Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
Science Magazine Highlight:Moving Towards Near Zero Platinum Fuel Cells
Piotr Zelenay
Co-Authors
Gang Wu, Hoon Chung, Christina Johnston, Patrick Turner, Zhongfen Ding, Jerzy Chlistunoff, Nate Mack, Mark Nelson
Los Alamos National Laboratory Los Alamos, New Mexico 87545, USA
11Fuel Cell Technologies Program Webinar – April 25, 2011DOE Fuel Cell Technologies Webinar – April 25, 2011
Outline
• Introduction:─ rationale─ recent developments in non-precious metal oxygen reduction
reaction (ORR) catalysis( ) y
• Low-temperature Oxygen Reduction Reaction (ORR) catalysts (PPy-Co-C)
• Catalysts obtained by heat treatment of organic and transition-metal precursors:─ polyaniline-derived catalysts as a best combination of activitypolyaniline derived catalysts as a best combination of activity
and stability─ cyanamide-based catalysts─ ORR activity vs. catalyst structure and composition
• Summary
Fuel Cell Technologies Program Webinar – April 25, 2011 2
• Acknowledgements
The Catalyst Cost Challenge
Stack Cost - $26/kW
Analysis:Analysis:• Scaled to high-volume production
of 500,000 units/year • Assumed Pt cost of $1100/ozChallenges:• Platinum cost representing ~34%
of total stack cost• Catalyst durability in need of
improvement
James et al., DTI, Inc., 2010 DOE Hydrogen Program Review, Washington, DC, June 9, 2010
Main strategies to address catalyst cost challenge:
g , g , , ,
• Reduction in the platinum group metal (PGM) content
• Improvements to Pt catalyst utilization and durability
• Pt alloy catalysts with comparable performance to Pt but costing less
N i l l i h i d f d d bili
Fuel Cell Technologies Program Webinar – April 25, 2011 3
• Non-precious metal catalysts with improved performance and durability
Platinum Challenge
Grand challenge: High platinum cost, price volatility, and resource concentration in virtually one location in the world (South Africa)
Fuel Cell Technologies Program Webinar – April 25, 2011 4
concentration in virtually one location in the world (South Africa)
Non-Precious Metal ORR Catalysts
• Catalysts derived from macrocycle precursors, e.g. metal porphyrins,phthalocyanine, etc.Jasinski, 1968; CWRU (Yeager, Savinell; 1980s-1990s); 3M, 2000s; UNM, 2000s;Tokyo Tech, 2000s;
• Materials synthesized from non-aromatic precursors by heat treatment, e.g.ammonia, cyanamide, ethylenediamineINRS, 1990s-2000s; University of Calgary, 2000s; University of South Carolina et al.y g y y2000s; Michigan State University, 2000s; LANL, 2000s
• Catalysts obtained via pyrolysis of aromatic precursors, including polymers(phenantroline, polypyrrole, polyaniline, etc.)INRS, 1990s-2000s; 3M, 2000s; LANL, 2000s
• Inorganic ORR catalysts obtained by high-temperature treatment, e.g.oxides, oxynitrides, yYokohama National University et al., 2000s; Dalhousie University, 2000s
• Catalysts produced without heat treatmentMIT 2000 LANL 2000 D ih t 2000 I di I tit t f T h l 2000
Fuel Cell Technologies Program Webinar – April 25, 2011 5
MIT, 2000s; LANL, 2000s; Daihatsu, 2000s; Indian Institute of Technology, 2000s
Non-Precious ORR Catalysis – Concept 1: Metal Not Participating in ORR
Nallathambi et al J Power Sources 183 34 2008Nallathambi et al, J. Power Sources 183, 34, 2008 Sidik et. al, J. Phys. Chem. B 110, 1787, 2006 Niwa et al, J. Power Sources, 187, 93, 2009
Nanostr ct red carbon doped ith nitrogen (CN )
Fuel Cell Technologies Program Webinar – April 25, 2011 6
Nanostructured carbon doped with nitrogen (CNx) often viewed as ORR active site
Non-Precious ORR Catalysis – Concept 2: Metal Participating in ORR
Lefèvre et al, Science, 324, 71, 2009
MeN4/C (Me: Co or Fe) species embedded in carbon micropores an alternative ORR active site concept
Fuel Cell Technologies Program Webinar – April 25, 2011 7
micropores – an alternative ORR active site concept
Turnover Frequency of Various ORR Catalysts
ORR turnover frequency: f (s-1)= i / (e N)i current density (Acm-2) e electron charge (1 6 10-19 C) N active site density (cm-2)i – current density (Acm 2), e – electron charge (1.6 10 19 C), N – active site density (cm 2)
Fuel Cell Technologies Program Webinar – April 25, 2011 8
Gasteiger and Markovic, Science 324, 48 2009
DOE-EERE-Funded Advanced Cathode Catalysts Project (2006-2010)
• Catalysts with ultra-low Pt content, e.g. non-precious-metal core catalysts; mixed-metal shells for higher ORR activity (Brookhaven National Laboratory, Radoslav Adzic, PI)
• Non precious metal catalysts obtained by high temperature• Non-precious metal catalysts obtained by high-temperature treatment of various precursors of carbon, nitrogen, and transition metals (Los Alamos National Laboratory)
Fuel Cell Technologies Program Webinar – April 25, 2011 9
DOE Cathode Catalyst Performance Targets
• Platinum group metal loading: 0.2 mgPGM/cm2 (both electrodes)
• Activity of PGM catalysts:Activity of PGM catalysts:
mass activity 0.44 A/mgPt at 0.90 ViR-free
area specific activity 720 μA/cm2 at 0.90 ViR-free
• Activity of non-PGM catalysts (per catalyst volume):
> 130 A/cm3 at 0.80 ViR-free (2010); 300 A/cm3 at 0.80 ViR-free (2015)
• Durability with cycling: 5,000 hours at T ≤ 80ºC, 2,000 hours at T > 80ºC
• Electrochemical surface area (ESA) loss: < 40%; Cost: < 5 $/kW
Fuel Cell Technologies Program Webinar – April 25, 2011 10
Low-Temperature Non-Precious Metal Composites: Co-PPy-XC72
• Hypothesis: CoN2 (CoN4) sites claimed to act as ORR active sites ( i l d C h i )(e.g. in pyrolyzed Co porphyrins)
• Objective: Generate ORR active sites without destroying the ordered structure of the polymer catalyst
• Approach: Heteroatomic polymer as a matrix for entrapping and• Approach: Heteroatomic polymer as a matrix for entrapping and stabilizing non-precious metal
• Choice: Cobalt-polypyrrole-carbon composite (Co-PPy-XC72)
Pyrrole + XC72
PPy-XC72
(i) C(i) Co precursor(ii) reduction
Co-PPy-XC72
Fuel Cell Technologies Program Webinar – April 25, 2011 11
Bashyam and Zelenay, Nature 443, 63, 2006
Co-PPy-XC72 Catalyst: Molecular Modeling
DO-O = 1.21 Å DO-O = 1.47 Å
+ 2H+ + 2e- →dioxygen
DO-O = 1.56 Å DO-O = 1.99 Å
+ 2e- →
[(tetrapyrrole) H2O Co-OH-HO-Co H2O (tetrapyrrole)]2-
Dioxygen interaction with two Co centers significantly weakening O-O bond
Fuel Cell Technologies Program Webinar – April 25, 2011 12
relative to the O2/H2O2 reference system
Low-Temperature Non-precious Metal Composites: Co-PPy-XC72
cm-2
)
0.20
0.25
10%Co-PPy-XC72
H2 / air; 80°C
V)
0.8
1.0
Powe0.12
0.1580°C
nt D
ensi
ty (A
0.10
0.15
0%Co y C(0.2 mg cm-2 Co)
PyrolyzedCoTPP
( 2)Cel
l Vol
tage
(V
0.4
0.6
er Density (W0.06
0.09
0 50 100 150 200
Cur
re
0.00
0.05
0.4 V
(2 mg cm-2)
Current Density (A cm-2)0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70
C
0.0
0.2
W cm
-2)
0.00
0.03
10%Co-PPy-XC72; H2 / O2
Time (hours)Current Density (A cm 2)
Bashyam and Zelenay, Nature 443, 63, 2006
• Significant improvement in the stability over the state of the art
• ORR activity in dire need of improvement
Fuel Cell Technologies Program Webinar – April 25, 2011 13
PolyanilineCyanamide Polypyrrole
High-Temperature Synthesis
NC
H
H
N
Transition metalCarbon support
First heat treatment at up to 1100oC in inert atmosphere
0.5 M H2SO4 leach at 80-90oC
S
ORR f l ti
Second heat treatment at 900ºC in inert atmosphere
Fuel Cell Technologies Program Webinar – April 25, 2011 14
ORR performance evaluation; characterization
Polyaniline-Derived Catalysts
Fuel Cell Technologies Program Webinar – April 25, 2011 15
Schematic Representation of Polyaniline-Derived Catalyst Synthesis
Metal Addition Polymerization
Adsorption
Heat Treatment
Acid leachN
E 2nd Heat Treatment
Fuel Cell Technologies Program Webinar – April 25, 2011 16
Final Catalyst
E 2 Heat Treatment
PANI Catalysts: Evolution of ORR Activity and 4e- Selectivity (RRDE)
Rotating Ring Disk Electrode (RRDE) Data
1000
Rotating Ring Disk Electrode (RRDE) Data
improving activity improvingselectivity
d (%
)
60
80
100
Pyrolyzed carbon
Carbon
y (m
A/c
m2 )
2
-1
0
PANI-Fe-C
H2O
2 yie
l
10
15
20 PANI-C
PANI-Co-Crren
t den
sity
-3
-2
Pyrolyzed carbon
PANI-C
PANI-Co-C
E-TEK Pt/C
Carbon
Potential (V vs RHE)0.2 0.3 0.4 0.5 0.6 0.7 0.8
0
5PANI-Fe-C
Potential (V vs RHE)0.2 0.4 0.6 0.8 1.0
Cur -4
PANI C60 μg-Ptcm-2
• Best-performing PANI-derived catalyst trailing Pt/C reference catalyst by no more than 80 mV at E½
Fuel Cell Technologies Program Webinar – April 25, 2011 17
• H2O2 generation reduced to ~ 0.5%
Composite catalysts derived from heteroatomic organic precursors (e.g. polyaniline - PANI, transition metals, and carbon; heat-treated at 600ºC -
Nanostructure of PANI-Derived Catalysts: HR-TEM Images
1100ºC; and subjected to post-synthesis purification and activation steps
10 nm
500 nm
20 nm
500 nm
Onion-like filled and hollow turbostraticcarbon layers, disordered nanofiber and metal-containing nanoparticles visible after pyrolysis in PANI-derived catalyst
Fuel Cell Technologies Program Webinar – April 25, 2011 18
PANI-Fe-C: Effect of Heat-Treatment Temperature on Activity and Speciation
0 50
ty (m
A/cm
2 )
-2
-1
0
PANI-Fe-C
Carbon
d (%
)
30
40
50400ºC 600ºC 800ºC 850ºC 900ºC 950ºC
urre
nt d
ensi
t
-4
-3
400ºC 600ºC 800ºC 850ºC 900ºC 950ºC 1000ºC
H2O
2 yie
l
10
20
1000ºC
Potential (V vs RHE)0.2 0.4 0.6 0.8 1.0
Cu
-5
1000ºC
El t l tifi ti b XPS
Potential (V vs NHE)0.2 0.3 0.4 0.5 0.6 0.7 0.8
0
• Catalysts synthesized at 900ºC showing the highest ORR activity and the lowest H2O2 yield (~1.0%)en
t (at
%)
60
80
100
C 1s
tent
(at%
)
4 0
6.0
8.0
10.0
N 1s
Elemental quantification by XPS
2 2 y ( )
• Gradual loss in the total nitrogen content with heat-treatment temperature representing the most notable result from XPS analysis of
Ele
men
t con
te
0
10
20
60
O 1s
Ele
men
tal c
ont
0 0
0.3
0.6
0.9
4.0
Fe 2p
Fuel Cell Technologies Program Webinar – April 25, 2011 19
notable result from XPS analysis of the PANI-Fe-C catalyst
Temperature (ºC)0 200 400 600 800 1000
0
Temperature (ºC)0 200 400 600 800 1000
0.0
Polymer Electrolyte Fuel Cell (PEFC)
e-e-
1.5 O2 or 0.5 O2 (air)e
H2OCH3OH or H2
H2O
1.5 O2 or 0.5 O2 (air)e
H2OCH3OH or H2
H2O
H+
CO +
CH3OH
Electroosmotic
2
H+
CO +
CH3OH
Electroosmotic
2
ANODE CATHODE
CO2 +6 H+ + 6 e-
drag2 H+ + 2 e-
MEMBRANE3 H2O or H2O
ANODE CATHODE
CO2 +6 H+ + 6 e-
drag2 H+ + 2 e-
MEMBRANE3 H2O or H2O
Fuel Cell Technologies Program Webinar – April 25, 2011 20
X-Ray Tomography: MEA Image
Fuel Cell Technologies Program Webinar – April 25, 2011 21
Fuel Cell Testing of PANI-derived ORR Catalysts: Performance
1.0
(V)
0 6
0.8
H2-O2/2.8-2.8 bar
PANI-Fe-C
Vol
tage
(0.4
0.6
PANI-FeCo(3:1)-C
0 0 0 5 1 0 1 5 2 0
V
0.0
0.2PANI-C PANI-Co-C
Current density (A/cm2)0.0 0.5 1.0 1.5 2.0
• Relative performance observed in RDE experiments reproduced in fuel cell testing (except for relatively lower PANI-Fe-C performance, possibly due to limited stability)
• PANI-FeCo(3:1)-C showing the best activity
Fuel Cell Technologies Program Webinar – April 25, 2011 22
Fuel Cell Testing of PANI-derived ORR Catalysts: Performance Stability
0.5
(A/c
m2 )
0.4PANI-FeCo(3:1)-C
t den
sity
0.2
0.3
PANI-Fe-C
Cur
rent
0.1
PANI-Co-C
H2-air/2.8-2.8 bar
Time (hours)0 200 400 600
0.0
• PANI-FeCo(3:1)-C showing the best combined activity/durability, never before observed with non-precious metal catalysts in fuel cell testing
• Co likely to responsible for improved activity, Fe for stability
Fuel Cell Technologies Program Webinar – April 25, 2011 23
y p p y, y
ORR ActivityORR Activityvs.
Catalyst Structure and Composition
Fuel Cell Technologies Program Webinar – April 25, 2011 24
PANI-FeCo(3:1)-KJ and PANI-Fe-MWNT: Final Catalysts
PANI-FeCo(3:1)-KJ: Three images, same spot
• The two most durable catalysts showing notable similarities in morphology
• Significant number layered
0.1 μm
graphene “bubbles” formed in PANI-FeCo(3:1) and PANI-Fe-MWNT (→), co-located with the Fe(Co)Sx
TEM HAADF-STEM SEM
regions/particles (→); MWNT still present (→)
• BF-STEM images of PANI-Fe-MWNT showing graphitic
PANI-Fe-MWNT: Three images, same spot
carbon particles that surround/encapsulate FeSx
• Relationship of graphenesheets (e.g. their increased
metal-rich phase
graphene
( ghydrophobicity) to durability requiring further exploration
MWNT
Fuel Cell Technologies Program Webinar – April 25, 2011 25
TEM SEMHAADF-STEM0.1 μm
1.4
1.6
1.4
1.6
PANI-Fe-C XAS: Comments on Interpretation of Spectra
• Multi-component nature of samples rendering analysis of EXAFS region of Fe K-edge spectra virtually
Fe Metal/Fe Oxides Standards
0 6
0.8
1.0
1.2
norm
mu(
E)
F l0 6
0.8
1.0
1.2
norm
mu(
E)
F l
g g p yimpossible – only XANES region proving useful
• Linear combination of XANES spectra of standards providing mole fraction of Fe species in coordination environments similar to those of the given standards
0.0
0.2
0.4
0.6
7090 7110 7130 7150 7170 7190
Fe metalFeOFe3O4Fe2O3
0.0
0.2
0.4
0.6
7090 7110 7130 7150 7170 7190
Fe metalFeOFe3O4Fe2O3
environments similar to those of the given standards
• XAFS edge step and/or destructive ICP-OES analyses yielding total loading of Fe in materials
Fit of Spectrum by Linear
1.61.6
7090 7110 7130 7150 7170 7190Energy (eV)
7090 7110 7130 7150 7170 7190Energy (eV)
1.21.2Fe Sulfides Standards Examples of
Fe-Complexes Standards
Fit of Spectrum by Linear Combination of Standards
0 8
1.0
1.2
1.4
mu(
E)
0 8
1.0
1.2
1.4
mu(
E)0.6
0.8
1.0
norm
mu(
E)
FeS0.6
0.8
1.0
norm
mu(
E)
FeS
0.2
0.4
0.6
0.8
norm
Porphine(Fe3+)
Fe(3+)-pc (tetra)
Fe(2+)-pc
Ferrocene(Fe2+)0.2
0.4
0.6
0.8
norm
Porphine (Fe3+)
Fe(3+)-Pc-O2
Fe(2+)-Pc
Ferrocene (Fe2+)0.0
0.2
0.4
7090 7110 7130 7150 7170 7190
FeS2
0.0
0.2
0.4
7090 7110 7130 7150 7170 7190
FeS2
Fuel Cell Technologies Program Webinar – April 25, 2011 26
0.07090 7110 7130 7150 7170 7190
Energy (eV)
0.07090 7110 7130 7150 7170 7190
Energy (eV)
Energy (eV)Energy (eV)
PANI-Fe-C: Effect of Heat-Treatment on Activity and Fe Speciation
cm2 ) 0
Den
sity
(mA
/c
-3
-2
-1
400oC 600oC 800oC 850oC
Correlation between activity and presence of Fe-Nx-type coordination observed; Fe-Cx-type and Fe oxides
Potential (V vs RHE)0.2 0.4 0.6 0.8 1.0
Cur
rent
-5
-4850oC 900oC 950oC 1000oC
x ypcorrelating to a lesser degree
2.01 0Total FeN Species O t lli (F C ) M t lli (F F )
Fe X-Ray Absorption Spectroscopy Results
Best Match with Activity Fe-Cx and Fe-Nx Other Fe Species
wt%
Fe
0.5
1.0
1.5
t% F
e in
Sul
fides
0.5
1.0
0.5
1.0Total FeNx Species Organometallic (Fe-Cx)
Porphine (Fe-N4)
Metallic (Fe-Fe)
Oxides (Fe-Ox)
mA/
mg
0.5
1.0
1.5
2.0
0.5
1.0
0.5
1.0
wt%
Fe
or 0
.1 w
t
ORR Activity (at 0.80 V) Fe-N6
Phthalocyanine (Fe-N4)
Sulfides (Fe-Sx)
Sulfate (Fe-SO4)
Fuel Cell Technologies Program Webinar – April 25, 2011 27
500 600 700 800 900 1000500 600 700 800 900 10000.0
Heat-Treatment Temperature (°C)500 600 700 800 900 1000
0.0
Heat-Treatment Temperature (°C)
PANI-Fe-C: Effect of Heat-Treatment Temperature on Surface Area and Pores
• Sample heat-treated at 900ºC exhibiting the highest BET surface area and micropore volume/surface areamicropore volume/surface area
• Clear correlation observed between mass activity and micropore surface area
1.00
10.00
y at
0.8
0 V
A/g)
900°C
800°C
850°C950°C
0.01
0.10
Mas
s Ac
tivity
(RD
E, A 800 C
1000°C
Fuel Cell Technologies Program Webinar – April 25, 2011 28
10 100 1000
Micropore Surface Area
PCA: Correlation of Combined Characterization Data and Activity
0 3
0.4
Fe o ide
Fe sulfate
Fe s lfideFe disulfide 900°C
Notes on Principle Component Analysis (PCA)
0.1
0.2
0.3
%)
Fe oxideFe(II) phthalocyanine
Fe sulfide
Fe organometallic
Total Fe
C=C
T t l S
ORR Activity
1000°C950°C 15.6-19.9 Å 7.3-8.9 Å
(PCA)
Samples and variables with mean composition, activity, BET or pore
‐0.1
0
PCA2 (32.99% Total S
Quaternary N
850°C800°C
N-Fe/pyridinic N
Fe carbide
9.5–10.9 Å
Fe(III) phthalocyanine
size would lie on the intersection of axes on this biplot.
The further the variables or samples
‐0.3
‐0.2
P
FeN6
Fe porphine
XANESXPS
Pore size
400°C600°C
Amines
11.7-12.4 Å
Fe(III) phthalocyaninefrom the intersection in any direction (vertical, horizontal or diagonal), the more different they are from the average
‐0.4
‐0.5 ‐0.4 ‐0.3 ‐0.2 ‐0.1 0 0.1 0.2 0.3 0.4
PCA1 (36.01%)
Heat-Treatment TAmines
Total Oare from the average values.
(As-synthesized powders examined.)
( )
• Fe-N-type coordination, correlated with ORR activity and ORR-active samples (900°C, 850°C, 800°C), remaining a primary candidate for active site
• The most active, 900°C catalyst having the highest BET surface area, a
Fuel Cell Technologies Program Webinar – April 25, 2011 29
bimodal pore distribution, high initial content of Fe carbide, Fe in oxide-like coordination environment, and Fe in phthalocyanine-like coordination
Cyanamide-Fe-C Catalyst
Fuel Cell Technologies Program Webinar – April 25, 2011 30
2 45
Cyanamide-Fe-C Catalyst: RRDE Performance
0
% H
2O2
123
y (m
A cm-2
) -1
0
CM-Fe-C(2)E1/2 0.77 V
rren
t Den
sity
-3
-220 μgPt cm-2
E1/2 0.84 V
Limiting current calculated fromLevich equation for 900 rpm atLos Alamos' latitude:3.5 mA cm-2
Potential (V vs RHE)0.2 0.4 0.6 0.8 1.0
Cur
-4
3.5 mA cm
( )
• ~ 70 mV difference in E1/2 between CM-Fe-C catalyst and Pt reference catalyst (20 μgPt cm-2)
Fuel Cell Technologies Program Webinar – April 25, 2011 31
Pt• H2O2 generation: ~ 1%
V)
1.0
Cyanamide-Fe-C Catalyst: Performance
1.0 pO2 2.4 bar pO2 1.0 bar
l Vol
tage
(V
0 8
0.9165 A cm-3
ltage
(V)
0.6
0.8
iR-corrected
iR-F
ree
Cel
l
0.7
0.8
Cel
l Vol
0.2
0.4 iR-corrected
Improved Version 2Improved Version 1
Volumetric Current Density (A cm-3)0.1 1 10 100 1000
i
0.6
Current Density (A/cm2)0.0 0.5 1.0 1.5
0.0
Improved Version 1April 2009 Version
• Version 1 achieved with change in carbon support and adjustment of precursor ratios
• Version 2 generated by including additional sulfur-containing precursor
0.39 A cm-2 – measured per MEA surface area at 0.80 V (iR-corrected)60 A cm-3 – measured per electrode volume at 0.80 V (iR-corrected)165 A cm-3 – extrapolated per electrode volume at 0.80 V (iR-corrected)
• CM Fe C to Pt performance ratio at 0 60 V (standard Pt loading of 0 2 mg/cm2): ~ 0 65
Fuel Cell Technologies Program Webinar – April 25, 2011 32
• CM-Fe-C to Pt performance ratio at 0.60 V (standard Pt loading of 0.2 mg/cm2): ~ 0.65
Performance Summary: Non-PGM Catalysis Research at Los Alamos
200
/cm
3 )
100
150
200
DOE 2010 Target (130 A/cm3)
.80
V iR-fr
ee (A
/
12
50measured (with mass-transport losses)mass transport loss-free(Tafel extrapolation)
j V at
0.
0
4
880°C, pO2
= 1.0 bar
Year2007 2008 2009 2010
0
• Fuel cell performance improved by more than 100× since 2008
• DOE 2010 activity target of 130 A/cm3 at 0.80 V achieved
Fuel Cell Technologies Program Webinar – April 25, 2011 33
• PANI-FeCo-C catalyst (possibly the most promising non-PGM catalysts to date):
Summary
PANI FeCo C catalyst (possibly the most promising non PGM catalysts to date):
(i) high activity(ii) respectable performance durability(iii) excellent selectivity for four electron reduction process
Gang Wu, Karren L. More, Christina M. Johnston, Piotr ZelenayScience, 443-447, 332, 2011
( ) y p
• Open cell voltage (OCV) of 1.04 V and volumetric ORR activity of 165 A/cm3electrode
(after mass-transport correction) achieved with CM-Fe-C catalyst in fuel cell testing
• High durability demonstrated with PANI-based catalysts to potential holding at OCVHigh durability demonstrated with PANI based catalysts to potential holding at OCV and 0.4 V in fuel cell testing; much of the performance loss at 0.60 V recoverable with reduced humidity
• ORR activity of PANI-derived catalysts correlated to microporosity and Fe-N y y p ycoordination; improved durability linked to graphene sheet formation (results of advanced spectroscopic and microscopic characterization)
• Immediate future research:(i) active ORR site determination(ii) improvements to stability and activity
Fuel Cell Technologies Program Webinar – April 25, 2011 34
Potential
Non-Precious Metal Oxygen Reduction Catalysts vs. Platinum
CatalystPotential
at 0.10 mA/cm2
(V vs. RHE)
EDA-Co-C 0.81
PANI-Fe-C 0.94
PANI-Fe/EDA-Co-C
0.96EDA-Co-C
20 wt% Pt/C(60 µg cm-2)
1.00
• ORR activity: At least 40 mV needed, especially at low overpotential (η); low Tafel slope of help at higher η values
• Stability: In spite of a major progress, still not there yet (factor of ~10)Stability: In spite of a major progress, still not there yet (factor of 10)• Selectivity: Sufficient with the best performing catalysts• Novel electrode structures required to accommodate high non-precious
catalyst volume and prevent O2 mass-transfer loss
Fuel Cell Technologies Program Webinar – April 25, 2011 35
y p 2
Acknowledgements
U.S. Department of Energy for funding of “Advanced Cathode Catalysts” project through Office of FuelCathode Catalysts project through Office of Fuel Cell Technologies (EERE)
Los Alamos National Laboratory for initial funding of non-precious metal catalysis research through Laboratory Directed Research and Development (LDRD) Program
Deborah Myers, Magali Ferrandon, Jeremy Kropffor X-ray absorption spectroscopy characterization
Karren More for characterization by electronKarren More for characterization by electron microscopy
Kateryna Artyushkova for principal component l i
Fuel Cell Technologies Program Webinar – April 25, 2011 36
analysis
Science Paper Reference
Gang Wu, Karren L. More, Christina M. Johnston, Piotr ZelenayScience, 443-447, 332, 2011
Fuel Cell Technologies Program Webinar – April 25, 2011 37
Backup Slides
Fuel Cell Technologies Program Webinar – April 25, 2011 38
Raman Spectra: Carbon Black vs. PANI-derived Catalyst
Raman spectra of (a) Ketjenblack support and (b) PANI Fe C
G PeakD Peak Distortion(a) (b)
Raman spectra of (a) Ketjenblack support and (b) PANI-Fe-C catalysts subject to the single and double heat-treatment at 900ºC
sity
(a.u
.)
As-receivedKetjenblack
sity
(a.u
.)
G PeakD Peak
Distortion(C5-ring or
heteroatom)
Inte
ns Ketjenblack, AH1
Ketjenblack, AH2
Inte
ns PANI-Fe-C, AH1
PANI-Fe-C, AH2
Raman shift (cm-1)600 800 1000 1200 1400 1600 1800 2000
Raman shift (cm-1)600 800 1000 1200 1400 1600 1800 2000
Both the support and polymer-derived carbon contributing to the catalyst morphology
Fuel Cell Technologies Program Webinar – April 25, 2011 39
RDE: 0.6 mg cm-2; 0.5 M H2SO4; 900 rpm; Cycling:50 mV/s, 0.0-1.0 V in 0.5 M N2-saturated H2SO4
Anode: 0.25 mg cm-2 Pt; Cathode: 2.0 mg cm-2 PANI-Fe-C; Membrane: NRE-212; Cell: 80°C; H2-O2 /1.0-1.0 bar; Cycling: 0.6-1.0 V, 50 mV/s, N2 at 100% RH
PANI-Fe-C: More on Durability (Cycling)A/
cm2 )
-1
0
0.8
1.0After cycling at 80°C
(fuel cell)After cycling at 60°C(aqueous electrolyte)
0.8
0.9Kinetic Range
t Den
sity
(mA
-3
-2 20 cycles500 cycles1K cycles5K cycles10K cycles Vo
ltage
(V)
0.4
0.6
1K cycles5K cycles10K l
0.0 0.1 0.20.6
0.7
P t ti l (V RHE)0.2 0.4 0.6 0.8 1.0
Cur
rent
-5
-4
10K cycles
C t D it (A/ 2)0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
V
0.0
0.210K cycles20K cycles30K cyclesInitial
cycling increasing
• Only ∼10 mV loss in E½ at RDE testing after 10,000 potential cycles at 60ºC
• After 30,000 cycles in the fuel cell, performance increase observed with PANI-Fe-C cathode in a H -O cell at lower voltages than 0 65 V; ca 25% loss in current density
Potential (V vs RHE) Current Density (A/cm2)
cathode in a H2-O2 cell at lower voltages than 0.65 V; ca. 25% loss in current density observed at 0.80 V (kinetic range)
• Increase in mass-transport controlled performance in fuel cell experiments highlighting the need to understand changes in catalyst and/or ionomer structure, including porosity
Fuel Cell Technologies Program Webinar – April 25, 2011 40
• ICP results showing 30-35% Fe lost after 7,000 cycles (0.40-1.0 V); by XAS, mostly FeSremoved, whereas most of the Fe in Fe-N4 coordination retained
Ox + e- ⇔ Red kfwd = ks exp[-αF(E-Eº)/RT]kbw = ks exp[(1-α)F(E-Eº)/RT]
00.40.8
A)
Mechanistic Analysis of ORR Kinetic Data for PANI-Fe-C Catalyst
Red + O2 + H+ → Ox + (HO2) * rds (kchem - rate constant)
bw s p[( ) ( ) ]
(HO -)* + 3e- + 3H+ 2H O fast reaction
-0.8-0.4
0
0 0.2 0.4 0.6 0.8 1 1.2
i (m
A
E (V) vs. RHE
Background CV in argon
2
3
ensi
ty (A
cm
-3))
Loading 0.7 mg cm-2
Fitting equation:log (i) = log(nFAΓk) + log{{exp[-αF(E-Eº)/RT]/
(HO2-)* + 3e- + 3H+ → 2H2O fast reaction
0
1
met
ric C
urre
nt D
e
Loading 0.03 mg cm-2
log (i) = log(nFAΓk) + log{{exp[-αF(E-E )/RT]/{exp[-αF(E-Eº)/RT] + (k/ks) + exp[(1-α)F(E-Eº)/RT]}}Γ - surface concentration of mediator sites (Γ = Γred + Γox)k = kchemCH+CO2; A - real catalyst surface area; n - number of electrons exchanged in ORR
-2
-1
0 0.2 0.4 0.6 0.8 1 1.2
Log
(Vol
um
g
Average parameter values obtained:α = 0.25; k/ks = 12.6; Eº = 0.662 V (Eº measured for the reversible surface system - 0.646 V)
Steady-state ORR
E vs. RHE (V)
• Variable Tafel slopes in RDE experiments unrelated to catalyst-layer porosity; ORR Tafel 80-90 mV/decade at low overpotential
• Intrinsic catalytic properties of the PANI catalyst responsible for the Tafel plot curvature
Fuel Cell Technologies Program Webinar – April 25, 2011 41
Intrinsic catalytic properties of the PANI catalyst responsible for the Tafel plot curvature• ORR likely mediated by a fast red-ox system on the catalyst surface
Fe0
Fe metal
PANI-Fe-C XAS: XANES Iron Standards
Fe metal
Fe2+ compoundsFe0Fe2+ phthalocyanine (at right)Fe NFe2NFe4NFeSO4 anhydrousFeSO4 7H2OFe II acetylacetonate
di(cyclopentadienyl) iron (ferrocene)
iron(II) phthalocyanine
High degree of symmetry of Fe2+ square planarFerrocene (at right)
FeSFeS2 tris(2,2'-bipyridine) Iron(II)
hexafluoro-phosphate (at right)
( )(organometallic)
Fe2 square planar environment (e.g., Fe-pc) causing unique spectral feature near 7115 eV; inclusion of this standard required to fit data wellhexafluoro-phosphate (at right)
1,10-phenanthroline Iron(II) sulfate complex (at right)
Fe3+/2+ compoundsF O
q
Fe3O4
Activity of non-precious catalysts attributed by many groups to FeN4–type or FeN type sites
tris(2,2'-bipyridine) 1,10-phenanthroline
iron(II) sulfate complex(F N )
Fuel Cell Technologies Program Webinar – April 25, 2011 42
type or FeN2+2–type sites iron(II) hexafluoro-phosphate
(FeN6)
(FeN6)
PANI-Fe-C XAS: XANES Iron Standards
Fe3+ compoundsFeCl3 6H2OFe III acetylacetonateFe(NO3)3 9H2O 2 3 7 8 12 13 17 18-octaethyl-Fe2O3Fe porphine (at right)Fe3+ phthalocyanine (at right)modified Fe+ porphine
(FeTPPS, below)
2,3,7,8,12,13,17,18-octaethyl-21H,23H porphine iron(III)
acetate (porphine)
(FeTPPS, below)
iron(III) phthalocyanine-4,4′,4′′,4′′′-tetrasulfonic acid, compound with
oxygen monosodium salt
5 10 15 20 t t ki (4 lf t h l) 21H 23H hi i (III) hl id
Fuel Cell Technologies Program Webinar – April 25, 2011 43
5,10,15,20-tetrakis(4-sulfonatophenyl)-21H,23H-porphine iron(III) chloride (FeTPPS)