Effects of Fuel and Air Impurities on PEM Fuel Cell Performance

Fernando GarzonLos Alamos National Lab

5/16/2007FCP#17

This presentation does not contain any proprietary, confidential, or otherwise restricted information

2

Technical Staff

Los AlamosRod BorupEric BroshaJohn DaveyFernando GarzonBryan PivovarTommy RockwardTom SpringerFrancisco UribeIdoia UrdampilletaJudith Valerio

Case Western Reserve University

Thomas ZawodzinskiBrian Kienitz

3

Overview

• Project start -FY07• Status- ongoing

• The cost of fuel cells limits their use– Fuel and air impurity removal systems add

cost

– Higher Pt loading required to maintain performance in the presence of impurities increases cost

• Durability may decrease in the presence of impurities

• Fuel cell performance is decreased by impurity effects

Timeline

Budget

• Funding in FY06: $800 K• Funding for FY07: $1200 K• Non-cost shared

Barriers

• USFCC• ASME• ASTM• SAE • ISO • FCTESQA

• OEMs• Xradia Corp

Partners

4

ObjectivesOverall Objective: Contribute to the understanding of the effects of fuel and air impurities on fuel cell performance

Specific Objectives:•Test fuel cell performance under simulated multi-component hydrogen impurity gas mixtures

•Investigate effects of impurities on catalysts and other FC components

•Understand the effect of catalyst loadings on impurity tolerance

• Investigate the impacts of impurities on catalyst durability

•Develop methods to mitigate negative effects of impurities

•Develop models of fuel cell-impurity interactions

•Collaboration with USFCC, Fuel Cell Tech Team, Industry and other National Laboratories to foster a better understanding of impurity effects

5

Approach

• Impurities effect fuel cells in many ways:– Electrocatalyst poisoning e.g. H2S, CO and SO2

adsorption onto Pt catalysts– Reduce ionomer conductivity- Na+, Ca++, NH3– GDLs become hydrophilic and flood at high current

densities

Electrode kinetics

Ionic transport

Mass transport

Properties of gas diffusion layer (GDL)

(Cl- hydrophobicity changes)

Protonic conductivity of catalyst layer and membrane

(H+ exchange by other cations)

Electrode catalyst activity(S, CO poisoning)

Impurityand

Effect

6

Approach• Fabricate and operate fuel cells under

controlled impurity gases– Multi-gas mixing manifolds and FC

test stations– Pre-blend impurity gases– Measure performance– Steady state and Drive cycle

conditions• Understand degradation

mechanisms• Study mitigation approaches

• Develop analytical tools for studies– Electroanalytical methods – In situ diagnostics– Sub PPM gas analysis

• Analyze and model data

7

Technical Accomplishments/ Progress/Results

• Hydrogen impurity mixture used to characterize fuel cell performance for different anode loadings– 2007 and 2010 anode Pt loadings characterized. Milestone

accomplished– Little effect of Pt catalyst loading on impurity performance degradation

• Experimental evidence for hydrogen sulfide crossover from anode to cathode as a cathode poisoning mechanism

• DOE drive cycle durability in the presence of impurities measurements commenced

• New high resolution X-ray CT (Computed Tomography) method develop for studying impurity impacts on durability

• Particulate effects studies commenced• Sulfur adsorption on Pt electrode modeling commenced• Cation impurity modeling commenced with Case Western Reserve

University• Wet electrochemical cell experiments validate fuel cell results

8

H2S Effects in Fuel Cells

• Past results show relative large performance losses due to ppb quantities of H2S introduced from hydrogen fuel.

– Stripping voltammetry indicated partial coverage of the anode~ 40% of the sites for 10 ppb exposure 1000 hrs

– Not enough sites blocked to account for >10% performance drop

– Poisoning the cells at low operating voltages resulted in much more performance loss than at high cell voltages

• Is impurity crossover to cathode also responsible for performance loss?

– Cathode electrocatalyst kinetics are much more sensitive to site blocking than anodes due to lower reaction rates.

• Experiments designed to study the possibility of hydrogen sulfide crossover effects

Cell performance recovery (at 0.5 V) afteranode exposure to H2S at different voltages

5 cm2 cell

9

H2S Crossover Studies

• H2 /H2S mixture was passed through the anode. The cathode was purged and then the CV was run

• CV indicates a strong adsorbate– Not stripped until high potentials

achieved

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

0 0.2 0.4 0.6 0.8 1 1.2 1.4

0.5V_Cathode CV_clean (4th cycle)0.5V_Cathode CV_after cathode poisoning (1st cycle)

Voltage (V)C

urre

nt (A

)

Cathode poisoning with 1ppm of H2S for 2.5h

-0.2

0

0.2

0.4

0.6

0.8

0 0.2 0.4 0.6 0.8 1 1.2 1.4

TF673, 50cm2Anode poisoning with 2ppm of H2S for 2h

0.5V_Cathode CV_after anode poisoning (1st cycle)0.5V_Cathode CV_clean (4th cycle)

Voltage (V)

Cur

rent

(A)

• Air/ H2S mixture was passed through the cathode. The cathode was purged and then the CV was run

• CV indicates a similar strong adsorbate

10

Impurity Effects: Electrochemical Cell Studies

• Pt anodes, Pt metal and Pt 20%- XC 72 were pre-poisoned with sodium sulfide, then placed in sulfuric acid cells

• Severe reduction in hydrogen adsorption sites by S poisoning for both cases• Note the more pronounced poisoning of the supported fuel cell catalyst than the bulk Pt

metal catalyst– Demonstrates the importance of studying the behavior of the actual carbon supported fuel cell

catalyst

-4

-2

0

2

4

6

8

0 0.2 0.4 0.6 0.8 1 1.2 1.4

Baseline5mM Na

2S

2.5mM Na2S

1.25mM Na2S

0.6mM Na2S

Voltage / V

Cur

rent

/ μA

1st cycle

-40

-20

0

20

40

60

80

100

0 0.2 0.4 0.6 0.8 1 1.2 1.4

Baseline0.6mM Na

2S

1.25mM Na2S

2.5mM Na2S

5mM Na2S

Cur

rent

/ μA

Voltage / V

Pt metal Pt -XC72

11

Hydrogen Impurity Mixture

Component Level LANL TestHydrogen > 99.9 95-99 *Sulfur (as H2S) 10 ppb 10 ppbCO 0.1 ppm 0.1 ppmCO2 5 ppm 5 ppmNH3 1 ppm 1 ppmNMHC 100 ppm 50 ppm ethyleneParticulates Conform to ISO

14687not included in

first test

* Includes dilution due to inert gas in stock mixtures

FreedomCAR Fuel Cell Tech Team proposed hydrogen impurity spec.

12

Impurity Effects: Low Pt loadings

• Similar losses for all Pt loadings• Impurities caused 100 mV performance loss after 800 hrs• H2S partial poisoning detected at the anode by CV• Membrane conductivity also affected as indicated by increase of HFR

0.5

0.6

0.7

0.8

0.9

1

30

35

40

45

50

0 100 200 300 400 500 600 700 800

0.2 mg Pt/cm2 (neat H2)

0.2 mg Pt/cm2

0.1 mg Pt/cm2

0.06 mg Pt/cm2

current

Time / hour

Voltage

Normalized Voltage losses at 0.8 A/cm2

Frac

tion

of in

itial

per

form

ance

-1

-0.5

0

0.5

1

0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45

0.20 mg Pt/cm2 (neat H2)

0.20 mg Pt/cm2

0.10 mg Pt/cm2

Cur

rent

(A)

Voltage (V)

Anode CV after 800+ hrs with impurities mixture

13

Impurity Impacts On Durability

• Impurities may impact durability of PEMFC fuel cells

• Electrocatalyst growth may be accelerated

• Ionomer lifetimes may decrease

• GDL properties may change faster in the presence of impurities

• DOE drive cycles with and without start and stop used to test durability under non steady-state conditions

14

Impurity Effects Durability Tests Commenced

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.0 10.0 20.0 30.0 40.0 50.0 60.0Current / Amp

0.000

0.002

0.004

0.006

0.008

0.010

0.012

0.014

0.016

0.018

0.020

HFR

/ O

hnm

s

voltage (V)HFR (ohm)

0.60

0.61

0.62

0.63

0.64

0.65

0.66

0.67

0.68

0.69

0.70

0 50 100 150 200 250Time / hr

0.000

0.001

0.002

0.003

0.004

0.005

0.006

0.007

0.008

0.009

0.010

HFR

/ O

hms

Watt-hrs / cycleHFR (Ohms)

Started 10 ppb H2S injection

•Baseline established and test started•Initial anode CV for drive-cycle test

•Initial polarization curve for drive-cycle test

•Drive Cycle testing started•10 ppb H2S used as impurity for first tests•Baseline degradation rate established

15

Cathode impurities: Particulates

To date we have:-Successfully produced particulates under 10 microns-Modified existing fuel cell hardware to inject particulate matter into a pressurized anode feed

16

XRadiaXRadia nanoXCT 8nanoXCT 8--50Z Laboratory System 50Z Laboratory System For Imaging FC For Imaging FC MEAsMEAs

• 8 keV Cu target x-ray source• 3-D resolution: 50 x 50 x 80 nm• Automated x-ray tomography• Negative Zernike phase-contrast for

imaging low contrast samples• Samples run in air

17

HighHigh--resolution Xresolution X--ray Imaging for Impurityray Imaging for Impurity--Durability StudiesDurability Studies

• Xradia Corp-LANL collaboration• Minimal sample prep needed for CT scans

•FC membrane went through DOE Drive cycle test until failure• Large Pt particles (a few hundred of nanometers) were formed in the electrode.• Smaller Pt particles were found in the membrane near the membrane-electrode interface• Non-uniform distribution in the membrane•Their size appeared larger near interface than further into the membrane•Tool will be used to characterize impurity-durability samplesMembrane Electrode

Interface

Gold particles

This region has high concentration coarse (large) Pt particles

High concentration but fine Pt particles

Text

ure

diffe

renc

e

18

Impurity Effects Modeling- Electrodes• Surface/speciation model in development

– Modification of USGS Parkhurst PHREEQE codes• Predominate sulfur species are H2S, S-Pt, PtS, PtS2 and

HSO4-• Predicts decreased stability of Pt nanoparticles to S

chemisorption as compared to bulk Pt• Predicts Pt sulfur coverage at -0.15 volts with increasing

coverage as anode potential is raised for 1 ppb H2S• Predicts that the oxidation cleaning mechanism is

inhibited by kinetics not thermodynamics

• Surface speciation model will be coupled to fuel cell electro-kinetics model

Sθ - Pt + 4H2O = HSO4 - + Pt + 7H + + 6e -

19

Future Work• Continued contaminant crossover studies:• Fundamental electrokinetic measurements of poisoned

electrodes• Lower cathode loading impurity studies• Impurity effects on durability studies

– humidity dependence• Refine and validation of electrode impurity modeling

efforts• Salt impurity modeling commencing• Development of impurity tolerant electrode materials• Future key milestones

– Impurity effects on durability studies

20

Summary

• Results– Decreasing the fuel cell anode loading is not having a great impact on

the performance degradation behavior of PEMFCs. – Sulfur species adsorb very strongly on Pt for a wide range of potentials

and concentrations– Crossover effects of impurities need to be considered. Hydrogen sulfide

crossover from anode to cathode may be occurring– Oxygen reduction at the cathodes is easily affected by impurities– Carbon particulates will negatively impact GDL properties

• Contact: Fernando Garzon, Los Alamos National Laboratory garzon@lanl.gov

•Relevance•Trace impurities do impact fuel cell performance and degradationand cost

•Approach•Expose fuel cells to common fuel and air impurities and measure the impact on performance and durability