Transcript MS PowerPoint

A Most Efficient Oxygen
Reduction Electrode Is this realizable?
1
INTRODUCTION
Electrons
Air/O2
Importance of electrochemical
reduction of oxygen
 Fuel Cells
 Metal-Air batteries
 Industrial electrolytic processes
H+
H+
H+
H+
H+
Water
H+
Oxygen reduction reaction (ORR) is
a four electron process:
O2 + 4 H+ + 4 e- → 2 H2O
Eo = 1.229 V vs. NHE
Cathode
2
Reaction pathways for oxygen reduction reaction
Path A – direct pathway, involves four-electron reduction
O2 + 4 H+ + 4 e-  2 H2O ; Eo = 1.229 V
Path B – indirect pathway, involves two-electron reduction followed
by further two-electron reduction
O2 + 2 H+ + 2 e-  H2O2 ; Eo = 0.695 V
H2O2 + 2 H+ + 2 e-  2 H2O ; Eo = 1.77 V
Halina S. Wroblowa, Yen-Chi-Pan and Gerardo Razumney, J. Electroanal. Chem., 69 (1979) 195
3
Essential criteria for choosing an electrocatalyst for oxygen reduction
 Reversible
 Oxygen adsorption capacity
 Structural stability during oxygen adsorption and reduction
 Stability in electrolyte medium and also in suitable potential window
 Ability to decompose H2O2
 Good conductivity
 Low cost
4
Why Pt ?
 High work function ( 4.6 eV )
 Ability to catalyze the reduction of oxygen
 Good resistance to corrosion and dissolution
 High exchange current density
Volcano plot
J. J. Lingane, J. Electroanal. Chem., 2 (1961) 296
5
Difficulties with Pt
 Slow ORR due to the formation of –OH species at +0.8 V
O2 + 2 Pt  Pt2O2
Pt2O2 + H+ + e-  Pt2-O2H
Pt2-O2H  Pt-OH + Pt-O
Pt-OH + Pt-O + H+ + e-  Pt-OH + Pt-OH
Pt-OH + Pt-OH + 2 H+ + 2 e-  2 Pt + 2 H2O
 Scarce and expensive
Cyclic voltammograms of the Pt electrode in
helium-deaerated () and O2 sat. (- - -) H2SO4
Charles C. Liang and Andre L. Juliard, J. Electroanal. Chem., 9 (1965) 390
6
Objective(s)
 Reduction of Pt loading in the electrodes
 Replacement of Pt by non-noble metal based electrocatalysts
Strategy
Cathode (Oxygen Reduction) catalysts
Pt based catalysts
Non-Pt based catalysts
Different sized Pt nanoparticles loaded carbon
RuxSey/CDX975 catalysts
Pt-M (M = Fe, Co and Cr) alloys loaded carbon
Carbon supported Pd alloys;
Pd-Co-M(7:2:1)/CDX975 (M = Mo and Au)
Pyrolyzed Fe and Co macrocycles supported
7
on CDX975
Salient features of carbon black support (CDX975)
(A)
(B)
Particle size: 80-100 nm
100 nm
BET specific surface area is ~300 m2/g
Dominant pore diameter is in the
range of 2–50 nm
(A) SEM image (B) TEM image (C) XRD pattern and (D) pore size distribution
of the carbon black particles (CDX975)
8
Synthesis, Characterization and Oxygen Reduction Activity of Carbon
Supported Pt Nanoparticles Prepared by Polyol Reduction Method
0.5 mmol Pt(acac)2 + 2 mmol 1,2-hexadecanediol + 25 ml solvent
Heated to 100 oC
N2
heated to 100 oC under N2 atm
0.5 mmol capping agents
Thermometer
heated and refluxed for different
Capping agents
time intervals
cooled to room temperature
Black suspension
Heated and refluxed
for 30 min
Reactants:
Pt(acac)2, 1,2-hexadecanediol and solvent
H
R
H
C
CH2
OH
OH
where R = C14H29
- H2O
O
Pt
R
C
H
C
R
H
H
O
O
H
C
C
C
C
+2
H
H
R
Pt0
9
Purification
25 ml of the suspension + 100 ml of ethanol (EtOH)
centrifuged for 15 min at 7200 rpm
black precipitate
40 ml of hexane, 20 ml of EtOH, small amount
of capping agents
centrifuged for 15 min at 7200 rpm
Black suspension
50 ml of EtOH
centrifuged for 15 min and dried
Pt nanoparticles
10
10 nm
TEM images of Purified Pt nanoparticles
The Pt 4f region of the XPS spectrum of the
Pt nanoparticles loaded carbon support (CDX975)
 TEM shows the monodispersed Pt nanoparticles of size ~3.8 nm
Pt 4f spectrum shows a doublet with a low-energy peak (Pt 4f7/2) and
a high-energy peak (Pt 4f5/2) at 71.2 and 74.7 eV, respectively. These peaks
indicated that Pt is present in metallic state, Pt(0)
20 nm
11
Reactants and conditions employed in polyol reduction method for the
preparation of Pt nanoparticles of different sizes
Reactants
Temp (oC)
and time
(min)
Size of Pt
(nm)
Pt(acac)2, 1,2-hexadecanediol, oleic
acid, oleyl amine, octyl ether
290-300
30 min
1.7-1.9 (Pt1)
,,
290-300
45 min
2.3-2.6 (Pt2)
,,
290-300
60 min
3.2-3.4 (Pt3)
Pt(acac)2, 1,2-hexadecanediol,
nonanoic acid, nonyl amine,
diphenyl ether
250-260
30 min
3.8-4.0 (Pt4)
250-260
45 min
4.6-4.9 (Pt5)
250-260
60 min
5.7-6.1 (Pt6)
,,
,,
12
Synthesis of 20 wt.% Pt/carbon black(CDX975)
Carbon black (CDX975) dispersed in hexane
required amount of Pt nanoparticles
dispersed in hexane
ultrasonic agitation
20 wt.% Pt/CDX975
Carbon black
support
Ultrasonication
Pt sol
Pt
Pt
Carbon black
Carbon black
Ultrasonic
agitation
solvent rich
adsorption layer
13
(111)
(200)
(220)
(311)
(222)
XRD patterns of the as-synthesized Pt/CDX975 catalysts (A) Pt1C
(B) Pt2C (C) Pt3C (D) Pt4C (E) Pt5C and (F) Pt6C
14
A
B
C D
E
F
A
B
C D
E
F
TEM images of the as-synthesized Pt/CDX975 catalysts (A) Pt1C (B) Pt2C
(C) Pt3C (D) Pt4C (E) Pt5C and (F) Pt6C
A
B
C
D
F
E
EDX spectra of the as-synthesized Pt/CDX975 catalysts (A) Pt1C (B) Pt2C
(C) Pt3C (D) Pt4C (E) Pt5C and (F) Pt6C
15
Electrochemical measurements
Electrode fabrication
Catalyst dispersed in isopropanol (5 mg/5 ml)
ultrasonicated for 10 min
20 µl pipetted onto the GC disk
dried in flowing Ar at room temperature
10 µl Nafion was pipetted onto the catalyst deposited GC disk
dried at room temperature
Electrochemical Conditions
Electrolyte: O2 saturated 0.5 M H2SO4
Working electrode: Catalyst deposited GC Disk
Reference electrode: Ag/AgCl, 3.5 M KCl (+0.205 V vs. NHE)
Counter electrode: Pt
Scan rate: 5 mV/sec
16
Linear sweep voltammograms of the as-synthesized Pt/CDX975
catalysts in Ar- and O2-saturated 0.5 M H2SO4
17
Comparison of the oxygen reduction activity of Pt nanoparticles of different sizes
Catalyst
Pt
metal loading
(wt%)
Particle size
(nm)
Onset potential
for oxygen
reduction
(mV vs. NHE)
ORR activity
at
+0.7 V vs.
NHE
(mA cm-2)
Mass
activity at
+0.7 V vs.
NHE
(A g-1)
Pt1C
19.3
1.7-1.9
+935
1.9
34
Pt2C
19.5
2.3-2.6
+930
2.5
44
Pt3C
19.3
3.2-3.4
+930
4.3
76
Pt4C
19.1
3.8-4.0
+920
3.4
60
Pt5C
19.3
4.6-4.9
+890
2.8
50
Pt6C
19.6
5.7-6.1
+880
2.2
39
Commercial Pt/C (E-TEK)
19.8
3.5-3.9
+920
3.1
55
Optimum size of Pt for oxygen reduction is 3 - 4 nm
18
Synthesis, characterization and oxygen reduction activity of
polyol reduced Pt alloy nanoparticles loaded carbon black
Why Pt alloys are more active for oxygen reduction ?
Inhibition of –OH species on Pt
Shortening of Pt-Pt interatomic distance
Increased d-band vacancies
Order of oxide formation:
Pt > Pt-Ni > Pt-Fe > Pt-Co > Pt-Cr
Order of oxygen reduction activity:
Pt < Pt-Ni < Pt-Fe < Pt-Co < Pt-Cr
Catalyst
Pt-Pt
distance
(Å)
Pt d-band
vacancies
Pt
Pt53Ni47
Pt51Fe49
Pt49Co51
Pt50Cr50
2.77
2.66
2.69
2.71
2.73
0.370
0.378
0.390
0.390
0.401
S. Mukerjee, S. Srinivasan, M. P. Soriaga, and J. McBreen, J. Electrochem. Soc.,
142 (1995) 1409
19
PtCr/C
PtNi/C
PtCo/C
PtFe/C
Pt/C
Performance of DMFC MEAs operating at 80 oC
Performance of PEMFC MEAs operating at 80 oC
 Pt alloys offer a performance gain of 50 mV compared to Pt/C
T. R. Ralph and M. P. Hogarth, Platinum Met. Rev., 46 (2002) 3
20
Proposed mechanism for oxygen reduction on Pt alloys
 Increase of 5d vacancies led to an increased 2 electron donation from O2 to
surface Pt and weaken the O-O bond
 As a result, scission of the bond must occur instantaneously as electrons are back
donated from 5d orbitals of Pt to 2* orbitals of the adsorbed O2
T. Toda, H. Igarashi, H. Uchida and M. Watanabe, J. Electrochem. Soc., 146 (1999) 3750
21
Synthesis of Pt-Fe alloy nanoparticles by polyol reduction method
0.5 mmol Pt(acac)2 + 0.5 mmol FeCl2 + 2 mmol 1,2-hexadecanediol + 25 ml Ph2O
heated to 100 oC under N2 atm
0.5 mmol nonanoic acid and
0.5 mmol nonylamine
heated to 250-260 oC and refluxed
for 30 min
cooled to room temperature
Black suspension
Similarly, Pt-Co(1:1) and Pt-Cr(1:1) alloy nanoparticles were synthesized.
22
Purification
25 ml of the suspension + 100 ml of ethanol (EtOH)
centrifuged for 15 min at 7200 rpm
black precipitate
40 ml of hexane, 20 ml of EtOH, small amount
of capping agents
centrifuged for 15 min at 7200 rpm
Black suspension
50 ml of EtOH
centrifuged for 15 min and dried
Pt alloy nanoparticles
23
B
A
C
TEM images of as-synthesized (A) Pt-Fe (B) Pt-Co and (C) Pt-Cr nanoparticles
A
B
C
TEM images of carbon supported Pt alloy catalysts: (A) Pt-Fe(1:1)/CDX975
(B) Pt-Co(1:1)/CDX975 and (C) Pt-Cr(1:1)/CDX975
24
A
B
C
EDX spectra of the (a) Pt-Cr(1:1)/CDX975 (b) Pt-Co(1:1)/CDX975 and
(c) Pt-Fe(1:1)/CDX975
25
(a) Commercial Pt/C
(b) Pt-Cr(1:1)/CDX975
(c) Pt-Co(1:1)/CDX975
(d) Pt-Fe(1:1)/CDX975
Structural parameters
Catalyst
Maximum
2θ of Pt(111)
(˚)
Lattice
parameter
afcc (Å)
Pt-Cr(1:1)/CDX975
40.48
3.857
Pt-Co(1:1)/CDX975
40.52
3.852
Pt-Fe(1:1)/CDX975
40.58
3.848
Pt/C (E-TEK)
39.8
3.920
XRD patterns of the catalysts
The diffraction peaks are characteristic of a face-centered cubic (FCC) lattice,
but the reflections are shifted to higher angles in Pt alloys compared to that of
Pt metal, indicating a contraction of the lattice due to alloy formation.
26
Electrochemical measurements
Electrode fabrication
Catalyst dispersed in isopropanol (5 mg/5 ml)
ultrasonicated for 10 min
20 µl pipetted onto the GC disk
dried in flowing Ar at room temperature
10 µl Nafion was pipetted onto the catalyst deposited GC disk
dried at room temperature
Electrochemical Conditions
Electrolyte: O2 saturated 0.5 M H2SO4
Working electrode: Catalyst deposited GC Disk
Reference electrode: Ag/AgCl, 3.5 M KCl (+0.205 V vs. NHE)
Counter electrode: Pt
Scan rate: 5 mV/sec
27
Cyclic voltammograms of the carbon supported
Pt alloys and commercial Pt/C (E-TEK) catalysts
in Ar-saturated 0.5 M H2SO4
Linear sweep voltammograms of the carbon
supported Pt alloys and commercial Pt/C
catalysts in Ar- and O2-saturated 0.5 M H2SO4
 High ORR activity of Pt alloy catalysts is attributed to the inhibition of –OH species
on Pt by the alloying element
28
Comparison of ORR activities of Pt alloys and commercial Pt/C catalysts
Particle size
(nm)
Onset potential
for oxygen
reduction
(mV vs. NHE)
ORR activity
at +0.7 V vs
NHE
(mA/cm2)
Catalyst
Metal loading
(wt%)
20% Pt-Cr(1:1)/CDX975
Pt – 15.3;
Cr – 4.1
5.1-5.3
+990
5.2
20% Pt-Co(1:1)/CDX975
Pt – 15.1;
Co – 4.5
4.9-5.2
+970
5.0
20% Pt-Fe(1:1)/CDX975
Pt – 15.4;
Fe – 4.4
6.4-6.8
+960
4.2
20% Pt/C (E-TEK)
Pt – 19.8
3.5-3.9
+925
3.1
Order of oxygen reduction activity:
Pt/C (E-TEK) < Pt-Fe(1:1)/CDX975 < Pt-Co(1:1)/CDX975 ≈ Pt-Cr(1:1)/CDX975
29
10%
22%
24%
33%
Current density-time plots of as-synthesized Pt-M(1:1)/CDX975 and commercial Pt/C
catalysts at +0.7 V vs. NHE in O2-saturated 0.5 M H2SO4
30
Synthesis, characterization and oxygen reduction activity of
carbon supported Pd and Pd alloy catalysts prepared by
Reverse Microemulsion Method
Why Pd alloys are more active than metal alone for oxygen reduction?
(1) Induced change in the density of states (DOS) at the Fermi level of Pd sites by alloy formation.
Ota’s group considered that the decreased DOS induced by the electron transfer from Co, Ni, or Cr
to Pd may weaken the chemisorption bonds between Pd and reactants such as O2, O/OH, O2−, and
H2O2, reducing their blocking effect in the O2 reduction process. As a result, the ORR kinetics may
be enhanced. For example, an alloyed Pd showed a reduced ORR overpotential by 50mV at 0.2 mA
cm−2 compared to that catalyzed by pure Pd. [K. Lee, O. Savadogo, A. Ishihara, S. Mitsushima, N.
Kamiya, K.-I. Ota, J. Electrochem. Soc. 153 (1) (2006) A20.]
(2) Induced change in ORR Gibbs free energy. Wang and Balbuena explained that if a catalyst
consists of two metals, one with a low occupancy of d-orbitals (such as Co, Ni, Cr, or V) and the
other with fully occupied d-orbitals (such as Pd, Au, and Ag), the d-orbital coupling effect between
them can significantly decrease the Gibbs free energy of the electron transfer steps in ORR, resulting
in an enhancement in ORR kinetics. [Y.X. Wang, P.B. Balbuena, J. Phys. Chem. B 109 (2005)
18902.]
31
Comparison of steady-state polarization curves of the various heat treated carbon supported Pd alloy
catalysts for ORR in single-cell PEMFC with that of the commercial (JM) Pt catalyst
The samples with Pd:Co:Mo/Au = 70:20:10 exhibit higher catalytic activity than the
Pd:Co = 80:20 sample (without Mo/Au) suggesting that the incorporation of Mo/Au
improves the ORR activity significantly.
The Pd-Co-Mo/C catalyst heat treated at 500 °C (24 nm) exhibits better
performance with lower polarization loss than the commercial Pt (3.8 nm) and PdCo-Au catalysts (13 nm) at the same loading.
Pd-Co-Mo/C and Pd-Co-Au/C shows an open circuit voltage of 0.9 V, which is
close to that found with the commercial (JM) Pt catalyst.
32
Synthesis of 20 wt% Pd-Co-Mo(7:2:1)/CDX975
Step 1
H2PdCl4 + CoCl2 + MoCl5 + Triton-X-100 + 2-propanol + cyclohexane
stirred for 30 min followed by ultrasonication
for 15 min
Microemulsion 1
Step 2
Hydrazine + Triton-X-100 + 2-propanol + cyclohexane
stirred for 30 min followed by ultrasonication
for 15 min
Microemulsion 2
33
Microemulsion 1
+
Microemulsion 2
ultrasonication for 1 hr
carbon black (CDX975)
stirred for 2 hr
filtered and washed with acetone
and water
dried for 2 hr
heat treatment under Ar-H2 atm for
1 hr at dfiferent temp
20% Pd-Co-Mo(7:2:1)/CDX975
Similarly, 20% Pd/CDX975 and 20% Pd-Co-Au(7:2:1)/CDX975 catalysts were prepared.
34
(a) Pd/CDX975
(b) as-syn Pd-Co-Mo(7:2:1)/CDX975
(c) Pd-Co-Mo(7:2:1)/CDX975 at 700 ˚C
(d) Pd-Co-Mo(7:2:1)/CDX975 at 800 ˚C
(e) Pd-Co-Mo(7:2:1)/CDX975 at 900 ˚C
(a) Pd/CDX975
(b) as-syn Pd-Co-Au(7:2:1)/CDX975
(c) Pd-Co-Au(7:2:1)/CDX975 at 700 ˚C
(d) Pd-Co-Au(7:2:1)/CDX975 at 800 ˚C
(e) Pd-Co-Au(7:2:1)/CDX975 at 900 ˚C
XRD patterns of the carbon supported Pd and Pd alloy catalysts
The diffraction peaks are characteristic of a face-centered cubic (FCC) lattice, but
the reflections are shifted to higher angles compared to that of Pd metal, indicating
a contraction of the lattice due to alloy formation.
35
Structural parameters of Pd/CDX975 and Pd-Co-M/CDX975 (M =Mo and Au) catalysts
Catalyst
Maximum
2θ of
Pt(111) (˚)
Lattice
parameter
afcc Å)
Crystallite
size (nm)
Pd/CDX975
As-synthesized Pd-Co-Mo(7:2:1)/CDX975
Pd-Co-Mo(7:2:1)/CDX975 at 700 ˚C
Pd-Co-Mo(7:2:1)/CDX975 at 800 ˚C
Pd-Co-Mo(7:2:1)/CDX975 at 900 ˚C
40.03
40.04
40.46
40.59
40.68
3.898
3.897
3.858
3.847
3.838
1.2
1.6
7.1
9.4
13.5
As-synthesized Pd-Co-Au(7:2:1)/CDX975
Pd-Co-Au(7:2:1)/CDX975 at 700 ˚C
Pd-Co-Au(7:2:1)/CDX975 at 800 ˚C
Pd-Co-Au(7:2:1)/CDX975 at 900 ˚C
40.05
40.51
40.67
40.84
3.896
3.854
3.839
3.824
2.1
6.3
8.7
10.4
Debye-Scherrer’s equation
Crystallite size,
0.89
L
1 2 cos 
where λ is the wavelength of the X-ray (1.5406 Å),
θ is the angle at the position of the peak maximum
and β1/2 is the FWHM (in radians).
36
TEM image and EDX spectrum of the 20% Pd-Co-Mo/CDX975 (700)
TEM image and EDX spectrum of the 20% Pd-Co-Au/CDX975 (800)
37
Electrochemical measurements
Electrode fabrication
Catalyst dispersed in isopropanol (5 mg/5 ml)
ultrasonicated for 10 min
20 µl pipetted onto the GC disk
dried in flowing Ar at room temperature
10 µl Nafion was pipetted onto the catalyst deposited GC disk
dried at room temperature
Electrochemical Conditions
Electrolyte: O2 saturated 0.5 M H2SO4
Working electrode: Catalyst deposited GC Disk
Reference electrode: Ag/AgCl, 3.5 M KCl (+0.205 V vs. NHE)
Counter electrode: Pt
Scan rate: 5 mV/sec
38
Cyclic voltammograms of the carbon supported Pd and Pd alloy catalysts
in 0.5 M H2SO4
39
Linear sweep voltammograms of the carbon supported Pd and Pd alloy
catalysts in Ar- and O2-saturated 0.5 M H2SO4
40
Comparison of ORR activities of Pd based catalysts with commercial Pt/C
EDX
composition
Pd:Co:M
(M = Mo, Au)
Onset potential for
oxygen reduction
(mV vs. NHE)
ORR activity
at +0.7 V vs
NHE
(mA/cm2)
20% Pd/CDX975
100: - : -
+860
1.6
20% Pd-Co-Mo(7:2:1)/CDX975 at 700˚C
77.7:12.3:10.0
+925
4.1
20% Pd-Co-Mo(7:2:1)/CDX975 at 800˚C
78.3:12.6:9.1
+905
3.0
20% Pd-Co-Mo(7:2:1)/CDX975 at 900˚C
78.7:12.5:8.8
+870
2.6
20% Pd-Co-Au(7:2:1)/CDX975 at 700˚C
70.2:11.2:18.6
+895
3.3
20% Pd-Co-Au(7:2:1)/CDX975 at 800˚C
71.3:11.6:17.1
+905
3.9
20% Pd-Co-Au(7:2:1)/CDX975 at 900˚C
71.8:11.3:16.9
+905
2.9
Commercial 20% Pt/C (E-TEK)
---
+920
3.1
Catalyst
41
14%
4%
2%
17%
23%
12%
22%
22%
Current density-time plots of carbon supported Pd and Pd alloy catalysts at +0.7 V vs.
NHE in O2-saturated 0.5 M H2SO4
42
RuxSey/C catalysts for Oxygen Reduction – A Reverse
Microemulsion Method of Fabrication of Electrode Material
Why RuxSey catalysts?
MoxRuySz, RhxRuySz, RexRuySz,
MoxRuySez
• Ru
O
X = S, Se, Te
MoxRhySz, MoxOsySz, WxRuySz
RuxSy, RuxSey, RuxTey
O2 + 4 H++ 4 e-  2 H2O
Carbon supported catalysts
Crystal structure of RuxXy catalysts
Characteristic features
 Metal cluster - reservoir of electronic charge carriers
 Capacity to provide neighbouring binding sites for reactants and intermediates
 Volume and bond distances are flexible
 Good conductivity
N. Alonso Vante, W. Jaegerman, H. Tributsch, W. Honle and K. Yvon, J. Am. Chem. Soc., 43
109
(1987) 3251
Influence of selenium
 High current densities
 Inhibition of formation of Ru oxides
 Lower amount of H2O2 production (< 3 vol%)
Tafel plots for the ORR, as obtained from RDE
experiments in O2 saturated 0.5 M H2SO4
 Enhanced stability towards electrochemical
oxidation
Ru
A: 0 Mol% Se
B: 10.01 Mol% Se
C: 14.3 Mol% Se
RuOx
XRD-spectra of catalysts prepared with
different amounts of selenium
A: 14.3 mol% Se
B: 5.27 mol% Se
C: 0 mol% Se
D: metallic Ru
Tafel slopes and over potentials for Ru-based
cluster catalysts with different Se contents
Mol%
Se
Tafel slope
/mV dec-1
Overpotential //
mV at 10 A cm-2
14.3
10.0
5.3
1.8
0
96.6
101.5
120.0
128.4
146.2
330.0
322.5
317.5
327.0
342.5
Potential dependent hydrogen peroxide production of Ru
based cluster catalysts with different selenium content
M. Bron, P. Bogdanoff, S. Fiechter, I. Dorbandt, M. Hilgendorff, H. Schulenburg and
H. Tributch, J. Electroanal. Chem., 500 (2001) 510
44
Preparation methods for transition metal chalcogenide catalysts for oxygen reduction
Method
Starting
chemicals
Operating
conditions
Typical
catalysts
Particle
size
(nm)
Reference
Low temp
pyrolysis
method
Ru3(CO)12,
chalcogen (X)
300 ˚C, sealed
glass ampoule
RuxXyOz
~5
S. Duron et al., 2000
Colloidal
method
RuCl3, Se
THF solvent
with
N(C8H17)4BEt3
RuxSeyOz
3-4
H. Tributsch et al.,
2001
Impregnation
method
Ru3(CO)12,
H2SeO3
300 ˚C, aq.
Solvent
RuxSey
4-6
M. Hilgendorff et al.,
2003
Aqueous
medium
method
RuCl3, SeO2,
NaBH4
80 ˚C, aq.
Solvent
RuxSe
~4
A. Stephen, US
patent 2004
Pyrolysis
method
Ru4Se2(CO)11
220 ˚C, He atm
Ru2Se
clusters
~2
W. Vogel et al., 2007
Alloying
method
RuCl3, SeO2
200 ˚C, H2 atm
RuxSey
4-7
L. Colmenares et al.,
2007
45
Synthesis of 20 wt% RuxSey/CDX975 catalysts
Step 1
RuCl3.xH2O + H2SeO3 + AOT + water + heptane
stirring followed by ultrasonication for 20 min
Microemulsion 1
Step 2
NaBH4 + AOT + water + heptane
stirring followed by ultrasonication for 20 min
Microemulsion 2
46
Microemulsion 1
+
Microemulsion 2
ultrasonication for 1 hr
carbon black (CDX975)
stirred for 2 hr
filtered and washed with acetone
and water
dried for 2 hr
20% RuxSey/CDX975
where x = 1 and y = 0-1
47
XRD patterns of as-synthesized RuxSey/CDX975 catalysts; (a) Ru/CDX975 (b) Ru1Se0.2/CDX975
(c) Ru1Se0.4/CDX975 (d) Ru1Se0.6/CDX975 (e) Ru1Se0.8/CDX975 and (f) Ru1Se1/CDX975
(inset shows the slow scan XRD spectra for the (110) peak of RuxSey/CDX975 catalysts)
As-synthesized RuxSey/CDX975 catalysts show the peaks at 2θ values around 38˚, 42˚, 44˚, 58˚,
69˚, 78˚ and 85˚ corresponding to the (100), (002), (101), (102), (110), (103) and (112) planes of
48
ruthenium respectively. These characteristic peaks can be assigned to hcp ruthenium.
Structural parameters
Se content
C (Å)
a (Å)
0.866 a2c (Å3)
Ru/CDX975
4.27
2.713
27.21732
Ru1Se0.2/CDX975
4.263
2.708
27.07264
Ru1Se0.4/CDX975
4.257
2.700
26.87504
Ru1Se0.6/CDX975
4.251
2.697
26.77755
Ru1Se0.8/CDX975
4.245
2.694
26.6803
Ru1Se1/CDX975
4.239
2.687
26.50432
Variation of lattice parameters with Se content
Both the lattice parameters (a, c) and the cell volume of RuxSey decrease linearly with
increasing Se content (obeying Vegard’s law) indicate the formation of RuSe solid solution.
49
a
b
1 µm
5 µm
(a) SEM image (inset shows the high magnification image) and (b) EDX spectrum
of Ru1Se0.6/CDX975
50
TEM images of (a) Ru1Se0.6/CDX975 at low magnification (b) Ru1Se0.6/CDX975
at high magnification (c) histogram of Ru1Se0.6 particles on CDX975 and
51
(d) commercial Pt/C (E-TEK)
Electrochemical measurements
Electrode fabrication
5 mg of catalyst + 0.5 ml of 5wt% Nafion + 0.5 ml of isopropanol
ultrasonicated for 20 min
5 µl pipetted onto the GC disk
dried in flowing Ar at room temperature
Electrochemical Conditions
Electrolyte: O2 saturated 0.5 M H2SO4
Working electrode: Catalyst deposited GC Disk
Reference electrode: Ag/AgCl, 3.5 M KCl (+0.205 V vs. NHE)
Counter electrode: Pt
Scan rate: 5 mV/sec
52
LSVs of O2 reduction on 20 wt% RuxSey/CDX975
and commercial Pt/C (E-TEK) catalysts in 0.5 H2SO4;
Scan rate – 5 mV sec-1
(Empty and full symbols corresponding to the LSVs
in Ar- and O2- saturated 0.5 M H2SO4 respectively)
CVs of (■) Ru/CDX975 and
(●) Ru1Se0.6/CDX975 catalysts in
Ar-saturated 0.5 H2SO4;
Scan rate – 20 mV sec-1
 ORR activity exhibits a maximum for the Ru1Se0.6/C catalyst
 ORR activity is due to stabilization of Ru active sites by Se against blocking as a
result of (hydr)oxide formation.
53
Elemental composition, Se/Ru atomic ratio, crystallite size, onset potential for oxygen
reduction and ORR activities of 20 wt% RuxSey/CDX975 (x = 1 and y = 0-1) and commercial
20 wt% Pt/C (E-TEK) catalysts
Catalyst
Elemental
composition by
EDX
Se/Ru atomic
ratio
Crystallite size
from XRD (nm)
Onset potential
(mV) for oxygen
reduction
ORR activity at
+0.65 V vs. NHE
(mA/cm2)
Ru/CDX975
Ru1Se0.2/CDX975
Ru1Se0.4/CDX975
Ru1Se0.6/CDX975
Ru1Se0.8/CDX975
Ru1Se1/CDX975
Pt/C (E-TEK)
100:87.7:12.3
76.6:23.4
68.5:31.5
62.2:37.8
56.2:43.8
-
0.0
0.18
0.38
0.59
0.78
1.00
-
3.0
3.0
3.1
3.1
3.1
3.1
+850
+875
+890
+905
+885
+870
+930
1.3
2.1
3.0
4.2
1.6
1.4
4.0
 ORR activity exhibits a maximum for the
Ru1Se0.6/CDX975 catalyst
ORR activity of Ru1Se0.6/CDX975 was
comparable with that of Pt/C (E-TEK)
catalyst
Se/Ru atomic ratio vs. ORR current density
of as-synthesized RuxSey/CDX975 catalysts
54
Catalyst
Degradation of
activity
(%)
Ru/CDX975
Ru1Se0.2/CDX975
Ru1Se0.4/CDX975
Ru1Se0.6/CDX975
Ru1Se0.8/CDX975
Ru1Se1/CDX975
Pt/C (E-TEK)
80
33
14
2
48
50
23
Current density vs. time curves of as-synthesized 20 wt% RuxSey/CDX975 catalysts;
(a) Ru/CDX975 (b) Ru1Se0.2/CDX975 (c) Ru1Se0.4/CDX975 (d) Ru1Se0.6/CDX975
(e) Ru1Se0.8/CDX975 (f) Ru1Se1/CDX975 and (g) commercial Pt/C (E-TEK)
measured in oxygen saturated 0.5 M H2SO4 at +0.65 V vs. NHE
Order of stability:
Ru/CDX975 < Ru1Se1/CDX975 < Ru1Se0.8/CDX975 < Ru1Se0.2/CDX975 < Pt/C (E-TEK)
55
< Ru1Se0.4/CDX975 < Ru1Se0.6/CDX975
MN4Cx clusters (M = Fe and Co) – A potential oxygen
reduction electrodes for PEMFC applications
Transition metal macrocycles
 Square planar complexes with the central
metal atom symmetrically surrounded by
four nitrogen atoms
 Structural analogues of metabolic systems
 Delocalization of ‘’ electrons
– high conductivity
 Stability in both acidic and basic media
56
Why Fe- and Co- containing macrocycles appear to be the best for
oxygen reduction ?
ORR activity (V vs. SCE)
Volcano plot
. ..
FeTPP
Oxygen reduction activities
of various catalysts
catalyst
Mass activity at
0.7 V vs. NHE
(mA/mg)
FeTPP
CoTPP
FePc
CoPc
RuPc
RuTPP
MnTPP
OsTPP
CrTPP
CoTAA
0.06
0.08
0.07
0.05
0.04
0.02
0.01
0.007
0.007
0.005
CoOEP
CoTPP
Redox potential (V vs. SCE)
Jose H. Zagal, Coord. Chem. Rev., 119 (1992) 89
57
Adverse effect of H2O2 on catalytic activity
x
x
+ H2O2, + O2
M
-x
x
M
M
-M
x
Mechanism of the disintegration of metal macrocycle
K. Weisener, Electrochimica Acta, 31 (1986) 1073
58
How to increase the oxygen reduction activity ?
 Pyrolysis of the carbon supported metal macrocycles
Remarkable oxygen reduction activities of pyrolyzed Fe- and Co- based catalysts
catalyst
Metal loading
(wt%)
FeTPP/Vulcan XC72R heat treated at 600oC
CoTPP/Vulcan XC72R heat treated at 600oC
FePc/Vulcan XC72R heat treated at 500oC
CoPc/Vulcan XC72R heat treated at 600oC
FeTMPP-Cl/BP heat treated at 800oC
FeTPP/CoTPP heat treated at 600oC
2.0
2.0
2.0
1.9
2.0
2.0
ORR activity at 0.7 V vs.
NHE
2
(mA/cm )
(mA/mg)
3.9
3.1
4.0
3.1
5.1
3.0
102 (0.06)
98 (0.08)
78 (0.07)
58 (0.05)
127 (0.11)
69 (----)
# The values shown in bracket are the activities of non-heat treated catalysts
Visualization of the reaction of the porphyrin with the carbon during heat treatment
Active species for oxygen reduction --- MN4Cx (M = Fe, Co)
59
Criteria for oxygen reduction by MN4
Orbital wave functions have to be of same symmetry
Orbital energies have to be matched for suitable activation - Reduction
Electron transfer should be facile
3u*
1g*
2p
2p
1u
3g
2u*
2s
2s
2g
Molecular Orbital diagram of O2 molecule
 Oxygen reduction takes place by the transfer of electron from HOMO of MN4 to
60
the anti-bonding * of O2 molecule
Methodology
 Single point energy – DFT calculations by Gaussian98
 B3LYP LANL2DZ – Basis set
Model systems : MN4 (M = Fe, Co) and O2
N
O
M
N
N
N
Fe-N distance: 2.00 Å
Co-N distance: 2.01 Å
N-M-N
: 109o47'
O
O-O distance : 1.26 Å
61
Percentage atomic orbital contributions to HOMO and LUMO of
MN4 and O2
Model
system
FeN4
CoN4
Model
system
O2
s
M = Fe, Co
p
d
s
p
HOMO: -7.56
0.0
0.0
30.8
0.0
69.1
LUMO: -7.00
0.0
0.0
0.0
0.0
100.0
HOMO: -7.78
2.3
0.0
21.1
1.1
75.4
LUMO: -7.01
0.0
1.6
3.2
1.6
93.4
E (eV)
E (eV)
N
O
s
O
p
s
p
HOMO: -8.18
(* level)
0.0
50.0
0.0
50.0
LUMO: +6.41
28.3
21.6
28.3
21.6
 Suitable energies of HOMO of MN4 and * orbital of O2 causes the
facile reduction of oxygen
62
Electron density maps
HOMO of FeN4
HOMO of CoN4
 * orbital of O2
63
Energy level diagram
Mode of activation of O2 by FeN4
 Presence of ‘N’ in MN4 plays an important role in attaining directional
wave functions there by leading to facile electron transfer.
64
Synthesis of Iron Tetraphenylporphyrin (FeTPP)
40 ml anhydrous pyridine + 5 ml triethylamine + 0.5 ml TBP
10-2 mol FeCl2
5 x 10-3 mol H2TPP
refluxed under Ar for 2 hr
filtered and dried at 75 oC
FeTPP
65
Preparation FeTPP/CDX975
0.17 g FeTPP + 60 ml anhydrous Pyridine
0.54 g CDX 975
refluxed overnight under Ar
filtered and washed with H2O
dried at 75 C
FeTPP/CDX975
- The prepared catalyst was made into fine powder and heat
treated at different temperatures ranging from 100 – 900 oC
66
Synthesis of Iron Phthalocyanine (FePc)
2.54 g FeCl2
10.2 g Phthalonitrile
5 ml TBP
100 ml ethylene glycol
refluxed for 2 hr under Ar
filtered and dried at 75 oC
150 ml methanol + 8 ml HCOOH
refluxed for 30 min under Ar
dried at 75 oC
FePc
67
Preparation FePc/CDX975
0.12 g FePc + 60 ml anhydrous Pyridine
0.5 g CDX 975
refluxed for 8 hr under Ar
filtered and washed with H2O
dried at 75C
FePc/CDX975
- The prepared catalyst was made into fine powder and heat
treated at different temperatures ranging from 100 – 900 oC
68
Electrochemical measurements
Electrode fabrication
16 mg catalyst + 0.4 ml of H2O + 0.4 ml of 5 % Nafion
Ultrasonicated for 10 min
10 µl pipetted onto the GC disk
Dried in air at 75 °C
Electrochemical Conditions
Electrolyte: O2 saturated 0.5 M H2SO4
Working electrode: Catalyst deposited on GC
Reference electrode: Ag/AgCl electrode
Counter electrode: Pt
Scan rate: 10 mV/sec
69
Heat treatment
temp (oC)
Wt % of ‘Fe’ by
redox titration
method
Wt % of ‘N’
by Kjeldahl
method
Oxygen reduction activity
at +0.7 V vs NHE (mA/cm2)
untreated
1.96
2.3
0.0
100
NM
2.2
0.0
200
NM
2.14
0.0
300
NM
2.1
0.08
400
NM
1.87
1.1
500
NM
1.74
1.8
600
1.97
1.7
3.81
700
NM
1.65
2.72
800
NM
1.3
1.6
900
NM
1.04
1.2
(NM = not measured)
 heat-treated FeTPP/CDX975 at 600 oC showing higher activity
70
Heat treatment
temp (oC)
Wt % of ‘Fe’ by
redox titration
method
Wt % of ‘N’
by Kjeldahl
method
Oxygen reduction activity
at +0.7 V vs NHE (mA/cm2)
untreated
1.86
1.94
0.0
100
NM
1.62
0.0
200
NM
1.58
0.0
300
NM
1.5
0.13
400
NM
1.48
0.4
500
NM
1.4
2.4
600
1.85
1.2
2.2
700
NM
1.1
0.44
800
NM
0.4
0.31
900
NM
0.18
0.18
(NM = not measured)
 heat-treated FePc/CDX975 at 500 oC showing higher activity
71
Synthesis of Iron and Cobalt Tetramethoxyphenylporphyrins
56 ml of N,N'-dimethylformamide
refluxed under Ar
0.67 g H2TMPP
0.22 g metal salt (10% extra)
refluxed under Ar for 2 hr
cooled in an ice bath and a portion of
ice-water mixture was added
filtered and dried at 110 oC
M-TMPP (M = Fe and Co)
Finally it was purified by column chromatography using benzene-chloroform (1:1) as the eluant.
Elemental composition of the synthesized complexes
Complex
FeTMPPCl
(C48H36N4O4FeCl)
CoTMPP
(C48H36N4O4Co)
Elemental analysis
calculated (wt%)
C, 66.95; H, 4.18; N, 6.51;
Fe, 6.49
C, 72.82; H, 4.55; N, 7.08;
Co, 7.46
found (wt%)
C, 66.81; H, 4.07; N, 6.43;
Fe, 6.57
C, 71.90; H, 4.41; N, 6.97;
Co, 7.49
72
Modification of carbon support (CDX975) with HNO3
1 g of carbon, CDX975 + 20 ml of 70 wt % HNO3
refluxed for 7 hr
filtered, washed with deionized water and methanol
dried at 70 °C
CDX975(T)
Preparation M-TMPP (M = Fe & Co)/CDX975(UT&T) and MN4Cx catalysts
M-TMPP
(M = Fe and Co)
+
Acetone
required amount of CDX 975(UT & T)
ultrasonicated for 30 min
solvent removal under vacuum
M-TMPP/CDX975(UT &T)
heat treated at 800 °C in Ar atm for 2 hr
MN4Cx (M = Fe & Co)
(Metal loading: 2 wt%)
The untreated M-TMPP adsorbed on as-received (CDX1) and oxidized (CDX2) carbon were designated as CDX1-MTMPP(UT)
and CDX2-MTMPP(UT) respectively. The heat treated M-TMPP adsorbed on as-received (CDX1) and oxidized (CDX2) carbon 73
were designated as CDX1-MTMPP(HT) and CDX2-MTMPP(HT) respectively. Here ‘M’ represents Fe and Co.
Electrochemical measurements
Electrode fabrication
Electrochemical Conditions
16 mg catalyst + 0.4 ml of H2O + 0.4 ml of 5 % Nafion
Electrolyte: O2 saturated 0.5 M H2SO4
Working electrode: Catalyst deposited on GC
Ultrasonicated for 10 min
Reference electrode: Ag/AgCl, 3.5 M KCl
(+0.205 V vs. NHE)
10 µl pipetted onto the GC disk
Counter electrode: Pt
Dried in air at 75 °C
Scan rate: 10 mV/sec
Single cell PEMFC measurements
Gas diffusion electrodes were prepared by a combined filtration/brushing procedure.
A homogeneous catalyst suspension consisted of 12.9 mg of catalyst, 0.5 ml of H2O
and 0.3 ml of 5 wt % Nafion solution blended ultrasonically for 1 h was applied on the
teflonized carbon cloth substrate by layer wise.
Both anode and cathode electrodes were then placed in a vacuum oven at 75 C for 1 h.
The anode (20 wt% Pt/C from E-TEK) consists of 0.4 mg Pt/cm2. The cathode
(prepared Fe and Co catalyst) consists of 0.2 mg metal/cm2.
The membrane-electrode assembly (MEA) was obtained by sandwiching the Nafion115
membrane between the cathode and anode by hot pressing at 140 °C and 50 kg cm−2 for 1 min.
The fuel cell testing was carried out at an operating temperature of 80 °C and 1 atm
74
pressure with humidified hydrogen and oxygen gas reactants.
 CDX1
 CDX2
CVs of untreated and treated CDX975
in Ar- saturated 0.5 M H2SO4
As-received carbon (CDX1) and oxidized
carbon (CDX2) exhibit well-defined redox
peaks at ~0.55 V vs. NHE which are
corresponding to quinone/hydroquinone
groups on the carbon surface.
75
Linear sweep voltammograms (LSVs) of untreated and heat treated Fe(III)TMPP-Cl
supported on as-received and oxidized carbon catalysts in Ar- and O2 -saturated 0.5 M H2SO4
76
Linear sweep voltammograms (LSVs) of untreated and heat treated CoTMPP supported on asreceived and oxidized carbon catalysts in Ar- and O2 - saturated 0.5 M H2SO4.
77
B
A
100 nm
100 nm
D
C
100 nm
100 nm
TEM images of (A) CDX1-FeTMPPCl(HT), (B) CDX2-FeTMPPCl(HT),
(C) CDX1-CoTMPP(HT) and (D) CDX2-CoTMPP(HT)
78
Particle size of catalysts analyzed in TEM
Catalyst
CDX1-FeTMPPCl(HT)
CDX2-FeTMPPCl(HT)
CDX1-CoTMPP(HT)
CDX2-CoTMPP(HT)
Particle size (nm)
25-37
8-15
35-50
10-18
Higher dispersion of catalyst particles on
oxidized carbon (CDX2) compared to the
un-oxidized carbon (CDX1).
Estimated metal loading, onset potential for oxygen reduction and oxygen
reduction activities of Fe- and Co-based catalysts as well as commercial Pt catalysts
Catalyst
CDX1-FeTMPPCl(UT)
CDX2-FeTMPPCl(UT)
CDX1-FeTMPPCl(HT)
CDX2-FeTMPPCl(HT)
CDX1-CoTMPP(UT)
CDX2-CoTMPP(UT)
CDX1-CoTMPP(HT)
CDX2-CoTMPP(HT)
Pt/C (E-TEK)
Metal
(Fe or Co)
loading (wt%)
Onset potential for
oxygen reduction
(mV) vs. NHE
ORR activity at +0.7 V
vs. NHE (mA cm-2)
2.01
2.03
1.96
1.97
1.98
1.97
1.89
1.93
2.07
+810
+830
+840
+870
+740
+760
+795
+860
+910
0.2
0.35
4.2
4.9
0.16
0.24
3.2
4.5
4.9
(UT = untreated and HT = heat-treated)
By increasing the number of surface oxygen groups (quinones) on the carbon support,
79
the dispersion of the catalysts, as well as their ultimate performance, is increased.
Δ CDX1-FeTMPPCl(HT) -- 40%
▲CDX2- FeTMPPCl(HT) – 33%
□ CDX1-CoTMPP(HT) ---- 50%
■CDX2-CoTMPP(HT) ----- 30%
 Pt/C (E-TEK) ------ ------- 30%
Current density vs. time curves of catalysts
in oxygen saturated 0.5 M H2SO4 at +0.7 V vs. NHE
Δ CDX1-FeTMPPCl(HT)
▲CDX2- FeTMPPCl(HT)
□ CDX1-CoTMPP(HT)
■ CDX2-CoTMPP(HT)
 Pt/C (E-TEK)
H2/O2 PEM fuel cell polarization curves
with catalysts for oxygen reduction at 80 C
Single-cell PEMFC performance of Fe- and Co-based catalysts as well as Pt catalysts
Catalyst
Performance at +0.7 V vs. NHE
(mA cm-2)
CDX1-FeTMPPCl(HT)
CDX2-FeTMPPCl(HT)
CDX1-CoTMPP(HT)
CDX2-CoTMPP(HT)
Pt/Vulcan XC72R (E-TEK)
125
171
82
104
180
80
Salient features:
In the case of Untreated and heat treated catalysts, the ORR activity was found to be,
CDX1-FeTMPPCl(UT)<CDX2-FeTMPPCl(UT)<CDX1-FeTMPPCl(HT)<CDX2-FeTMPPCl(HT)
CDX1-CoTMPP(UT)<CDX2-CoTMPP(UT)<CDX1-CoTMPP(HT)<CDX2-CoTMPP(HT)
It implies that (i) the active species, MN4Cx species generated by the heat treatment increases the
ORR activity and stability and (ii) The increased number of surface oxygen functionalities
introduced by nitric acid treatment improve the ORR performance as well as stability under the
experimental conditions.
TEM measurements show that the increased dispersion of active species in the case of oxidized
carbon (CDX2) compared to the as-received carbon (CDX1). EDX confirms the presence of
metal (Fe and/or cobalt), nitrogen and carbon in the prepared electrocatalysts.
In both Fe and Co based catalysts, there is approximately +30 mV or more positive shift of onset
potential for oxygen reduction in the case of oxidized carbon containing electrocatalysts
compared to that of the as-received carbon containing electrocatalysts. It implies that the
overpotential for oxygen reduction was reduced by means of oxidized carbon support.
The increased order of ORR activity as well as stability of heat treated catalysts and commercial
Pt/C was found to be,
CDX1-CoTMPP(HT)<CDX2-CoTMPP(HT)≈CDX1-FeTMPPCl(HT)<CDX2-FeTMPPCl(HT)
≈commercial Pt/C.
81
Conclusion
82
Scanning electrochemical microscopy (SECM)
was based on simple thermodynamic principles
that involve the pairing of a good oxygen-bond
cleaving metal with a good oxygen-reducing
metal.
SECM images of oxygen reduction activity measured on a binary Pd-Ti array in 0.5 M H2SO4
by scanning a 25 µm tip generating oxygen above an array of different compositions at
different potentials. Tip-substrate distance = 30 µm, tip current = -171 nA, and scan rate = 50
µm each 0.2 s. Potentials are vs. the hydrogen reference electrode. The image is a color-map
graph of substrate current vs. tip position (X, Y).
This screening technique is based upon producing an array of bimetallic or trimetallic
catalyst spots with different compositions on a glassy carbon (GC) support. A small
ultramicroelectrode tip is used to generate oxygen at a constant current as the tip is
scanned above the array of spots, with the GC held at different potentials, so that each
spot registers a current proportional to the rate of oxygen reduction at that location (GC,
which is a poor oxygen reduction catalyst, does not produce a significant response at these
83
potentials).
Why Pd or Pt alloys are more active than metal alone for oxygen reduction?
(1) Induced change in the density of states (DOS) at the Fermi level of Pd sites by alloy formation.
Ota’s group considered that the decreased DOS induced by the electron transfer from Co, Ni, or Cr
to Pd may weaken the chemisorption bonds between Pd and reactants such as O2, O/OH, O2−, and
H2O2, reducing their blocking effect in theO2 reduction process. As a result, the ORR kinetics may
be enhanced. For example, an alloyed Pd showed a reduced ORR overpotential by ∼50mV at
0.2mAcm−2 compared to that catalyzed by pure Pd. [K. Lee, O. Savadogo, A. Ishihara, S.
Mitsushima, N. Kamiya, K.-I. Ota, J. Electrochem. Soc. 153 (1) (2006) A20.]
(2) Induced change in ORR Gibbs free energy. Wang and Balbuena explained that if a catalyst
consists of two metals, one with a low occupancy of d-orbitals (such as Co, Ni, Cr, or V) and the
other with fully occupied d-orbitals (such as Pd, Au, and Ag), the d-orbital coupling effect between
them can significantly decrease the Gibbs free energy of the electron transfer steps in ORR, resulting
in an enhancement in ORR kinetics. [Y.X. Wang, P.B. Balbuena, J. Phys. Chem. B 109 (2005)
18902.]
(3) Pd lattice compression induced by alloy formation. A metal d-band center can be altered by orbital
overlap. The formation of the alloy can induce Pd lattice compression (or reduction of bond lengths
between metals) through the modification of the electronic structure and orbital overlap. This will
cause the shift of the d-band center, resulting in a change in the surface activity of Pd sites. The
density functional theory calculations made by Norskov’s group confirmed that the compression of a
Pd lattice in alloys could downshift the d-band center energy. The Pd lattice compression can
enhance the catalyzed ORR. [B. Hammer, J.K. Norskov, Adv. Catal. 45 (2000) 71.]
84
Typical polarization curve
EOC
1.21 V

Activation
losses
Ohmic
losses
Mass transport
losses
Cell E
(i)
(ii)
(iii)
Current
Typical polarization curve for a fuel cell: voltage
drops
due to: (i) surface reaction kinetics; (ii) electrolyte
resistance and (iii) reactant/product diffusion rates
Activation losses – Energy barrier
associated with catalytic reactions at
the electrodes
Ohmic losses - ionic resistance in the
electrolyte and electrodes
electronic resistance in the
electrodes, current collectors and
interconnects, and contact
resistances.
Mass transport losses – Limit of
getting the reactants to the active
catalyst surface
At low current densities (jo < 1 mA cm-2), electrodes gives a larger Rtr
and therefore overpotential, η should be greater than 400 mV (at RT).
An extremely active electrocatalyst is needed to overcome this initial voltage
drop in the E – I curve.
Best electrocatalysts – Pt based materials
85
86
87
88
89
The higher ORR activity in the case of as-synthesized catalyst (4 nm) compared to the
commercial catalyst (2.8 nm) is due to,
As the particle size increases, there should be an increase in the number of Pt atoms on various
crystal facets (order of ORR activity in aq. H2SO4 is, Pt(111) < Pt(100) < Pt(110)) increases with
respect to those at edges and corners. Under the operating conditions of fuel cell, especially in the
presence of electric potential there should be a drastic change in the geometries, charges and
adsorption energies of the species on a catalyst surface and preferential exposure of Pt in
particular orientation. As a result, population of the antibonding 2π* orbitals in oxygen molecule
increases and facilitates the reduction of oxygen to water.
Hammer and Norskov concept: Metal reactivity changes through the changes in adsorbate–
surface interaction energy due to the attraction of the local d-band position relative to the Fermi
level. Since the Pt particles are anchored on the surface of carbon in the case of as-synthesized
catalyst (evident from TEM images), restructuring of particles which can induce steric and
electronic differences among the catalytic sites was more favorable and facilitates the oxygen
reduction.
The anion (sulfate) effect on the specific activity at a given potential is comparatively stronger for
the smaller (d<3 nm) Pt nanoparticles. This effect is generally ascribed to the impeding effect of
specific anion adsorption on different crystal faces, the distribution of which changes with Pt
particle size. As a result, there should be a less availability of oxygen at the surface Pt atoms (mass
transfer limitations) and consequently lower oxygen reduction activity on small size Pt particles
present in the commercial Pt/C (2.8 nm) catalyst compared to that of the as-synthesized catalyst (4
nm).
90
Inhibiting or shifting the onset potential (approximately 800
mV vs RHE for Pt) of Pt-OH formation, providing free sites
for molecular oxygen adsorption, is generally expected to lower
the overpotential losses. A number of prior reports have provided
indirect evidence to the possibility of inhibiting the formation
of anodic activation of water for Pt-OH formation.
Shifting the onset potential of OH formation on Pt is dependent
on (a) the ability of the alloying elements to modify the Pt
electronic and short-range atomic order for inhibiting activation
of H2O and (b) the ability of the alloying element to attract and
hold H2Oads more strongly than the surrounding surface Pt
atoms.
91
The Pt d-band vacancies were derived from analysis of the Pt L3 and L2 white
lines of XANES. The L2 and L3 excitations are due to the transition of 2p1/2 and
2p3/2 electrons to the empty Fermi level. Dipole selection rules restrict the orbital
angular quantum number transitions to (1. Hence a transition of 2p3/2 to a 5dorbital is favored. An increase in the L3 and L2 white line peak is related to
changes in d-band vacancy and reflects the extent of d-band occupancy. A
detailed description of this methodology is provided in ref 2 and references
therein. The Pt d-band occupancies evaluated from the Pt L3 and L2 edge
curves are given in Table 2 along with the Pt-M atomic ratio. Comparison at 0.54
V vs RHE shows an increase in d-band vacancies per atom on alloying, this
follows the trend Pt/C < PtFe/C < PtCo/C. This increase in d-band character of Pt
follows closely with the electron affinity of the alloying metal. XANES also
offers the ability to monitor the extent of alloying and the nature of the active
surface.
(J. Phys. Chem. B 2004, 108, 11011-11023).
92
d-band vacancies
• The XANES region of the Pt LIII and LII absorption
edges can be used to determine the fractional d-electron
occupancy of the Pt atoms in the catalyst sample by a
so-called white line analysis.
• comparison of the white line intensities of a sample
with those of a reference metal foil provides a measure
of the fractional d-electron vacancy, fd, of the absorber
atoms in the sample. fd is defined as follows:
where A3,r represents the area under the white line
at the LIII edge and A2,r represents the area at the
LII edge of the reference foil spectrum
• fd can then be used to calculate the total number
of unoccupied d-states per Pt atom in the samples
as follows:
93
Since the intensity of the L3 and to a lesser
extent the L2 peaks increase with Pt d-band vacancy, they can
been used to make quantitative determination of Pt d-band
occupancy. On the basis of this premise, it has been shown3'
that the difference in areas under the Pt L3 and L2 absorption
edges between the test sample (WC and Pt alloy electrocatalyst)
and a pure Pt reference foil provides the fractional change in
the number of d-band vacancies relative to the reference material
(fd) using the relation
where AA3 and AA2 can be expressed as
Here A2 and A3 refer to the area under the L3 and L2 absorption
edges of the sample (s) and reference (r) material. The total
number of unoccupied d-states, characterized by the total angular
moment J ((hJ)total = h3/2 + hy2) for pure Pt, has been evaluated
from band structure calculations to be 0.3.29330 Therefore the
d-band vacancies of Pt in the sample can evaluated using the expression:
94
Why carbon materials are used as an electrocatalyst support ?
Electrochemical properties
- Wide electrochemical potential window
Chemical properties
- Good corrosion resistance
Electrical properties
- Good conductivity
Mechanical properties
- Dimensional & mechanical stability
- Light weight & adequate strength
95
Oleic acid
Pale yellow or brownish yellow
oily liquid
C18H34O2 (or) CH3(CH2)7CH=CH(CH2)7COOH)
(9Z)-octadec-9-enoic acid
Oleylamine
C18H37N
96