Low Pt Loading Fuel Cell Electrocatalysts · j / mAcm-2 E / V RHE rpm 1600 Au 2/3ML /Pt/C intial...

1

Low Pt Loading Fuel Cell Electrocatalysts

Radoslav Adzic (P.I.), Junliang Zhang, Kotaro Sasaki, Miomir Vukmirovic, Jia Wang, Minhua Shao

Chemistry Department Brookhaven National Laboratory, Upton, NY 11973-5000

(This presentation does not contain any proprietary or confidential information.) Project ID # FC09

DOE Hydrogen Program Review, May 16-19, 2006

2

Overview

• Project start date: 06.02.• Project end date: Multi-year• Percent complete

Barriers addressed

1. Precious metal loading

2. Electrocatalysts’ Activity

3. Electrocatalysts’ Durability

Target 2010: 0.3 g/kW

Target 2010: 5000hr

• Total project funding: $1314K • DOE share: $1314K• Funding received in FY05:

$330K• Funding for FY06: $360K

Budget

Timeline

Collaborations• Los Alamos National Laboratory (Fuel cell tests - F. Uribe, P. Zelenay)• Battelle Memorial Institute (CRADA Scale-up of synthesis - J. Sayre and A. Kawczak ) • 3M (PdCo catalyst, exploratory activities – R. Atanasoski)• Plug Power – Test of the PtRu20 anode catalyst: 2390 hr showed a small loss in

activity (B. Do).• General Motors Co. planning stage (F. Wagner)

3

--- ComprehensiveTo assist the DOE in the development of fuel cell technologies by providing low-platinum-loading electrocatalysts.

--- During the current year• To demonstrate the stability of Pt mixed metal monolayer (Pt80Ir20/Pd/C) and

Pt/Au/Ni/C electrocatalysts in fuel cell tests.

• To explore a novel class of electrocatalysts for O2 reduction consisting of Pt monolayer on noble metal -non-noble metal core-shell nanoparticles.

• Further studies of the electrocatalysts with low, or no Pt.

• Scale – up of synthesis based on displacement of a Cu monolayer.

• Addressing the problem of Pt dissolution under potential cycling regimes.

• Testing the effects of Au clusters on Pt stability.

Objectives

4

Why Pt monolayer catalysts?

• Complete Pt utilization (of all atoms that are not blocked by Nafion®)

• Ultimate reduction of Pt loading

• Increased activity

(Six patents pending)

APPROACH

For 3-5nm nanoparticles ~25% of atomsare on the surface,

~75% are not available for catalysis.

1. Pt on Pd nanoparticles

Improved activity

Improved durability

2. Mixed-metal Pt on Pd nanoparticles

Even higher activity

Durability tests at LANL

and Battelle

3. Pt on noble/non-noble core-shell nanoparticlesYet higher activityper mass of noble metal

Durability to be tested

Pt

Pd

Pt

PdIr

PtAuNi

Three types of Pt electrocatalysts:

Pt monolayer electrocatalysts

5

Compared to commercial MEA baselines (0.6 mg Pt/Ru/cm2): Higher than neat H2 baseline; lower than reformate baseline.

Total run time: 2394 hours

0.022 mgPt/cm2

Long-term test (one segment) of the anode PtRu20electrocatalyst at PLUG POWER

6

Segregation at elevated T

ML of Cu by UPD

Cu displaced by Pt

Cu ML Pt ML, or mixed Pt-M ML

1-2 ML of Au, Pd, or PtCo or Ni

New class of catalysts: Pt monolayer on noble metal - non-noble metal core-shell nanoparticles

New class of electrocatalysts that facilitate a further decrease of noble metal content while possessing a very high activity.

Au, Pd, or

Pt

Synthetic route

7

O2 Reduction Kinetics and Mass Activities of PtML/PtCo5/C and PtML/PdCo5/C

0.4 0.6 0.8 1.0

-6

-4

-2

0

Pt/C PtML/PdCo5/C

1600 rpm

j / m

A.c

m-2

E/V vs RHE

0.4 0.6 0.8 1.0

-6

-4

-2

0

Pt/C PtML/PtCo5/C

1600 rpm

j / m

A.c

m-2

E/V vs RHE 0.8V 0.85v0

5

10

15

20

-j k mA

/μg

Pt

Pt10/C PtML/AuNi10/C PtML/PdCo5/C PtML/PtCo5/C

Total noble metal mass activity

Pt mass activity

0.8V 0.85v0

1

2

3

4

5

6

-j k mA

/μg

nobl

e m

etal

Pt10/C PtML/AuNi10/C PtML/PdCo5/C PtML/PtCo5/C

Pt

Pd

2- to 5-fold increase in noble metal mass activity with Pt on core-shell nanoparticles

8

-1

-0.8

-0.6

-0.4

-0.2

0

0.2

0 0.2 0.4 0.6 0.8 1 1.2

Pt/NbO2/C, 5 μg Pt/cm2

E1/2

= 863 mV

commercial Pt/C, 5 μg Pt/cm2

E1/2

= 778 mV

commercial Pt/C, 10 μg Pt/cm2

E1/2

= 815 mV

I, m

A

E, V(RHE)

0.1 M HClO4 +O

2

1600 rpm, 10 mV/s

O2 Reduction Kinetics and Mass Activities of PtML/NbO2/C and Pt/C/C

PtML/NbO2/C has 3 times higher mass activity than the commercial Pt/C

0

0.2

0.4

0.6

0.8

0.4 0.6 0.8 1 1.2

Pt/NbO2/C

Pt/C

(Δμ

- Δμ 0.

42V)/Δ

μ 0.42

V

E , V(RHE)

relative changes in Pt L3 edge

XANES shows a small oxidation of Pt/NbO2/C compared to Pt/C.

Pt is not observed by XRD in Pt/NbO2/C

0

1

2

3

4

j k, mA

μg-1

0.80 V

Pt/NbO2/C

Pt/C

0.85V

Pt/NbO2/C

Pt/C

Pt/NbO2/C

5 μg Pt/cm2

9

Formation of Au clusters on Pt electrodes

0.0 0.2 0.4 0.6 0.8 1.0

-80

-40

0

40

80 Pt(111); Au / Pt(111)

j / μ

Acm

-2

E / V RHE

0.1M HClO4

50mV/s

0.0 0.4 0.8 1.2-2

-1

0

1

j / m

Acm

-2

E / V RHE

Pt/C Au/Pt/C

0.1M HClO450 mV/s

PtAu

2D Au deposited on Pt(111); 3D formed after several cycles to 1.2V in O2 saturated solution

115 nm x 115 nm 120 nm x 120 nm

No effect on H ads. Inhibition of OH ads.

Model of Au clusters on Pt nanoparticle

10

0.0 0.2 0.4 0.6 0.8 1.0 1.2

-6

-4

-2

0

Pt (111)

j / m

Acm

-2

E / V RHE

Au2/3ML/Pt (111)

rpm1600

0.0 0.2 0.4 0.6 0.8 1.0 1.2

-6

-4

-2

0

j / m

Acm

-2

E / V RHE

1600rpm

Au2/3ML/Pt/C Pt/C

Effect of submonolayer of Au clusters on O2 reduction on Pt(111) and Pt/C

Small inhibition of O2 reduction by Au, not in proportion to Au coverage

11

0.0 0.4 0.8 1.2

-6

-4

-2

0

j / m

Acm

-2

E / V RHE

rpm 1600

Pt/C intial 30,000 cycles

30,000 cycles from 0.6 to 1.1V at 25°C in O2-saturated solution

0.0 0.4 0.8

-0.6

-0.3

0.0

0.3

j / m

Acm

-2E / V RHE

Pt10/C20mV/s, 0.1MHClO4

Initial After 30,000 cycles

Effect of potential cycling on the activity of Pt/C in the ORR

A negative shift of 40mV in E1/2 and a loss of surface area of 45%

12

Stabilization of Pt/C by a submonolayer of Au clusters under a potential cycling regime

0.0 0.4 0.8 1.2

-6

-4

-2

0

j / m

Acm

-2

E / V RHE

rpm 1600

Au2/3ML/Pt/C intial 30,000 cycles

30,000 cycles from 0.6 to 1.1V at 25°C in O2-saturated solution

0.0 0.4 0.8 1.2-2

-1

0

1

intial After 30,000 cycles

j / m

Acm

-2

E / V RHE

Au0.67ML/Pt10/C50mV/s, 0.1M HClO4

A negligible shift in E1/2 and a loss of surface area

13

0.0 0.4 0.8 1.2

-6

-4

-2

0 rpm 1600

j / m

Acm

-2

E / V RHE

PtML/Pd/C initial 30,000 cycles

0.0 0.4 0.8 1.2

-6

-4

-2

0 rpm 1600

j / m

Acm

-2

E / V RHE

Au2/3ML/PtML/Pd/C

initial 30,000 cycles

30,000 cycles from 0.6 to 1.1V at 25°C in O2-saturated solution

Stabilization of PtML/Pd/C by a submonolayer of Au nanoparticles under a potential cycling regime

A negative shift of 26mV in E1/2 A negligible shift in E1/2

14

Pd3Fe

Compressive strain-induced downshift of the d - band center decreases the activity of Pd and the blocking effect of OH, O2, O2

-, H2O2.

The optimal Pd-Pd distance is 0.273 nm.

The activity of Pd3Fe/C is as high as that of commercial Pt/C.

Pd3Fe alloy O2 reduction electrocatalyst

-2.4 -2.2 -2.0 -1.8

-2.2

-2.0

-1.8

-1.6 Pd skin on PdFe

Pd skin on Pd3Fe

Pd3Fe

4.7%compressed Pd

2.3%compressed Pd

Ado

sorp

tion

ener

gy (e

V/O

2)

εd-εf (eV)

Pd

15

Comparison between PtML/Pd3Fe/C, PtML/Pd/C and Pt/C

0

2

4

6

8

0

2

4

6

8

0.85 V

Pt/C PtML/Pd/C PtML/Pd3Fe/C

j k / m

A μg

-1

0.8 V

Pd/C particle size of 9 nm.

0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1

-6

-4

-2

0 Pt/C (Pt: 51 nmol cm-2) Pt/Pd/C (Pd: 51 nmol cm-2; Pt: 5.2 nmol cm-2) Pt/Pd

3Fe/C (Pd: 51 nmol cm-2; Pt: 8.67 nmol cm-2)

j / m

A c

m-2

E / V vs RHE

Total noble metal mass activity is about 5 times that of the commercial Pt catalyst.

Total noble mass activity

Fine tuning of the Pt-substrate interaction: Very high activity of PtML/Pd3Fe/C

E1/2 = 0.880 V

PtML/Pd3Fe/C has 0.14g/kW, which meets 0.15g/kW,i.e., 1/2 of 0.3g/kW, the DOE’s target for 2010

16

XANES data indicates pronounced electronic effects in Pd-Co and an increased stability of Pd, its activity being close to that of Pt(10%)/C.

BNL#101-103 are from one sample obtained by different acid wash procedure.

Pd2Co

MeOH Concentration (M)0 1 2 3 4 5 6

i @ 0

.4 V

(mA

/cm

2 )

0

20

40

60

80

100

120

140

160BNL #101BNL #102BNL #103Chalcogenide

Time (h)0 50 100 150 200

Cur

rent

Den

sity

(A/c

m2 )

0.00

0.05

0.10

0.15

0.20

DMFC, 0.5 M, 80°C, 0.3 V

Fuel cell methanol tolerance test by P. Zelenay et al., LANL

Current Density (A/cm2)0.00 0.05 0.10 0.15 0.20 0.25 0.30

Cel

l Vol

tage

(V)

0.0

0.2

0.4

0.6

0.8

1.0BNL #101BNL #102BNL #103Chalcogenide

Methanol tolerance of the Pd2Co electrocatalyst

17

Dual-Pathway Kinetic Equation for the Hydrogen Oxidation Reaction on Pt Electrodes

The Butler-Volmer equation, jk = j0 , appears to be inappropriate for describing the kinetics of the H2 oxidation reaction (HOR) on a Pt electrode (1,2).

1. Gasteiger, et al., Power Sources, 127 (2004), 162.2. Chen, Kucernak, J. Phys. Chem. B 108 (2004) 13984.3. J. X. Wang, T. E. Springer, R. R. Adzic, J. Electrochem.

Soc. Submitted.

)( /3.2/3.2 bb ee ηη −−

Measured (symbols, from ref [2]) and fitted (solid lines using the Eq.) polarization curves for the HOR on Pt microelectrodes

)()1( )2/(2/0

/0

RTFRTFH

RTFTk eejejj βηηβη +−− −+−=

Based on the Tafel-Heyrovsky-Volmer mechanism, we derived a new equation, ref. 3, to describe the HOR on Pt over the entire potential region:

Calculated kinetic current (solid line), components, jTV (dashed line) and jHV (dotted line), the kinetic current from the Butler-Volmer equation (dotted-dashed line).

A fast, inverse exponential rising of kinetic current at small overpotentials through the Tafel-Volmer pathway, jTV, and a gradual rise at η > 50 mV through the Heyrovsky-Volmerpathway, jHV. The anode behavior in a PEM fuel cell can now be well understood. New basis for fuel cell modeling, optimization, and diagnosis.

18

Future WorkRemainder of FY2006

• PtML/Pd/C electrocatalyst- Post fuel cell tests Z-contrast TEM, XANES.

• Pd2Co electrocatalyst- Stability studies; and fuel cell tests.

• Mixed-metal Pt monolayer electrocatalysts- Stability studies. Effects of Au clusters in MEA’s under potential cycling regimes.

FY 2007

• Stabilization effects of Au clusters:- RDE and MEA studies under potential cycling conditions with Pt/Pd/C; Pt/PdCo5/C.

• Pt/AuNi/C electrocatalyst- Segregation of Pt, Au; Stability tests; Fuel cell tests.

• Further reduction of Pd content: Pt monolayers on core-shell nanoparticles- Basic in situ surface science and electrochemical studies; stability and fuel cell tests.

19

BNL ANODE PtRu20/C electrocatalyst:Durability and CO tolerance confirmed in a 2400h test at Plug Power. The performance of 0.063 gPt/kW meets the 2010 target of 0.15g/kW, i.e.,1/2 of 0.5g/kW (Ru not counted).

BNL CATHODE electrocatalysts:1. Novel class of electrocatalysts developed: Pt monolayer on

noble metal - non-noble metal core-shell nanoparticles. The performance of 0.14g/kW meets the target for 2010 of 0.15g/kW.

2. MEA test of Pt/Au/Ni/C failed, attributable to inadequate synthesis.

3. Scale up of catalysts synthesis to 1g/batch accomplished.

4. Au cluster submonolayer has a pronounced stabilization effect on Pt under potential cycling regime. It is a highly promising solution to that problem.

5. The Pd2Co and Pd3Fe have activities similar to that of Pt; Methanol tolerance of Pd2Co confirmed in a fuel cell test at LANL.

6. NbO2 is a promising support for Pt monolayer catalysts.

Summary: Pt monolayer electrocatalysts

20

1. Include repeated potential cycling tests as part of durability screening of promising candidates… --- The tests have been started.

2. Do small-scale development of the processes needed to make 50cm2 MEA and, eventually, short-stack quantities of catalysts through the UPD-replacement process. ---The synthesis has been scaled up to make 1 g of the catalyst per day.

3. Should try to reduce the amount of Pd sooner in timeline (2 reviewers) -- Pt monolayer on noble metal -non-noble metal core-shell nanoparticles provide such a possibility.

4. A more appropriate industry collaboration would be with companies that make catalysts,--- BNL is open for collaboration.

Responses to Previous Year Reviewers’ Comments

21

Publications

1. J. Zhang, F.H.B. Lima, M. H. Shao, K. Sasaki, J.X. Wang, J. Hanson, R. R. Adzic, Platinum monolayer on non-noble metal - noble metal core-shell nanoparticles electrocatalysts for O2 reduction, J. Phys. Chem. B, 109 (2005) 22701-22704.

2. M.H. Shao, P. Liu, R.R. Adzic, Superoxide is the intermediate in the oxygen reduction reaction on platinum electrode. J. Am. Chem. Soc. (submitted).

3. J. Zhang, M. B Vukmirovic, K. Sasaki, F. Uribe, R. R. Adzic, Platinum monolayerelectrocatalysts for oxygen reduction: effect of substrates, and long-term stability, J. Serb. Chem. Soc., 70 (2005) 513-525 (75th Anniversary issue ).

4. J. X. Wang, F. A. Uribe, and R. R. Adzic, Parameterizing H2/air-PEMFC polarization curves, and quantifying cell performance and major voltage losses”, Electrochem. Solid-State Lett. (submitted).

5. M.H. Shao, K. Sasaki, R.R. Adzic, Pd-Fe nanoparticles as electrocatalysts for oxygen reduction. J. Am. Chem. Soc. 128 (2006) 3536.

Presentations

Five papers at national and three at international meetings,

22

• Stability of Au clusters on Pt surfaces

• Stability of surface segregation induced by annealing

• The extent of mixing in bimetallic systems at 80-100°C

• Electronic effects vs. strain effects

Critical Assumptions and Issues