A nonelectrochemical reductive deposition of ruthenium adatoms onto nanoparticle platinum: Anode catalysts for a series of direct methanol fuel cells

Article abstract of DOI:10.1016/S0013-4686(03)00542-5

The surface of Pt nanoparticles was cleaned and saturated with hydrogen by treatment first with a 3% aqueous solution of H₂O₂ and then with hydrogen gas under water at room temperature. Reaction between the surface hydrogen and aqueous RuCl₃ deposited 0.18 surface equivalents of Ru_ad onto the Pt nanoparticles. The deposition was repeated several times, with each reaction depositing ～0.18 surface equivalents more Ru_ad onto the Pt-Ru_ad nanoparticles. The resulting Pt-Ru_ad nanoparticles were analysed using cyclic voltammetry, CO stripping voltammetry, and as catalysts for electrooxidation of MeOH in three-electrode experiments and in prototype direct methanol fuel cells. The optimum surface coverage (θ_Ru) for electrooxidation of MeOH was ～ 0.33 under these conditions.

Full text of DOI:10.1016/S0013-4686(03)00542-5

Electrochimica Acta 48 (2003) 4021ꢀ4031
/
www.elsevier.com/locate/electacta
A nonelectrochemical reductive deposition of ruthenium adatoms
onto nanoparticle platinum: anode catalysts for a series of direct
methanol fuel cells
Dianxue Cao, Steven H. Bergens *
Department of Chemistry, University of Alberta, Edmonton, Alta, Canada T6G 2G2
Received 1 June 2003; received in revised form 23 June 2003; accepted 24 June 2003
Abstract
The surface of Pt nanoparticles was cleaned and saturated with hydrogen by treatment first with a 3% aqueous solution of H2O2
and then with hydrogen gas under water at room temperature. Reaction between the surface hydrogen and aqueous RuCl3
deposited 0.18 surface equivalents of Ruad onto the Pt nanoparticles. The deposition was repeated several times, with each reaction
depositing ꢀ
analysed using cyclic voltammetry, CO stripping voltammetry, and as catalysts for electrooxidation of MeOH in three-electrode
experiments and in prototype direct methanol fuel cells. The optimum surface coverage (uRu) for electrooxidation of MeOH was ꢀ
/
0.18 surface equivalents more Ruad onto the Ptꢀ
/
Ruad nanoparticles. The resulting PtꢀRuad nanoparticles were
/
/
0.33 under these conditions.
# 2003 Elsevier Ltd. All rights reserved.
Keywords: Platinum; Ruthenium; Adatom; Methanol; Fuel cell; Anode electrocatalyst
1. Introduction
composition (ratio of moles surface Pt atoms (Ptsurface
to moles Ruad deposited) that is difficult to achieve with
)
We report a nonelectrochemical, aqueous method to
deposit submonolayer amounts of Ru adatoms (Ruad)
onto Pt nanoparticles that is controlled by the stoichio-
other catalyst preparations. The first PtꢀRuad catalyst
/
system was prepared in 1975 by Watanabe and Motoo
[6,7] through electrochemical deposition of Ruad onto
polycrystalline Pt. Since then, the reported preparations
metry of a surface reduction reaction. The resulting Ptꢀ
/
Ruad nanoparticles were studied by CO stripping
voltammetry, and their activity as electrocatalysts for
the potentiostatic electrooxidation of MeOH in aqueous
H2SO4 electrolyte was determined. They were then used
as anode catalysts in polymer electrolyte direct MeOH
of Ptꢀ
of Ruad onto single-crystal Pt (Pt(h k l)) [8ꢀ
polycrystalline Pt [6,7,18ꢀ20], evaporative deposition
of Ru atoms onto Pt(h k l) [11,21ꢀ26], and chemical
/
Ruad systems include electrochemical deposition
/16] or
/
/
vapour deposition onto Pt(h k l) [27,28].
fuel cells. First reported during the mid-1960s [1ꢀ
/
3], Ptꢀ
/
Perhaps less conventional among preparations carried
out in solution (thereby being readily applied to
nanoparticle substrates) are the often termed sponta-
neous depositions and the hydrogenation of organome-
tallic Ru compounds over Pt. By far the most studied
and used of the spontaneous depositions is the adsorp-
tion of Ru species from aged perchloric acid solutions of
RuCl3 onto Pt, followed by electrochemical reduction of
the Ru deposits to generate strongly adsorbed, mainly
metallic Ruad. This deposition method, developed by
Ru nanoparticle systems are the most studied, and
among the most active anode catalysts for the electro-
oxidation of MeOH and reformate (H2/CO2 with trace
CO) in fuel cells [4,5]. The preparation of catalyst
systems by deposition of submonolayer amounts of
Ruad onto Pt substrates provides a level of control over
real surface area (number of active sites) and surface
* Corresponding author. Tel.: ꢁ
/
1-780-492-9703; fax: ꢁ1-780-492-
/
Wieckowski and co-workers [10,27ꢀ
/
31], has been car-
8231.
E-mail address: steve.bergens@ualberta.ca (S.H. Bergens).
ried out using single crystal Pt (vide infra), polycrystal-
0013-4686/03/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0013-4686(03)00542-5

4022
D. Cao, S.H. Bergens / Electrochimica Acta 48 (2003) 4021ꢀ4031
/
line Pt [29,32], and nanoparticle Pt as substrates
[27,28,33ꢀ35]. Studies using Pt(h k l) show that the
rate of Ru deposition [32,36], the degree of preference
tion is self-limiting: further deposition beyond ꢀ
surface equivalents of Ruad is poisoned by adsorbed CO.
The resulting PtꢀRuad nanoparticles were used as the
/
0.1
/
/
for two-dimensional over three-dimensional Ruad island
anode catalyst in a prototype direct MeOH fuel cell,
showing that adatom nanoparticle catalysts can survive
heat-pressing into a Nafion†-based membrane electro-
lyte assembly (MEA) without significant loss of Ruad.
A self-limiting, directly reductive, aqueous deposition
using RuCl3 as Ruad source was developed by Janssen
and Moolhuysen in 1975 [44,45]. A related version was
later studied by Szabo´ and Bakos [46,47]. Janssen et al.
electrochemically cleaned a Pt substrate (polycrystalline
or nanoparticle Pt) in aqueous H2SO4, saturated the Pt
surface with adsorbed hydrogen at 0.05 V versus RHE,
and then transferred the surface to an aqueous solution
of RuCl3. The hydrogen adsorbed onto the Pt surface
reduced an undetermined amount of RuCl3(aq.) to
growth [36ꢀ
sition and activity for MeOH electrooxidation are
surface structure-dependent. Optimised Pt(1 1 1)ꢀRuad
is more active towards MeOH electrooxidation than
optimised Pt(1 1 0)ꢀRuad and Pt(1 0 0)ꢀRuad [8ꢀ10,32].
This is a rare example of an adatom deposition used to
prepare PtꢀRuad nanoparticles. PtꢀRuad nanoparticles
/39], as well as the optimum surface compo-
/
/
/
/
/
/
prepared in this way have been studied by EC-NMR
spectroscopy [27,28,34,40], by voltammetry, by CO
stripping voltammetry, and as catalysts for the electro-
oxidation of MeOH and formic acid [27,28,33ꢀ35]. A
/
spontaneous-type deposition developed by Vielstich and
co-workers [22,23] involves reaction of aqueous RuCl3
with clean Pt. This deposition is proposed to proceed via
prior adsorption of chloride onto the Pt surface with
dissolved Ru3ꢁ ions acting as counterions for the
adsorbed chloride. Electrochemical reduction was then
deposit Ruad onto the surface. The resulting PtꢀRuad
/
surface was investigated by cyclic voltammetry and it
was used to electrooxidise MeOH in aqueous H2SO4.
Barbier and co-workers [48ꢀ
/
56] prepared a series of Mꢀ
Mad catalysts for nonelectrochemical reactions (e.g. Ptꢀ
Auad, PtꢀCuad, PdꢀPtad, PtꢀRead, and RuꢀCuad) using
a similar methodology.
/
?
used to generate Ptꢀ/Ruad and to desorb the chloride
/
ions from the surface. This deposition has been carried
out using Pt(1 1 1) as substrate, and it is reportedly
prone to ageing effects on the order of minutes. The
/
/
/
/
We now report a nonelectrochemical method to clean
and prepare nanoparticle Pt black, to saturate its surface
with hydrogen, and to carry out repeated reductive
resulting Pt(1 1 1)ꢀRuad systems have been studied by
/
cyclic voltammetry and electrooxidation of MeOH
[22,23]. Both these spontaneous-type depositions are
two-step processes involving deposition of a Ru pre-
cursor followed by electrochemical reduction to gener-
ate Ruad. These depositions are also self-limiting: they
cease after a specific submonolayer amount of Ruad
precursor is deposited onto the Pt surface. The amount
of Ruad precursor adsorbed onto Pt when the deposition
ceases depends upon the properties of the deposition
and the substrate. These depositions can be repeated to
obtain incrementally higher amounts of Ruad than those
obtained from single depositions.
depositions of Ruad using aqueous RuCl3×
/
xH2O (xꢃ
/
1ꢀ3) as the adatom source. The amount of Ruad
/
deposited during each deposition will be quantified,
and we will show that the electrochemical properties
(voltammetry, CO stripping voltammetry, MeOH elec-
trooxidation) of PtꢀRuad nanoparticles depend not only
/
upon their surface stoichiometry, but also upon the
method used to prepare them. We also report the
relative activities of a series of direct MeOH fuel cells
(DMFCs) operating with the PtꢀRuad nanoparticles as
/
anode catalyst.
A nonelectrochemical deposition that directly gener-
ates Ruad in reduced form uses an organometallic Ru
compound, Ru(COD)(h3-C3H5)2 (COD is 1,5-cyclooc-
tadiene) (1), that is hydrogenated over Pt in hexane
2. Experimental
under mild conditions (ꢂ
/
40 to ꢂ
/
15 8C, 1 atm H2) to
2.1. General
generate PtꢀRuad, cyclooctane, and propane. This
/
deposition, developed by Bergens and co-workers [41],
has been carried out using blacked Pt gauzes and Pt
The Ar (Praxair, pre-purified), O2 (Praxair), and CO
(Matheson, ultrahigh purity) were used as received. The
hydrogen gas was generated using a Peak ScientificTM
Hydrogen Generator, Model H-300. The water was
deionised, distilled, and then distilled again from alka-
line permanganate. MeOH (Sigma-Aldrich, ACS HPLC
grade, 99.93%), H2SO4 (Alfa Aesar, 99.9999%), H2O2
nanoparticles as substrates. The PtꢀRuad gauzes were
/
studied by CO stripping voltammetry and by electro-
oxidation of MeOH [42]. Unlike the spontaneous
depositions, this deposition is not self-limiting; it con-
tinuously deposits Ruad onto the Pt substrate either until
1 is consumed, or until the Ptꢀ/Ruad surface is physically
(ACP Chemicals, Inc., Reagent ACS, 30%), RuCl3×
/
removed from solution when the desired surface stoi-
chiometry has been obtained. A related deposition uses
Ru4H4(CO)12 to deposit Ruad and CO onto Pt nano-
particles under 60 atm H2 in hexanes [43]. This deposi-
xH2O (Aithaca Chemical Corp.), KOH (BDH, Inc.,
ACS), and K2S2O8 (Fisher Scientific Company, Certi-
fied Reagent) were used as received. The Pt black was
Johnson Matthey fuel cell grade with a specific surface

D. Cao, S.H. Bergens / Electrochimica Acta 48 (2003) 4021ꢀ
/4031
4023
area of 27 m2/g (both as quoted by J.M. and measured
independently in this lab).
exhaustive anodic stripping of Ruad from the Ptꢀ
/
Ruad
surface in 1.0 M aqueous NaOH using a 9-V battery as
the power source, followed again by ICP analysis. The
results are shown in Table 1.
Electrochemical experiments were performed in a
typical three-electrode electrochemical cell using a Pine
Bipotentiostat Model AFCBP1 controlled with Pine-
Chem 2.00 software. Electrolytes were purged with Ar
for at least 15 min prior to use and protected under an
atmosphere of Ar at bubbler pressure during experi-
ments. The reference electrode was a standard hydrogen
electrode. The counterelectrode was a blacked Pt gauze
behind a D-porosity glass frit. A constant temperature
2.3. MeOH electrooxidation
To perform MeOH electrooxidation over the nano-
particle catalysts (e.g. Pt, Ptꢀ
immobilised on the surface of a shiny Pt foil (20 mmꢄ
10 mmꢄ0.18 mm) electrode using the following proce-
/Ruad), the catalyst was
/
/
bath (Brinkmann, Lauda Model RK20, 9
used to maintain the temperature of the electrochemical
cell. Inductively coupled plasma (ICP) spectroscopy was
/
0.03 8C) was
dure. The Pt foil was cleaned with aqua regia, piranha
solution, and then with ultrapure water. A catalyst ink
was prepared by ultrasounding known amounts of the
catalyst in ultrapure water (typically 5 mg/ml) for 30 min
(Branson ultrasonic bath, model 1210). An aliquot of
the suspension (typically 200 ml) was placed upon the Pt
foil to deposit a known amount of catalyst (typically 1
mg) on the foil. The suspension was air-dried to yield a
uniform catalyst layer on the electrode. The amount of
catalyst deposited onto the Pt foil surface was deter-
mined from the volume of the suspension and it was
cross-checked by weight. These immobilised nanoparti-
cle catalyst electrodes were physically stable enough to
withstand the operations involved in cyclic voltammetry
and the electrooxidation reaction. The use of bare
particle surfaces resulted in good contact between the
particles and the Pt foil conductor, thereby minimising
diffusion and catalyst utilisation problems. The surface
area and MeOH electrooxidation activity of the Pt foil
were very small and subtracted as background.
carried out using a PerkinꢀElmer Optima equipped with
/
an atomic emission detector. The performance of the
DMFCs was evaluated using a 890 Series computer-
controlled fuel cell test load unit (Scribner Associates,
Inc.).
2.2. Preparation of PtꢀRuad catalysts
/
The deposition reactor was a 3-dram 21mmꢄ
/
50 mm
vial containing a 4 mmꢄ14 mm Teflon-coated mag-
/
netic stir bar and capped with a rubber septum. Pt
substrates were first cleaned with 3% H2O2 (80 mg Pt in
ꢀ1 ml of H2O2 solution) and then maintained in the
/
remaining water in the reactor. The reactor was then
flushed with Ar for 30 min to remove any O2, then
flushed with Ar-diluted hydrogen for 15 min, and with
pure hydrogen for another 15 min to reduce the oxides
on the Pt surface, and to saturate the Pt surface with
hydrogen. The system was then flushed with Ar to
eliminate the excess hydrogen gas. An Ar-purged,
freshly made 0.05 M RuCl3 aqueous solution was
cannulated into the reactor. The ratio of Pt (mg) to
The electrode potential was set to 0.075 V upon
immersion into the Ar-purged electrolyte ([H2SO4]ꢃ1.0
/
M). Continuous potential sweeps were carried out
between 0 and 0.70 V at 10 mV/s. No obvious changes
in the voltammograms were observed between the first
and subsequent sweeps. The potential was then held at
0.075 V while the deoxygenated MeOH was added to the
RuCl3 aqueous solution (ml) was ꢀ
was stirred under Ar for 1 h to complete the deposition.
The PtꢀRuad was isolated by vacuum filtration in air,
/10:1. The solution
/
cell to bring the final [MeOH]ꢃ
/
[H2SO4]ꢃ1.0 M. The
/
then thoroughly washed with ultrapure water, and dried
under vacuum in a desiccator for 24 h. The deposition
solution was stirred for 5 min to ensure the equilibration
of both concentration and temperature prior to electro-
chemical data acquisition. The potentiodynamic cycling
could be repeated as many times as needed to obtain Ptꢀ
/
Ruad with high surface equivalents of Ruad.
To determine the amount of Ruad deposited onto the
was started after the stirring (sweep rateꢃ
/
10 mV/s,
sweep range 0ꢀ0.60 V) with the first sweep in the
/
Pt surfaces, the Ptꢀ
solution of K2S2O8 in 4.0 M KOH. The ratio of Ptꢀ
/
Ruad was suspended in a saturated
negative direction. The potential was then held at 0.075
V for 5 min and then the potentiostatic oxidation was
started by stepping the potential to 0.45 V at room
temperature, or to 0.40 V at 60 8C.
/
Ruad (mg) to the alkaline solution (ml) was 3:5. The
solution was stirred for 1 h in a sealed container, and
then it was left overnight without stirring at room
temperature [57]. It was found that all the adsorbed
Ruad and some of the surface Pt are oxidised from such
surfaces to form soluble oxides under these conditions.
2.4. CO stripping voltammetry
CO was adsorbed on catalyst surfaces by bubbling
CO at 1 atm through the stirred electrolyte (H2SO4, 1.0
M) for 25 min while holding the potential at 0.075 V.
The excess CO was then flushed from the electrolyte
with Ar gas for 25 min, the stirring was stopped, and the
UVꢀVis spectra of the resulting solutions indicated that
/
the Ru ions primarily consist of ruthenate. ICP analysis
was carried out on the resulting solution to measure the
amount of Ru. This amount was cross-checked by

4024
D. Cao, S.H. Bergens / Electrochimica Acta 48 (2003) 4021ꢀ4031
/
Table 1
Nanoparticle PtꢀRuad catalysts prepared by surface reductive deposition of Ru on nanoparticle Pt black
/
Number of depositions (n)
Surface equivalents Ruad
Ruad/n
1
2
3
5
7
0.18
0.18
0.38
0.19
0.57
0.19
0.81
0.17
1.31
0.19
potential was cycled between 0 and 0.70 V at 10 mV/s.
The first sweep was in the positive direction.
even distribution of reactants. PTFE gaskets were
inserted to prevent the cell from leaking. The fuel cells
were operated with a 1.0-M aqueous MeOH solution
pumped through the anode compartment at 4.0 ml/min
and zero back-pressure from a reservoir at ambient
temperature, and with dry O2 from cylinders passed
through the cathode compartment at 400 standard cubic
centimeters per minute (sccm) and 20 psig back pressure.
The fuel cells were conditioned for 3 days before taking
polarisation data using the following procedure: the
DMFC was heated to 60 8C at open circuit with MeOH
solution circulating through the anode and O2 flowing
through the cathode, then the cell was operated at 20
mA/cm2 for 4 h. The cell temperature was then raised to
90 8C, and the cell was operated at 100 mA/cm2 for
another 4 h. The cell was then shut down by turning off
the load, heating, MeOH and O2 supply, and left
overnight at room temperature. The performance of
the DMFCs stabilised after such conditioning. The
reported cell potentials are not IR compensated.
2.5. Preparation of the membrane electrode assembly
The MEAs of geometrical area 5 cm2 were fabricated
using the decal transfer method developed at the Los
Alamos National Laboratory [58]. PtꢀRuad and fuel cell
/
grade Johnson Matthey Pt black (27 m2/g) were used as
the anode and cathode catalyst, respectively, both with a
loading of 2 mg/cm2. To fabricate an MEA, catalyst/
ionomer inks were prepared by suspending the catalyst
in water (e.g. 13 mg catalyst in 150 mg H2O), and
agitating in an ultrasonic bath for 30 min to thoroughly
wet and disperse the catalyst. Enough 5% Nafion†
solution (ElectroChem, Inc.) was then added to the
mixture to give a dry ink composition of 80 wt.%
catalyst with 20 wt.% Nafion† ionomer. The mixtures
were ultrasounded for another 2 h at room temperature
to obtain uniformly dispersed inks. The catalyst inks
were painted onto 5 cm2 Teflon† decals and dried at
70 8C, and further baked in an oven at 135 8C for 5 min
to cure the catalyst/Nafion† composites. Nafion†-117
membranes were cleaned and converted into the acid
form by boiling in 3% H2O2 for 1 h, followed by boiling
in 0.5 M H2SO4 for 2 h, and finally boiling in ultrapure
water for 2 h with the water being changed every 30 min.
The cleaned membrane was stored in ultrapure water,
and dried on a vacuum table at 60 8C for 45 min before
use. The catalyst inks were transferred from the Teflon†
3. Results and discussion
3.1. Preparation of nanoparticle PtꢀRuad
/
We used commercially available, unsupported Pt
nanoparticles (Johnson Matthey, Fuel Cell Grade,
measured specific surface area ꢀ
27 m2/g) as substrate.
/
decals to the Nafion† membrane by hot-pressing (125ꢀ
127 8C, 1450ꢀ1550 psi for ꢀ2.5 min).
/
We found the following to be an effective chemical
method to clean and prepare the surface of the as-
received Pt nanoparticles. First, the reaction between the
Pt nanoparticles and a 3% aqueous solution of H2O2
was allowed to go to completion at room temperature.
This procedure both cleaned the surface of the nano-
particles and it left the surface protected as oxides under
pure water. The surface oxides were then reduced by
reaction with hydrogen gas (1 atm) under water. We
found that the nanoparticles were quite reactive towards
hydrogen gas after they were cleaned with peroxide. In
fact, reaction of the cleaned nanoparticles with hydro-
gen gas in the absence of water resulted in sintering of
the nanoparticles and thereby a loss of specific surface
area (we believe the water acts as a heat sink). After
reduction of the surface oxides, the excess hydrogen gas
in the reactor was replaced by Ar, and the nanoparticles
/
/
2.6. Operation of direct MeOH fuel cells
The MEAs were mounted into commercial fuel cell
hardware (ElectroChem, Inc., FC05-01SP). The current
collectors consisted of low-porosity, high-purity gra-
phite blocks with machined serpentine flow fields for the
MeOH and O2 feeds. The cell was held together between
two gold-plated copper contact plates using a set of
retaining bolts positioned around the periphery of the
cell. Electrical heaters were placed behind each of the
copper plates in order to heat the cell to the desired
operating temperature. Separate voltage connectors and
thermocouple wells are also provided in each cell block
for accurate measurement and temperature control.
Teflon†-treated carbon papers (ElectroChem, Inc., 30
wt.% PTFE) were used as backing layers to allow for
were reacted with aqueous RuCl3×
/
xH2O for 1 h at room
temperature to complete the deposition. Leaving the

D. Cao, S.H. Bergens / Electrochimica Acta 48 (2003) 4021ꢀ
/4031
4025
reaction for longer than 1 h did not result in further
deposition of Ruad.
The amount of Ruad deposited on Pt was determined
by exhaustive electrochemical or chemical (using perox-
ydisulfate [57]) oxidative stripping in aqueous hydrox-
ide, and quantification of the Ru ions in solution using
surface, as well as upon Ptsurf. Otherwise, the deposition
would stop once ꢀ1 surface equivalent of Ruad was
deposited.
Fig. 1 shows the stabilised voltammograms for a
series of PtꢀRuad nanoparticles prepared by the reduc-
tive deposition recorded in 1.0 M aqueous H2SO4. There
was little difference between the first and subsequent
sweeps. This observation is consistent with the expecta-
tions that Ruad was not deposited in a high oxidation
/
/
ICP. This procedure showed that
ꢀ0.18 surface
/
equivalents (ratio of moles Ruad deposited to moles
Ptsurf) of Ruad were deposited onto the Pt nanoparticles
after the deposition ceased. This amount is less than the
0.33 surface equivalents expected by the theoretical
stoichiometry of the deposition reaction (Eq. (1)). The
reasons why the deposition stopped at
state (Eq. (1)), and that deep oxidation of the PtꢀRuad
/
nanoparticles did not occur during preparation of the
electrode (see Section 2). Fig. 1 also shows the voltam-
mogram for the Pt nanoparticles. The upper potential
limit of the voltammograms was limited to 0.7 V (versus
RHE) to avoid dissolution of Ruad [63]. The voltammo-
grams show that as the coverage by Ruad increases, the
3Ptsurf ꢂHad ꢁRuCl3(aq:)
0 (Ptsurf )3 ꢂRuad ꢁ3Clꢂ ꢁ3Hꢁ
(1)
peaks in the hydrogen region (ꢀ
less resolved, and the currents in the double layer region
(ꢀ0.4 to 0.6 V) increase. These are typical observations
for PtꢀRu polycrystalline alloy and adatom systems
(e.g. [6,7,32,42,45,63,64]). The morphologies of the
voltammograms for 0.81 and 1.31 equivalent PtꢀRuad
nanoparticles are more characteristic of polycrystalline
/0.05 to 0.4 V) become
ꢀ/0.18 rather than 0.33 surface equivalents Ruad have
not been determined. Perhaps a portion of the surface
sites were blocked by adsorbed chloride during the
deposition, or perhaps not all the adsorbed hydrogen
was available to reaction with RuCl3(aq.). A more likely
/
/
/
explanation is that RuCl3×
/
xH2O often contains signifi-
cant amounts of Ru(IV) species [59], which would
decrease the stoichiometry of the deposition from that
shown in Eq. (1).
Ru [63,65,66] than are the lower equivalent Ptꢀ
nanoparticle surfaces.
There is a peak at ꢀ
PtꢀRuad nanoparticles prepared by more than two
/Ruad
/0.07 V in the voltammograms of
We found that the specific surface area of the Pt
nanoparticles after either electrochemical or chemical
/
reductive depositions. The intensity of this peak in-
creased with the number of depositions. A similar peak
was observed in voltammograms for Ruad-decorated
Pt(1 1 1) surfaces prepared by bubbling hydrogen
through perchloric acid solutions of RuCl3 [22]. A
similar peak is also observed in voltammograms of
stripping of the Ruad remained at ꢀ
27 m2/g, even after
/
several depositions of Ruad (vide infra). This result is a
good indication that significant changes in specific
surface area of the Pt nanoparticle substrate do not
occur during the reductive deposition reaction. Based
upon this result, we used the initial specific surface area
of the Pt substrate to normalise the activity of the
polycrystalline Ru [66] and Ru(0 0 0 1) [67ꢀ
that the presence of this peak in voltammograms for the
PtꢀRuad nanoparticles indicates some degree of three-
/71]. We infer
resulting Ptꢀ/Ruad nanoparticles towards electrooxida-
/
tion of MeOH.
The deposition can be repeated as many times as
dimensional growth of the Ruad deposits during the
reductive deposition.
We note that the hydrogen region of voltammograms
required to obtain Ptꢀ
higher than ꢀ0.18 equivalents Ruad. We found it was
not necessary to repeat the cleaning procedure using
H2O2 between the depositions. Instead, the PtꢀRuad
nanoparticles were washed with purified water in air
between each deposition. We expect that the PtꢀRuad
/
Ruad with a surface stoichiometry
/
for PtꢀRuad nanoparticles prepared using spontaneous
/
deposition by Wieckowski and co-workers [33] have
significantly smaller changes in morphology upon Ruad
/
deposition than do the PtꢀRuad nanoparticles prepared
/
/
by this reductive deposition. This interesting observa-
tion requires further investigation to be understood, but
surface formed surface oxides upon exposure to air,
thereby raising the potential of the surface into the range
it does show that the electrochemical properties of Ptꢀ
/
where chloride ions do not strongly adsorb [60ꢀ
/
62]. The
Ruad nanoparticle catalysts depend not only upon their
surface stoichiometries, but also on their method of
preparation.
surface oxides were then reduced by hydrogen (1 atm,
under water) for the next deposition. Exhaustive strip-
ping of Ruad followed by analysis with ICP showed that
each subsequent reaction deposited ꢀ
equivalents more Ruad, regardless of the amount of
Ruad initially on the surface (Table 1). That ꢀ1.31
equivalent PtꢀRuad nanoparticles can be made by this
deposition method proves that Ruad must at least to
some extent be deposited upon Ruad already on the
/
0.18 surface
3.2. CO stripping voltammetry
/
CO stripping voltammetry is a common method to
/
study Ptꢀ
32,34,72ꢀ77]). Fig. 2 shows the CO stripping voltam-
mograms of a series of PtꢀRuad nanoparticles and over
/Ru electrocatalysts (e.g. [8,11,13,14,22,
/
/

4026
D. Cao, S.H. Bergens / Electrochimica Acta 48 (2003) 4021ꢀ4031
/
Fig. 1. Cyclic voltammograms for (a) Pt and (bꢀ
/
f) PtꢀRuad with different surface equivalents Ruad. The voltammograms were recorded in 1.0 M
/
H2SO4 at 10 mV/s and room temperature. The numbers in brackets indicate the estimated Ru coverage (see text).
the commercial Ptꢀ
/
Ru catalyst. The CO was adsorbed
uted to electrooxidation of CO adsorbed onto Pt sites in
the vicinity of Ruad islands and of CO adsorbed onto
Ruad. The peak at high potentials is attributed to CO
adsorbed onto Pt sites not in the vicinity of Ruad. That
at 0.075 V, the excess CO in the cell was replaced by Ar,
and the first sweep was carried out in the positive-going
direction. The absence of peaks in the hydrogen region
of the first sweep shows that the surface was saturated
with adsorbed CO under these conditions. Whereas two
two CO stripping peaks are obtained with the ꢀ
/0.18
equivalents PtꢀRuad nanoparticles therefore indicates
/
CO stripping peaks were obtained with the ꢀ
equivalents (single deposition) PtꢀRuad nanoparticles,
only a single stripping peak was obtained with the
higher coverage (multiple deposition) PtꢀRuad nano-
/
0.18
that Ruad is deposited as islands rather than a homo-
geneous distribution among the atoms on the substrate
surface. That a single CO stripping peak is obtained
/
/
with higher equivalents PtꢀRuad nanoparticles indicates
/
particles. While the observations of two CO stripping
peaks have mostly been reported for single-crystal
that Pt sites not in the vicinity of Ruad no longer exist on
these surfaces. This result shows that two-dimensional
growth is favoured (but not exclusively) over three-
dimensional growth of the Ruad deposits.
In contrast, Wieckowski and co-workers [27,28,34]
reported that two CO stripping peaks are observed in
Pt(h k l)ꢀ
co-workers [27,28,34] recently observed two stripping
peaks in slow scans using PtꢀRuad nanoparticles pre-
/Ruad systems [15,26,62,78], Wieckowski and
/
pared by spontaneous deposition. Evidence of a second
peak was observed in CO stripping voltammograms of
slow scans for all in a series of PtꢀRuad nanoparticles
/
low equivalents Ptꢀ
nation of 1 [42], and a second peak was observed in the
CO stripping voltammogram recorded using the Ptꢀ
Ruad gauze obtained from the reaction with
Ru4H4(CO)12 [43]. The presence of two peaks is often
/
Ruad gauzes prepared by hydroge-
prepared by spontaneous deposition with surface Ruad
packing densities (surface equivalents) ranging from
0.14 to 0.52. This difference, although not fully under-
stood at this time, again illustrates that the properties of
/
PtꢀRuad nanoparticles depend in part on the method of
/
interpreted as arising from two regions on the PtꢀRuad
surface distinguishable by CO stripping voltammetry
[15,26,34,62,78]. The peak at lower potentials is attrib-
/
preparation, not only surface stoichiometry.
Fig. 3 shows a plot of CO stripping peak potential
against equivalents Ruad recorded at room temperature

D. Cao, S.H. Bergens / Electrochimica Acta 48 (2003) 4021ꢀ
/4031
4027
tures of the PtꢀRuad nanoparticles prepared by the
/
reductive deposition and by spontaneous deposition
have different initial surface structures.
3.3. MeOH electrooxidation
Fig. 4 shows plots of current versus time for the
electrooxidation of MeOH over a series of Ptꢀ
nanoparticles prepared by the reductive deposition. The
curves were recorded at 22 8C (Eꢃ0.45 V) and 60 8C
(Eꢃ0.40 V). As discussed above, the currents are
normalised using the specific surface area of the Ptꢀ
Ruad nanoparticles (27 m2/g). At both temperatures,
the ꢀ0.38 equivalent PtꢀRuad nanoparticles were the
most active, with the ꢀ0.57 equivalent nanoparticles
being the second most active. An important observation
is that the 1.31 equivalents PtꢀRuad surface was more
/Ruad
/
/
/
/
/
/
/
active towards electrooxidation of MeOH than Pt at
22 8C. Since pure Ru surfaces are inactive towards
electrooxidation of MeOH at room temperature [82],
the activity of the 1.31 equivalents PtꢀRuad nanoparti-
/
cles towards MeOH shows there must be exposed Ptsurf
remaining on the surface. This conclusion is consistent
with the observation that the reductive deposition
deposits Ru onto Ruad as well as onto Pt. The surface
equivalents Ruad deposited by the reductive deposition
therefore overestimates the coverage of Ptsurf by Ruad.
Further, the CO stripping experiments indicate that two-
dimensional growth of Ruad islands is favoured (but not
exclusively) over three-dimensional growth, and the
reductive deposition deposits ꢀ0.18 equivalents Ruad,
/
regardless of surface composition (vide supra). Based on
these observations, we propose that a more accurate
estimation than given by the surface equivalents for the
coverage of Ptsurf by Ruad (uRu) is obtained by assuming
the probabilities of depositing Ru onto Ruad and Ptsurf
Fig. 2. CO stripping cyclic voltammograms for (a) Pt and (bꢀ
/
e) Ptꢀ
/
Ruad with different surface equivalents Ruad. The voltammograms
were recorded in 1.0 M H2SO4 at 10 mV/s and room temperature. The
numbers in brackets indicate the estimated Ru coverage (see text).
of a PtꢀRuad nanoparticle are similar. The true coverage
remains to be determined experimentally, but this rough
/
for a series of Ptꢀ
reductive deposition. The optimum activity is in the
range of ꢀ0.38 to 0.57 surface equivalents Ruad. This
result is in good agreement with those reported for
smooth PtꢀRu alloy surfaces [73,79ꢀ81], but it differs
from that of a previous study in this group using
polycrystalline PtꢀRuad gauzes prepared by hydrogena-
tion of 1 [42]. The maximum difference in CO stripping
peak potential between PtꢀRuad gauzes prepared by
hydrogenation of 1 and a pure Pt gauze occurred at ꢀ
/
Ruad nanoparticles prepared by the
/
/
/
/
/
/
0.05 equivalents Ruad. Addition of more Ruad did not
shift the peak potential significantly further. A similar
dependence of CO stripping potential on equivalents
Ruad was reported by van Veen and co-workers [20]
using PtꢀRu systems prepared by coelectrodeposition of
/
Ru and Pt film electrodes. More investigations are
required to determine the origins of these differences
in behaviour between these various types of Ptꢀ
/
Ru
systems, but these results show that the surface struc-
Fig. 3. Dependence of CO stripping peak potential on surface
.
equivalents Ruad

4028
D. Cao, S.H. Bergens / Electrochimica Acta 48 (2003) 4021ꢀ4031
/
Fig. 4. Potentiostatic electrooxidation of MeOH ([MeOH]ꢃ
/
1.0 M,
Fig. 5. Real activities of PtꢀRuad nanoparticles toward MeOH
electrooxidation. The data were obtained after 20 min.
/
[H2SO4]ꢃ1.0 M) over PtꢀRuad with the following surface equivalents
/
/
of Ruad: (a) 0, (b) 0.18 (0.18), (c) 0.38 (0.33), (d) 0.57 (0.45), (e) 0.81
(0.63), and (f) 1.31 (0.75). The numbers in brackets indicate the
estimated Ru coverage (see text).
estimation predicts that the coverage of Ptsurf by Ruad
(uRu(estimated)) at maximum activity for electrooxidation
of MeOH at 22 and 60 8C is 0.33. uRu(estimated) for other
MeOH, are those by Wieckowski and co-workers
[27,28,33] using spontaneous deposition from aged
perchloric acid solutions of RuCl3. They reported that
the optimum packing density (equivalents Ruad) for
electrooxidation of MeOH carried out at 0.4 V versus
RHE (conditions: 0.5 M MeOH, 0.5 M H2SO4, room
temperature) was 0.5, using data collected after 20 h.
They also reported the optimum packing density for
Ptꢀ
oxidation of MeOH are 0.75, 0.63, 0.45, and 0.18 for the
1.31, 0.81, 0.57, and 0.18 equivalents PtꢀRuad nano-
/Ruad nanoparticles in the series used for the electro-
/
particles, respectively. Fig. 5 shows plots of current per
unit real surface area versus uRu(estimated) for the
electrooxidations. The data were collected after 20 min
of MeOH electrooxidation. The reported optimum
surface compositions for electrooxidation of MeOH
electrooxidations carried out at 0.3 V was ꢀ
difference in optimum surface composition (0.38 equiva-
lents Ruad, uRu(estimated) 0.33 versus packing densityꢃ
/0.3. This
ꢃ
/
/
over PtꢀRuad systems varies with the nature of the
substrate, the deposition procedure, the method used to
determine uRu, and the conditions of the electrooxida-
/
0.5 at 0.4 V) again illustrates that the initial electro-
chemical properties (and thereby the surface structure)
of PtꢀRuad nanoparticles depend on the method of
/
tion experiment (e.g. [6,8ꢀ
The values range from uRu
prepared by hydrogenation of 1 [42]) to ꢀ
crystalline Pt decorated with electrodeposited Ruad
[6,17]). The only other reports that we are aware of
that use nanoparticle Pt as substrate, and that evaluate
/
11,17,18,22,23,27,32ꢀ
0.10 (PtꢀRuad gauzes
0.50 (poly-
/
34,42]).
preparation. We note that although the optimum sur-
face compositions are different, the activities per active
site of the two optimised catalysts are similar. These
interesting results require further investigation to be
understood.
ꢀ
/
/
/
We notice that the activity of the 0.18 equivalents Ptꢀ
/
the resulting PtꢀRuad catalysts for electrooxidation of
/
Ruad system increases less rapidly than the higher

D. Cao, S.H. Bergens / Electrochimica Acta 48 (2003) 4021ꢀ
/4031
4029
equivalent surfaces upon increasing the temperature to
60 8C. This behaviour is consistent with the observations
of Gasteiger et al. [82] using well-defined, shiny, poly-
fuel that used Ptꢀ
anode catalysts.
/
Ruad and Ptꢀ
/
Pdad nanoparticles as
Fig. 6 shows the polarisation curves for the series of
cells prepared for this investigation after they were
conditioned, and operated at 60 8C with 1 M MeOH
anode feed, and pure O2 (1 atm) cathode feed. Pure O2
was used to maximise the activity of the cathode and
thereby minimise cathode effects on the relative activ-
ities of the cells in this series. Although the cells are not
optimised, every effort was made to be as consistent as
possible during their assembly in order to obtain a true
comparison of anode catalyst activity. The currents are
normalised per gram of anode catalyst.
crystalline Ptꢀ
low uRu (ꢀ0.10) were the most active at 25 8C (0.4 V),
whereas alloys with higher uRu (ꢀ0.30) were more
/Ru alloys. They reported that alloys with
/
/
active at 60 8C [63,82]. This phenomenon was explained
by the ensemble effect. As Ru is inactive towards MeOH
near room temperature, ensembles of adjacent Ptsurf are
required to activate and dehydrogenate MeOH. It is
proposed that the activation and dehydrogenation of
MeOH is rate-limiting at room temperature, and so
surfaces with low uRu are the most active. As Ru is
active towards MeOH at 60 8C or higher, it is proposed
that the reaction between COad and OHad becomes rate-
limiting, and so surfaces with higher uRu are more active
The polarisation curves show that the relative order of
activity of the PtꢀRuad nanoparticles obtained in the
three-electrode experiments carried out at 60 8C is the
same as the relative order of activity obtained when the
/
than surfaces with lower uRu
.
Ptꢀ
operated at 60 8C. This result is consistent with our
earlier work using PtꢀRuad nanoparticles prepared with
/
Ruad nanoparticles are incorporated into DMFCs
3.4. Direct MeOH fuel cells
/
Ru4H4(CO)12 in which we showed that significant
surface segregation or dissolution resulting in loss of
Ruad did not occur during construction and initial (days
long) operation of a DMFC. A more detailed study of
these cells including in-cell cyclic voltammetry, longevity
studies, and temperature dependence will be presented
in the next paper in this series.
A series of prototype DMFCs were constructed using
the PtꢀRuad nanoparticles made by the reductive
deposition as anode catalysts. The cells were constructed
Ruad nanoparticles into a
Nafion†-117 membrane using the decal transfer method
developed by Gottesfeld and co-workers [58]. We used
untreated substrate Pt nanoparticles as the cathode
catalyst. The loadings on both sides of the cell were
/
by hot-pressing the Ptꢀ
/
ꢀ
/
2 mg/cm2. This construction is the same as we used
previously to prepare a DMFC using ꢀ0.10 equivalents
PtꢀRuad nanoparticles synthesised using Ru4H4(CO)12
as the adatom source. Wieckowski and co-workers [35]
recently reported fuel cells operating on formic acid as
4. Conclusions
/
/
This work describes a nonelectrochemical reductive
deposition to prepare PtꢀRuad nanoparticles with
/
reasonable control over surface stoichiometry. This level
of control can be used to optimise the activity of the
nanoparticles towards electrooxidation of MeOH both
in three-electrode experiments and in prototype
DMFCs. As such, this method compliments the sponta-
neous deposition developed by Wieckowski et al. The
voltammograms, the CO stripping behaviour, and the
optimum composition for electrooxidation of MeOH of
the PtꢀRuad nanoparticles prepared by the two methods
/
are different. These differences show that development
of alternative syntheses of nanoparticle catalysts is not
only driven by practical concerns, but alternate synth-
eses also result in alternate catalyst surface structures.
Acknowledgements
We thank Dr. Youbin Shao and Professor Gary
Horlick at the University of Alberta for their kind,
expert assistance with inductively coupled plasma spec-
trophotometry measurements. We thank the Natural
Sciences and Engineering Research Council of Canada
and the University of Alberta for supporting this work.
Fig. 6. Voltageꢀ
different PtꢀRuad anode catalysts operating at 60 8C. The numbers in
brackets indicate the estimated Ru coverage (see text). Anode: 2 mg/
cm2 PtꢀRuad catalyst, 1.0 M MeOH at 4.0 ml/min. Cathode: 2 mg/cm2
Pt black catalyst, 20 psi dry O2 at 400 sccm.
/
current density polarisation curves for DMFCs with
/
/

4030
D. Cao, S.H. Bergens / Electrochimica Acta 48 (2003) 4021ꢀ4031
/
[34] Y.Y. Tong, H.S. Kim, P.K. Babu, P. Waszczuk, A. Wieckowski,
E. Oldfield, J. Am. Chem. Soc. 124 (2002) 468.
[35] C. Rice, S. Ha, R.I. Masel, A. Wieckowski, J. Power Sources 115
(2003) 229.
References
[1] A.N. Frumkin, B.I. Podlovchenko, Ber. Akad. Wiss. USSR 150
(1963) 34.
[36] A. Crown, A. Wieckowski, Phys. Chem. Chem. Phys. 3 (2001)
3290.
[2] J.O.’M. Bockris, H. Wroblowa, J. Electroanal. Chem. 7 (1964)
428.
[37] E. Herrero, J.M. Feliu, A. Wieckowski, Langmuir 15 (1999) 4944.
[38] A. Crown, C. Johnston, A. Wieckowski, Surf. Sci. 506 (2002)
L268.
[3] O.A. Petry, B.I. Podlovchenko, A.N. Frumkin, H. Lal, J.
Eectroanal. Chem. 10 (1965) 253.
[4] X.M. Ren, P. Zelenay, S. Thomas, J. Davey, S. Gottesfeld, J.
Power Sources 86 (2000) 111.
[39] A. Crown, I.R. Moraes, A. Wieckowski, J. Electroanal. Chem.
500 (2001) 333.
[5] B.D. McNicol, D.A.J. Rand, K.R. Williams, J. Power Sources 83
(1999) 15.
[40] Y.Y. Tong, A. Wieckowski, E. Oldfield, J. Phys. Chem. B 106
(2002) 2434.
[6] M. Watanabe, S. Motoo, J. Electroanal. Chem. Interf. Electro-
chem. 60 (1975) 267.
[41] C.E. Lee, P.B. Tiege, Y. Xing, J. Nagendran, S.H. Bergens, J. Am.
Chem. Soc. 119 (1997) 3543.
[7] M. Watanabe, S. Motoo, J. Electroanal. Chem. Interf. Electro-
chem. 60 (1975) 275.
[42] C.E. Lee, S.H. Bergens, J. Phys. Chem. B 102 (1998) 193.
[43] D. Cao, S.H. Bergens, J. Electroanal. Chem. 533 (2002) 91.
[44] M.M.P. Janssen, J. Moolhuysen, Electrochim. Acta 21 (1976) 861.
[45] M.M.P. Janssen, J. Moolhuysen, Electrochim. Acta 21 (1976) 869.
[46] S. Szabo´, I. Bakos, J. Electroanal. Chem. 230 (1987) 233.
[47] S. Szabo´, I. Bakos, F. Nagy, J. Electroanal. Chem. 271 (1989) 269.
[48] P. Del Angel, J.M. Dominguez, G. Del Angel, J.A. Montoya, J.
Capilla, E. Lamy-Pitara, J. Barbier, Top. Catal. 18 (2002) 183.
[49] P. Del Angel, J.M. Dominguez, G. Del Angel, J.A. Montoya, E.
Lamy-Pitara, S. Labruquere, J. Barbier, Langmuir 16 (2000) 7210.
[50] J.M. Dumas, S. Rmili, J. Barbier, J. Chim. Phys. PCB 95 (1998)
1650.
[8] E. Herrero, K. Franaszczuk, A. Wieckowski, J. Electroanal.
Chem. 361 (1993) 269.
[9] W. Chrzanowski, A. Wieckowski, Langmuir 14 (1998) 1967.
[10] G. Tremiliosi-Filho, H. Kim, W. Chrzanowski, A. Wieckowski,
B. Grzybowska, P. Kulesza, J. Electroanal. Chem. 467 (1999) 143.
[11] W.F. Lin, M.S. Zei, M. Eiswirth, G. Ertl, T. Iwasita, W. Vielstich,
J. Phys. Chem. B 103 (1999) 6968.
[12] S. Cramm, K.A. Friedrich, K.-P. Geyzers, U. Stimming, R.
Vogel, Fresenius’ J. Anal. Chem. 358 (1997) 189.
[13] K.A. Friedrich, K.-P. Geyzers, A. Marmann, U. Stimming, R.
Vogel, Z. Phys. Chem. 208 (1999) 137.
[14] K.A. Friedrich, K.-P. Geyzers, U. Linke, U. Stimming, J.
Stumper, J. Electroanal. Chem. 402 (1996) 123.
[15] K.A. Friedrich, K.-P. Geyzers, A.J. Dickinson, U. Stimming, J.
Electroanal. Chem. 524 (2002) 261.
[51] C. Micheaud-Especel, D. Bazin, M. Guerin, P. Marecot, J.
Barbier, React. Kinet. Catal. Lett. 69 (2000) 209.
[52] C. Micheaud, P. Marecot, M. Guerin, J. Barbier, Appl. Catal. A
171 (1998) 229.
[16] G. Samjeske´, X.Y. Xiao, H. Baltruschat, Langmuir 18 (2002)
4659.
[53] C.L. Pieck, P. Marecot, J. Barbier, Appl. Catal. A 134 (1996) 319.
[54] C.L. Pieck, P. Marecot, J. Barbier, Appl. Catal. A 141 (1996) 229.
[55] C.L. Pieck, P. Marecot, J. Barbier, Appl. Catal. A 143 (1996) 283.
[56] J.M. Dumas, C. Geron, H. Hadrane, P. Marecot, J. Barbier, J.
Mol. Catal. 77 (1992) 87.
[17] M. Watanabe, Y. Genjima, K. Turumi, J. Electrochem. Soc. 144
(1997) 423.
[18] F. Vigier, F. Gloaguen, J.M. Leger, C. Lamy, Electrochim. Acta
46 (2001) 4331.
[57] A. van der Wiel, Chem. Weekblad. 48 (1952) 597.
[58] X. Ren, M.S. Wilson, S. Gottesfeld, J. Electrochem. Soc. 143
(1996) L12.
[19] T. Frelink, W. Visscher, J.A.R. van Veen, Surf. Sci. 335 (1995)
353.
[20] T. Frelink, W. Visscher, J.A.R. van Veen, Langmuir 12 (1996)
3702.
[59] J.A. Rard, Chem. Rev. 85 (1985) 1.
[60] C.A. Lucas, N.M. Markovic, P.N. Ross, Phys. Rev. B 55 (1997)
7964.
[21] T.D. Jarvi, T.H. Madden, E.M. Stuve, Electrochem. Solid State
Lett. 2 (1999) 224.
[61] N. Li, J. Lipkowski, J. Electroanal. Chem. 491 (2001) 95.
[62] H. Massong, H. Wang, G. Samjeske´, H. Baltruschat, Electro-
chim. Acta 46 (2000) 701.
[22] T. Iwasita, H. Hoster, A. John-Anacker, W.F. Lin, W. Vielstich,
Langmuir 16 (2000) 522.
[23] H. Hoster, T. Iwasita, H. Baumgartner, W. Vielstich, Phys. Chem.
Chem. Phys. 3 (2001) 337.
[63] H.A. Gasteiger, N. Markovic, P.N. Ross, Jr., E.J. Cairns, J. Phys.
Chem. 97 (1993) 12020.
[24] J.C. Davies, B.E. Hayden, D.J. Pegg, Electrochim. Acta 44 (1998)
1181.
[64] B.D. McNicol, R.T. Short, J. Electroanal. Chem. 92 (1978) 115.
[65] S. Swathirajan, Y.M. Mikhail, J. Electrochem. Soc. 138 (1991)
1321.
[25] J.C. Davies, B.E. Hayden, D.J. Pegg, Surf. Sci. 467 (2000) 118.
[26] J.C. Davies, B.E. Hayden, D.J. Pegg, M.E. Rendall, Surf. Sci. 496
(2002) 110.
[66] S. Hadzijordanov, H. Angerstein-Kozlowska, M. Vukovic, B.E.
Conway, J. Phys. Chem. 81 (1977) 2271.
[27] C. Lu, C. Rice, R.I. Masel, P.K. Babu, P. Waszczuk, H.S. Kim, E.
Oldfield, A. Wieckowski, J. Phys. Chem. B 106 (2002) 9581.
[28] P. Waszczuk, G.Q. Lu, A. Wieckowski, C. Lu, C. Rice, R.I.
Masel, Electrochim. Acta 47 (2002) 3637.
[67] W.F. Lin, P.A. Christensen, A. Hamnett, Phys. Chem. Chem.
Phys. 3 (2001) 3312.
[68] W.F. Lin, P.A. Christensen, A. Hamnett, J. Phys. Chem. B 104
(2000) 12002.
[29] W. Chrzanowski, A. Wieckowski, Langmuir 13 (1997) 5974.
[30] W. Chrzanowski, H. Kim, A. Wieckowski, Catal. Lett. 50 (1998)
69.
[69] W.F. Lin, P.A. Christensen, A. Hamnett, M.S. Zei, G. Ertl, J.
Phys. Chem. B 104 (2000) 6642.
[70] W.B. Wang, M.S. Zei, G. Ertl, Phys. Chem. Chem. Phys. 3 (2001)
3307.
[31] H. Kim, I.R. de Moraes, G. Tremiliosi-Filho, R. Haasch, A.
Wieckowski, Surf. Sci. 474 (2001) L203.
[71] W.F. Lin, M.S. Zei, Y.D. Kim, H. Over, G. Ertl, J. Phys. Chem. B
104 (2000) 6040.
[32] A. Crown, H. Kim, G.Q. Lu, I.R. de Moraes, C. Rice, A.
Wieckowski, J. New Mater. Elect. Syst. 3 (2000) 275.
[33] P. Waszczuk, J. Solla-Gullon, H.S. Kim, Y.Y. Tong, V. Montiel,
A. Aldaz, A. Wieckowski, J. Catal. 203 (2001) 1.
[72] W.F. Lin, T. Iwasita, W. Vielstich, J. Phys. Chem. B 103 (1999)
3250.

D. Cao, S.H. Bergens / Electrochimica Acta 48 (2003) 4021ꢀ
/4031
4031
[73] H.A. Gasteiger, N. Markovic, P.N. Ross, Jr., E.J. Cairns, J. Phys.
Chem. 98 (1994) 617.
[78] G.Q. Lu, P. Waszczuk, A. Wieckowski, J. Electroanal. Chem. 532
(2002) 49.
[74] P. Waszczuk, A. Wieckowski, P. Zelenay, S. Gottesfeld, C.
Coutanceau, J.M. Leger, C. Lamy, J. Electroanal. Chem. 511
(2001) 55.
[79] M. Krausa, W. Vielstich, J. Electroanal. Chem. 379 (1994) 307.
[80] A. Kabbabi, R. Faure, R. Durand, B. Beden, F. Hahn, J.-M.
Leger, C. Lamy, J. Electroanal. Chem. 444 (1998) 41.
[81] K. Wang, H.A. Gasteiger, N.M. Markovic, P.N. Ross, Jr.,
Electrochim. Acta 41 (1996) 2587.
[75] H.N. Dinh, X. Ren, F.H. Garzon, P. Zelenay, S. Gottesfeld, J.
Electroanal. Chem. 491 (2000) 222.
[76] N.M. Markovic, P.N. Ross, Jr., Surf. Sci. Rep. 45 (2002) 117.
[77] C.L. Green, A. Kucernak, J. Phys. Chem. B 106 (2002) 1036.
[82] H.A. Gasteiger, N. Markovic, P.N. Ross, Jr., E.J. Cairns, J.
Electrochem. Soc. 141 (1994) 1795.