2010 DOE Annual Merit Review

Nanosegregated Cathode Catalysts with Ultra-Low Platinum Loading

DE-PS36-08GO98010

PI: Nenad M. Markovic

Co-PI: Vojislav R. Stamenkovic

Announcement No:Topic: 1A

Materials Science DivisionArgonne National Laboratory

Project ID#FC008

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

Timeline

• Project end: 9/2012

Budget• Total Project funding $ 3.6M

• Received in FY09: $ 300KBarriers

Overview

• The main limitations: CATHODE

• Project start: 9/2009

~ 30-40% (!!!)Cathode kinetics

1) Durability (Pt dissolves: power loss)

2) High content of Pt = High Cost ($75/g)

3) Poor activity: Pt/C = Pt-poly/10

Subcontractors:

• Oak Ridge National Laboratory – Karren More• Jet Propulsion Laboratory – S.R. Narayan• Brown University – Shouheng Sun• Indiana University Purdue – Goufeng Wang• 3M Company – Radoslav Atanasoski

• DOE share: 80 %• Contractor share: 20%

• Funding for FY10: $ 564.4K

2

DOE Technical Targets

• Durability w/cycling (80oC): 5000 hrs

• Cost*: $ 30/kWe

• Mass activity @0.9V: 0.44 A/mgPt

• PGM Total content: 0.2 g/kW

• Specific activity @0.9ViR-free: 720 µA/cm2

• Electrochemical area loss: < 40%

• Catalyst support loss: < 30%

• PGM Total loading: 0.2 mg/cm2electrode

*based on Pt cost of $450/troy ounce

The main focus of ongoing DOE Fuel Cell Hydrogen Program is fundamentalunderstanding of the oxygen reduction reaction on PtM bimetallic and PtM1M2 (M1=Co,Ni;M2=Fe, Mn, Cr, V, Ti etc) ternary systems that would lead to the development of highly-efficient and durable real-world nanosegregated Pt-skin catalysts with low-Pt content

Objectives-Relevance

ANLTechnical Targets

• Mass activity @ 0.9ViR-free2015 DOE target x 3

• PGM Total content< 0.1g/kW

• Specific activity @ 0.9ViR-free2015 DOE target x 3

• Electrochemical area loss2015 DOE target

3

Materials-by-design approach - developed by ANL will be utilized to design,characterize, understand, synthesize/fabricate and test nanosegregated multi-metallicnanoparticles and nanostructured thin metal films

REAL NANOPARTICLESMODEL NANOPARTICLESEXTENDED Multi-M SURFACES

Approach

Nanosegregated Profile

1st Layer

2nd Layer

3rd Layer

4th Layer

Pt=100 at.%

Pt=48 at.%

Ni=52 at.%

Pt=87 at.%

Ni=13 at.%

Pt=75 at.%Ni=25 at.%

Pt[111]-Skin surface

Pt3Ni((111)-Skin is the most active catalyst for the oxygen reduction reaction, and it is ~100 times more active than the state-of-the-art Pt/C catalysts. Design of real-world bi-/multi-metallic catalysts with this activity is our goal.

Well-Defined Systems

d-band center [eV]2.63.03.4

Spec

ific

Act

ivity

: ik @

0.9

V [m

A/c

m2 re

al]

0

1

2

3

4

5

Pt3TiPt3V

Pt3Fe

Pt3CoPt3Ni

Pt-polyPt-skin surfacesPt-skeleton surfaces

Act

ivity

impr

ovem

ent f

acto

r vs

. Pt-p

oly

1

2

3

17

18

19

Target Activity Pt3Ni(111)

(a)

Pt/C ---

Advanced Nanoscale Catalyst

4

Approach / Milestone

1.1 Resolved electronic/atomic structure and segregation profile (85%)1.2 Confirmed reaction mechanism of the ORR (95%)

3.3 MEA testing (50 cm2)

2.1 Physical methods: TM films (5-10 layers), nanoparticles (5-300 nm) (90%)2.2 Established chemical methods: colloidal and impregnation synthesis (90%)

1.3 Improved specific and mass activity (95%)

3.1 New PtM1M2 catalyst to increase catalytic activity and decrease dissolution

2.3 Characterization: Ex-situ (UHV, TEM) and in-situ (SXS, EC) (90%)2.4 Theoretical modeling (DFT, MC) methods (90%)

3.2 Carbon support vs. nanostructured thin film catalysts

3.4 Short stack testing verification (>300cm2)

Milestone 1. Fundamental understanding (2009/10) (Accomplished)

Milestone 2. Synthesis and characterization (2009/10)

Future Milestone 3. Fabrication and testing

(Go-No Go Decision Met)

5

Relevant Prior Work

V.Stamenkovic, B.Flower, B.S.Mun, G.Wang, P.N.Ross, C.Lucas, N.M.Markovic Science, 315(2007)493

V.Stamenkovic, B.S.Mun, M. Arenz, K.J.J.Mayerhofer, C.Lucas, G.Wang, P.N.Ross, N.M.Markovic Nature Materials, 6(2007)241

V.Stamenkovic, B.S.Mun, K.J.J.Mayrhofer, P.N.Ross, N.M.MarkovicJ. Am.Chem.Soc., 128(2006)8813

V.Stamenkovic, B.S.Mun, K.J.J.Mayrhofer, P.N.Ross,N.M.Markovic, J.Rossmeisl, J.Greeley, J.K. NorskovAngew.Chem.Int.Ed., 45(2006)2897

H.A. Gasteiger, N.M.MarkovicScience, 3124(2009)48

Nanosegregated particle would be the best catalysts for the ORR

6

Colloidal deposition approach is used to synthesize (a) monodispersed Pt3Co bimetallic NPs with diameter of 3, 4.5, 6 and 9nm

Technical Accomplishments: Particle Size Effect

Particle size is determined by analyzing TEM images (b-e)

Particle size effect applies to bimetallic NPs

0.1 M HClO4

25 C°

Specific Activity increases withparticle size: 3 < 4.5 < 6 < 9nm

Mass Activity decreases with particle size: optimal size ~5nm

Specific surface area is determined from ΘHupd

0.1 M HClO4

0.1 M HClO4

7

As Prepared 400oC

500oC 800oC

Temperature Induced Segregation: apply 400 to 500°C to optimize catalytic activity of the ORR on PtCo bimetallic NPs

MC simulations at 400oC confirmed segregation profile of Pt3Co NPs, which was experimentally observed for extended surfaces

External View Cross Section

Annealing above 500oC provides small increase in specific activity but significant decrease in mass activity of Pt3Co NPs

Technical Accomplishments: Temperature-induced Segregation

TEM: no agglomeration below 500oC

0.1 M HClO4

60C°0.1 M HClO4

8

Segregation of Pt at 400oC is not complete in PtxNiy NPs, which induces dissolution of Ni

1 nm

0.22 nm

1 nm1 nm

Pt3Ni PtNi PtNi2 PtNi3

Technical Accomplishments: Composition Effect

Colloidal method is used to synthesize PtxNiy NPs with 3:1, 1:1, 1:2 and 1:3 atomic ratio

~5nm PtxNiyLast step

annealing at 400oC

0.1 M HClO425 C°

0.1 M HClO425 C°

Maximum activity obtained for as-synthetized NPs with 1:1 Pt to Ni atomic ratio

In acidic environment, atomic % of Ni in PtxNiy NPs decreases due to dissolution of Ni surface atoms

Pt3Ni Pt85Ni15 PtNi Pt75Ni25 PtNi2 Pt80Ni20 PtNi3 Pt90Ni10PtNi transforms into Pt3Ni skeleton

9

As-prepared PtNi

Inte

nsity

(a. u

.)

Position (nm)43 5210

Position (nm)43 5210

Inte

nsity

(a. u

.)

PtNi

400oC

To be published 2010

Temperature annealing protocol used to transform PtNi1-x skeletons to PtNi/Pt core/shell NPs with 2-3 atomic layers thick Pt shell

PtNi1-y Skeleton

HRTEM characterization

Line profile of NP

Models

Monte CarloSimulations

PtNi/Pt core/shell

Technical Accomplishments: Tailoring of Nanosegregated Layers

10

0.01 0.1 1 100.85

0.90

0.95

1.00

6 nm Pt/C acid leached PtNi/C acid leached/annealed PtNi/C

E (V

vs.

RHE

)

jk (mA/cm2)

0.0

0.2

0.4

0.6

0.8

1.0

1.2

PtNi acid leached& annealed

PtNiacid leached

Surfa

ce A

rea

(cm

2 )

Spec

ific

Activ

ity (m

A/cm

2 )

Pt6 nm

0

10

20

30

40

50

x 4

x 7

0.6 0.7 0.8 0.9 1.0

-1.5

-1.0

-0.5

0.0

6 nm Pt/C acid leached PtNi/C acid leached/annealed PtNi/C

I (m

A)

E (V vs. RHE)

Technical Accomplishments: Tailoring the Catalytic Activity

To be published 2010

0.95 V is more accurate to use for ik (mA/cm2)

0.1 M HClO4

25 C°

PtNi/Pt core/shell catalyst has 7 fold improved activity over corresponding (similar size) Pt/C

0.1 M HClO4

25 C°

Activity/Stability trend is controlled by surface coverage of spectator OHad species formed from dissociation of water:

Pt < Pt3Ni (skeleton) < Pt3Ni/Pt core/shell

Pt and PtNi NPs

The same Tafel slope

Multi-layered Pt-shell protects Ni in the core: Pt3Ni/Pt NPs do not suffer from the decay in activity or surface area after 30,000 cycles between 0.5 to 1.1V

11

Technical Accomplishments: Catalytic Trends

0.0 0.2 0.4 0.6 0.8 1.0

-1.5

-1.0

-0.5

0.0

Curr

ent (

mA)

E (V vs. RHE)

-0.1

0.0

0.1

Curr

ent (

mA/

cm2 Pt

)

Pt3Fe Pt3Co Pt3Ni

0.1 1 100.85

0.90

0.95

1.00

E (V

vs.

RHE

)

jk (mA/cm2)

0

500

1000

1500

2000

Mas

s Ac

tivity

(A/g

)

Pt3NiPt3CoPt3FePt

0

1

2

3

4

5

Spec

ifc A

ctiv

ity (m

A/cm

2 )

0

20

40

60

80

Spec

ifc S

urfa

ce A

rea

(m2 /g

)

Specific and Mass Activity trends for the ORR of Pt-bimetallic NPs is the same as electrocatalytic trends established on extended Pt3M well-defined surfaces

The same particle size

Pt < Pt3Fe < Pt3Ni < Pt3Co

Similar Hupd coverage

Similar active surface area

The same Tafel slope

ORR Activity Trend

Reaction mechanism for the ORR is the same on extended and real-world nanosegregated Pt3M catalysts; the 4e- serial pathway

Adsorption Trend

The same mechanism

0.1 M HClO4

25C°

0.1 M HClO4

25C°5nm

Activity is determined by electronic properties of the Pt surface atoms, i.e., by surface coverage of blocking non-reactive OHad species - not by the energetics of reaction intermediates

12

Technical Accomplishments: Ternary Alloy Catalysts

0

1

2

3

4

5

6

Spec

ifc A

ctiv

ity (m

A/cm

2 )

0

20

40

60

80

100

Spec

ifc S

urfa

ce A

rea

(m2 /g

)

0

500

1000

1500

2000

Mas

s Ac

tivity

(A/g

)

PtFeNiPtCoNiPtFeCoPt0.0 0.2 0.4 0.6 0.8 1.0

-1.5

-1.0

-0.5

0.0

Curr

ent (

mA)

Potential (V vs. RHE)

-0.1

0.0

0.1

PtCoNi PtFeCo PtFeNi

Curr

ent (

mA)

2 4 6 8 100.80

0.85

0.90

0.95

1.00

E (V

vs.

RHE

)

jk (mA/cm2)

Oleylamineoleic acid

~ 200oCTM2+

TM = Fe, Co, and/or Ni

PtM1N2 (NM = Fe, Co, and/or Ni)

To be published 2010

Specific and Mass Activities for ORR of ternary alloys confirmed potential of utilizing TM in further reducing of Pt content

Ternary alloys could provide additional tunability towards activity and stability PtNiCo the most active catalyst

Segregation trends in ternary alloys are still under investigation

Synthesis of

Activity

13

Collaborations

• Oak Ridge National Laboratory – HRTEM• Jet Propulsion Laboratory – Alloying and Combinatorial Approach• Brown University – Chemical Synthesis• Indiana University Purdue – Theoretical Modeling• 3M – Testing

PARTNERS

• GM – Collaboration to utilize highly active Pt-alloy catalysts• Argonne National Laboratory – Nanoscale fabrication (CNM)

TECHNOLOGY TRANSFER

14

Future Work

• Fundamental understanding of ternary alloy catalysts• Resolving electronic and atomic structures

• Characterization and evaluation of activity and stability trends• Syntheses, characterization and laboratory testing

• Composition screening

FY 2010

FY 2011

• Fabrication and testing in MEA

• Characterization and evaluation of activity and stability trends• Syntheses, characterization and laboratory testing

• Composition validation

15

S u m m a r y

1.1 Resolved electronic/atomic structure and segregation profile 1.2 Confirmed the reaction mechanism of the ORR

2.1 Physical methods: TM films (5-10 layers) were used to confirm EC properties2.2 Chemical methods: colloidal solvo-thermal approach in NPs synthesis

1.3 Mass activity and durability improvement are obtained for Pt-alloy NPs

2.5 Novel multimetallic catalyst with Pt increased activity and improved stability

2.3 Characterization: Ex-situ (UHV, TEM) and in-situ (EC, Combinatorial)2.4 Theoretical modeling (DFT, MC) methods

Task 1. Fundamental understanding

Task 2. Synthesis and characterization

16

Supplemental Slides

E [V vs. RHE]

0.0 0.2 0.4 0.6 0.8 1.0

Cur

rent

den

sity

[mA

/cm

2 ]

-150

-100

-50

0

50

100

E [V vs. RHE]

0.0 0.2 0.4 0.6 0.8 1.0

Cur

rent

den

sity

[mA

/cm

2 ]

-6

-5

-4

-3

-2

-1

0

1

1 ML Pt on PtNi3 ML Pt on PtNi7 ML Pt on PtNi

Pt Poly

- Pt

1stPt ML2ndPt ML3rdPt ML

- Ni

Thin Film Deposition in UHV

Pt thin film thickness: 1-10 ML

E [V vs. RHE]

0.88 0.90 0.92 0.94 0.96

log

Cur

rent

den

sity

[mA

/cm

2 ]

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

1.2

Thin Films of Pt over PtNi substrate

0.1 M HClO4

0.1 M HClO4

25 C°

The most active surface is PtNi/Pt-3ML which corresponds to core/shell PtNi/Pt

ORR

Supplemental Slide 1