Electrocatalytic dioxygen reduction on underpotentially deposited Pb on Au(111) studied by an active site blocking strategy

Article abstract of DOI:10.1016/S0021-9517(02)00011-8

Electrochemical measurements and in situ scanning tunneling microscopy (STM) were carried out to establish a structure-reactivity correlation for peroxide or dioxygen reduction on underpotentially deposited (upd) Pb on Au(111) in 0.1 M HClO₄. STM imaging revealed the presence of Pb islands with height of 0.25 ± 0.05 nm at the potential of highest catalytic activity toward the O₂ and H₂O₂ reduction. The full Pb monolayer formed at - 0.03 v vs. NHE showed about half the activity of the Pb islands. Ethanethiol (EtSH) considerably, but not completely, inhibited H₂O₂ reduction activity of the Pb island structure. EtSH introduction resulted in the formation of a 0.13-nm-high terrace along the edges of the Pb islands, which was assigned to EtSH bound to the Au surface near the Pb islands with the alkyl chain oriented roughly perpendicular to the surface. These results showed that edge sites around the Pb island were the active site of catalysis, though the sites atop the Pb islands might also take part in catalytic O₂ reduction by Pb upd on Au(111).

Full text of DOI:10.1016/S0021-9517(02)00011-8

Journal of Catalysis 213 (2003) 17–22
www.elsevier.com/locate/jcat
Electrocatalytic dioxygen reduction on underpotentially deposited Pb
on Au(111) studied by an active site blocking strategy
Ilwhan Oh,a Andrew A. Gewirth,b,∗ and Juhyoun Kwak a,∗
a
Department of Chemistry, Korea Advanced Institute of Science and Technology (KAIST), 373-1 Guseong-dong, Yuseong-gu,
Daejeon 305-701, Republic of Korea
b
Department of Chemistry and the Frederick Seitz Materials Research Laboratory, University of Illinois, Urbana, IL 61801, USA
Received 22 February 2002; revised 20 August 2002; accepted 24 August 2002
Abstract
Electrochemical measurements and in situ scanning tunneling microscopy (STM) are performed to establish a structure–reactivity
correlation for peroxide or dioxygen reduction on underpotentially deposited (upd) Pb on Au(111) in 0.1 M HClO . While STM imaging
4
reveals the presence of Pb islands with height of 0.25 0.05 nm at the potential of highest catalytic activity toward the O and H O
2
2 2
reduction, the full Pb monolayer formed at −0.03 V vs. NHE shows about half the activity of the Pb islands. Ethanethiol (EtSH) significantly
but not completely inhibits the H O reduction activity of the Pb island structure. STM shows that EtSH introduction leads to the formation
2
2
of a 0.13-nm-high terrace along the edges of the Pb islands, which is assigned to EtSH bound to the Au surface near the Pb islands with
the alkyl chain oriented roughly perpendicular to the surface. These results show that edge sites around the Pb island are the active site of
catalysis, though the sites atop the Pb islands may also take part in catalytic O reduction by Pb upd on Au(111).
2
2003 Elsevier Science (USA). All rights reserved.
Keywords: Electrocatalysis; Oxygen reduction; Fuel cell; Underpotential deposition; Scanning tunneling microscopy
1. Introduction
catalysis of O2 reduction forms the basis of applications such
as fuel cells, metal–air batteries, and corrosion, and so has
been extensively studied [4–6].
Surfaces modified with monolayers of foreign metals of-
ten exhibit unique and interesting properties that are absent
in the unmodified bare surfaces. In an electrochemical en-
vironment, the metal-monolayer-modified electrode surface
can be fabricated through the process of underpotential de-
position (upd), in which the foreign metal atoms are elec-
trodeposited up to monolayer coverage prior to the bulk de-
position [1–3]. In particular, upd submonolayers of Bi, Tl,
and Pb on Au(111) are all known to catalyze the reduction
of O2. It is well accepted that the electroreduction of O2 on
most cathodes proceeds via a H2O2 intermediate:
One important goal in the study of O2 electrocatalysis
has been to correlate surface structure with catalytic activity
at the atomic level. In upd systems exhibiting an open
adlayer structure, catalytic activity was ascribed to the
local heterobimetallic geometry or to the electronic effect
imposed by the heterometal. For example, Bi upd on
Au(111) [7] exhibits a (2 × 2) open adlayer structure in
the potential region of catalytic activity, possibly implicating
the lone Au site in the unit cell as the locus of catalytic
activity. And for Tl upd on Au(111), the island edges are
understood to be the origin of the catalytic enhancement [8].
Alternatively, there are indications that other mechanisms
involving dissolution of the active upd admetal followed by
a solution redox event and redeposition of the metal may be
the case in some systems [9].
O2 + 2H+ + 2e− → H2O2,
(1)
(2)
H2O2 + 2H+ + 2e− → 2H2O.
The two-electron reduction of H2O2 in Eq. (2) is often the
rate-determining step in the whole reaction. The electro-
Pb is another upd metal which exhibits catalytic ac-
tivity toward dioxygen reduction. Pb upd on Au(111) has
been extensively studied by several experimental techniques
[10–14]. Figure 1 shows the cyclic voltammogram of a
*
Corresponding authors.
E-mail addresses: agewirth@uiuc.edu (A.A. Gewirth),
jhkwak@cais.kaist.ac.kr (J. Kwak).
0021-9517/03/$ – see front matter
PII: S0021-9517(02)00011-8
2003 Elsevier Science (USA). All rights reserved.

18
I. Oh et al. / Journal of Catalysis 213 (2003) 17–22
Oxide formation and stripping voltammetry of the surface in
pure electrolyte were found to closely match those reported
in the literature for Au(111) [16].
Voltammetric data were collected using a Pt wire coun-
terelectrode and a saturated Hg/Hg2SO4 reference electrode
connected to the electrochemical cell via a capillary salt
bridge to minimize contamination from the reference elec-
trode. All potentials in this paper are reported relative to
the normal hydrogen electrode (NHE). The solutions were
purged with Ar prior to use, and an atmosphere of Ar was
maintained in the cell during all electrochemical measure-
ments. Potential control and sweeps were established using
an Autolab potentiostat (Eco Chemie, Netherlands). Rotat-
ing disk electrode (RDE) measurements were obtained us-
ing a BAS model RDE-1 rotator (Bioanalytical Systems, IN,
USA) equipped with a collet, which holds the Au single
crystal to form a hanging meniscus with electrolyte.
STM images were obtained in constant current mode
with a Topometrix TMX2000 which was calibrated against a
highly ordered pyrolytic graphite (HOPG) surface in air for
in-plane dimensions, and against monatomic Au(111) steps
for dimensions normal to the surface. An electrochemically
etched Pt/Ir wire (Molecular Imaging, AR) coated with
Apiezon wax was used as the STM tip.
2+
Fig. 1. Cyclic voltammogram in 1.0 mM Pb + 0.1 M HClO on Au(111)
4
with scan rate of 20 mV/s. The solution is purged and blanketed with Ar
gas before measurement.
Au(111) electrode in an O2-free solution containing 1.0 mM
Pb2+ and 0.1 M HClO4, which is consistent with ones previ-
ously reported [10–14]. Initially, the electrode is poised at a
positive potential where only a bare Au surface is present.
As the electrode potential is swept negatively, no distinct
voltammetric features are observed until the cathodic peak C
at −0.03 V. On the reverse scan, the corresponding peaks A1
and A2 are observed. In previous studies using STM [12–14]
and AFM [10,11], it was observed that a number of Pb is-
lands grew at potentials positive of peak C, which then co-
alesce to the full Pb monolayer at potentials negative of
peak C.
Because Pb upd shows enhanced catalytic activity toward
H2O2 and O2 reduction only in the potential region positive
of peak C, we postulated that the Pb island structure is
responsible for the enhanced catalytic activity, a result now
strongly suggested from a thermodynamic analysis of the
catalytic behavior [15].
The working electrode for STM imaging was Au evap-
orated onto glass (Metallhandel Schroer GmbH, Germany)
or onto mica (Molecular Imaging) and annealed following
a published procedure [17]. A clean Pb wire served as the
reference electrode, while a Pt wire was used for the coun-
terelectrode. Images were obtained in height mode and typ-
ically took 1 min to complete. All images are presented un-
filtered.
3. Results and discussion
3.1. Electroreduction of peroxide on Pb
upd-modified Au(111)
In this study, we adapt an active site blocking strategy,
in which a poison species is deliberately introduced to Pb
upd catalyst to inhibit the catalytic active site. The structure
of the thus-formed poison–catalyst composite is investigated
with STM and electrochemical techniques and is discussed
in terms of the structure–reactivity correlation.
In order to establish a structure–reactivity correlation for
Pb upd, we first measured the electrocatalytic activity of the
Pb upd system toward H2O2 and O2 reduction. Figure 2A
shows the current–potential graph from a Au(111) electrode
rotating at 400 rpm in a solution containing 10 mM H2O2,
1.0 mM Pb2+, and 0.1 M HClO4 which is purged of
dissolved O2 and compares this with a Au(111) electrode in
a solution containing only 10 mM H2O2 and 0.1 M HClO4.
The solution without Pb2+ evinces a very small current
associated with peroxide reduction throughout the potential
range considered in these measurements. Au(111) alone in
acidic media is known to be a poor substrate for peroxide
electroreduction.
2. Experimental
Electrochemical solutions were prepared from ultrapure
water (Modulab, US Filter, MA, > 18 Mꢀ) and PbO
(Aldrich, 99.999%) with 0.1 M HClO4 (Aldrich, double-
distilled) as supporting electrolyte and ethanethiol (97%,
Aldrich) introduced as a poison. The working electrode for
cyclic voltammetric and chronoamperometricmeasurements
was a Au(111) single crystal (MaTeck, Germany) with a
diameter of 0.95 cm and a nominal area of 0.71 cm2. The
crystal was annealed for 3 min in hydrogen flame prior to
use and quenched in ultrapure water or slowly cooled in air.
The situation is different in the Pb2+-containing solution.
In the potential region positive of 0.4 V, virtually no current
is measured. As the potential is moved to more negative
values, the current due to the H2O2 reduction begins to

I. Oh et al. / Journal of Catalysis 213 (2003) 17–22
19
active and the coverage of Pb atoms at these edges is smaller
than the unity coverage used in this calculation. By way of
contrast, the bare Au surface exhibits negligible reactivity at
the same overpotentials. The Pb upd adlayer thus functions
as a catalyst for peroxide electroreduction.
Comparing the voltammograms in Figs. 1 and 2, we
find that the potential of the maximum catalytic activity in
Fig. 2A (−0.07 V) does not precisely coincide with the
peak C in Fig. 1 but is shifted by ca. 40 mV. This is most
likely caused by the iR drop due to the high level of catalytic
current in Fig. 2A. This kind of potential shift was also
observed in the Bi and Tl upd systems [8,18].
3.2. Electroreduction of dioxygen on Pb upd-modified
Au(111)
Figure 2B shows the current–potential graph from a
Au(111) electrode rotating at 400 rpm in a solution con-
taining 1.0 mM Pb2+ and 0.1 M HClO4, which is saturated
with O2. The voltammetric behavior of the Pb upd for the
electrocatalytic reduction of the dissolved O2 is similar to
that for H2O2 in Fig. 2B in that virtually no current is ob-
served in the potential region positive of ca. 0.4 V, while it
begins to increase as the potential becomes more and more
negative. However, even in the potential region negative of
−0.07 V, where the full Pb layer is expected to exist, cat-
alytic current associated with O2 reduction is maintained.
By comparison with the corresponding peroxide results, the
current in this region must therefore be associated with the
two electron reduction of oxygen to peroxide (Eq. (1)). This
result suggests that the two-electron reduction of dioxygen
to peroxide is relatively facile on both the island and full
monolayer Pb structures. Alternatively, the subsequent two-
electron reduction of peroxide to water is slower on the full
monolayer.
Fig. 2. Linear sweep voltammogram on Au(111) rotating at 400 rpm (A) in
2+
10 mM H O + 1.0 mM Pb + 0.1 M HClO , which is purged with Ar
2
2
4
2+
gas, and (B) in 1.0 mM Pb + 0.1 M HClO saturated with O . Scan
4
2
rate = 20 mV/s. Inset: Levich plot for the O reduction (solid circle)
2
in the O -saturated electrolyte ([O ] = ca. 1 mM) and for the H O
2
2
2 2
reduction (hollow circle: [H O ] = 1.0 mM). Peak current is measured at
2
2
the maximum of the catalytic current at each rotation speed.
The inset to Fig. 2B shows Levich plots for the O2 and
H2O2 reductions. While the Levich plot for the O2 shows the
linear relation expected for a mass-transfer-limited process,
that for the equivalent concentration of H2O2 exhibits a
small and saturated current level, indicating that the kinetics
of H2O2 reduction on Pb upd is very slow. (However, it
is still very much greater than that afforded by bare Au.)
Note in addition that the maximum current level of H2O2
reduction in Fig. 2A is almost the same as that of O2
reduction, despite the nearly 10-fold greater concentration
of H2O2 (10 mM) relative to that of O2 (ca. 1 mM) [20].
These results indicate that H2O2 reduction on the Pb upd is
limited by the reaction rate on the electrode rather than by
mass transfer.
increase. The catalytic current reaches its maximum at
−0.07 V and then suddenly decreases at more negative
potentials. However, there remains a substantial amount
of residual current even at this quite negative potential.
Considering the fact that the full Pb layer exists in this
potential region, it seems that even the full Pb layer has some
activity toward peroxide reduction. For comparison, the full
Tl monolayer in the Tl upd system shows almost no activity
toward H2O2 electroreduction [8,19].
The maximum reaction rate is estimated to be 7.8 ×
1015 s−1 for the H2O2 reduction by the Pb upd on Au(111).
Also, assuming that Pb adatoms cover the whole surface
at the maximum catalytic activity (from Ref. [10], the
interatomic distance of Pb adsorbate is 0.35 nm) and that the
whole Pb adatoms work as the active site, the turnover rate is
estimated to be 11 s−1 per active site. This number is a lower
bound to the turnover rate since only the island edges are
Finally, both voltammograms of the O2 and H2O2 reduc-
tions on the Pb upd usually shows a noisy response, es-
pecially at the high current level. This kind of oscillatory
voltammetric response was observed in other system and
was attributed to the presence of the surface hydroxide [21].

20
I. Oh et al. / Journal of Catalysis 213 (2003) 17–22
islands becomes larger (Fig. 4A). This result reproduces that
from previous AFM and STM studies, which reported the
existence of the Pb island structure at potentials positive of
the upd peak, C [10,12–14]. In Fig. 4B is shown the line
profile across one of the islands as indicated in Fig. 4A. The
height of the island with respect to the Au(111) terrace is
measured to be 0.25 0.05 nm. In our previous report using
AFM [10], we reported two values for the island height:
the Pb island is initially one Pb adatom high (0.30 nm) but
rapidly grows in height to 0.53 nm through association of the
Pb with solvent or hydroxide. The discrepancy between the
heights measured by STM and AFM probably arises from
the different imaging mechanisms of the two techniques;
while the Pb aggregation with solvent/hydroxide would lead
to the higher topology when probed with the AFM tip, the
Pb aggregate is likely to be less conductive than the pure Pb
island and attract the STM tip toward the surface, resulting
in the lower height.
Fig. 3. Current–time response at E = −0.04 V on Au(111) rotating at
2+
400 rpm in 10 mM H O + 1.0 mM Pb +0.1 M HClO , which is purged
2
2
4
with Ar gas. A small amount of EtSH is injected at the moment indicated
by an arrow.
Because the electrolyte solution was saturated with at-
mospheric O2 while obtaining the STM image in Fig. 4A, the
Pb adlayer structure shown in Fig. 4A is the catalyst struc-
ture while the reaction is proceeding rather than the one in
which no reactant is present. We have performed STM imag-
ing in an Ar-filled dry box to rid the solution of O2 and could
observe no significant change in the Pb island structure. This
indicates that the presence of reactant O2 in the electrolyte
solution exerts little effect to the structure of the Pb upd cat-
alyst. Recently, Ocko and co-workers reached an equivalent
conclusion in their X-ray scattering study of the Bi-upd sys-
tem [28].
When the electrode potential is moved negative of peak C,
the Pb island structure is transformed to a full Pb monolayer,
as has been previously described [10,22]. Note that the RDE
measurement of H2O2 reduction in Fig. 2A shows that the
catalytic activity toward peroxide reduction is significantly
retarded with this structural transformation, though the full
Pb layer still has residual catalytic activity.
Figure 4C shows a STM image of the Pb upd that is
poisoned with 30 µM EtSH at E = 0.03 V. On the bare
Au(111) surface absent the Pb adsorbate, EtSH is reductively
desorbed below ca. 0.2 V [23]. Thus, at potentials lower than
ca. 0.2 V, EtSH is prevented from sticking to the uncovered
Au area, greatly simplifying the STM image analysis. While
the island structure is maintained, it is observed that the
step edge of the Pb island is decorated with an additional
feature. The line profile across the one of the decorated
island edge is shown in Fig. 4D. The height of the terrace T1
is measured to be 0.25 0.06 nm relative to the Au surface,
which is identical to the value for the bare Pb island. So the
terrace T1 can with no doubt be assigned to the Pb island
structure which existed before adding EtSH. The height of
the additional terrace T2 is measured to be 0.13 0.05 nm.
This terrace feature can be assigned to EtSH aggregate at
the step edge of the Pb island because it is not observed
without EtSH (Fig. 4A) and its height is too small to be
assigned to Pb or Au. In a STM study of thiol adsorption
3.3. Effect of added poison on peroxide electroreduction
activity
Figure 3 shows the current–time graph obtained from
an Au(111) electrode rotating at 400 rpm in a solution
containing 10 mM H2O2, 1.0 mM Pb2+, and 0.1 M HClO4,
which is purged of O2. The electrode potential is set at
−0.04 V, where a large steady-state current due to the
electrocatalytic H2O2 reduction keeps flowing. At the time
indicated by the arrow, a small quantity of ethanethiol
(EtSH) is injected by a syringe into the slowly stirred
electrochemical cell. The final concentration of EtSH in
the cell is 0.39 mM. The volume of the injected EtSH
solution is less than 1% of that in the cell, so the change
in concentrations of H2O2, Pb2+, and HClO4 is small and
negligible.
With addition of EtSH, the catalytic current decreases
rapidly. However, the catalytic activity of the Pb upd is not
completely inhibited as there remains a substantial amount
of residual current. Taking into consideration that the con-
centration of EtSH is rather high, this is an unexpected
and interesting result. For comparison, catalytic activity was
almost completely inhibited by this high concentration of
EtSH in the case of the Tl upd system [8]. The existence of
the residual catalytic current is related to the selective block-
ing of the active site by EtSH, as will be verified by STM
imaging described below.
3.4. STM imaging
STM was used to examine the surface structure of the
Pb upd electrocatalyst both with and without poisoning. At
electrode potentials positive of the upd peak C in Fig. 1, the
STM image shows the bare Au(111) terraces and steps. As
the potential is moved to more negative values, a number of
small islands begin to appear on the Au(111) terraces and,
when the potential becomes more negative, the size of the

I. Oh et al. / Journal of Catalysis 213 (2003) 17–22
21
Fig. 4. (A) 40 nm × 40 nm STM image of the Pb upd on Au(111) at E = 0.03 V; I = 1 nA; E
= 700 mV. (B) Line profile along the line indicated in
tip
bias
part A. (C) 138 nm × 138 nm STM image of the Pb upd poisoned with 30 µM EtSH at E = 0.07 V; I = 1 nA; E
= 500 mV. (D) Line profile along the
tip
bias
line indicated in part C.
on Au(111) in ultrahigh vacuum [24], HS(CH2)6OH exhibits
apparent heights of 0.08 nm relative to the Au surface when
the alkyl chains lie flat on the surface and 0.18 nm when they
align with the surface normal. The length of the alkyl chain
contributes negligibly to the apparent height measured by
STM. Also, in our previous report on the Tl upd poisoned
with EtSH, the height of the EtSH terrace was measured
to be 0.15 nm, a height which is virtually identical to that
reported here.
The adsorption of EtSH exclusively at the step edge of
the Pb island and the accompanied inhibition of the catalytic
activity indicates that the step edge site is the active site for
the electroreduction of peroxide. The Au atoms adjacent to
the Pb island must be positively polarized because EtSH
binds on the Au site near the Pb island in spite that the
EtSH is expected to desorb at this negative potential. This
result coincides with that found for Tl upd and is expected
considering the resemblance of the two upd systems [8].
In this sense, the Au surface adjacent the upd island acts
like Auδ+, a system whose catalytic efficiency in other
contexts, especially CO oxidation, have long been noted
[25,26]. Localized differences in surface potential are well
understood to occur at step edges and other heterogeneities
on metal surfaces. The upd islands produced here apparently
provide another heterogeneity whereby charge transfer to the
upd metal leaves a net positive charge on the surrounding
terrace.
We note that, from the catalysis inhibition measurement
in Fig. 3, there remains a significant residual current even
with the high concentration of EtSH. This indicates that
there still are some sites unblocked with EtSH in the catalyst
structure. From Fig. 4C we note that significant areas of
the island edges are left undecorated by the thiol poison. In
addition, the presence of significant catalytic current in the
full monolayer region may imply that some of Pb terrace
sites themselves are also active. Alternatively, it may well be
that the Pb full monolayer is not strictly homogeneous, and
there may be active sites formed at step edges and other such
locations.
The edge decoration feature depends on the EtSH con-
centration and the electrode potential. If [EtSH] is lower than
20 µM, STM imaging shows no EtSH terraces. In contrast,
when [EtSH] is as high as 0.30 mM, the entire surface be-
comes covered with myriad vacancy islands, which indicates
that EtSH adsorbs not only at the step edge site but on the Au
terrace area in addition. The electrode potential also affects
the EtSH terrace feature. When the potential is moved posi-
tively, the edge decoration feature becomes less and less dis-

22
I. Oh et al. / Journal of Catalysis 213 (2003) 17–22
tinct. At very positive potentials, the whole surface becomes
covered with a number of vacancy islands. This observation
indicates that EtSH binds not only at the Pb island edges but
to the whole surface at this positive potential [27].
[2] T.D. Jarvi, E.M. Stuve, in: J. Lipkowski, P.N. Ross (Eds.), Electro-
catalysis, Wiley–VCH, New York, 1998, p. 75.
[3] R. Adzic, in: J. Lipkowski, P.N. Ross (Eds.), Electrocatalysis, Wiley–
VCH, New York, 1998, p. 197.
[4] M.R. Tarasevich, A. Sadkowski, E. Yeager, in: B.E. Conway, J.O.M.
Bockris, E. Yeager, S.U.M. Kahn, R.E. White (Eds.), Oxygen Electro-
chemistry, Plenum, New York, 1983, p. 301.
[5] D.M. Kolb, in: H. Gerischer, C.W. Tobias (Eds.), Advances in
Electrochemistry and Electrochemical Engineering, Wiley, New York,
1978, p. 125.
4. Conclusion
We performed electrochemical and STM measurements
in order to establish a structure–reactivity correlation for
electrocatalytic H2O2 and O2 reduction by Pb upd on
Au(111) in the acid electrochemical environment. At the
potential of maximal catalytic activity, a Pb island structure
is found. Introduction of EtSH at this potential leads to
significant but incomplete inhibition of electroreduction
activity. STM images show that EtSH adsorbs exclusively
on Au at the edge sites of the Pb island at potentials at
which EtSH is expected to reductively desorb from the
Au(111) terrace. These results suggest that the Au atoms
near the Pb islands are positively polarized by the adjacent
Pb atoms and this heterobimetallic assemblage at the edge
site of the Pb island is probably the catalytic site of peroxide
electroreduction activity.
[6] S.M. Sayed, K. Juttner, Electrochim. Acta 28 (1983) 1635.
[7] C.-H. Chen, A.A. Gewirth, J. Am. Chem. Soc. 114 (1992) 5439.
[8] I. Oh, A.A. Gewirth, J. Kwak, Langmuir 17 (2001) 3704.
[9] R.R. Adzic, J.X. Wang, J. Phys. Chem. B 104 (2000) 869.
[10] C.-H. Chen, N. Washburn, A.A. Gewirth, J. Phys. Chem. 97 (1993)
9754.
[11] B.K. Niece, A.A. Gewirth, J. Phys. Chem. B 102 (1998) 818.
[12] M.P. Green, K.J. Hanson, D.A. Scherson, X. Xing, M. Richter, P.N.
Ross, R. Carr, I. Lindau, J. Phys. Chem. 93 (1989) 2181.
[13] M.P. Green, K.J. Hanson, R. Carr, I. Lindau, J. Electrochem. Soc. 137
(1990) 3493.
[14] M.P. Green, K.J. Hanson, Surf. Sci. Lett. 259 (1991) L743.
[15] S.-J. Hsieh, A.A. Gewirth, Surf. Sci. 498 (2002) 147.
[16] H. Angerstein-Kozlowska, B.E. Conway, A. Hamelin, L. Stoicoviciu,
J. Electroanal. Chem. 228 (1987) 429.
[17] T. Will, M. Dietterle, D.M. Kolb, in: A.A. Gewirth, H. Siegenthaler
(Eds.), Nanoscale Probes of the Solid–Liquid Interface, Vol. 228,
Kluwer Academic, Dordrecht, 1995, p. 137.
[18] I. Oh, M.E. Biggin, A.A. Gewirth, Langmuir 16 (2000) 1397.
[19] R.R. Adzic, J.X. Wang, B.M. Ocko, Electrochim. Acta 40 (1995) 83.
[20] K. Kinoshita, in: Electrochemical Oxygen Technology, Wiley–
Interscience, New York, 1992, p. 8.
[21] S. Nakanishi, Y. Mukouyama, K. Karasumi, A. Imanishi, N. Furuya,
Y. Nakato, J. Phys. Chem. B 104 (2000) 4181.
[22] N.J. Tao, J. Pan, Y. Li, P.I. Oden, J.A. DeRose, S.M. Linday, Surf. Sci.
Lett. 271 (1992) L338.
[23] H. Hagenstrom, M.A. Schneeweiss, D.M. Kolb, Langmuir 15 (1999)
2435.
Acknowledgments
This work was supported in part by the Korea Science
and Engineering Foundation through the MICROS center at
KAIST and by the Ministry of Information and Communica-
tion (Grant IMT 2000-B3-2). A.A.G. acknowledges the NSF
(CHE-9820828) for partial support of this research.
[24] G.E. Poirier, E.D. Pylant, Science 272 (1996) 1145.
[25] M. Valden, S. Pak, X. Lai, D.W. Goodman, Catal. Lett. 56 (1998) 7.
[26] G.K. Bethke, H.H. Kung, Appl. Catal. A Gen. 194 (2000) 43.
[27] G.E. Poirier, Chem. Rev. 97 (1997) 1117.
References
[1] S.-G. Sun, in: J. Lipkowski, P.N. Ross (Eds.), Electrocatalysis, Wiley–
VCH, New York, 1998.
[28] K. Tamura, B.M. Ocko, J.X. Wang, R.R. Adzic, J. Phys. Chem. B 106
(2002) 3896.