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