Electrochemically fabricated tin-palladium bimetallic electrocatalyst for oxygen reduction reaction in acidic media

Article abstract of DOI:10.1149/1.3183803

In the present paper, we explore the electrocatalytic activity of the glassy carbon (GC) electrode modified with Sn-Pd bimetallic electrocatalysts, i.e., Sn was electrochemically deposited on the GC electrode premodified by electrodeposition of Pd, toward

Full text of DOI:10.1149/1.3183803

B1142
Journal of The Electrochemical Society, 156 ͑10͒ B1142-B1149 ͑2009͒
0013-4651/2009/156͑10͒/B1142/8/$25.00 © The Electrochemical Society
Electrochemically Fabricated Tin–Palladium Bimetallic
Electrocatalyst for Oxygen Reduction Reaction in Acidic Media
Md. Rezwan Miah,a,z Muhammad Tanzirul Alam,
,z
*
Takeyoshi Okajima, and Takeo Ohsaka
Department of Electronic Chemistry, Interdisciplinary Graduate School of Science and Engineering, Tokyo
Institute of Technology, Midori-ku, Yokohama 226-8502, Japan
In the present paper, we explore the electrocatalytic activity of the glassy carbon ͑GC͒ electrode modified with Sn–Pd bimetallic
electrocatalysts, i.e., Sn was electrochemically deposited on the GC electrode premodified by electrodeposition of Pd, toward the
oxygen reduction reaction ͑ORR͒. The results demonstrate that the in situ underpotential deposition of Sn onto the GC electrode,
premodified by an optimized amount of electrodeposited Pd, enhance the electrocatalytic activity of the electrode to such an extent
that an exclusive one-step four-electron ORR takes place at a considerably lower overpotential. The mechanistic idea for the
observed enhanced ORR at the Sn–Pd bimetallic electrocatalyst is also presented based on the ligand effect of Sn. The experi-
mental investigations were performed using cyclic, steady-state voltammetric, and open-circuit potential measurement techniques.
© 2009 The Electrochemical Society. ͓DOI: 10.1149/1.3183803͔ All rights reserved.
Manuscript submitted December 11, 2008; revised manuscript received April 13, 2009. Published July 31, 2009.
Electrocatalytic oxygen reduction reaction ͑ORR͒ is greatly im-
Sn–Pd bimetallic electrocatalyst has not been investigated so far for
the ORR. Besides, most of the bimetallic electrocatalysts of Pd have
been synthesized by the chemical technique, which, as compared
with the electrochemical technique, is more expensive because of
the consumption of several chemicals, and time-consuming as it
involves a series of tedious synthetic procedures.
The aim of this paper is to investigate the electrocatalytic effect
on the ORR of Sn deposits on the bare GC electrode and GC elec-
trode that was previously modified by electrodeposition of Pd. The
effect of the loading amount of predeposited Pd is also examined.
The mechanism of the observed enhancement of the activity of the
Sn–Pd bimetallic electrocatalyst is presented. The bare GC elec-
trodes modified by Sn and Pd separately are hereinafter denoted as
the Sn/GC and Pd/GC electrodes, respectively, while the Pd/GC
electrode modified by Sn is denoted as the Sn–Pd/GC electrode. The
experimental measurements were performed using cyclic and hydro-
dynamic voltammetric and open-circuit potential measurement tech-
niques.
portant for the electrochemical energy conversion in fuel cells,
metal–air batteries, corrosion, and some other industrial
processes.1-6 It continues to be a challenging research target because
of its complex kinetics and the need for better electrocatalysts that
support the vital four-electron reduction process. The mechanism of
the ORR largely depends on the electrode materials, their surface
structures, and the nature of the electrolytic solutions.7-10 Modifica-
tion of electrode surface by metal adatoms/nanoparticles has long
been well known to potentially catalyze the ORR.11-22 Pt has been
widely used as an electrocatalyst for the ORR because of its high
catalytic activity. However, its high price has encouraged the devel-
opment of alternate platinum-free electrocatalysts for the ORR. Pd is
less expensive, relatively abundant, and has a valence shell elec-
tronic configuration and lattice constant similar to Pt. Pd electrocata-
lysts have an activity comparable to Pt toward the ORR and more
importantly they are highly selective to the ORR in the presence of
methanol. Therefore, Pd catalysts have been largely investigated for
the ORR to replace Pt catalysts.22-28 As compared with other metals,
Sn electrocatalysts have so far been employed less for the ORR.
Only a few reports in this regard have been available. For example,
Zinola et al.29,30 reported that the underpotential deposition ͑UPD͒
of Sn adatoms onto the Pt electrode surface changes the four-
electron ORR to a two-electron process producing bulk H2O2. Re-
cently, Jeyabharathi et al.31 reported that chemically prepared Pt–Sn
bimetallic nanoparticles supported on carbon substrate possess an
efficient electrocatalysis for the four-electron ORR. We have also
reported very recently that the underpotential deposited Sn-adatom-
modified polycrystalline gold ͓Au ͑poly͔͒ electrode functions as a
promising electrocatalyst for the exclusive four-electron ORR form-
ing H2O as the final product.32 We are increasingly interested in
evaluating the electrocatalytic behavior of Sn electrocatalysts, sup-
ported on various substrates, toward the ORR. In this paper, we
address the catalytic effect of the in situ electrochemically deposited
Sn onto the glassy carbon ͑GC͒ electrode without and with its ͑GC͒
premodification by the electrodeposition of an optimized amount of
Pd.
Experimental
Chemicals
.— Palladium chloride ͑PdCl2, 99%͒, tin͑II͒ sulfate
͑SnSO4, 96%͒, hydrogen peroxide ͑H2O2, 35.5%͒, and sulfuric acid
͑H2SO4͒ ͑Wako Pure Chemical Industries, Japan͒ were purchased
and used as received. Milli-Q water was used as a solvent. The
electrolytic solution was purged with N2 or O2 gas ͑purity 99.99%,
Nippon Sanso Co. Inc., Japan͒ to obtain their saturated solutions and
the gases were flushed over the cell solution during measurements.
Apparatus and procedures.— For cyclic voltammetric measure-
ments, GC electrodes ͑in the form of a disk, ␾ = 1.0 mm and sealed
in a Teflon jacket͒ with an exposed surface area of 7.85
ϫ 10−3 cm2 were used as working electrodes. A spiral Pt wire and
Ag͉AgCl͉NaCl ͓sat. ͑saturated͔͒ were the counter and reference
electrodes, respectively. A conventional two-compartment Pyrex
glass cell was used. Before measurements either N2 or O2 gas was
bubbled into the cell for 30 min to obtain N2- or O2-saturated solu-
tion. The gases were flashed over the cell solution during the mea-
surements. All the measurements were performed at 25 Ϯ 1°C. The
GC electrodes were polished with no. 2000 emery paper and then
with aqueous slurries of successively finer alumina powder ͑down to
0.06 ␮m͒ with the help of a polishing microcloth to a mirror finish.
The GC electrodes were then sonicated for 10 min in Milli-Q water
and electrochemically pretreated in 0.05 M H2SO4 solution by re-
peating the potential scan in the range of Ϫ0.2 to 1.5 V vs
Ag͉AgCl͉NaCl ͑sat.͒ at 0.1 V s−1 for 20 cycles. The currents were
normalized using the geometric surface area of the GC electrode.
For the steady-state voltammetric measurements, a rotating ring-disk
electrode ͑RRDE͒, consisting of GC disk ͑␾ = 6.0 mm͒ and Pt ring
Bimetallic electrocatalysts often show an improved activity to-
ward the ORR because of the increased interaction of O2 molecules
with the catalysts. Pd has been widely used with a large number of
other metallic elements such as Ti,33 Au,34 Co,35,36 Fe,24 Ni,37 Mo,38
40
Cr,39 and Cu with a view to preparing bimetallic electrocatalysts
of improved catalytic activity for the ORR. To our knowledge, the
*
Electrochemical Society Active Member.
a Permanent address: Department of Chemistry, School of Physical Sciences,
Shahjalal University of Science and Technology, Sylhet 3114, Bangladesh.
z E-mail: mrmche@yahoo.com; ohsaka@echem.titech.ac.jp
Downloaded on 2014-07-12 to IP 129.173.72.87 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 156 ͑10͒ B1142-B1149 ͑2009͒
B1143
1.25
1.6
0.8
0
0
i
h
–2
(A)
–4
b
–0.4 –0.2
0
0.2
d
E
/ V vs. Ag / AgCl / NaCl (sat.)
0
a
c
b
a
g
f
e
c
(B)
d
–1.25
–0.8
0
E
0.5
1
1.5
–0.8
–0.4
0
/ V vs Ag / AgCl / NaCl (sat.)
E
/ V vs. Ag / AgCl / NaCl (sat.)
Figure 1. CVs obtained at the ͑a͒ GC and ͓͑b͒–͑d͔͒ Pd/GC electrodes in
N2-saturated 0.5 M H2SO4 solution. Potential scan rate of 0.1 V s−1. Elec-
trodeposition of Pd onto the GC electrode was carried out in N2-saturated 0.5
M H2SO4 solution containing 0.1 mM PdCl2 using a chronoamperometric
technique by stepping the potential from an initial value of 1.0 V to a final
potential of 0 V for various deposition times: ͑b͒ 10, ͑c͒ 40, and ͑d͒ 300 s.
Figure 2. ͑A͒ CVs obtained at the GC electrode in N2-saturated 0.5 M
H2SO4 solution containing ͑a͒ 0 and ͑b͒ 1.0 mM SnSO4. ͑B͒ CVs obtained at
the ͑c͒ GC and ͓͑d͒–͑g͔͒ Pd/GC electrodes in N2-saturated 0.5 M H2SO4
solutions containing ͑c͒ 0 mM and ͓͑d͒–͑g͔͒ 1.0 mM SnSO4. The lower
potential limits are ͑d͒ Ϫ0.45, ͓͑c͒ and ͑e͔͒ Ϫ0.5, ͑f͒ Ϫ0.55, and ͑g͒ Ϫ0.6 V.
Potential scan rate of 0.1 V s−1. Inset shows the CVs obtained at the Pd/GC
electrode in N2-saturated 0.5 M H2SO4 solutions in the ͑h͒ absence and ͑i͒
presence of 1.0 mM SnSO4. Potential scan rate of 0.1 V s−1
.
͑␾ = 7.0 mm and ␾ = 9.0 mm͒, was used. Steady-state voltam-
in
out
mograms were obtained using a rotary system from Nikko Keisoku
͑Japan͒ coupled with an electrochemical analyzer ͑ALS CHI-832͒.
For the steady-state voltammetric measurements, a compartment of
200 cm3 was used to eliminate the possible depletion of O2 concen-
tration and O2 gas was flashed over the cell solution during the
measurements. Electrochemical deposition of Sn and the ORR were
carried out in 0.5 M H2SO4 solution. Solution containing Sn2+ was
prepared by dissolving the required amount of SnSO4. The neces-
sary amount of H2O2 was added to the solution whenever necessary.
1.0 mM stock solution of PdCl2 was prepared in 0.5 M H2SO4
solution and stored at room temperature. The stock solution was
shaken well each time to dissolve the precipitated PdCl2 before its
use. Atomic force microscopy ͑AFM͒ images were taken ex situ
using a nanoscale hybrid microscope ͑model VN-8010M, from Key-
ence Corp., Japan͒. The microscope was used in the damping force
mode using a silicon cantilever ͑force constant of 40 N/m͒ with a
resonant frequency of 250 kHz to analyze the topography of the
samples.
geometric surface area ͑7.85 ϫ 10−3 cm2͒ of the GC electrode was
used to normalize the amount of Pd loading͔. The thus-prepared
Pd/GC electrodes were electrochemically characterized by measur-
ing cyclic voltammograms ͑CVs͒ in N2-saturated 0.5 M H2SO4 so-
lution in the potential range of Ϫ0.2 to 1.5 V at a scan rate of
0.1 V s−1. The typical CVs are presented in Fig. 1. The voltammet-
ric features of curves ͑b͒–͑d͒ as compared with curve ͑a͒ obviously
indicate the presence of the deposited Pd on the GC electrode. The
characteristics of the CVs are similar to those reported by Salvador-
Pascual et al.22 for the chemically prepared nanosized Pd catalyst
supported on the GC substrate. The peak currents at about Ϫ0.1 V
are assigned to the adsorption/desorption of hydrogen at the Pd par-
ticles. The increase in hydrogen adsorption/desorption current with
increasing the deposition time represents that more Pd was depos-
ited at a longer deposition time. The anodic peak at 0.7 is attributed
to the oxidation of Pd to its oxide, and the cathodic peak at 0.45 V
is attributed to the reduction of the produced Pd oxide.24,37 Bulk Pd
also produced the anodic peak for the oxide formation and the ca-
thodic peak for its reduction at comparable potentials. The voltam-
metric charge associated with the Pd oxidation and its reduction
gradually decreased with the potential cycling, implying that Pd
underwent the anodic stripping from the Pd/GC electrode. The po-
tential limit for the ORR at the Pd/GC electrode was therefore set
below 0.5 V to avoid the Pd stripping, and the electrode response
obtained for the ORR was highly stable. The sharp rise of current at
1.5 V is attributed to the generation of O2 from the electrolysis of
H2O.
Results and Discussion
Electrodeposition of Pd on the GC electrode.— Electrodeposition
of Pd onto various types of substrates is well known.41-47 In the
present study, the electrodeposition of Pd onto the GC electrode was
carried out in N2-saturated 0.1 mM PdCl2 solution, which was pre-
pared by dilution of the stock solution in the appropriate quantity of
0.5 M H2SO4. Cyclic voltammetric measurements ͑not shown͒
showed that the deposition of Pd onto the GC electrode in this PdCl2
solution commenced at about 0.6 V vs Ag/AgCl/NaCl ͑sat.͒. There-
fore, the electrodeposition of Pd was performed using a chrono-
amperometric technique by stepping the potential from an initial
value of 1.0 to a final value of 0 V. The loading of Pd was controlled
by choosing the deposition time in the range of 3–300 s. The amount
of the deposited Pd was estimated using the charge ͑after back-
ground correction͒ passed during the amperometric measurements
for the deposition of Pd considering two-electron reduction of Pd2+
to form Pd0. The amounts of Pd deposited in the deposition of 3, 10,
Electrochemical deposition of Sn on the GC and Pd/GC elec-
trodes.— The electrochemical deposition of Sn onto various sub-
strates has been well demonstrated in Ref. 29, 30, 32, and 48-50. In
this section we introduce the electrodeposition of Sn onto the GC
and Pd/GC electrodes. Figure 2A shows the CVs obtained at the GC
electrode in N2-saturated 0.5 M H2SO4 solution containing ͑a͒ 0 and
͑b͒ 1.0 mM SnSO4. The electrodeposition of Sn onto the bare GC
electrode requires a very high overpotential and appears as a single
cathodic wave at Ϫ0.63 V. This peak is attributed to the formation of
20, 40, 60, and 300 s were calculated as 1.29 ϫ 10−7, 1.91 ϫ 10−7
3.25 ϫ 10−7, 5.4 ϫ 10−7, 7.5 ϫ 10−7, and 3.5 ϫ 10−6 g cm−2 ͓the
,
Downloaded on 2014-07-12 to IP 129.173.72.87 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

B1144
Journal of The Electrochemical Society, 156 ͑10͒ B1142-B1149 ͑2009͒
a
b
a
+
b
I
–2
0.4 mA cm
c
d
c
+
1µm
1µm
I'
Figure 3. AFM images in air of the Pd/GC electrode ͑a͒ without and ͑b͒
with the deposition of Sn. ͑c͒ and ͑d͒ show the cross-section profiles through
the horizontal lines ͑white lines͒. The Pd/GC electrode was prepared by the
deposition of Pd for 40 s and keeping other conditions the same as in Fig. 1.
Sn was deposited onto the Pd/GC electrode in O2-saturated 0.5 M H2SO4
solution containing 1.0 mM SnSO4 using a single potential scan as 0.2 →
d
II'
−0.5 V at a potential scan rate of 0.1 V s−1
.
–1
–0.5
0
0.5
E / V vs. Ag / AgCl / NaCl (sat.)
Figure 4. CVs obtained at the GC electrode in ͓͑a͒ and ͑c͔͒ N2- and ͓͑b͒ and
͑d͔͒ O2-saturated 0.5 M H2SO4 solutions containing ͓͑a͒ and ͑b͔͒ 0 and ͓͑c͒
bulk Sn onto the GC electrode, while the anodic peak at Ϫ0.48 V is
assigned to the stripping of the bulk Sn. The electrodeposition of
Sn2+ was also performed onto the Pd/GC ͑Pd deposition time of 40
s͒ electrode in the same media, and the corresponding results are
presented in Fig. 2B. As compared with the bare GC electrode, the
overpotential of the deposition of Sn at the Pd/GC electrode is de-
creased and two cathodic waves for the deposition appear at Ϫ0.40
and Ϫ0.52 V. The first wave is assigned to the UPD of Sn2+ at the
Pd particles, while the second one is assigned to the deposition of
bulk Sn both at the bare fraction of the GC electrode and the UPD
layer of Sn on Pd particles. CVs obtained for the deposition of Sn2+
onto the Pd/GC electrode for various lower potential limits ͑curves
d–g͒ show that the anodic stripping of Sn corresponding to the first
cathodic wave is broad ͑curves d and e͒, while that corresponding to
the second wave is sharp and intense ͑curves g and f͒. The potential
of the intense anodic peak at the Pd/GC electrode is comparable to
that at the bare GC electrode and is assigned to the stripping of the
bulk Sn deposited at the bare fraction of the GC electrode and the
UPD layer of Sn on Pd particles. A CV was also recorded using the
Pd/GC electrode in N2-saturated 0.5 M H2SO4 solution in the ab-
sence of SnSO4 ͑curve h in the inset͒. The result shows a tremen-
dous evolution of H2 at the Pd/GC electrode at E Ͻ −0.2 V. There-
fore, the interesting feature observed in the presence of SnSO4 is the
large suppression of the evolution of H2 due to the formation of the
metallic Sn onto the Pd/GC electrode, which advantageously ex-
tended the lower potential limit to a more negative value, being
essentially important for the present purpose. An increase in H2
evolution overpotential by Sn has also been reported
elsewhere.30,32,48
and ͑d͔͒ 1.0 mM SnSO4. Potential scan rate of 0.1 V s−1
.
section profile ͑d͒ taken through the horizontal white line in image
͑b͒ represents that the density of the particles increases, while their
dimension remains almost unaltered. The larger density of particles
obviously indicates the formation of Sn on the electrode surface.
Previously deposited Pd functions as seed nuclei for the deposition
of Sn as well as Sn deposition take place at the bare fraction of the
GC electrode.
ORR at the GC and Sn/GC electrodes.— ORR at the GC elec-
trode was investigated in O2-saturated 0.5 M H2SO4 solution both in
the absence and presence of SnSO4, and the corresponding results
are presented in Fig. 4. A single cathodic peak ͑designated as I͒ at
Ϫ0.5 V was achieved at the GC electrode in the absence of SnSO4.
The GC electrode is well known to possess a low activity toward the
ORR as compared with metallic electrodes, typically Au, Ag, Pt, etc.
We have shown in our previously published articles that the ORR at
the GC electrode in acidic,51 alkaline,52 and neutral media20 pro-
ceeds through a two-electron pathway producing long-lived ͑bulk͒
H2O2. Peak I is therefore assigned to the two-electron reduction of
O2 producing stable H2O2 according to Eq. 1
O2 + 2H+ + 2e− = H2O2
The ORR in the presence of SnSO4 produced two cathodic waves at
about Ϫ0.45 V ͑designated as I ͒ and Ϫ0.70 V ͑designated as II ͒.
͓1͔
Ј
Ј
AFM images of the Pd/GC electrodes.— The Pd/GC electrode
was prepared by the electrodeposition of Pd onto the GC substrate
for 40 s under the same conditions as described earlier. The AFM
image of the Pd/GC electrode was taken by scanning an 8
ϫ 6 ␮m surface of the electrode, presented in Fig. 3a. The image
clearly shows the existence of the Pd particles on the GC substrate.
Cross-section profile ͑c͒ taken through the horizontal white line in
image ͑a͒ shows that the Pd particles are 10–25 nm high. The image
looks similar to that previously reported by Diculescu et al.43 for the
Pd particles deposited on the highly oriented pyrolytic graphite. The
Sn–Pd/GC electrode was prepared by the electrodeposition of Sn2+
onto the Pd/GC electrode ͑Pd deposition time of 40 s͒ by a single
linear sweep voltammetric scan of 0.2 → −0.5 V under the condi-
tions described in Fig. 2B, ͑curve e͒. The AFM image of the thus-
prepared Sn–Pd/GC electrode was also recorded under the same
conditions as image ͑a͒ and is presented in Fig. 3b. The cross-
Peak I is comparable with peak I ͑curve b͒ both in its intensity and
Ј
potential. It can be seen that peak I appears at a distinctly far
Ј
positive potential as compared with the onset potential of the depo-
sition of Sn2+ ͑curve c͒. Therefore, peak I ͑although obtained in the
Ј
presence of SnSO4͒ is also assigned to the two-electron ORR at the
bare GC electrode surface according to Eq. 1. However, the inter-
esting feature achieved in the presence of SnSO4 is the appearance
of the second consecutive peak II at a potential comparable to that
Ј
of the electrodeposition of Sn2+ with a greater intensity as compared
with the peak current for the Sn2+ electrodeposition ͑compare with
curve c͒. Unlike the case at the bare GC electrode in the absence of
Sn2+, the H O electrogenerated during peak I in Sn2+-containing
Ј
2
2
solution is converted into H2O in two ways. First, a fractional
amount of the produced H2O2 is consumed in the chemical oxida-
32
tion of Sn2+ to Sn4+ according to Eq. 2
Downloaded on 2014-07-12 to IP 129.173.72.87 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 156 ͑10͒ B1142-B1149 ͑2009͒
B1145
0
–1
–2
–3
a
a
0
–1
–2
c
b
b
f
I'''
e
d
IIV
I''
c
–0.4
–0.2
0
0.2
–1
–0.5
0
0.5
E/ V vs. Ag / AgCl / NaCl (sat.)
E / V vs. Ag / AgCl / NaCl (sat.)
Figure 5. CVs obtained at the ͓͑a͒–͑d͔͒ Pd/GC ͑Pd deposition time of 40 s͒
and ͑e͒ Pt electrodes in ͓͑a͒ and ͑c͔͒ N2- and ͓͑b͒, ͑d͒, and ͑e͔͒ O2-saturated
0.5 M H2SO4 solutions containing ͓͑a͒, ͑b͒, and ͑e͔͒ 0 and ͓͑c͒ and ͑d͔͒ 1.0
mM SnSO4. Potential scan rate of 0.1 V s−1. ͑f͒ CV is the same curve
presented in Fig. 4b.
Figure 6. CVs obtained at the Pd/GC electrode in ͑a͒ N2- and ͓͑b͒ and ͑c͔͒
O2-saturated 0.5 M H2SO4 solutions containing 1.0 mM SnSO4 and ͓͑a͒ and
͑b͔͒ 0 and ͑c͒ 2.0 mM H2O2. Potential scan rate of 0.1 V s−1
.
dergo its electrochemical reduction also at the same potential as the
ORR. Therefore, the ORR was also examined in the presence of
H2O2 ͑initially 2.0 mM͒, and the obtained results are shown in Fig.
6. Sn2+ undergoes chemical oxidation in the presence of H2O2 in
acidic media according to Eq. 2, as mentioned earlier.32 Therefore,
the results were obtained within as short an experimental time as
possible to minimize the effect of Eq. 2. Curve c shows that the
addition of H2O2 into the solution largely increased the cathodic
peak current at Ϫ0.34 V, although the cathodic current does not
correspond quantitatively to 2.0 mM H2O2, suggesting that the
added H2O2 also underwent the reduction at the same potential as
the ORR. Therefore, a one-step four-electron reduction of O2 to
H2O at the Sn–Pd/GC is quite reasonable. More support in this
regard is given later. The remarkable finding in the present study is
that the Sn–Pd/GC electrode drives the ORR at a lower overpoten-
tial as compared with the bare GC and Sn/GC electrodes through a
one-step four-electron pathway forming H2O as a final product,
which is essentially important for the efficient performance of fuel
cells.
Sn2+ + H2O2 + 2H+ = Sn4+ + 2H2O
͓2͔
and second, the remaining H2O2 is catalytically electroreduced to
H O during peak II according to Eq. 3 by virtue of the concurrent
Ј
2
electrodeposition of Sn on the GC electrode surface, that is, at the
Sn/GC electrode
H2O2 + 2H+ + 2e− = 2H2O
͓3͔
O can also undergo the reduction to H O during peak II . Electro-
Ј
2
2
catalytic reduction of H2O2 by other concurrently electrodeposited
metal adlayers on various substrates is also well known.53-56 Two
consecutive cathodic peak currents, the first one assigned to the
reduction of O2 to H2O2 and the second one to further reduction of
the electrogenerated H2O2 to H2O, have also been reported previ-
ously by our group using gold nanoparticle-electrodeposited poly-
crystalline gold electrode in acidic media.57 Further details of the
ORR at the Sn/GC electrode are beyond the scope of the present
paper and will be presented elsewhere.
The onset potential of the ORR at the Sn–Pd/GC electrode is the
same as that of the UPD of Sn2+ at the Pd/GC electrode ͑curve c in
Fig. 5͒, and the current of the ORR begins to increase concurrently
with the formation of Sn. Therefore, the catalytic activity of the
Pd/GC electrode for the ORR is a consequence of the formation of
Sn on the Pd/GC electrode surface. The remarkable feature of curve
d ͑Fig. 5͒ is that it shows a crossover point, that is, the current
intensity in the potential zone of Ϫ0.275 to Ϫ0.145 V is unusually
high during the anodic sweep as compared with the intensity in the
same zone of potential during the cathodic sweep. This is explained
as follows: In the cathodic sweep, the ORR current begins to grow
only after the UPD of Sn2+ onto the Pd/GC electrode commences.
During the anodic sweep, Sn underwent a partial stripping at E Ͼ
−0.275 V, i.e., the electrode surface remained partially covered by
the deposited Sn that catalyzed the ORR, resulting in an increase in
the current up to a potential of Ϫ0.145 V. The hydrodynamic volta-
mmograms for the ORR ͑not shown͒ also indicate that the half-wave
potential of the ORR during the anodic sweep is about 100 mV more
positive than that during the cathodic sweep. Therefore, the remain-
ing Sn deposits on the electrode surface catalyze the ORR up to a
considerable positive potential during the anodic sweep.
ORR at the Sn–Pd/GC electrode.— ORR was carried out at the
Pd/GC electrode in the presence of SnSO4, that is, at the Sn–Pd/GC
electrode; the results are presented in Fig. 5. An intense single ca-
thodic peak ͑designated as I ͒ appears at Ϫ0.34 V, which is about
Љ
160 and 110 mV more positive as compared with peaks I and I
obtained at the bare GC and Sn/GC electrodes, respectively ͑curves
Ј
b and d in Fig. 4͒. The intensity of peak I is almost twice that of
Љ
peaks I and I . Peak I may therefore be assigned to the catalytic
Ј
Љ
single-step four-electron reduction of O2 to H2O at the Sn–Pd/GC
electrode according to Eq. 4
O2 + 4H+ + 4e− = 2H2O
͓4͔
The Pt electrode is well known as a highly active material for the
single-step four-electron ORR. Therefore, for comparison, the cur-
rent density for the ORR was obtained at the Pt electrode in
O2-saturated 0.5 M H2SO4 ͑curve e͒. It can be seen that the intensity
of peak I is comparable in its intensity to peak IIV, suggesting that
Љ
the ORR at the Sn–Pd/GC electrode proceeded through single-step
consumtion of four electrons and produced H2O as the final product
of the reaction. If the ORR proceeds through a four-electron path-
way at the Sn–Pd/GC electrode, then H2O2 can be expected to un-
The ORR was also carried out at the Pd/GC electrode in the
Downloaded on 2014-07-12 to IP 129.173.72.87 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

B1146
Journal of The Electrochemical Society, 156 ͑10͒ B1142-B1149 ͑2009͒
6
4
2
0
absence of SnSO4; the result is shown in Fig. 5b. It shows that a
cathodic peak ͑designated as I ͒ assigned to the ORR appears at
ٞ
0.14 V with a current intensity approximately half of peaks I and
Љ
IIV. Pd electrocatalysts possess an activity for the ORR comparable
to Pt or Pt-containing electrocatalysts.22-24,26-28,33-40 However, the
activity of the Pd catalysts largely depends on their synthetic proce-
dures, combination with other metals, and, more importantly, on the
amount of Pd loading in the catalysts. In the present case, the Pd/GC
electrode possesses considerably less activity for the ORR as com-
pared with the bulk Pt electrode as well as many other reported Pd
electrocatalysts22,24,26 probably because of a small amount of Pd
loading. For example, the amount of the deposited Pd onto the GC
electrode in the present study was 102–104 times less as compared
with the highly active Pd electrocatalysts reported by Shao et al.24,26
and Salvador-Pascual et al.22 A minor cathodic peak at the Pd/GC
electrode is observed at about Ϫ0.05 V and it could be attributed to
the further reduction of the produced H2O2. The second minor ca-
thodic peak grew more prominently with increasing the amount of
Pd loading onto the GC electrode ͑discussed later͒. A comparable
behavior of the ORR was also reported by Lin et al.58 for chemically
synthesized Pd–carbon nanotube catalyst supported on the GC elec-
trode in 1.0 M H2SO4 solution. In the present research the addition
0.36
0.34
0.32
0.3
–2
–1
0
–1
log
ν
/ (V s
)
0
0.4
ν 1/2 / (V s
0.8
1.2
–1
1/2
)
of SnSO into the solution completely diminished peak I for the
ٞ
4
Figure 7. Linear plot of cathodic peak current vs the square root of scan rate.
Inset shows the plot of peak potential vs log ␯. The corresponding peak
currents and peak potentials were derived from CVs obtained at the Pd/GC
electrode in O2-saturated 0.5 M H2SO4 solution containing 1.0 mM SnSO4
over the scan rate range of 0.005–0.8 V s−1 ͑not shown͒.
ORR in exchange of the appearance of an intense and sharp peak I
Љ
as described above. The disappearance of peak I in the presence of
ٞ
SnSO4 is attributed to the adsorption of ionic Sn2+ onto the Pd
particles of the Pd/GC electrode. The adsorbed Sn2+ ions blocked
the active sites of the electrode and thereby very efficiently inhibited
the ORR in the potential region of peak I . The ORR was also
ٞ
͑r2 = 0.9995͒, suggesting that the reaction was solely controlled by
the mass transfer of oxygen to the electrode surface. The CVs for the
ORR were repeatedly recorded at several potential scans in the same
bath. Highly reproducible results were obtained with a relative stan-
dard deviation ͑RSD͒ of Ͻ1% ͑n = 5͒. The cathodic peak potentials
were also derived from the CVs and plotted as a function of log ␯.
The inset of Fig. 7 shows that the values at ␯ Ն 0.2 V s−1 fall on a
straight line ͑r2 = 0.9994͒. From the slope of the straight line the
cathodic electron transfer coefficient ͑␣c͒ was calculated as 0.41 by
considering the four-electron ORR.
The Pd/GC ͑Pd deposition time of 40 s͒ was modified by the
deposition of Sn2+ by scanning the potential from 0.2 to Ϫ0.5 V in
0.5 M H2SO4 solution containing 1.0 mM SnSO4 solution at a po-
tential scan rate of 0.1 V s−1. The thus-prepared Sn–Pd/GC elec-
trode was taken from the solution, washed thoroughly with Milli-Q
water, and used to investigate the ORR in O2-saturated 0.5 M
H2SO4 solution in the absence of SnSO4; the obtained results are
shown in Fig. 8. A CV obtained at the first cycle ͑curve a͒ shows a
peak current for the ORR very similar to that obtained in
SnSO4-containing solution ͑curve d in Fig. 5͒. However, with re-
peated potential scanning, the intensity of the peak current gradually
decreased and the peak potential shifted to the positive direction of
the potential. A CV obtained at the ninth cycle ͑curve i͒ is similar to
that obtained at the Pd/GC electrode in SnSO4-free solution. With
the repeated potential cycle in SnSO4-free solution, the Sn under-
went an irreversible anodic stripping and disappeared from the elec-
trode surface, and consequently the electrode lost its catalytic activ-
ity for the four-electron ORR. For further support, CVs were also
obtained at the Sn–Pd/GC electrode in N2-saturated 0.5 M H2SO4
solution under the same conditions as those used for measuring
curves a–i. The typical CVs are presented in the inset of Fig. 8. The
investigated at the Pd electrode in O2-saturated 0.5 M H2SO4 solu-
tion in the presence of SnSO4, and it was observed that the peak
current for the ORR decreased drastically because of the adsorption
of Sn2+ onto the electrode surface.32 A similar decrease in the ORR
current at the Pt electrode was also reported by Zinola et al.29,30
The spontaneous adsorption of Sn2+ onto the Pd/GC electrode
was confirmed by measuring the open-circuit potential ͑EOCP
͒
change, which is useful for investigating the adsorption of various
species onto the electrode surface.59-61 For this purpose, the Pd/GC
electrode was held in the N2-saturated 0.5 M H2SO4 solution under
the open-circuit condition and then 1.0 mL of 0.02 M SnSO4 solu-
tion was injected into the solution ͑the final concentration of SnSO4
was 1.0 mM͒. It was observed that the EOCP of the electrode shifted
instantaneously to the negative direction by about 400 mV, indicat-
ing that Sn2+ undergoes a rapid adsorption onto the Pd/GC
electrode.59-61 The same experiment was also repeated using the
bare GC electrode and only a slight negative shift in EOCP ͑ca. 50
mV͒ was observed, suggesting very small adsorption of Sn2+ onto
the bare GC electrode. Spontaneous adsorption of Sn2+ onto the
Pd/GC electrode and thereby a large negative shift in EOCP implies
that the adsorption was triggered by electron donation from Sn2+
ions to Pd.59-61 For further support, the Pd/GC electrode was re-
moved from the SnSO4-containing solution and thoroughly washed,
and the potential was scanned as 0.2 → −0.5 → 0.2 V in
N2-saturated 0.5 M H2SO4 solution containing no SnSO4. A ca-
thodic peak similar to that at Ϫ0.4 V in Fig. 2e was clearly ob-
served. The observed cathodic peak is therefore considered to cor-
respond to the UPD of Sn2+, which underwent adsorption onto the
Pd/GC electrode surface during the EOCP measurement. The sponta-
neous adsorption of Sn2+ has also been reported elsewhere for other
metallic electrodes.30,32,62 Therefore, the disappearance of peak I
ٞ
for the ORR is obviously related to the adsorption of Sn2+ and the
blocking of the active sites of the Pd/GC electrode.
CV obtained at the first cycle ͑curve a ͒ clearly shows a cathodic
Ј
peak at about Ϫ0.4 V assigned to the UPD of Sn2+ ͑discussed
above͒. The peak, however, diminished gradually with repeated po-
tential scanning, suggesting that Sn underwent anodic stripping and
disappeared from the electrode surface. H2 evolution current also
supports the disappearance of the Sn deposit from the electrode
surface. The Sn adlayer at the Pd electrode was also highly unstable
under potential cycle in 0.5 M H2SO4 solution ͑not shown͒. The Sn
The ORR was investigated at the Sn–Pd/GC electrode at poten-
tial scan rates ranging from 0.005 to 0.8 V s−1. The cathodic peak
currents were derived from the obtained CVs ͑not shown͒ and plot-
ted ͑after the correction of the background current corresponding to
the upd of Sn2+͒ as a function of ␯1/2. Figure 7 shows that the peak
currents fall nicely on a straight line passing through the origin
Downloaded on 2014-07-12 to IP 129.173.72.87 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 156 ͑10͒ B1142-B1149 ͑2009͒
B1147
0
–1
–2
i
0
–0.8
–1.6
h
j
g
e
b
c
0.1
0
f
a
0
–0.4
–0.8
e
e'
b'
g' h' i'
j'
d
a'
c
–0.1
–0.2
d
b
a
0
0.5
E / V vs. Ag / AgCl / NaCl (sat.)
–0.4
0
0.4
E/ V vs. Ag / AgCl / NaCl (sat.)
–0.4
–0.2
0
0.2
–0.4
0
0.4
E / V vs. Ag / AgCl / NaCl (sat.)
E / V vs. Ag / AgCl / NaCl (sat.)
Figure 9. CVs obtained at the Pd/GC electrodes in O2-saturated 0.5 M
H2SO4 solution containing 1.0 mM SnSO4. Pd was electrodeposited onto the
GC electrode under the same conditions as in Fig. 1 except for the deposition
times of ͑a͒ 3, ͑b͒ 10, ͑c͒ 20, ͑d͒ 40, and ͑e͒ 60 s. Inset shows the CV
obtained at the Pd/GC electrode ͑Pd deposition time of 60 s͒ for the ORR in
Figure 8. CVs obtained at the ͓͑a͒–͑i͔͒ Sn–Pd/GC and ͑j͒ Pd/GC electrodes
in O2-saturated 0.5 M H2SO4 solution ͑containing no SnSO4͒ at different
potential cycles: ͓͑a͒ and ͑j͔͒ first, ͑b͒ second, ͑c͒ third, ͑d͒ fourth, ͑e͒ fifth,
͑f͒ sixth, ͑g͒ seventh, ͑h͒ eighth, and ͑i͒ ninth cycles. Potential scan rate of
0.1 V s−1. CVs for the background responses were obtained under the same
experimental conditions as ͑a͒–͑i͒ except the solution was N2-saturated. For
O2-saturated 0.5 M H2SO4 solution. Potential scan rate of 0.1 V s−1
.
clarity, CVs obtained at the ͓͑a ͒ and ͑j ͔͒ first, ͑b ͒ second, ͑e ͒ fifth, ͑g ͒
Ј
Ј
Ј
Ј
Ј
seventh, ͑h ͒ eighth, and ͑i ͒ ninth cycles are shown in the inset.
Ј
Ј
sence of SnSO4; the result is shown in the inset of Fig. 9. The
consecutive cathodic waves at 0.26 and 0 V, assigned to the O2
reduction to H2O2 and further reduction of the produced H2O2 to
H2O, respectively, were clearly observed.58 Therefore, the two-step
ORR at the Pd/GC electrode became prominent with increasing the
amount of deposited Pd ͑compare with Fig. 5b͒. On the contrary, the
same Pd/GC electrode showed only a single broad cathodic peak at
around 0 V for the ORR in SnSO4-containing solution ͑curve e͒. The
peak potential in the presence of SnSO4 shifted to the negative po-
tential by 260 mV and the peak intensity was 40% less as compared
with that obtained in the absence of SnSO4. The decrease in the
ORR current and its shifting to a higher overpotential is a conse-
quence of the adsorption of Sn2+ onto the Pd/GC electrode, as dis-
cussed in the previous section.
adlayer at the Pd/GC electrode is more stable as compared with that
at the Au ͑poly͒ electrode surface in the same media.32 The stability
of the Sn adlayer is influenced by the nature of the substrates and pH
of the electrolytic solutions,63,64 but a highly stable adlayer, particu-
larly in acidic media, is not favored. The ORR in the present study
was therefore carried out in Sn2+-containing solution for stable elec-
trode response.
Effect of Pd loading of the Sn–Pd/GC electrode on the
ORR.— The peak current intensity and peak potential of the ORR at
the Sn–Pd/GC electrode were remarkably influenced by the amount
of Pd predeposited onto the GC electrode surface. To clarify the
effect, Pd was electrodeposited onto the GC electrode for ͑a͒ 3, ͑b͒
10, ͑c͒ 20, ͑d͒ 40, and ͑e͒ 60 s, keeping other conditions the same as
those mentioned in Fig. 1. The corresponding amounts of deposited
Pd were calculated as ͑a͒ 1.29 ϫ 10−7, ͑b͒ 1.91 ϫ 10−7, ͑c͒ 3.25
ϫ 10−7, ͑d͒ 5.40 ϫ 10−7, and ͑e͒ 7.5 ϫ 10−7 g cm−2. The ORR
was investigated at all of the thus-prepared Pd/GC electrodes in
O2-saturated 0.5 M H2SO4 solution in the presence of 1.0 mM
SnSO4; the results are shown in Fig. 9. The results clearly show that
the peak current for the ORR gradually increased and the peak po-
tential shifted to the positive direction of the potential with increas-
ing the amount of Pd when it was less than 5.40 ϫ 10−7 g cm−2
͑deposition time of 40 s͒. The greater amount of Pd loading ͑depo-
sition time longer than 40 s͒ gave results which are far different
from those obtained for various amounts of Pd loading in the range
of 1.29 ϫ 10−7 to 5.40 ϫ 10−7 g cm−2 ͑deposition time of 3–40 s͒.
For example, a typical result obtained for the Pd loading equal to
7.5 ϫ 10−7 g cm−2 ͑deposition time of 60 s͒ shows several distinct
features: ͑i͒ Tremendous H2 evolution at E Ͻ −0.1 V, ͑ii͒ the ap-
pearance of a broad minor cathodic peak at around 0 V assigned to
the ORR, ͑iii͒ the appearance of an anodic peak attributed to the
desorption of H2 from the Pd particles at about 0.1 V, and ͑iv͒
overlapping of the Sn deposition with the H2 evolution reaction,
thus obscuring the four-electron ORR. The ORR was also investi-
gated at the Pd/GC electrode with the Pd loading of 7.5
ϫ 10−7 g cm−2 in O2-saturated 0.5 M H2SO4 solution in the ab-
RRDE
measurements
for
the
ORR
at
the
Sn–Pd/GC.— RRDE measurement is the key to study for under-
standing the pathway of the ORR. Accordingly, the measurements
were performed using the Pd/GC-disk ͑modified by predeposition of
Pd for 40 s͒ and Pt-ring RRDEs in O2-saturated 0.5 M H2SO4 so-
lution containing 1.0 mM SnSO4 at electrode rotation rates of ͑a͒
1000, ͑b͒ 2000, ͑c͒ 3000, and ͑d͒ 4000 rpm at a potential scan rate of
0.01 V s−1. The Pt-ring electrode was operated at 1.0 V, which is
sufficiently high for the oxidation of the produced H2O2 ͑if any͒ at
the Sn–Pd/GC-disk electrode of the RRDE. The results clearly
showed almost a zero ring current all over the potential window
under investigation ͑Fig. 10͒. The percentage of the electrogenerated
65
H2O2 ͑XH O ͒ was calculated using Eq. 5
2
2
2IR/N
ID + ͑IR/N͒
XH O
=
2
ϫ 100
͓5͔
2
where IR and ID are the ring and disk currents, respectively, and N is
the collection efficiency of the RRDE and was used as
0.40 Ϯ 0.02.9 The obtained values of XH O vary in the range of
2
0–3% all over the examined potential ra2nge, suggesting that the
ORR at the Sn–Pd/GC electrode proceeded exclusively through a
single-step four-electron pathway forming H2O as a final product.
Detailed analysis of the ring-disk electrode and RRDE data is cur-
rently underway with a view of evaluating various kinetic param-
Downloaded on 2014-07-12 to IP 129.173.72.87 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

B1148
Journal of The Electrochemical Society, 156 ͑10͒ B1142-B1149 ͑2009͒
which is required for the four-electron ORR. The ORR at the GC
IR x 5
electrode therefore generated bulk H2O2 as the two-electron reduc-
tion product. The produced H2O2 did not undergo further reduction
because of the very slow kinetics of the reaction at the GC electrode.
Unlike the bare GC electrode, the ORR at the Sn/GC electrode, that
is, in the presence of SnSO4, proceeded through a two-step four-
electron pathway, forming H2O as the final product. The first ca-
thodic peak current appeared before the onset potential of deposition
of Sn2+ and thus remained unaltered in both its intensity and poten-
tial in the absence and presence of SnSO4. Interestingly, the pro-
duced H2O2 in the presence of SnSO4 underwent further reduction
in the potential region of the formation of Sn deposits. Produced
H2O2 may undergo adsorption onto the Sn deposits forming a
GCSnOOH2͑ads͒ abduct which may result in the advantageous cleav-
0
a
b
ID
–5
c
age of the oxygen–oxygen bond in H2O2 required for its reduction
to H2O ͑Scheme 1B͒.53-56 Premodification of the GC electrode with
Pd decreased the overpotential of deposition of Sn2+ and conse-
quently led to a radical improvement of the catalytic activity toward
an exclusive single-step four-electron ORR as compared with the
GC and Sn/GC electrodes. The improved activity of the Sn–Pd/GC
electrode may be correlated with an electronic interaction between
Pd and Sn. The electronic structures of Pd͑46͒ and Sn͑50͒ are
͓Kr͔4d104f0 and ͓Kr͔4d105s25p2, respectively. The value of elec-
tronegativity of Pd ͑2.2͒ is larger than that of Sn ͑1.96͒. Sn, being
less electronegative, can show a so-called ligand effect,66,67 that is,
Sn can partially donate its electron to the neighboring Pd, as shown
in Scheme 1C. The EOCP measurements showed that Sn2+ ͑even in
the 2+ oxidation state͒ can donate its electron to Pd and adsorb
strongly on it. Therefore, partial donation of an electron from the
metallic Sn to Pd is reasonable. Very recently we reported that the
donation of electron from Sn to the Au atom of the Au ͑poly͒ sub-
strate largely increases the activity of the electrode toward the
ORR.32 Donation of an electron from Snx+ ͑in SnOx͒ to a Au atom
and enhancement of the ORR in acidic media has also been reported
elsewhere.66,67 The donation of an electron to Pd makes the Sn
electron deficient and thus O2 molecule may interact more strongly
with Sn adatoms ͑which become electron deficient as a result of
donation of electrons to the Pd͒ through the electron donation from
the 2␲ orbitals of O2 to the 4dz2 orbitals of Sn adatoms and the back
donation of an electron from the 4dxz or 4dyz orbital of Sn adatoms
d
–0.6
–0.3
/ V vs. Ag / AgCl / NaCl (sat.)
0
E
Figure 10. Steady-state voltammograms obtained at the Pd/GC-disk ͑Pd
deposition of 40 s͒ and Pt-ring RRDEs in O2-saturated 0.5 M H2SO4 solution
containing 1.0 mM SnSO4 at various electrode rotation rates: ͑a͒ 1000, ͑b͒
2000, ͑c͒ 3000, and ͑d͒ 4000 rpm. The Pt-ring electrode was operated at 1.0
V. Potential scan rate of 0.01 V s−1
.
eters of the ORR at the Sn–Pd/GC electrode and will be reported
elsewhere.
Mechanism of the ORR at the Sn–Pd/GC electrode.— GC elec-
trode is well known to be active typically only for the two-electron
reduction of O2 to H2O2.20,51,52 The poor activity of the GC elec-
trode toward the ORR is ascribed to its poor ability to adsorb O2.
Poorly adsorbed O2 molecules onto the GC electrode may have an
end-on orientation, as shown in Scheme 1A. The end-on adsorption
mode is not suitable for the cleavage of the oxygen–oxygen bond,
*
to the 2␲ orbitals of O2, resulting in a decrease in the bond order in
H2O
H2O2
an O2 molecule ͑Scheme 1C͒. Such an orbital interaction of an O2
O
O
H2O2 H2O
O
O
molecule with the active sites for ORR is also considered elsewhere
Sn
32
×
͑e.g., Sn-adatom-modified Au
and Pt alloys with Fe-group
A
C
GC
GC
B
metals68͒. Pd is known to have a high binding strength to O2.26,28
Donation of an electron from Sn can slightly reduce the adsorption
affinity of Pd for an O2 molecule, but still O2 can be adsorbed on it
because of its very high O2 binding strength, in the same fashion as
shown for Sn. To possess the maximum activity for the ORR, the
catalyst should have a “moderate” O2 binding strength, as a high
binding strength can lead to the adsorption of harmful oxyspecies
produced as intermediates in the ORR.26,28 Therefore, a decrease in
the binding strength of O2 to Pd as a result of the ligand effect of the
neighboring Sn can advantageously improve the activity of the Sn–
Pd/GC electrode for the ORR. It has also been reported that alloying
with Fe results in a decrease in the binding strength of Pd for O2 but
improves the ORR activity.26,28 O2 molecules on the Sn–Pd bime-
tallic surface can thus undergo adsorption having a parallel orienta-
tion, as shown in Scheme 1D. The oxygen–oxygen bond in such an
adsorbed O2 molecule with reduced bond order can undergo cleav-
age efficiently, which is required for the four-electron ORR.
2e−
2e−
2e−
O2
O2
2π
2π∗
backdonation
donation
Sn
Sn
2
4dyz
4dz
H2O
O
O
Pd
Pd
GC
Sn
2
4dz
D
4e−
Scheme 1. ͑A͒ End-on orientation of the adsorbed O2 molecule on the GC
electrode and the two-electron ORR forming H2O2 which does not undergo
further reduction. ͑B͒ End-on orientation of the adsorbed O2 molecule on the
GC electrode and the two-electron ORR forming H2O2 which undergoes
further reduction catalyzed by the Sn deposit. ͑C͒ Orbital illustration of
donation of electron from O2 to Sn and from Sn to Pd and backdonation from
the Sn to O2. ͑D͒ Parallel orientation of the adsorbed O2 molecule on the
Pd–-Sn bimetallic surface and four-electron ORR forming H2O.
Conclusions
The results demonstrate that the Pd–Sn bimetallic electrocatalyst,
which was electrochemically fabricated onto the GC electrode, pos-
sesses a unique catalytic activity toward an exclusive single-step
four-electron ORR forming H2O as the final product. The preloading
level of Pd in the catalyst showed a remarkable effect on the ORR,
Downloaded on 2014-07-12 to IP 129.173.72.87 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 156 ͑10͒ B1142-B1149 ͑2009͒
B1149
that is, an optimized amount of Pd equal to 5.40 ϫ 10−7 g cm−2 led
to the maximum catalytic effect for the ORR. The observed en-
hanced activity of the Pd–Sn bimetallic electrocatalyst was attrib-
uted to the ligand effect of Sn, that is, the donation of an electron
from the Sn to Pd, which increased the O2 adsorption affinity of Sn.
Such a ligand effect may slightly decrease the O2 adsorption affinity
of Pd but O2 still can undergo adsorption on it as it has a very high
O2 binding strength. The O2 molecules therefore undergo adsorption
onto the bimetallic surface with a parallel orientation which facili-
tates the cleavage of the oxygen–oxygen bonds required for the
single-step four-electron ORR forming H2O.
25. S. Szunerits and R. Boukherroub, C. R. Chim., 11, 1004 ͑2008͒.
26. M. H. Shao, P. Liu, J. Zhang, and R. Adzic, J. Phys. Chem. B, 111, 6772 ͑2007͒.
27. M. Arenz, T. J. Schmidt, K. Wandelt, P. N. Ross, and N. M. Markovic, J. Phys.
Chem. B, 107, 9813 ͑2003͒.
28. M. H. Shao, T. Huang, P. Liu, J. Zhang, K. Sasaki, M. B. Vukmirovic, and R. R.
Adzic, Langmuir, 22, 10409 ͑2006͒.
29. C. F. Zinola and J. Rodríguez, J. Solid State Electrochem., 6, 412 ͑2002͒.
30. C. F. Zinola, J. Rodriguez, and G. Obal, J. Appl. Electrochem., 31, 1293 ͑2001͒.
31. C. Jeyabharathi, P. Venkateshkumar, J. Mathiyarasu, and K. L. N. Phani, Electro-
chim. Acta, 54, 448 ͑2008͒.
32. Md. R. Miah and T. Ohsaka, J. Electrochem. Soc., 156, B429 ͑2009͒.
33. J. L. Fernandez, V. Raghuveer, A. Manthirum, and A. J. Bard, J. Am. Chem. Soc.,
127, 13100 ͑2005͒.
34. M. Nie, P. K. Shen, and Z. Wei, J. Power Sources, 167, 69 ͑2007͒.
35. X. Li, Q. Huang, Z. Zou, B. Xia, and H. Yang, Electrochim. Acta, 53, 6662 ͑2008͒.
36. W. Wang, D. Zheng, C. Dua, Z. Zoua, X. Zhang, B. Xia, H. Yang, and D. L. Akins,
J. Power Sources, 167, 243 ͑2007͒.
Acknowledgments
This work was supported by the Grant-in-Aid for Scientific Re-
search ͑no. 12875164͒ and Scientific Research ͑A͒ ͑no. 19206079͒
to T. Ohsaka from the Ministry of Education, Culture, Sports, Sci-
ence and Technology ͑MEXT͒, Japan and also by the New Energy
and Industrial Technology Development Organization ͑NEDO͒, Ja-
pan. Md. R. Miah thanks the Government of Japan for a MEXT
scholarship.
37. G. Ramos-Sánchez, H. Yee-Madeira, and O. Solorza-Feria, Int. J. Hydrogen En-
ergy, 33, 3596 ͑2008͒.
38. A. Sarkar, A. V. Murugan, and A. Manthiram, J. Phys. Chem. C, 112, 12037
͑2008͒.
39. O. Savadogo, K. Lee, K. Oishi, S. Mitsushima, N. Kamiya, and K.-I. Ota, Electro-
chem. Commun., 6, 105 ͑2004͒.
40. X. Wang, N. Kariuki, J. T. Vaughey, J. Goodpaster, R. Kumar, and D. J. Myers, J.
Electrochem. Soc., 155, B602 ͑2008͒.
41. C. Batchelor-McAuley, C. E. Banks, A. O. Simm, T. G. J. Jones, and R. G. Comp-
ton, Analyst (Cambridge, U.K.), 131, 106 ͑2006͒.
42. M. A. Ghanem, H. Hanson, R. G. Compton, B. A. Coles, and F. Marken, Talanta,
Tokyo Institute of Technology assisted in meeting the publication costs of
this article.
72, 66 ͑2007͒.
References
43. V. C. Diculescu, A.-M. Chiorcea-Paquim, O. Corduneanu, and A. M. Oliveria-
Brett, J. Solid State Electrochem., 11, 887 ͑2007͒.
44. A. L. N. Pinheiro, M. S. Zei, M. F. Luo, and G. Ertl, Surf. Sci., 600, 641 ͑2006͒.
45. H. Naohara, S. Ye, and K. Uosaki, J. Electroanal. Chem., 473, 2 ͑1999͒.
46. M. E. Quayum, S. Ye, and K. Uosaki, J. Electroanal. Chem., 520, 126 ͑2002͒.
47. L. A. Kibler, M. Kleinert, V. Lazarscu, and D. M. Kolb, Surf. Sci., 498, 175
͑2002͒.
48. V. A. Vicente and S. Bruckenstein, Anal. Chem., 44, 297 ͑1972͒.
49. Z. Qiao, W. Shang, and X. Zhang, Anal. Bioanal. Chem., 381, 1467 ͑2005͒.
50. R. L. Broggi, G. M. Deoliveira, L. L. Barbosa, E. M. J. A. Pallone, and I. A.
Carlos, J. Appl. Electrochem., 36, 403 ͑2006͒.
1. R. Adzic, A. Tripkovic, and R. Atanasoski, J. Electroanal. Chem., 94, 231 ͑1978͒.
2. A. Kuzume, E. Herrero, and J. M. Feliu, J. Electroanal. Chem., 599, 333 ͑2007͒.
3. M. M. Walczak, C. A. Alves, B. D. Lamp, and M. D. Porter, J. Electroanal. Chem.,
396, 103 ͑1995͒.
4. C. L. Perdriel, A. J. Arvia, and M. Ipohorski, J. Electroanal. Chem., 215, 317
͑1986͒.
5. J. Gomez, L. Vasquez, A. M. Baro, C. L. Perdriel, and A. J. Arvia, Electrochim.
Acta, 34, 619 ͑1989͒.
6. E. Higuchi, H. Uchida, and M. Watanabe, J. Electroanal. Chem., 583, 69 ͑2005͒.
7. J. Hernández, J. Solla-Gullón, and E. Herrero, J. Electroanal. Chem., 574, 185
͑2004͒.
8. S. Strbac and R. R. Adzic, J. Electroanal. Chem., 403, 169 ͑1996͒.
9. M. S. El-Deab and T. Ohsaka, Electrochem. Commun., 5, 214 ͑2003͒.
10. E. R. Vago, K. de Weldige, and M. Rohwerder, Fresenius’ J. Anal. Chem., 353,
316 ͑1995͒.
11. Md. R. Miah and T. Ohsaka, Electrochim. Acta, 52, 6378 ͑2007͒.
12. M. B. Vukmirovic, J. Zhang, K. Sasaki, A. U. Nilekar, F. Uribe, M. Mavrikakis,
and R. R. Adzic, Electrochim. Acta, 52, 2257 ͑2007͒.
51. M. M. Saleh, M. I. Awad, T. Okajima, K. Suga, and T. Ohsaka, Electrochim. Acta,
52, 3095 ͑2007͒.
52. M. S. El-Deab, T. Sotomura, and T. Ohsaka, Electrochem. Commun., 7, 29 ͑2005͒.
53. S.-J. Hsieh and A. A. Gewirth, Surf. Sci., 498, 147 ͑2002͒.
54. S. M. Sayed and K. Juttner, Electrochim. Acta, 28, 1635 ͑1983͒.
55. R. Amadelli, N. Markovic, R. Adzic, and E. Yeager, J. Electroanal. Chem., 159,
391 ͑1983͒.
56. F. Matsumoto, S. Uesugi, N. Koura, and T. Ohsaka, J. Electroanal. Chem., 549, 71
͑2003͒.
13. S. Ben Aoun, Z. Dursun, T. Sotomura, and I. Taniguchi, Electrochem. Commun., 6,
747 ͑2004͒.
57. M. S. El-Deab and T. Ohsaka, Electrochem. Commun., 4, 288 ͑2002͒.
58. Y. Lin, X. Cui, and X. Ye, Electrochem. Commun., 7, 267 ͑2005͒.
59. Md. R. Miah and T. Ohsaka, J. Electrochem. Soc., 153, E195 ͑2006͒.
60. Md. R. Miah and T. Ohsaka, Electrochim. Acta, 54, 1570 ͑2009͒.
61. Md. R. Miah, J. Masud, and T. Ohsaka, Electrochim. Acta, 54, 316 ͑2008͒.
62. K. Shimazu, T. Kuwaguchi, and K. Tada, J. Electroanal. Chem., 529, 20 ͑2002͒.
63. A. Rodes, E. Herrero, J. M. Feliu, and A. Aldaz, J. Chem. Soc., Faraday Trans.,
92, 3769 ͑1996͒.
64. S. A. Campbell and R. Parsons, J. Chem. Soc., Faraday Trans., 88, 833 ͑1992͒.
65. M. S. El-Deab and T. Ohsaka, Electrochim. Acta, 52, 2166 ͑2007͒.
66. D. S. Gatewood, D. E. Ramaker, K. Sasaki, and K. E. Swider-Lyons, J. Electro-
chem. Soc., 155, B834 ͑2008͒.
14. T. Abe, G. M. Swain, K. Sashikata, and K. Itaya, J. Electroanal. Chem., 382, 73
͑1995͒.
15. R. R. Adzic, J. Wang, and B. M. Ocko, Electrochim. Acta, 40, 83 ͑1995͒.
16. F. Matsumoto, S. Uesugi, N. Koura, T. Okajima, and T. Ohsaka, J. Electroanal.
Chem., 505, 150 ͑2001͒.
17. M. S. El-Deab and T. Ohsaka, Electrochim. Acta, 47, 4255 ͑2002͒.
18. M. S. El-Deab and T. Ohsaka, J. Electroanal. Chem., 553, 107 ͑2003͒.
19. M. S. El-Deab, T. Sotomura, and T. Ohsaka, Electrochim. Acta, 52, 1792 ͑2006͒.
20. M. S. El-Deab, T. Okajima, and T. Ohsaka, J. Electrochem. Soc., 150, A851
͑2003͒.
21. T. Nishimura, T. Morikawa, M. Yokoi, C. Iwakura, and H. Inoue, Electrochim.
Acta, 54, 499 ͑2008͒.
67. W. S. Baker, J. J. Pietron, M. E. Teliska, P. J. Bouwman, D. E. Ramaker, and K. E.
22. J. J. Salvador-Pascual, S. Citalan-Cigarroa, and O. Solorza-Feria, J. Power Sources,
Swider-Lyons, J. Electrochem. Soc., 153, A1702 ͑2006͒.
172, 229 ͑2007͒.
68. T. Toda, H. Igarashi, H. Uchida, and M. Watanabe, J. Electrochem. Soc., 146, 3750
͑1999͒.
23. L. Cheng, Z. Zhang, W. Niu, G. Xu, and L. Zhu, J. Power Sources, 182, 91 ͑2008͒.
24. M. H. Shao, K. Sasaki, and R. R. Adzic, J. Am. Chem. Soc., 128, 3526 ͑2006͒.
Downloaded on 2014-07-12 to IP 129.173.72.87 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).