Fuel Cell Program

2004 Hydrogen and Fuel Cells Merit Review Meeting

Philadelphia Pa, May 24-27

Rod Borup

Los Alamos National Laboratory

Stack Durability on Hydrogen and Reformate

Michael Inbody

John Davey

David Wood

Fernando Garzon

FY2004: Funding: $900k

This presentation does not contain any

proprietary or confidential information.

Troy Semelsberger

Jose Tafoya

Kirk Weisbrod

Eric Brosha

Susan Pacheco

Dennis Guidry

Jian Xie

Francisco Uribe

Fuel Cell Program

• Identify and quantify factors that limit PEMFC Durability• Measure property changes in fuel cell components during long term testing

• Membrane-electrode durability

• Electrocatalyst activity and stability

• Gas diffusion media hydrophobicity

• Bipolar plate materials and corrosion products

• Develop and apply methods for accelerated and off-line testing

• Improve durability

• Component Technical Barriers Addressed:• Durability (Barrier P)

• Electrode Performance (Barrier Q)

• Stack Material & Manufacturing Cost (Barrier O)

• DOE Technical Target for Fuel Cell Stack System (2010)• Durability 5000 hours

• Precious metal loading (0.2 g/rated kW)

• Survivability (includes thermal cycling and realistic driving cycles)

Technical Objectives:Quantify and Improve PEM Fuel Cell Durability

Fuel Cell Program

Approach to Durability Studies

• PEM fuel cell durability testing

• 5 cm2, 50 cm2 and full size active area (200 cm2) / 12 cell stack

• Testing: simulated vehicle drive cycle and steady-state testing

• VIR / cell impedance

• catalyst active area

• effluent water analysis

• in situ and post-characterization of membranes, catalysts, GLDs

• SEM/EDAX / XRF / XRD / TEM / ICP-MS / neutron

scattering / H2 adsorption

• Develop and test with off-line and accelerated testing techniques

• Potential sweep methods

• Environmental/leachate chamber

• Corrosion tests

Fuel Cell Program

Fuel Cell Durability Testing Timeline

Project initiated in 2001 as Fuel Cell Stack Durability on Gasoline ReformateBeginning FY2004 concentration on PEM H2 Durability

2001 2002 2003 2004 2005

Start Modular

gasoline

fuel

processor

1000 hrs

testing on

gasoline FP

3500 hrs

PEM cell

S.S

PEM

drive

cycle

Off-line

testing

Reformate

impurity

analysis

Carbon

formation

during

start-up

FC

effluent

analysis

in situ XRD

real-time

particle size

analysis

correlate off-line

accelerated tests

to PEM tests

2004 MilestonesDec 03 Complete water analysis of impurities developed during testing.

Nov 03 Incorporate drive cycle into durability testing.

Jan 04 Initiate off-line durability accelerated testing procedure.

Jan 04 Incorporate Teledyne Stack into H2 durability testing.

Fuel Cell Program

Response to Reviewer Comments at

2003 DOE Review Meeting

Stack Durability on Hydrogen and Reformate and

Testing of Fuels in Fuel Cell Reformers2003 presentation concentrated on Fuel Effects on Fuel Reforming, somost comments not applicable

- Redirected to work on H2 PEM durability

Reviewer comments relevant after redirection:• The durability objective of this project is very important and I hope it will be

actively addressed.

• I especially like the proposal of operating the system in a duty cycle

operating mode.

• Introduction of drive cycle dynamics and start-up for next year is a plus …

• Need more fundamental work.

Fuel Cell Program

1000 hr Steady-State Test (5 cm2)

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

0 100 200 300 400 500 600 700 800 900 1000 1100

Time (hr)

Vo

lta

ge

(V

) /

Cu

rre

nt

De

ns

ity

(A/c

m2)

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.10

HF

Re

sis

tan

ce

(O

hm

*cm2

)

Voltage

Current Density

HFR

Initial Voltage (Linear Fit) = 0.642 V;

Linear Voltage Decay Rate = 53.9 ?V/hr;

= 57.7 ? W/cm2/hr @ 1.07 A/cm2

HFR Slope = -3.1

?? *cm2/hr

Cell Temp. = 80°C

Anode/Cathode Bubbler Temp. = 105/80°C

Anode/Cathode Inlet Relative Humidity = 255%/100%

Anode/Cathode Gas Pressure = 30/30 psig

Anode/Cathode Stoich. = 3.6/5.9 (133/550 sccm)

Voltage

HFR

Current Density

Constant current

Temperature = 80 oC

MEA geometric active area = 5.0 cm2

Anode catalyst: 20% Pt/C

Cathode catalyst: 20% Pt3Cr/C

Loadings of 0.20 ± 0.01 mg Pt/cm2

N112 membrane.

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.00 0.25 0.50 0.75 1.00 1.25 1.50Current Density (A/cm 2)

Vo

ltag

e (

V)

0 hr (Initial)

310 hr

504 hr

698 hr

905 hr

1000 hr

Cell Temp. = 80°C

Anode/Cathode humidifier Temp. = 105/80°C

Anode/Cathode Gas Pressure = 30/30 psig

Anode/Cathode Stoich. = 3.6/5.9 (133/550 sccm)

Comparison of

Polarization Data

During MEA 1000-

hr Durability Test

Fuel Cell Program

Analysis of Steady-State 1000-hr Test

-90

-70

-50

-30

-10

10

30

50

70

90

110

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

Voltage (V)

Curr

ent (m

A)

0 hr (Initial)102 hr203 hr310 hr399 hr504 hr599 hr698 hr809 hr905 hr1000 hr

CV Set Points:

Anode H2 Flow Rate = 133 sccm

Cathode N2 Flow Rate = 550 sccm

100 mV/s, 1 A/V, 20 data points/s

Cell Temp. = 80°C

Anode/Cath. humidifier Temp. = 105/80°C

Anode/Cathode Gas Pressure = 30/30 psig

Anodic

Cathodic

• During 1000-hr steady-state constant

current durability test

• Catalyst surface area decreases

• Hydrogen cross-over increases

10

14

18

22

26

30

34

0 100 200 300 400 500 600 700 800 900 1000Time (hr)

Tru

e P

t E

lectr

ocata

lyti

c S

urf

ace

Are

a (

m2/g

Pt)

Q_Desorption

Q_Adsorption

Average

Linear (Average)

CV Set Points:

100 mV/s

Cell Temp. = 80°C

Anode/Cath. Humid. Temp. = 105/80°C

Anode/Cathode Gas Pressure = 30/30 psig

Average Rate of Loss of True Pt

Electrochemically Active

Surface Area = 71.4 cm2/g-Pt/hr

Cathode catalyst layer

0

10

20

30

40

50

60

70

80

0 200 400 600 800 1000

Time / hr

Cro

ss-o

ver

Curr

ent / m

A

H2 Adsorption-Desorption CV Scans

Hydrogen cross-over Current

Fuel Cell Program

X-ray Maps of Tested MEA (Cathode)(Steady State Testing for ~ 1000 hrs)

Z-contrast Platinum

SulfurFluorine

• After life test, a layer approximately

50-100nm thick develops at the

interface of membrane and cathode

catalyst layer

• This layer is enriched in S and

depleted in F with respect to the rest

of the membrane

• The fresh MEA had a uniform S

and F composition across the

membrane/anode interface

Fuel Cell Program

3500 hrs Life Tests (50 cm2)

MEA1 shows little/no performance degradation (till crossover starts)

MEA2 shows gradual performance degradation

cross-over developed in both MEAs at about 3000 hours

MEA1 Degradation:

~ 0 microamps / hr - (for 3000 hrs)

MEA2 Degradation:

~ 2 microamps / hr - (for 3000 hrs)

Constant Voltage: 0.6 V

Pt/Pt: 0.2 mg/cm2

N112

Cell Temp. = 80°C

Anode/Cath Humid. Temp = 105/80 oC

Anode/Cath Gas Press. = 15/15 psig

Surface area ReductionMEA1:Anode: 0%Cathode: 14%MEA2Anode: 75%Cathode: 86 %Particle size same-0.6

-0.5

-0.4

-0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

0.4

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

Voltage / V

Cu

rre

nt

/ A

mp

s

Current - Initial Hydrogen Desorption

Current - Final Hydrogen Desorption

-0.4

-0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

0.4

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

Voltage / VC

urr

en

t

Current - Initial Hydrogen Desorption

Current - Final Hydrogen Desorption

0.3

0.4

0.5

0.6

0.7

0.8

0 500 1000 1500 2000 2500 3000 3500 4000

Time / hours

Cu

rre

nt

de

nsity /

Am

ps/c

m2

MEA 1 Current

MEA 2 Current

MEA1 MEA2

Fuel Cell Program

Fuel Cell Drive Cycle Testing

Voltage control profile:

Volt vs. Time (sec)

Power control profile

and

Power response profile

Watts vs. Time (sec)

1 cycle occurs over 20 minutes• Drive cycle ‘controls’ power

• Uses fuel cell VIR to calculate voltage for a power level• Actively controls voltage to get power from VIR

• Current hardware with Labview control• 50 cm2 single cell, Pt/Pt: 0.2 mg/cm2, N112, Cell Temp. = 80°C• constant humidification and constant anode/cathode flowrates

Fuel Cell Program

Initial/Final Drive Cycle Comparison

0

5

10

15

20

25

120284 120384 120484 120584 120684 120784 120884 120984 121084 121184 121284

Time / sec

Pow

er

/ W

att

Command Power

Read Power

0

5

10

15

20

25

916764 916964 917164 917364 917564 917764

Time / sec

Pow

er

/ w

att

Command Power

Read Power

~100 cycle

~ 3500 cycle

Blue is Control Power Cycle

Red is MEA Power Response

Power per cycle over 1200 hrs

Test

Stand

Shut-down

Reduction in H2 adsorption after testing:

Anode: 31%

Cathode: 57%

Fuel Cell Program

Fuel Cell Water Effluent Analysis(S.S. constant current testing / Pt/PtCr 5 cm2)

0

100

200

300

400

500

600

700

82 196 285 389 490

Run Time (hr)

Part

s p

er

Billio

n (

ng

/g)

S32 (ppb)

Pt195 (ppb*1000)

Cr52 (ppb*1000)

Na23 (ppb)

Al27 (ppb*10)

Si28 (ppb)

Ca44 (ppb)

0

50

100

150

200

250

300

350

400

450

0 100 200 300 400 500 600 700 800 900 1000

Run Time (hr)

Co

nc

en

tra

tio

n (

pp

b)

MEA S5, F- Conc.

MEA S5, (SO4)-2 Conc.

MEA S4, F- Conc.

MEA S4, (SO4)-2 Conc.

(NO 3)- conc. zero.

Sharp Increase

of ~2x in F-

Water supplied measured zero for

F- and 101 ppb for (SO 4)-2.

Cathode Effluent

F- and SO4-2 Species Concentrations:

Change in concentration of fluoride (F-) and

sulfate (SO4-2) anions

Sharp increase in F- may coincide with cross-

over formation

Change in pH also corresponds with increased

crossover

ICP-MS Analysis of Cathode Outlet

Water through ~500 hr

4.0

4.5

5.0

5.5

6.0

6.5

Blank

A

Blank

D

102 203 310 399 504 599 698 809 905 1000

Cell Run Time (hr) @ 1.07 A/cm 2

pH

Cell Temp. = 80°C

Anode/Cathode Bubbler Temp. = 105/80°C

Anode/Cathode Inlet Relative Humidity = 255%/100%

Anode/Cathode Gas Pressure = 30/30 psig

Anode/Cathode Stoich. = 3.6/5.9 (133/550 sccm)

Cathode Outlet Effluent pH

Fuel Cell Program

Off-line Testing:

MEA Potential Cycling

• Voltage cycling 0.1 V to 1.0, 1.2 V

• Tcell = 80 oC

• Anode humidifier = 105 oC

• Cathode humidifier = 80 oC

•Obtain predictive, accelerated life test of PEMFC MEA, electrocatalysts.

Within several hundred potential cycles of the MEA electrode,

electrocatalyst surface area is decreased, as is MEA performance

Characterizing CVs (@ 100 mV sec-1)

Voltage (mV)

0 200 400 600 800 1000

Ct (

A)

-200

-150

-100

-50

0

50

100

prior to CV cycling

after 300 CVs

after 600 CVs

after 900 CVs

after 1200 CVs

after 1500 CVs 0

20

40

60

80

100

120

0 500 1000 1500 2000

Potential cycles

Ele

ctr

ocata

lyst

Su

rface A

rea

/ m

Co

ul

60 C

80 C

Fuel Cell Program

Potential Cycling of MEAs

XRD: Pt crystallite sizeANODE: 4.8 nm

CATHODE: 7.6 nm

80ºC VIRsFuel Cell JD033004

Current Density (mA cm-2

)

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8

Vlt

()

0.0

0.2

0.4

0.6

0.8

1.0

1.2

prior to CV cycling

after 300 CVs

after 600 CVs

after 900 CVs

after 1200 CVs

after 1500 CVs

60ºC VIRsFuel Cell JD033004

Current Density (mA cm-2

)

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8

Vlt

()

0.0

0.2

0.4

0.6

0.8

1.0

1.2

prior to CV cycling

after 300 CVs

after 600 CVs

after 900 CVs

after 1200 CVs

after 1500 CVs

XRD: Pt crystallite sizeANODE: 2.3 nm

CATHODE: 7.4 nm

Fuel Cell Program

New Pt catalyst

Cathode catalyst

Anode catalyst

Electrocatalyst Size GrowthXRD analysis of electrocatalysts

0

1

2

3

4

5

6

7

8

Fresh

Catalyst

Prepared

MEA

900 hr

Steady

State

3500 hr

Steady

State

1200 hr

Drive

Cycle

1.2 V

Cycling at

60 C

1.2 V

Cycling at

80 C

Pt

Pa

rtic

le S

ize

/ n

m

• Electrocatalyst particle growth

•Z with time

•Z with drive cycle

•Z with potential cycling

•Z Temperature

Fuel Cell Program

Off-line Testing:

Enviromental / Leachate Chamber

• Simulate PEMFC operating conditions

• Temperature

• Chemical environment

• Isolation of components and separation of degradation effects

• GDL, MEA, bipolar plates, gaskets, electrocatalysts

• Obtain predictive, accelerated life test for prospective individual components.

• Correlate PEMFC effluent water with components found in the off-line testing

80

90

100

110

120

130

140

150

0 100 200 300 400 500

Time / hr

Co

nta

ct

An

gle

60 C

80 C

Linear (60 C)

Linear (80 C )

Gas Diffusion Media

Change in contact angle

shows decreasing relative

hydrophobicity

Inert atmosphere, DI water

Fuel Cell Program

Bipolar Plate Corrosion Test Cell

Bipolar Plate

Sample

Reference

Electrodes

Aluminum Plug

Teflon Faced

Polycarbonate

Blocks

Carbon

Backing

Layer

Pt Screen

Pt on

Carbon

Heater

H2

SupplyAir

Supply0.001 N H2SO4

3 ppm Fluoride

FUNCTION• Simulates the bipolar plate environment (Temperature, anode and cathode potentials and acidity)• Provides in-situ indication of contact resistance changes arising from corrosion film growth• Electrolyte samples indicate production of soluble ions.

STATUS• Developed in 1999 to2000 with DOE funding• Patented in 2002• Tested candidate bipolarplate materials for MikeBrady (ORNL)• Loaned, licensed cells toBallard (2001 to 2003).• Technology available forlicensing

Fuel Cell Program

Interactions/Collaborations

• National Technical Presentations/Publications

– Fuel Cell Seminar, ECS, JECS submission

• Fuel Cell Materials

– MEAs (3M, Gore, LANL)

– GDLs (Spectracorp, Toray, SGL, ETEK)

• Stack: Teledyne Energy Systems

• Characterization

– ORNL (Douglas Blom and Karren More)

– UNM (Plamen Atanassov)

– LANL - NMT Division (Dave Wayne), C Division (Pat Martinez),

LANSCE (Jaroslaw Majewski)

• Drive Cycle NREL (Tony Markel)

Fuel Cell Program

Project Safety

Management Safety Controls:Hazard Control Plan (HCP) - Hazard based safety review

Integrated Work Document (IWD) - Task based safety review

Integrated Safety Management (ISM)Define work Analyze Hazards Develop Controls Perform Work Ensure Performance

Engineering Controls:Hydrogen and carbon monoxide room sensors

Electrically and computer interlocked with the test stand power, the gas supplies

H2 sets off the CO sensors, (set at 30 ppm)

Limits H2 far from the explosive limit

Safety Related Lessons

There have been no safety related incidents ( & related projects).

Use of gas sensors, test stand interlocks limit hydrogen hazards.

Fuel Cell Program

• Steady-state and drive cycle testing of MEAs• MEA degradation quicker with drive cycle testing compared with S.S. testing

• H2 cross-over increases with time for both S.S. and cycling

• Electrocatalyst active surface area decreases

• Platinum particle size growth observed

• higher particle growth with cycling, time

• Change in conc. of fluoride (F-), sulfate (SO4-2) anions, pH

• coincides with increased cross-over (‘hole’) formation

•A layer 50-100nm thick developed at the cathode/membrane interface

• Layer is enriched in S and depleted in F in comparison to the membrane

• Off-line (accelerated) degradation techniques• High catalyst sintering during potential sweeps to high potentials

• Temperature effect on anode catalyst sintering

• GDL hydrophocity shows little change in DI water

• Neutron scattering shows promise for delineating PTFE/Nafion degradation

• Corrosion cell for bipolar plate testing

Summary/Findings

Fuel Cell Program

Remainder of FY 2004:

– correlate potential cycling tests to drive cycle testing

– correlate increase in F- and SO4-2 with cross-over in membrane

FY 2005:

• Membrane / MEA degradation

– examine Nafion bonding via neutron scattering

– simulate membrane cross-over by inducing penetrations

• Gas Diffusion Media

– continue off-line testing determining hydrophobicity degradation

– determine PTFE/graphite (GDL) bonding interaction changes

• Catalyst Durability / characterization

– examine some Pt alloys for particle size growth

– in situ XRD real-time particle size analysis during simulated fuel cell

operation

Future Plans