properties of carbon and may also beexpensive.

The sulfur content of Shawinigan Black, a purer carbon support, was alsoanalyzed. As shown in Table 3-56, the sulfur level is 160 ppm in theas-received material. Heat treatment to 1200°C (2200°F) and 1800°C (3300°F)reduced this level slightly.

3-198

—-• —•- WAESD-TR-85-0030

TABLE 3-55

SUMMARY OF ACID ADDITIONS TO NINE CELL STACKS, cm3

Stack Number

Addition No. E-009-01 E-009-02 E-009-04 E-009-06

1 25 12 10 28

2 - 40 43 -

3 - 20 54

Total 25 72 107 28

3-199

WAESD-TR-85-0030

TABLE 3-56

TOTAL AND FREE SULFUR CONTENTS OF CARBON SUPPORTS

Material Sulfur Content %Total Free

As-received Vulcan XC-72

Hot Phosphoric Acid (190°C)Leached Vulcan XC-72

900°C heat treated VulcanXC-72 (in N£ atmosphere)

1800°C heat treated VulcanXC-72 (in N2 atmosphere)

2500°C.heat treated Vulcan'XC-72 (in N£ atmosphere)

0.93 0.53

0.88 0.27

0.89 0.51

0.025 Not detectable*


As-received Shawinigan

1200°C heat treated(in N2 atmosphere)Shawinigan Acetylene Black

1800°C heat treated(in Ng atmosphere)Shawinigan Acetylene Black




*Less than 40 ppm.

3-200

. _....._ WAESD-TR-85-0030

4.0 ERC NINE CELL STACK TEST FACILITIES DEVELOPMENT

A pressurized test facility to accommodate a 12 x 17 in. nine cell stack wascompleted. This facility improved upon the past ERC facilities in thefollowing areas:

• capable of testing up to a nine cell stack.

• measurement of fuel, oxidant, and cooling flow pressure drops(bringing the total number of AP measurements to six).

• cooling inlet temperature control.

• redundant flow monitoring.

• fine control of electrical load.

• automatic failure mode control system.

t load transients with greater than 3 milliseconds switching time ableto be studied.

A schematic flow diagram for the improved facility is presented in Figure4-1. This facility was designed for the following range of operating

conditions:

Pressure Range: 0 to 1037 kPa (0 to 135 psig), pressurization rateof approximately 13.8 kPa (2 psi) per minute

Standard Gases: Hydrogen fuel, air oxidant, auxiliary inputs forspecial gases

Current Density Range: 50 to 460 mA/cm2 (47 to 430 A/ft2)

4-1

V

» z< oO oUI UI

Si

i

uiQ3

t

X" r | p

I

11

OllrSJvJ§i W£i

, > o

5

u.

WAESD-TR-85-0030

«a-O)

01

U) UJ UJ

UIW<O

4-2

WAESD-TR-85-0030

2Fuel Flow Range: Up to 80 percent utilization at 50 mA/cm

(47 A/ft2) 50 to 100 percent utilization at

460 mA/cm2 (430 A/ft2)

2Oxidant71ow Range: Up to 50 percent utilization at 50 mA/cm

(47 A/ft2) 20 to * 100 percent utilization at460 mA/cm2 (430 A/ft2)

Temperature Range: 30 to 250°C (86 to 480°F), independently controlled

end plate temperatures

Cooling: 100 scfm air, Ambient to 130°C (265°F)*

Unattended operation of the facility is made possible by the use of a FailureMode Control System (FMCS). This system continuously monitors the facilityfor conditions that could have a harmful effect on either the facility or thestack. The FMCS monitors several stack voltages, several temperatures, linepower, and hydrogen concentration in the vessel cover gas. The specificationsfor the FMCS are presented in Table 4-1.

Construction and checkout of the first nine cell stack pressurized test

facility, SE-1, was completed. The checkout of facility SE-1 was conducted by

testing Stack E-009-06 and when completed, the facility was put into service.

Stacks have been tested in this facility up to 825 kPa (120 psia) and400 mA/cm2 (372 A/ft2) using pure hydrogen as fuel with a 55°C (100°F)coolant AT. The verified operating condition ranges are presented inTable 4-2. The facility has accumulated 3,500 hours of testing.

Several recommended modifications to the original design were identified

during the testing of facility SE-1. These changes are outlined below:

*At blower element

4-3

WAESD-TR-85-0030

4-1.SPECIFICATIONS OF THE FMCS FOR 12 X 17 STACK PRESSURIZED TEST FACILITY

Problem

Stack Undervoltage(group of 2 or 3 cells)

Stack Over Temperature

Vessel Over Temperature

in Vessel Gas

Power Failure

Trip Point(Adjustable)

0.5 - 3.0 V

150 - 250°C

100 - 175°C

1% or 4%

Longer than 1 min.

Longer than 30* min.

Action

Remove Load

Remove Load

Remove LoadStop Process FlowsDepressurize System


Remove Load


*Due to loss of backup power and supply air.

4-4

WAESD-TR-85-0030

TABLE 4-2

VERIFIED OPERATING CONDITIONS (USING NINE CELL STACK

IN PRESSURIZED TEST FACILITY SE-1)

OPERATING CONDITIONS

Pressure Range:

Standard Gases:

Current Range:

14.7 to 120 psia(Pressurization Rate of^ 0.5 psi per min.)

\\2 fuel, air oxidantauxiliary inputs forother gases

At 14.7 psia-up to 150mA/cm2

At 70 psia-86 to 400

COMMENTS

Pressure above 120 psia nottested. Pressurized at83 mA/cm2 load, 50% H2utilization 40% oxidantutilization.

Modification for long termSRG planned

Fuel Flow Range: Up to 83% utilization at86 mA/cm2, 50 to 85%utilization at 325 mA/cm2

Oxidant Flow Range: Up to 50% utilization at86 mA/cm2, 40 to 65%utilization at 325 mA/cm2

Stack Temp Range: Ambient to 250°C (inde-pendently controlled foreach end plate)

Cooling AT: ^ 55°C (100°F) at325 mA/cm2 with 120°C(250°F) inlet manifoldtemperature

Coolant AT higher thanthe design value of 100°Fabove 400 mA/cm2. Modi-fication for increasedcoolant flow capacity sub-mitted to Westinghouse toreduce coolant AT from100 to 68°F.

4-5

WAESD-TR-85-0030

• Addition of a flow control/gas^blending system for extended operationon SRG. This system should allow incremental addition of the various

\ .f

component gases to investigate their effect on stack performance.Minor adjustments in gas compositions should also be possible.

• Addition of a temperature controller in the fuel inlet line to achievea more uniform thermal profile and to simulate the temperature of thefuel gas entering the stack from a reformer.

• Increase in cooling loop capacity. The program specification of 55°C(100°F) temperature rise in the cooling air system from inlet manifoldto outlet manifold was revised after the design of test facilitySE-1. The new specification of 38°C (68°F) rise and increasedpressure drop in the cooling plate, along with other factors, willrequire modification of the design to single pass cooling andinstallation of a dedicated compressed air system.

• Single pass cooling design to eliminate the blowers presently in use.The blowers are unreliable for long term continuous operation. Byreplacing them with a dedicated compressor system, overall systemreliability will increase.

• Oil-free compressors are recommended because oil decompositionproducts generated at elevated temperatures in the compression processmay poison fuel cell catalysts. Although filters can be used toremove oil particles, they do not remove vapors and gases. Periodiccompressor shutdowns required to replenish the oil also reduce thecompressor availability and may require unwarranted stack shutdowns.The air in these compressors is thoroughly mixed with the oil and afilter malfunction can introduce oil in the compressed air lines,requiring a long shutdown to decontaminate.

A comparison of the design for recirculation and single pass cooling ispresented in Figure 4-2.

4-6

0 t

z

0 r

ooCOCO<Q.

Ul_J

S L••CO

*•>m

H0u•

-•0 pJZ '-» ô 2O oio «(OZ 0)

^ QC "~ »<

2 -13 k-O 3-_ r ^CC °— *•* i0 LI111OC

^*<

I-au

I1.

cj*M

i

_

cJ»

J

rL(

(

<

L

Q[t

n

j

i—

^

n

ii

j

i i ,

—

-i r*-i

3 8

1 1_l L—J

0 El3 ED

I]j »-

J z3 <- 0

XO

'

•n i— ^

Ul W

3 O0 Ii I

_ —

31^j

uD1

0Oo

M +1111

r~i iS - 1< i§

lvl 1T 1' 1

. 1

[u?|^«J

• 111

111

1 1

5 'rf

§ 1

& 1VVr ' mH | t2h

. 1

WAESD-TR-85-0030

OC OC OCUl Ul Ul

r! H £ H-J — J UJ gr _Jo o Pi b ui oOC OC •" < I- CCH {I H Ul < |lz £ z u i > c c O r > _o 1 ! 0 - ^ ^ «w' ui O »- O 5;-i ^ _i 1- O ^ _i . oU - ? U , Z Z 5 U . Q.*^ *< — ^ OC O , ( 0 ^ - J - J C O S -C O n l C o S O O C O » -< 3 < X O O < - o2 u. 5 O 0 0 2 c

» - o i « ^ « > < D t ^ &(/>QJQ'

CO

4->(O

3OS-

u(U

C£

•of\\w

£ «=T CT

5 aOC </)ÛJ <UQ

0 £ g1

z ^.^b I m 1 a X ^2^^ *** r ^T^ ' ^ ^^ * x o

1 I OC CO °oUl Ul X 0) ...-1 tt _J nj C £i rs ui n ••- «^^^^^_ *j *•• ^J ' fo

m O DC H Q H z a.h° oc J " c t < o c < ^ o ,~T~ H .1- ui W 1- Ul = r-^> z • flc < H x î £ g

° U l x ^ o . O Q . b -O *~ . . 2 o Oo o1"* * ^ U l v J * ^ O t j

5 ill OC > O > O- .n -^J « I CL « -1 > «J10 rh ° i D o O «^|t«.l i m i_ _i m _, f« i

V U - J z l i - z u.z "*1 « < » « ~ ; ~ <uco , ;r • to -i co -J $-

c o j H S c o O z c o O =,1 < S x < o < < o •?z 2 u . O 2 O u . 2 O u -<2 t - t M c o ^ - m c r ^ e oXO

4-7

WAESD-TR-85-0030

Work on modification of the second nine cell stack pressurized test facilitySE-2 was initiated.

Test facility SE-3 is planned for construction in the next logical unit ofwork, After it is completed, facility SE-1 will be modified to the singlepass cooling scheme.

4-8

WAESD-TR-85-0030

5.0 REFERENCES

1. B. Ke, J. Polymer Sci., BI, 1967-70 (1963).

2. B. Wunderllch and E. Hellmuth, Dupont Thermogram, ]_, 5 (1964).

5-1

Reoort No.

NASA CR-175047

2. Government Accession .'.o. 3. Recipient's Catalog No.

4. Title and Subtitle

Gas Cooled Fuel Cell Systems Technology DevelopmentFinal Report for the Second Logical Unit of Work

5. Report Date

August 19866. Performing Organization Code

XAL-78010-AL7. Author(s)

J. M. Feret

8. Performing Organization Report No.

WAESD-TR-85-0030

10. Work Unit No.

9. Performing Organization Name and Address

Westinghouse Electric CorporationAdvanced Energy Systems DivisionP. 0. Box 10864Pittsburgh, PA 15236-0864

11. Contract or Grant No.

DEN3-290

12. Sponsoring Agency Name and Address

U. S. Department of EnergyMorgantown Energy Technology CenterMorgantown. WV 26505

13. Type of Report and Period CoveredContractor ReportMay 1983 - Mav 1QR4

14. Sponsoring Agency Code

DOE/NASA/0290-0115. Supplementary Notes

Prepared under Interagency Agreement DE-AI01Robert B. King, Energy Technology Division,Cleveland, Ohio 44135

-80ET17088. Project Manager,NASA Lewis Research Center,

16. Abstract

This report summarizes the work performed by Westinghouse Electric Corp.and Energy Research Corp. during the second phase of a planned multiphaseprogram to develop a Phosphoric Acid Fuel Cell (PAFC) for Electric UtilityPrototype Power Plant Application. The results of this effort include:.1) establishment of the final system level requirements for the Fuel Cell,Fuel Processing, Power Conditioning, Rotating Equipment, and Instrumentationand Control Systems, 2) advancement of fuel cell technology through innovativeimprovements in the areas of acid management, catalyst selection, electrodematerials, and quality assurance programs, and 3) demonstration of improvedfuel cell stack performance.

17. Key Wo-ds (Suggested by A.uthorls!)

Fuel Cells. Phosphoric Acid. EnergyConversion. Energy Systems. AdvancedPower. Carbon Corrosion. CarbonCompounds. Alloy Catalysts.

18. Distribution Statement

Unclassified - UnlimitedStar Category - 44DOE Category - UC-97d

19. Security Classif. (of this report)

Unclassif ied20. Security Classif. (of '.his page)

Unclassified] ;i. No. of Pares

279

22. Puce"

" Foi sale by the NV,l:onal Technical Intarcation Setvice v--i:ifiio!:! Virginia 22161

NASA-C-1IJ3 (Rw. 10-75)

t. SECURITY CLASSIFICATION:

REPORT SUMMARY FORM

N/A

2. REPORT NO.: WAESD-TR-85-0030

1. PROPRIETARY CLASS:

4. AVAILABILITY: AESD-WM-LIB

5. ORIGINATING SOURCE: AESP-L

6. CONTRACT NO.; DEN3-290

7. FUNDING SOURCE: U. S. Dept. of Energy/NASA Lewis Research Center

a . NOTES : N/A

9. AUTHQR(s); Feret, J. M.

10. TITLE: GAS COOLED FUEL CELL SYSTEMS TECHNOLOGY DEVELOPMENT

FINAL RFPORT FOR THE SECOND LOGICAL UNIT OF WORK

i i . DATE - - YYMMDD: "860800

12. PAGES. REF., 'ILLUS.: 279 P. 2 Ref. 69 Illus.

13. DOCUMENT TYPE: ReDOft

i 4 . .KEYWORDS : ' •• FUEL CELLS. PHOSPHORIC ACID. ENERGY CONVERSION. ENERGY

SYSTEMS. ADVANCED POWER. CARBON CORROSION. CARBON COMPOUNDS'. ALLOY CATALYSTS

IS. ABSTRACT- - PURPOSE. SCOPE, APPROACH. RESULTS, CONCLUSIONS. SIGNIFICANCE:( MAXIMUM: 200 WORDS)This report summarizes the work performed by Westinghouse Electric Corp.and Energy Research Corp. during the second phasp nf-a planned multiphaseprogram to develop a Phosphoric Acid Fuel Cell (PAFC) for Electric UtilityPrototype Power Plant Application. The results of this effort include:1) establishment of the final system level requirements for the Fuel Cell,Fppl Processing, Pnwer Conditioningt Rotating Equipment, and Instrumentationand Control Systems, 2) advancement of fuel cell technology through innovativejmpf nypp^pirl"^ •)'p tjh*? ^T1?^? 0^ ari'H man^nprnpirf rat'.flTy^t' ^plprf'.inn p1pr.f~.rnHp

materials, and quality assurance programs, and 3) demonstration of improvedfuel cell stack performance- — - —

16. THIS WAESD-TN SHOULD BE DESTROYED AFTER MONTH YEAR.

FOB* AESD-OOI30A

DOCUMENT DISTRIBUTION LIST

INTERNAL AESD

NAME, DEPT. LOCATION

See Fuel CellDistribution List

FULL

TEXT*

RSF

ONLY+

COPY

NO.*

EXTERNAL AESD

NAME AND ADDRESS

R, B, KINGPROJECT MANAGERMAIL STOP 301-5NASA LEWIS -RESEARCH CENTE21000 BROOKPARK RD,CLEVELAND, OHIO 44135

NO. OF

COPIES

*

COPY

N0.«

* INDICATE NUMBER OF COPIES OF FULL TEXT

4- REPORT SUMMARY FORM ONLY

• CONTROLLED DISTRIBUTION, IF APPLICABLE

FORM NO. AESD-00IJOB

DOE/NASA/0290-02 NASA CR-175047 · DOE/NASA/0290-02 NASA CR-175047 WAESD-TR-85-0030 FINAL REPORT...

Documents

Transcript of DOE/NASA/0290-02 NASA CR-175047 · DOE/NASA/0290-02 NASA CR-175047 WAESD-TR-85-0030 FINAL REPORT...