In-situ X-ray Spectroscopy and Scattering Diagnostic Studies of

PEFC Cathode Catalysts

D. Myers, M. Smith, A.J. Kropf, M. Ferrandon, and J. GilbertArgonne National Laboratory, Argonne, Illinois, United States

G. Wu, J. Chlistunoff, C. Johnston, and P. ZelenayLos Alamos National Laboratory, Los Alamos, New Mexico, United States

DIAGNOSTIC TOOLS FOR FUEL CELL TECHNOLOGIES

Trondheim, NorwayJune 23-24, 2009

2

Why don’t we have “two fuel cell cars in every garage”? Major hurdles to overcome

– Cost• 50% of cost of PEFC stack is due to Pt catalyst*

– Durability• Pt and Pt alloy cathode electrocatalysts lose

electrochemically-active surface area with time– Fuel storage, availability, and delivery

How can we get there?– Materials and engineering advances

• better utilization/performance• lower cost (e.g., PGM alternatives)

– Fundamental studies of materials• how they work• what limits their performance

AnodeAnode CathodeCathode

N2N2

N2N2

N2N2

H2H2

H2H2

H2H2

H2H2

H2H2

H2H2

O2O2

O2O2

O2O2H+H+

e-e- e-e-

O2O2N2N2

N2N2O2O2

O2O2

ElectrolyteElectrolyte

*2007 Status, Directed Technologies Incorporated Study, Feb. 2008

3

How can we get the necessary information? What’s needed for rational design of

catalysts: identity of active site; relationship between structure and degradation

Must “see” inside the fuel cell while it’s running with 0.1-10 nm “vision”

Probe must penetrate through flow field, gas diffusion layer, and ionomer to characterize catalyst on the atomic level

X-rays can penetrate through low atomic number materials and have wavelengths on the order of atomic dimensions

Synchrotron X-ray sources (high intensity, tunable wavelength), such as Argonne’s Advanced Photon Source, give us “X-ray vision”

4

X-ray Absorption Fine Structure (XAFS)

h

Oxidation state of absorbing atom Distances between atoms Number of neighboring atoms Identity of neighboring atoms Amount of absorbing material in

beam

5

Small-Angle X-ray Scattering (SAXS)

Gives information on particles 1 - 100 nm in size

Shape Mean Size Size Distribution

6

Examples of systems studied with in-situ and ex-situ X-ray techniques

Pt-based electrocatalyst degradation– Oxidation state and correlation of loss of Pt with voltage

• X-ray absorption in an aqueous environment– Oxide formation and Pt particle growth as a function of potential

cycling• Small angle X-ray scattering and anomalous small angle X-ray

scattering• Aqueous environment and MEA

Non-platinum group metal catalyst composition, structure, oxidation state, and amount of absorbing metal using X-ray absorption– During pyrolysis– Effect of post-pyrolysis acid treatment– As a function of potential in aqueous environment– In MEA during polarization

7

Cells for in situ X-ray studies of cathode catalysts

It

X-ray

I0

FluorescenceDetector If

APS

Wor

king

Ref

eren

ce

Cou

nter

Potentiostat

In SituElectrochemical

Cell

ItIt

X-ray

I0

FluorescenceDetector If

APS

Wor

king

Ref

eren

ce

Cou

nter

Potentiostat

In SituElectrochemical

Cell

300 m thick window machined over three channels of single serpentine flow field* (modified Fuel Cell Technologies Hardware)

*Based on published design: Principi, E.; Di Cicco, A.; Witkowski, A.; Marassi R. J. Synchrotron Rad., 2007, 14, 276.

8

Aqueous in-situ XAFS shows potential dependence of Pt loss and Pt oxidation state

10 mV/s

2 XAFSSCANS

10 mV/s

OpenCircuit

0.8 V

1.4 V

1.1 V

0.5 V

1.1 V

0.8 V 0.8 V

1.4 V

1.1 V

0.5 V

1.1 V

0.8 V

Potential cycling Š 1st cycle Potential cycling Š 2nd and subsequent cycles

Height of “white line” extent of oxidation of Pt

Height of Pt L3 absorption edge amount of Pt in electrode

Absorption edge loss over three cycles

0.8

0.9

1

1.1

1.2

1.3

1.4

1.5

1.6

11560 11565 11570 11575 11580 11585 11590

Nor

mal

ized

Abs

orba

nce

Energy (eV)

Pt L3-edge XANESPt L3-edge XANES

0.8

0.9

1

1.1

1.2

1.3

1.4

1.5

1.6

11560 11565 11570 11575 11580 11585 11590

Nor

mal

ized

Abs

orba

nce

Energy (eV)

0.8

0.9

1

1.1

1.2

1.3

1.4

1.5

1.6

11560 11565 11570 11575 11580 11585 1159011560 11565 11570 11575 11580 11585 11590

Nor

mal

ized

Abs

orba

nce

Energy (eV)

0.5 V

1.4 V

9

Platinum loss occurs during anodic and cathodic potential scans Greatest Pt loss observed in anodic step from 1.1 to 1.4 V

0.4

0.6

0.8

1

1.2

1.4Potential (V)

Potential Cycle

0

2

4

6

8

10

12

14

16 -2

-1.5

-1

-0.5

0

0.5

% L

oss

Edge

Ste

p H

eigh

t

(%

Loss)

10

XAFS shows platinum loss and oxide formation are linked

Pt loss is highest during oxide formation Approximately same extent of oxidation show different

Pt loss rates– Evidence against major role of oxide dissolution– Evidence for dissolution of metal– “Time-resolved” experiments are underway

Extent of Pt oxidation decreases with potential cycling -may be indicative of particle growth

11

SAXS studies shows Pt particle growth with cycling

20 wt% Pt/C

20 wt% Pt/C

40 wt% Pt/C

2

2.5

3

3.5

4

4.5

0 2 4 6 8 10 12 14 16

Par

ticle

siz

e (n

m)

Cycle Time (hrs)= 40 cycles

M.C. Smith et al., J. Am. Chem. Soc., 2008.

0102030405060708090

100

0 1 2 3 4 5 6 7 8Particle Size (nm)

Freq

uenc

y

1 hr2 hrs3 hrs5 hrs7 hrs10 hrs12 hrs14 hrs16 hrs

0

20

40

60

80

100

0 1 2 3 4 5 6 7 8 9 10Particle Size (nm)

Freq

uenc

y

SAXS AnalysisTEM Analysis

20 wt% Pt/C

20 wt% Pt/C

12

Non-platinum group metal electrocatalysts

Cobalt or iron either complexed with C-N polymer/molecule or pyrolyzed (J.P. Dodelet, Los Alamos NL, U. South Carolina, 3M, et al.)– Low cost

• (Co ~US$ 3 /oz, abundance 20,000-30,000 ppb in Earth’s crust vs 3-37 ppb for Pt)

– Promising oxygen reduction activity, but lower than platinum group metals

– Good durability, but longer testing and cycling tests are needed (>1000 hrs)

Issues:– Identity of the active site is unknown

• Metal center coordinated to pyridinicnitrogen

• Encapsulated metal catalyzes formation of active site

– Metal leaches from catalyst during operation

H

CN

Co

n

R. Bashyam and P. Zelenay, Nature, 2006.

Metal particle

13

XAFS analysis shows Co-polypyrrole (not pyrolyzed) catalyst changes with time/potential

Slow break in: possible formation of ORR sites during operation or removal of site-blocking species

Ex-situ XAFS data: as-prepared MEA contained a mixture of cobalt metal and a small oxide fraction

In-situ XAFS data: cobalt metal fraction is removed and/or converted to higher oxidation state

Three cobalt species observed in-situ:

H O/N Co

0.0

0.5

1.0

1.5

0 1 2 3 4

R (Å)

Mag

nitu

de

0.4 V, 0.3 V

0.2 V, 0.1 V

>0.3V, low RH

>0.1 V, high RH

2.83 Å

2.11 Å

1.92 Å

3.10 Å

2.05 Å

2.06-2.08 Å

0.2 V, low RH0.1 V, low RH

14

Los Alamos NL’s pyrolyzed polyaniline-Fe(Co)-C ORR catalysts

Pt/C

15

Aqueous cell in-situ data for pyrolyzed polyaniline-Fe-C system

XAFS shows reversible reduction of Fe3+

catalyst component between 0.64 and 0.44 V Fe is lost from the electrode with greatest loss

observed during this reduction step

0.0

1.0

2.0

3.0

Wt%

Fe

0.87 0.64 0.44 0.24 0.44 0.64 0.84 1.04 0.87

Potential (V vs. SHE)

FeS2

Fe2O3Fe3O4

FeO

FeSO4

Fe metal

Fe-phthalocyanine

0.0

0.4

0.8

1.2

1.6

2.0

7050 7100 7150 7200 7250Energy (eV)

Abs

orba

nce

0.87 V

0.64 V

0.44 V

0.24 V

1

2

3

4

5

6

8

16

Pyrolyzed polyaniline-Fe-C catalyst composition

Wt%

Fe

in

indi

cate

d co

ordi

natio

n en

viro

n.

MEA preparation:– Removes metal– Removes sulfides– Oxidizes Fe2+ to Fe3+

Fe2O3 Fe-pc FeS2 Fe3O4

0.0

0.5

1.0

1.5

MEA, 0.6 V for 200 h

Fresh MEA

Fe2O3 Fe-pc FeS2 Fe3O4

0.0

0.5

1.0

1.5

MEA, 0.6 V for 200 h

Fresh MEA

Wt%

Fe

in in

dica

ted

coor

dina

tion

envi

ron.

Fe m

etal

FeS

FeS

2

Fe-p

c

Fe2O

3

Fe3O

4

FeO

FeS

O4

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

MEA

Acid-TreatedPowder

Wt%

Fe

in in

dica

ted

coor

dina

tion

envi

ron.

Fe m

etal

FeS

FeS

2

Fe-p

c

Fe2O

3

Fe3O

4

FeO

FeS

O4

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

MEA

Acid-TreatedPowder

Fe is lost from MEA during long-term polarization at 0.6 V (approx. 50% loss)

Ratio of Fe2O3 to Fe-pc coordination is approx. unchanged

17

Summary

In-situ X-ray absorption and scattering techniques are powerful for diagnosing the state of PEFC catalysts during operation

New in-situ X-ray fuel cell block design allows XAFS studies in fluorescence mode– Enables study of very low loadings of low Z metals (e.g., Fe and Co)– Eliminates the need to modify flow field design– Allows the study of one electrode of a cell when the opposing electrode

contains the same metal (e.g., can study Pt in a Pt cathode with a Pt anode)

X-rays

Sample

SAXS Detector

XAFS Detector

X-rays

Sample

SAXS Detector

XAFS Detector

Future needs/experiments– Combination of scattering and absorption

experiments with microsecond time resolution– Simultaneous spatio-temporal resolved (micrometer

and microsecond) atomic, electronic, and particle size characterization for a wide range of metals (e.g., Pt and Co in Pt3Co catalyst)

18

Acknowledgements ANL – Chemical Sciences and Engineering Division

– Xiaoping Wang– Nancy Kariuki– Jennifer Mawdsley– Di-Jia Liu– Chris Marshall

ANL – Advanced Photon Source– Mali Balasubramanian– Sector 20 (PNC-CAT)– Nadia Leyarovska– Sönke Seifert– Sector 12 (BESSRC-CAT)

DOE, Office of Science, Basic Energy Sciences DOE, Office of Energy Efficiency and Renewable Energy, Hydrogen, Fuel Cells

& Infrastructure Technologies