Platinum Monolayer Electrocatalysts for Oxygen Reduction Reaction

Radoslav Adzic Co‐workers: Jia Wang, Miomir Vukmirovic, Kotaro Sasaki, Stoyan Bliznakov, Yun Cai, Yu Zhang, Kurian Kuttiyiel, Kuanping Gong, YongMan Choi, Ping Liu, Hideo Naohara1

Chemistry Department, Brookhaven National Laboratory, Upton, NY 11973 1Toyota Motor Corporation, Susono, Japan

Webinar June 19, 2012

Outline

- Introduction on fuel cells, electrocatalysis, existing developments and remaining obstacles to commercialization

- Platinum monolayer electrocatalysts, the main properties, synthesis, factor affecting the activity and stability- core – shell interactions.

- Tuning the activity and stability by core-shell interactions Several illustrations: Nanowires, Tetrahedral nanoparticles, Hollow Pd nanoparticles, Alloys as cores

- Fuel cell tests, long-term stability, self healing effects

- Conclusions

- Acknowledgements 2

Fuel Cells for Sustainable Energy Future

Fuel cells, an electrochemical power source that converts directly chemical energy of fuel into electrical energy, have several outstanding properties:

- High efficiency (theoretical, close to 100%, practical 50-60% )

- High power density

- Ideal for automotive application

Cathode: O2 reduction reaction (ORR) (slow)

O2 + 4H+ + 4e 2H2O E = 1.23V

Anode: H2 oxidation (HOR) (fast)

H2 2H+ + 2e- E = 0.0V

Anode: Methanol oxidation (MOR) (slow, CO strongly ads.)

CH3OH + H2O CO2 + 6H+ + 6e- E = 0.016V

Anode: Ethanol oxidation (EOR) (slow, partial oxidation)

C2H5OH + 3H2O 2CO2 + 12H+ + 12e- E = 0.084V

Main Reactions of Electrochemical Energy Conversion

3

                                      

         

         

                    

                   

     

http://www.ballard.com

With high efficiency and clean operation (H2O is the reaction product), fuel cells will prolong the availability of fossil fuels and improve quality of the environment.

Fuel Cells for Sustainable Energy Future Proton exchange membrane fuel cell (PEMFC)

0.0 0.2 0.4 0.6 0.8 1.00.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Cel

lvol

tage

/ V

Current density/ Acm-2

Theoretical Cell Potential (1.23 V at 25°C)

Kinetics losses (mostly at cathode)

Ohmic losses

Concentrationlosses

0.0 0.2 0.4 0.6 0.8 1.0 0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Cel

l vol

tage

/ V

Current density/ Acm-2

Theoretical Cell Potential (1.23 V at 25°C)

Kinetics losses (mostly at cathode)

Ohmic losses

Concentration losses

Obstacles caused by slow ORR kinetics: 1. Efficiency below theoretical, even for Pt,

the best catalyst 2. High Pt content in cathode; in addition to 3. Insufficient stability of Pt

4

The last two decades brought considerable advances in Fuel Cell Electrocatalysts by:

i) increasing activity ii) decreasing loadings and iii) increasing their stability

Some improvements are, however, needed to remove the remaining obstacles to their commercialization.

Several promising current approaches to address these problems include: 1. Segregated alloys (Markovic, Stamenkovic)

2. Nanostructured Pt films (Debe, Atanasoski)

3. Non-noble metal complexes (Zelenay)

4. Heat-treated macrocyclics (Dodelet)

Fuel Cells for Sustainable Energy Future

Our approach: Platinum Monolayer Electrocatalysts

5

   

Pt

Adzic, Zhang, Sasaki, Vukmirovic, Shao, Wang, Nilekar, Mavrikakis, Valerio, Uribe, Top. Catal. 46 (2007), 249

Pt Monolayer Electrocatalysts

M

PtML/M

Ultra-low Pt content

High utilization of Pt Tunable activity via strain and/or

electronic effects from the interaction between PtML and substrates

For ~5 nm Pt nanoparticles (NPs), ~25% of atoms are on the surface

This electrocatalyst is commercially available from N.E. ChemCat Co., based on four patents licensed by BNL to NE CC

6

C u / P d (1 1 1 )

Adzic, Zhang, Sasaki, Vukmirovic, Shao, Wang, Nilekar, Mavrikakis, Valerio, Uribe, Top. Catal. 46 (2007), 249

Pd Cu Pt

Cu UPD

Galvanic Replacement

Pt2+

Galvanic displacement of Cu ML deposited at underpotentials (UPD) - a ML-limited process)

Structure-controlled syntheses of Pt Monolayer Electrocatalysts

Brankovic et al. Surf. Sci., 477, L173 (2001)

0 .2 0 .4 0 .6 0 .8 1 .0

-0 .3

-0 .2

-0 .1

0 .0

0 .1

0 .2

0 .3

0 .4

j / m

A.c

m-2

E / V R H E

Cu / Pd(111) Pd(111)

UPDOPD

Z=78-Pt

Z=46-Pd

STEM, EELS evidence of a Pt ML shell

7

                                                

   

8

Pt/Ru(0001)

Compression

Pt/Pd(111)

Small compression

Pt/Au(111)

Expansion

Adzic, Zhang, Sasaki, Vukmirovic, Shao, Wang, Nilekar, Mavrikakis, Valerio, Uribe, Top. Catal. 46 (2007), 249 Zhang, Vukmirovic, Xu, Mavrikakis, Adzic, Angew. Chem. Int. Ed. 44 (2005), 2132

strain

d-band center shift

EOH

0.0 0.2 0.4 0.6 0.8 1.0 1.2

-6

-4

-2

0 rpm 1600 PtML/ Ru (0001) PtML/ Ir (111) PtML/ Rh (111) PtML/ Au (111) Pt (111) PtML/ Pd (111)

j / m

Acm

-2

E / V RHE

0

2

4

6

8

10

I ring /

uA

PtML/ Ru (0001) PtML/ Ir (111) PtML/ Rh (111) PtML/ Au (111) PtML/ Pd (111)

Core-induced surface strain

Trends in surface reactivity can be described by the position of d-band center (εd)

Small contraction of a Pt lattice (decreased reactivity) makes it more active for the ORR.

Ring

Disk

O2

H2O2

Factors affecting PtMLORR Activity

99

The concept of the OH –OH repulsion

Decreasing the OHads at Pt ML electrocatalysts;

0

0.2

0.4

0.6

0.4 0.6 0.8 1 1.2

Pt in PtIr/Pd/C

Pt/C

Ir in PtIr/Pd/C

(

- 0.

47V)/ 0

.47V

E / V RHE

-0.2 0.0 0.2 0.4 0.6 0.8 1.0

20

30

40

50 Re

Os Ir

Ru

Rh

Pt Pd

Au

M0.2Pt0.8/Pd(111)

-jk a

t 0.8

V m

A/cm

2

OH-OH or OH-O repulsion

Sketch of the OH-OH or OH – O repulsion

J. Am. Chem. Soc. Comm. 127 (2005) 12481

XANES evidence that Pt is stabilized against oxidation

A consensus: High OHads coverage from H2O reduces the ORR rate : Tarasevich 1977, Adzic 1989, Gottesfeld 1989

OH-OH, or OH-O repulsion from DFT

9

Activity vs. OH-OH, or OH-O repulsion for Pt-M mixed-monolayer (Pt0.2-M0.8).

            

Tuning the activity of PtML electrocatalysts by core ‐ shell interaction

1.Composition & structure

2.Shape and facets

core

PtML

Geometric and electronic effects: - to induce a small contraction in PtML; - small number of low-coordination sites - smooth core surfaces (111) oriented - weak bonding of OH

Recent findings: - Particle size-induced surface contraction of a top ML (affects the BEO; facet – dependent) - Coordination-dependent surface atomic contraction

10Wang, J. X., H. Inada, L. Wu, Y. Zhu, Y. Choi, P. Liu, W-P. Zhou, and R. R. Adzic, J. Am. Chem. Soc., 131, 17298 (2009).

High Activity and Stability of Pt ML Electrocatalysts

PtML/ Pd/C Pt-Ir (4:1)ML/ Pd/C Fuel Cell Tests

1 0

-1) 0.9 PtIr /Pd/C : 0.66 A mg -1 (0.24 A mgML Pt NM

-1 0.36 A/mgPt @ 0.9V -- PtML/Pd/C-1)Pt /Pd/C : 0.57 A mg -1 (0.21 A mg

ML Pt NM 0.8 -- 30% Pt/C -2

i, m

A/cm

2

E1/2 = 902 mV V0.7

Pt mass activity for ORR -3 0.6 0.12 A/mgPt @ 0.9V0.66 A mg -1 at 0.9 V

Pt

-4 0.5

0.1 M HClO4-5

1600 rpm 0.4

0.2 0.4 0.6 0.8 1 0 0.5 1 1.5 2

E, V RHE A・cm-2

3000 hour fuel cell test; H2/Air 0.6A/cm2, 80C Current density of 2 A/cm2 at 0.63 V; 0.36 membrane failure; 18% A/mgPt at 0.9V; commercial catalyst, Pt/C has

0.12 A/mgPt @ 0.9V performance loss at LANL

60,000 potential cycles - small loss of activity

11

12

Nanowires have smoother surface, less low-coordination sites, edges, more (111) facets.

Deposition of Pd on carbon surfaces in 0.1M HClO4with 1mM Pd2+.

The growth mechanism: Hupdin Pd acts as reducing promotor at terraces, while chlorides adsorb at low-coordination sites and block growth in that direction.

FFT of the TEM image showing (111) pattern of Pd(111)

Nanowires and nanorods as support Electrodeposition of Pd nanowires and nanorods on carbon

0.1V

0.1V 0.2V

0.05V

0.05V

Scale-up is simple: Cell for 25 cm2 electrode was constructed.

Bliznakov et al. JES, in press

The type of deposit: NPs, NWs or NRs, depend on the potential and Pd ion concentration

Polarization ORR curves measured on PtMLPdNW/C and PtMLPdNR/C

Polarization ORR curves measured on PtMLPdNR/C

Comparison of the mass and specific activities

Polarization ORR curves measured on PtMLPdNW/C

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

MA Total

PGM

MA Tota

l PGM

SA

MA

PtML/PdNW/COX PtML/ PdNR/COX

SA

MA

SA MA

MA

Mas

s Sp

ecifi

c Ac

tivity

(0.9

V),

A/m

g

SA

Pt/C 20 g/cm2

DOE Target 2011-20

Are

a Sp

ecifi

c Ac

tivity

(0.9

V),

mA/

cm2

13

Electrochemical deposition of PdAu NRs, PtML/Pd0.9Au0.1/GDL

PdAu_codep_m in330m V_300m C_1_03.m pr <I> vs . time dQ vs. time #

time/s 151050

<I >/

m A

-15

-20

-25

dQ / m

C

0

-50

-100

-150

-200

-250

-300

0.0

0.2

0.4

0.6

0.8

0.0

0.2

0.4

0.6

0.8 SA

MA

PtMLPdAu/CDL MEA measurements

SA

MA

Mas

s Sp

ecifi

c Ac

tivity

(0.9

V),

A/m

g

Pt NP/C (20 g/cm2) RDE measurements

Are

a Sp

ecifi

c Ac

tivity

(0.9

V),

mA/

cm2

Electrochemical deposition of NRDs and NWs cores with PtML deposition using galvanic displacement of Cu ML facilitates close to 100% utilization of Pt.

PGM content approx. 70μg/cm2

Current and charge transients

MEA test vs. RDE

0.0 0.5 1.0 1.5 2.0 0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

H2/O2

H2/Air

PtML/Pd0.9Au0.1/GDL- cathode

80 oC, 44 psi

Total PGM loading 70 g/cm2

Pt/C 20 wt%, 0.5 mg/cm2- anode Nafion 211

Pow

er d

ensi

ty, W

/cm

2

E (IR

free

), V

j, A/cm2

0.0 0.5 1.0 1.5 2.0

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

14

                      

           

Pd and Au precursors are combined with octadecylamine and a phase transfer catalyst in an organic solvent system.

Synthesis of the ultrathin bimetallic PdAu nanowires

With Koenigsmann and Wong The phase transfer catalyst dodecyltrimethyl ammonium bromide (DTAB) is used to allow for co-solubilization of NaBH4 into both the aqueous and organic phases.

15

16

Enhanced ORR kinetics on electrodeposited Pt / Pd nanowires

2.2 Ǻ

TEM image of toroidal particle formed by connecting wires’ ends. 2.2 Å is the interplanar distance of Pd(111).

FFT of the TEM image left showing the (111) pattern of Pd(111).

A very high activity of Pt ML on Pd nanowires is due to:

1. The dominant (111) pattern of Pd

2. Very high utilization of Pt

This result confirms the (111) facets of Pd are the best surface for the ORR.

Pt M GDL

Model for 100% utilization of Pt in PtML catalysts with cores electrodeposited on GDL

16

17

PtML/Pd/NiWNiW obtained by co-deposition of Ni and W on gas

diffusion layer (GDL)

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

0.0

0.2

0.4

0.6

0.8

1.0

1.2

initial after 1000 cycles

Pow

er d

ensi

ty, W

/cm

2

E(iR

free

), V

j, A/cm2

80 oC, H2/O2

Ni is partially displaced by Pd from the top layer of NiW. Electrode : 5cm2

SEM image after NiW deposition on GDL.

< 40 μgPt /cm2

Decreasing the content of Pd in cores Refractory metal alloys as cores - NiW

Pd

W

Ni

Model of NiW core with a partially displaced Ni by Pd

1:1 Ni:W ratio verified using EDS W max conc. 50%

     

        

 

Concave Tetrahedral Pd Nanocrystals

tetrahedral octahedral

SAED HRTEM

~ 30nm

J. Electroanal. Chem. (2011) In press

Dominant (111) and (110) facets

Pd(acac)2 + Formaldehyde + PVP (Pd salt) (Capping (100)) (Surfactant)

Hydrothermal preparation

Surfactant: poly(vinyl pyrrolidone) (PVP)

18

     

Concave Tetrahedral Pd Nanocrystals

0.1M HClO4, 10mV/s, 1600rpm

J. Electroanal. Chem. (2011) In press

19

Pt(111) PtML/TH Pd/C

Half-wave potential (V) 803 888

ECSA (m2/mg Pt) 205 15

Pt specific activity (mA/cm2

Pt)) 0.8 0.53

Pt mass activity (A/mg Pt)

1.6 0.82

20

TEM images of Pt(ML)-Pd20Au hollow particles fabricated using Ni nanoparticles as templates.

Core metals

SA [mA cm-2] Pt mass [A mg-1]

Pt+Pd+Au mass [A mg-1]

Pd20Au 0.85 1.62 0.57

Pd solid

Pt solid

0.50

0.33

0.96

0.16

0.25

0.16

Pt monolayer on hollow Pd nanoparticles

High activity is due to smooth surface morphology, and hollow-induced lattice contraction.

Scale-up synthesis is being developed using: 1.The cell for

electrodeposition of Pd NWs

2.The microemulsion method.

Zhang et al. Catalysis Today, htpp://dx.doi.org/10.1016/j.cattod.2012.03.040

       21

Fuel Cell Stability Tests of PtML/Pd/C

Angewandte Chemie International Edition, 49 (2010) 8602

Stability after 100,000 potential cycles

Mass

Specific

DOE target 2015: <40% loss after

30,000 cycles

Stability: 30 s at 0.7 and 0.9 V

membrane

Pd band

SEM image and EELS line scan analysis

    

Nature Chemistry, submitted

In-situ XAS

Based on the inhibition effect of Au on Pt (Science, 315 2007, 220), we synthesized PdAu (9:1) nanoparticles for Pt ML cores.

Significant retardation of Pd oxidation for Pd9Au/C compared with Pd/C.

Pd9Au/C nanoparticle cores for Pt monolayer

22

Fitting NAu-Au 2.4 (1.6) NAu-Pd 8.0 (1.1) NPd-Pd 9.2 (0.4) NPd-Au 0.9 (0.13)

Au-Au 2.771 Å Au-Pd (= Pd-Au) 2.760 Å Pd-Pd 2.756 Å

The fitting result indicates the Pd-Au (pseudo) solid-solution

PtML/Pd9Au/C contains 0.062 mgPt cm-2

Pt/C 0.102 mgPt cm-2

Fuel cell test of PtML/Pd9Au/C electrocatalyst

DOE target: Pt mass activity : < 40% loss of for 30,000 cycles; PtML/Pd9Au/C: 30% loss after 200,000 cycles; Pt/C: a terminal loss before 50, 000 cycles.

Core-protected core-shell electrocatalysts

The potential limits were 0.6 and 1.0 V; sweep rate of 50 mV s-1. 80C.

Stability after 200,000 potential cycles

23

       

0

100

200

300

400

0 50000 100000 150000 200000 250000

Pt/Pd9Au/C (0.105 mg Pt cm-2)

Pt(H)/Pd/C (0.099 mg Pt

cm -2)

30% Pt/C (0.102 mg Pt

cm -2)

mas

s ac

tivity

/ A

mg

Pt -1

at 0

.9V

number of potential cycles

Pd H Pt

H UPD

Galvanic Replacement

Pt2+

Pt Monolayer deposition via Hupd

24

25 25

The structure of Pt shells was almost retained after the tests

1. PtOH formation shifted positively. 2. Contraction of Pt and Pd lattices induced

by loss of Pd - self-healing effect; hollow may form 3. Cathodic protection effect.

Pd dissolution precludes dissolution of Pt, which would readily occur and a Pt ML would disappear.

Sasaki et al. Angew. Chem. Int. Ed., 49 (2010) 8602.

New mechanism of stability of core-shell electrocatalysts: Shell protected by the core and self-healing effect

Data supported by EXAFS, XANES, EELS, EDS, RDE, DFT results.

25

Stability of PtML/Pd9Au1 from DFT calculations

1. Model a sphere-like Pt ML on Pd9Au1 random alloy core (ca 1.7 nm) 2. Introduce a defect (vacancy) in the Pt ML at vertex (D3) 3. Calculate energy changes (E) when Au or Pd atom diffuses from

core to the defect site (EAu = -0.61 eV, EPd = -0.14 eV)

Au preferentially segregates on the surface inhibit further dissolution of Pd

This is in agreement with (Zhang, Sasaki, Sutter, Adzic et al., Science, 315 (2007) 220)

With Choi and Liu

26

                    

                                            

                                      

                 

     

                                     

Composition & structure

Shape and facets core

PtML

Conclusions

PtML electrocatalysts for ORR ‐‐‐ On the road to application and could be further improved!

- PtML/Pd9Au/C and PtML/Pd/C are practical electrocatalysts. - Self-healing-mechanism helps in providing stability of these catalysts. - Pd alloys with refractory metals provide stable and inexpensive cores. - Several new efficient syntheses include: electrochemical deposition of NWs, deposition of Pd

NWs using simple surfactant, using ethanol as a medium and reactant, using UPD H.

• EOH plays an vital important role in the activity and stability of PtML for ORR and can be tuned via the interaction between PtML and various substrates.

• Significant improvement of activity and stability over those of Pt/C. • Flexibility of the core‐shell structure enables the possibility of the further improvement.

Current Pt/C used in the laboratory tests: 400 μgPt/cm2;

PtML electrocatalysts: 40‐80 μgPt/cm2, 60‐100 μgPd/cm2;

For a 100‐kW fuel‐cell car, 1W/cm2: 4‐8g Pt + 10g Pd; current catalyst converter per car: ~ 5g Pt.

27

Acknowledgements

Collaborators

From BNL: W-P. Zhou, Y. Zhu, E. Sutter, C. Ma, C. Koenigsmann, S. Wong

Outside of BNL: M. Mavrikakis, U. Wisconsin, K. More, ORNL, N. Marinkovic, SCC

Funding

28

TOYOTA Motor Company

Electrocatalysis Group, BNL

29