Dependence of CH3OH oxidation activity for a wide range of PtRu alloys detailed analysis and new views

Article abstract of DOI:10.1149/1.1768545

Unsupported PtRu alloy powders of a wide range of compositions were prepared at low temperatures by carefully adjusting the preparation procedure. PtRu alloys of essentially the same surface, nominal and bulk composition, were formed up to ca. 46 atom % Ru content. Adsorbed CO stripping voltammetry and CH₃OH oxidation characteristics, namely, i-V curves and pseudo-steady-state current density values recorded at constant potentials, were the same as reported for corresponding bulk alloys, suggesting that the electrocatalytic activities of the powders can be compared to bulk alloys. CH₃OH oxidation activities obtained for PtRu alloy, Pt, and Ru powders showed the PtRu alloy of 70:30 atom % Pt:Ru composition to exhibit the highest activity independent of the temperature and potential tested (0.3 and 0.4 V vs. a reversible hydrogen electrode). The experimental pseudo-steady-state current density values for the CH₃OH oxidation reaction were found to show the same dependence on Ru content as theoretical values calculated assuming that an assembly of three neighboring Pt and one Ru site are involved in the oxidation of a CH₃OH molecule.

Full text of DOI:10.1149/1.1768545

Journal of The Electrochemical Society, 151 ͑8͒ A1269-A1278 ͑2004͒
A1269
0013-4651/2004/151͑8͒/A1269/10/$7.00 © The Electrochemical Society, Inc.
Dependence of CH3OH Oxidation Activity for a Wide Range
of PtRu Alloys
Detailed Analysis and New Views
,z
*
*
C. Bock, B. MacDougall, and Y. LePage
National Research Council of Canada, Ottawa, Ontario K1A 0R6, Canada
Unsupported PtRu alloy powders of a wide range of compositions were prepared at low temperatures by carefully adjusting the
preparation procedure. PtRu alloys of essentially the same surface, nominal and bulk composition, were formed up to ca. 46 atom
% Ru content. Adsorbed CO stripping voltammetry and CH3OH oxidation characteristics, namely, i-V curves and pseudo-steady-
state current density values recorded at constant potentials, were the same as reported for corresponding bulk alloys, suggesting
that the electrocatalytic activities of the powders can be compared to bulk alloys. CH3OH oxidation activities obtained for PtRu
alloy, Pt, and Ru powders showed the PtRu alloy of 70:30 atom % Pt:Ru composition to exhibit the highest activity independent
of the temperature and potential tested ͑0.3 and 0.4 V vs. a reversible hydrogen electrode͒. The experimental pseudo-steady-state
current density values for the CH3OH oxidation reaction were found to show the same dependence on Ru content as theoretical
values calculated assuming that an assembly of three neighboring Pt and one Ru site are involved in the oxidation of a CH3OH
molecule.
© 2004 The Electrochemical Society. ͓DOI: 10.1149/1.1768545͔ All rights reserved.
Manuscript submitted October 1, 2003; revised manuscript received February 10, 2004. Available electronically July 20, 2004.
Pt-Ru catalysts are of major interest as anode materials for low-
temperature fuel cells such as the direct methanol fuel cell. Platinum
catalyzes the methanol adsorption and dehydrogenation reactions,
shown in Eq. 1 ϩ 2. As a result, an intermediate ‘‘CO’’-type species
is formed that can be oxidized to CO2 with the assistance of -OH
type groups that are formed by the partial oxidation of H2O on
neighboring surface sites ͑M͒, Eq. 3 ϩ 4
mainly used PtRu catalyst systems that consisted of Ru particles/
islands deposited onto Pt surfaces prepared using a variety of meth-
ods, i.e., they used catalysts for which the distribution of Ru and Pt
sites is less random than for true alloy catalysts. This may explain
the independence of the CH3OH oxidation activity on the Pt:Ru
ratio for 10-30 atom % Ru contents. Other studies that involve dif-
ferent Pt:Ru ratios exist.2,6 Often these studies compare the CH3OH
oxidation activities per catalysts weight or geometrical electrode
area and not the true catalytic activity and/or use catalysts for which
the bulk and surface composition varies. However, for fundamental
and mechanistic studies, the use of catalysts of known and prefer-
ably random site distribution, i.e., true alloys, is essential. Indeed, it
is difficult to prepare well-defined PtRu alloys for which the bulk
and surface composition is known and the same. Typically, this in-
volves expensive methods such as sputter deposition, arc-melting
followed by surface polishing that leads to bulk metal alloys.3,4,7,8
In this work, the preparation and properties of a range of PtRu
alloy powders is discussed for Pt:Ru atomic ratios of up to ca. 45
atom % Ru. PtRu alloy powders that show essentially the same bulk
and surface compositions are used to obtain the CH3OH oxidation
activities as a function of the Pt:Ru atomic ratio for a broad potential
range. This includes the low-potential region which is relevant to
real fuel cell applications and where the -OH formation is believed
to be rate determining.9 The use of PtRu powders allows facile
preparation of electrodes of easily adjustable and sufficiently high
surface areas that permit reliable CH3OH oxidation current measure-
ments to be made, even at low potentials. The influence of tempera-
ture on the optimal Pt:Ru atom % for the CH3OH oxidation reaction
is also discussed.
Pt ϩ CH OH ϭ Pt- CH OH͒
ads
͓1͔
͓2͔
͓3͔
͑
3
3
Pt- CH OH͒ ϭ Pt- CO͒ads ϩ 4Hϩ ϩ 4eϪ
͑
͑
3
ads
M ϩ H2O → M-OH ϩ Hϩ ϩ 1eϪ
Pt- CO͒ads ϩ M Ϫ OH → Pt ϩ M ϩ CO2 ϩ Hϩ ϩ 1eϪ ͓4͔
͑
The -OH formation needs to take place at low potentials in order
for a fuel cell to generate high power outputs. Until now, bimetallic
Pt-Ru catalysts have exhibited some of the best catalytic activities
toward the electrochemical CH3OH oxidation reaction, likely due to
the ability of Ru to form active oxygen species at low potentials.1,2
Equation 4 suggests that the Pt and Ru site distribution at the atomic
level influences the CH3OH oxidation activity, as discussed in a
number of studies.3,4 Gasteiger et al.3 studied the CH3OH oxidation
reaction using well-characterized bulk PtRu alloys prepared by arc-
melting. Based on data collected for three different Pt:Ru ratios,
they concluded that a Pt:Ru composition of close to 90:10 atom %
shows the highest activity for the CH3OH oxidation reaction at room
temperature. Furthermore, the general view exists that an increase in
temperature results in a change of the optimal Pt:Ru ratio ͑toward
higher, namely, 50 atom % Ru, contents͒.3,5 This change in the op-
timal PtRu ratio has been based on an apparent activity of Ru to
dehydrogenate CH3OH at higher temperatures ͑Ͼ60°C͒, while Ru
does not show such an activity at room temperature.3 These conclu-
sions have been questioned by Iwasita et al.5 and Kabbabi et al.4
The latter suggested an optimal Ru content of 10-15 atom % inde-
pendent of the temperature; however, they reported experimental
current density data for the oxidation of CH3OH that were more
than one order of magnitude too high, thus casting some doubt on
the validity of their data. Iwasita et al. found very similar CH3OH
oxidation activities at room temperature for Pt:Ru ratios between
90:10 and 70:30 atom %. As pointed out by Iwasita et al.,5 they
Experimental
Catalyst powder preparation.—A range of Pt, Ru, and PtRu al-
loy powders were synthesized using a simple chemical reduction
method. The nominal atomic Pt to Ru ratio for the PtRu alloy pow-
ders was varied between 91 and 54 atom % Pt, i.e., between 9 and
46 atom % Ru. All powders were prepared at room temperature
using NaBH4 ͑Anachemia͒ as reducing agent. RuCl3 ͑99.99% pu-
rity, Alfa Aesar͒ and PtCl4 ͑99.9% purity, Alfa Aesar͒ were used as
precursor salts. Table I summarizes the amounts of Pt and Ru pre-
cursor salts used for the synthesis of the catalyst powders. In all
cases, the precursor salts were dissolved in 150 mL H2O and stirred
for 30 min. 75 mL of 0.2 M NaBH4 was added dropwise ͑within less
than 5 min͒ to the well-stirred solution for the catalyst powders of
less than 30 atom % Ru. For the catalysts made of more Ru, 75 mL
* Electrochemical Society Active Member.
z E-mail: christina.bock@nrc-cnrc.gc.ca
Downloaded on 2014-05-31 to IP 128.148.252.35 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

A1270
Journal of The Electrochemical Society, 151 ͑8͒ A1269-A1278 ͑2004͒
Table I. Amounts of precursor salts used for the powder synthe-
sis and Pt:Ru nominal and surface ratios.
Nominal Pt:Ru ratio
Surface Pt:Ru ratio
PtCl4
͑g͒
RuCl3
͑g͒
a
b
͑atom %͒
͑atom %͒
1.06
1.06
1.06
1.06
1.06
1.06
1.06
1.06
1.06
-
-
100:0
91:9
100:0
91:9
0.073
0.115
0.163
0.217
0.28
0.351
0.435
0.555
1.6
85:15
80:20
75:25
70:30
65:35
60:40
54:46
0:100
84:16
82:18
77:23
70:30
68:32
60:40
55:45
0:100
a Nominal Pt:Ru atom % ratio calculated from the Pt and Ru precursor
salts used for the powder preparation.
b Surface Pt:Ru atom % ratio determined from XPS measurements.
Figure 1. Typical XRD spectra for Pt, PtRu 70:30 atom %, and PtRu 54:46
atom % powders. An approximate amount of Si powder was homogeneously
mixed into the powders as an internal standard. The XRD spectra were
collected using a step size of 0.06° and an accusation time of 20 s. The
spectra were collected between 2␪’s of 20 and 120°; however, a zoomed-in
window of 25-85° is shown in Fig. 1.
of 0.4 M NaBH4 was added within less than 1 min to the precursor
salt solution ͑unless otherwise stated͒. Solid catalyst powder par-
ticles were formed immediately and the precursor salts were trans-
formed completely into solid catalyst powders. The formed Pt, PtRu,
and Ru catalyst powders were filtered and excessively washed with
H2O and subsequently dried in an air oven at 80°C for 30 min unless
otherwise stated. The precursor salts were dried in air at 135°C prior
to their use. Pt, Ru, and PtRu alloy powders were prepared using
different glassware. All glassware involved in the powder prepara-
tion was cleaned using aqua regia prior to its use.
Montreal͒ for the XPS analyses. For each catalyst, a survey spec-
trum was collected before high-resolution spectra of the C 1s, O 1s,
Pt 4f, Ru 3p, and Ru 3d core level regions were collected. Decon-
volutions of the XPS spectra were performed using a CasaXPS ver-
sion 2.1.34 ͑Neal Fairley͒. A Shimadzu UV-1201 S UV-visible spec-
trophotometer was also used.
Working electrode preparation.—The catalyst powders were
formed into electrodes by sonicating 13 mg of a particular powder
dispersed in solutions consisting of 300 ␮L Nafion solution ͑5 wt %
Nafion dissolved in lower alcohols, Fluka͒ and 2 mL H2O for 30
min. Appropriate amounts ͑1-20 ␮L͒ were pipetted onto ca. 0.25
cm2 Au foil electrodes ͑99.9% Au, 0.1 mm thick, Goodfellow͒ form-
ing thin catalyst layers. The electrodes were dried in air at room
temperature, cold-pressed for ca. 30 s applying 1 ton, and subse-
quently washed excessively with H2O. The Au foils were firmly
attached to Au wire electrodes, and Au not covered with the catalyst
powder was carefully wrapped with Teflon tape.
CO adsorbed (COads) stripping voltammetry.—CO was adsorbed
onto the Pt-based powder electrodes at 0.15 V by bubbling CO gas
͑Matheson purity, Matheson gas͒ through the 0.5 M H2SO4 solution
for 20 min. Solution CO was subsequently removed by argon gas
bubbling ͑air products͒ for 40 min, maintaining the potential at 0.15
V. The potential was then cycled at 10 mV sϪ1 starting at 0.15 V for
two complete oxidation/reduction cycles.
Solutions.—All CH3OH oxidation studies were carried out using
0.5 M CH3OH ϩ 0.5 M H2SO4 solutions. Prior to the electrochemi-
cal studies, the solutions were deoxygenated using high-purity argon
gas. A.C.S. grade chemicals and high resistivity 18 M⍀ water were
used.
Cells and electrodes.—Three-compartment cells, in which the
reference electrode was separated from the working and counter
electrode compartment by a Luggin capillary, were employed for the
electrochemical studies. The cells were equipped with water jackets
and condensers that allowed electrochemical studies under tempera-
ture control between 20 and 60°C. All potentials reported in this
paper are vs. the reversible hydrogen electrode ͑RHE͒ unless other-
wise stated. However, mercury sulfate electrodes ͑MSE, Ϫ0.68 V
vs. RHE10͒ were used as reference electrodes in the experiments.
Large-surface-area Pt gauzes served as counter electrodes.
Results and Discussion
Characterization of the catalyst powders XRD characteri-
zation.—XRD spectra were obtained for the as-prepared catalyst
powders as well as for the catalyst powders mixed homogeneously
with Si powder that served as internal standard. Figure 1 shows
typical XRD spectra for Pt, PtRu 70:30 atom %, and PtRu 54:56
atom % catalyst powders that were homogeneously mixed with Si.
The main peak of the hexagonal Si is clearly seen at 28.44͑29͒°.11
The peak position for Pt is seen to shift to more positive 2␪ values
with increasing Ru content. This indicates a decrease in the lattice
parameter of the face centered ͑fcc͒ Pt lattice, and hence, an incor-
poration of Ru into the Pt lattice. Diffraction peaks for Ru, such as
Ru metal or RuO2 phases, were not recognizable in the raw XRD
spectra. Figure 2 shows the dependence of the lattice parameter for
fcc Pt (aPt) determined using the Topas 2 program as a function of
the nominal atom % of Ru. Microstrain was taken into consideration
and the entire XRD spectra were used for the data-fitting procedure.
Particle size and microstrain values estimated using the Topas soft-
ware program are shown in Table II. Figure 2 shows that the lattice
parameter values obtained from the raw XRD data for the catalyst
powders prepared in this work decrease in a linear manner as the
atom % of Ru increases, in agreement with Vegard’s law.12 Lattice
parameter values reported for bulk PtRu alloys7 and values obtained
Techniques and instrumentation.—Electrochemical experiments
were performed using a Solartron SI 1287 electrochemical interface
͑Solartron Group, Ltd.͒ driven by Corrware software ͑Scribner
Assoc.͒. A Scintag XDS2000 system was employed using a Cu K␣
source to obtain X-ray diffraction ͑XRD͒ spectra for the as-prepared
catalyst powders. The scanning angle extended from 2␪ of 20 to
120°. The software program Topas 2 (DIFFRACPLUS Topas, Bruker
Axis, Inc.͒ was employed to extract lattice parameter constants from
the experimental XRD spectra. The entire XRD spectra ͑20-120°͒
were employed to analyze the data. Si powder ͑typically 1-20 ␮m,
99.9985% purity, Alfa Aesar͒ was used as the internal standard for
the XRD analyses. The powders were analyzed using a Si͑510͒ plate
͑GEM Dugout͒. X-ray photoelectron spectroscopy ͑XPS͒ spectra
were obtained using a Kratos Axis Ultra spectrometer equipped with
a monochromatized Al K␣ source. The as-prepared catalyst pow-
ders were attached to sticky Cu tape ͑3M copper tape, Soquelec,
Downloaded on 2014-05-31 to IP 128.148.252.35 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, 151 ͑8͒ A1269-A1278 ͑2004͒
A1271
carried out previously, focusing on the CO and -OH adsorption on Pt
and Ru sites,14 as well as calculating the lattice parameters as a
function of the Ru content for PtRu alloys.15
It is seen that the lattice parameter values for the powders pre-
pared in this work are close to the bulk alloy and ab initio calcula-
tion data. This indicates that the Ru content of the alloy is similar to
the nominal content for the catalysts prepared in this work. The
fitting of the XRD data of the higher Ru atomic composition ͑Ͼ35
atom %͒ required the addition of a hexagonal Ru phase. The Ru
amount in this phase was, however, very small compared to the
alloy, thus suggesting that the majority of the Ru is incorporated into
the fcc Pt lattice. Particle sizes estimated from the raw XRD data
suggests particle sizes of less than 25 nm ͑Table II͒. The particle size
is seen to decrease with increasing Ru content, consistent with pre-
viously reported results.16 It is clear, however, that these unsup-
ported and unstabilized catalyst powders consist of larger, agglom-
erated particles of ca. 150-200 nm size, as indicated by transmission
electron microscopy ͑TEM͒ measurements. Microstrain analyses
suggest stress to be considerable ͑in the percentage range͒ for all
powders. This is expected for these rapidly synthesized powders.
Figure 2. Pt lattice parameter (aPt) dependence on the nominal atom % of
Ru for fcc Pt. ͑ࡗ͒ The data obtained from fitting the experimental XRD
spectra for the Pt-based powders prepared in this work. A Si standard was
homogeneously dispersed between the powders for the analysis, and the
entire slow-scan spectra ͑2␪ between 20 and 120°͒ were used to calculate the
lattice parameter values using the Topas software program. The vertical lines
represent the errors of the lattice parameter in the y direction, while the bars
are only used to better locate the error lines. The solid line represents the
lattice parameter dependence obtained from ab initio calculations,13 while
the crosses represent the same, but for previously reported data obtained
from experimental XRD spectra for bulk PtRu alloys.7
XPS characterization.—The Pt-based powders were analyzed by
XPS to obtain the Pt to Ru atomic ratios as well as to identify the
oxidation states of the Pt and Ru surface components. The Pt:Ru
atomic ratios determined from the Pt 4f and Ru 3p spectra are shown
in Table I. The less intensive Ru 3p region was used instead of the
main Ru 3d region. The latter overlays with the C 1s peak, thus
complicating the deconvolution and interpretation of the Ru 3d
spectra. It is seen that for all catalyst powders, the surface atomic
ratio is close to the nominal value, thus suggesting that the surface
and bulk concentrations of these catalysts are essentially the same.
Figures 3a and b show deconvoluted XPS spectra for the Pt 4f and
Ru 3p region, respectively, for the PtRu 54:46 atom % catalyst. The
position of the C 1s peak, i.e., 284.6 eV, was used to correct the XPS
spectra for charging effects that were in the range of Ϯ0.1 eV. The
deconvolution of the Pt spectra indicates that a significant fraction
͑54%͒ of the Pt is present as Pt metal, indicated by Pt 4f7/2 and 5f5/2
peaks at 71.4 and 74.7 eV, respectively, and full widths at half maxi-
mum ͑fwhm͒ of 1.05 eV of each peak.17 39% of the Pt is suggested
to be present as PtO, indicated by Pt 4f7/2 and 5f5/2 peaks at 72.1 and
75.4 eV, respectively, and fwhm of 2.1 eV.17 A small fraction ͑ca.
6%͒ of the Pt is suggested to be present as Pt chlorides, indicated by
the Pt 4f7/2 and 5f5/2 peaks at 75 and 78.3 eV, respectively, and
fwhm of 2.3 eV.18 The presence of a small amount of ClϪ ͑that is
introduced into the catalyst systems by use of chloride-based pre-
cursor salts͒ was confirmed in the XPS survey spectra. The decon-
volution of the Ru 3p spectra suggest 70% of the Ru to be present as
Ru metal ͑Ru 3p1/2 and 3p3/2 peaks at 462 and 484 eV, respectively,
from ab initio calculations13 are also shown in Fig. 2. Ab initio cell
and structure optimizations were performed with VASP13 on the
known structure of Pt as well as on hypothetical cubic models for
the stoichiometric compositions Pt107Ru, Pt31Ru, Pt26Ru, Pt15Ru,
Pt7Ru, Pt3Ru, and PtRu with resulting symmetries Fm3m, Im3m,
or Pm3m. The cell edge for a hypothetical Fm3m model of Ru was
also optimized. Results of the ab initio computed cell edge (aPt) vs.
the atom % Ru content are seen ͑Fig. 2͒ to plot on a straight line
with a slope of Ϫ0.000133 nm, indicating that these cubic PtRu
alloys have a cell edge which follows Vegard’s law. We did not
attempt to establish the maximum Ru content of such alloys, but this
limit is experimentally known to be about 80 atom %. Hard-sphere
considerations based on experimental cell volumes for fcc Pt and
hexagonal close packed ͑hcp͒ Ru, and assuming a Vegard law type
of behavior of aPt vs. the atom % of Ru, actually give a slope of
Ϫ0.000137 nm, a result which is in excellent agreement with the
quantum results. Ab initio calculations for Pt-Ru systems have been
fwhm of 2.6 eV18,19͒ and 30% of the Ru between the metallic and
ϩIV state, i.e., as lower, reducible Ru oxides ͑Ru 3p1/2 and 3p3/2
peaks at 464.1 and 486.1 eV, respectively, fwhm of 3.5 eV18,19͒. It is
noteworthy that the XPS spectra shown in Fig. 4a and b are quali-
tatively representative for the PtRu powders of Pt:Ru atom % ratio
of 91:9 to 65:35 atom %, i.e., the catalyst particle surfaces are sug-
gested to mainly consist of Pt and Ru metals as well as lower,
reducible oxides. Similarly, the XPS data for the Ru powder showed
the surface to consist of mainly Ru metal and lower Ru oxides.
Table II. Parameters derived from raw XRD dataa for the as-
prepared Pt-based catalysts.
Nominal Pt:Ru ratio
Particle size
Microstrain
͑atom %͒
͑nm͒
͑%͒
100:0
91:9
27.5 Ϯ 3.4
13 Ϯ 1.2
1.1 Ϯ 0.063
2.6 Ϯ 0.15
1.7 Ϯ 0.17
1.1 Ϯ 0.063
1.8 Ϯ 0.08
2.6 Ϯ 0.26
2.7 Ϯ 0.5
COads stripping voltammetry.—COads stripping voltammetry is a
technique that can be used to obtain in situ information of catalyst
surfaces under electrochemical conditions.7,20 CO adsorbs on metal-
lic Pt sites at sufficiently negative potentials and the adsorbed CO
͑ca. 0.8-1 monolayer͒ can be anodically stripped from the Pt surface
to form CO2 .7 In the case of Pt-only catalysts, the COads stripping
charge (QCOads) can be used to obtain the electroactive Pt surface
area.7,21 For example, a charge-to-Pt area conversion factor of 420
␮C cmϪ2 is used for the case of the adsorption of one CO molecule
per Pt surface site. Figure 4a shows a COads stripping voltammo-
gram for a Pt powder catalyst electrode. The current axis is normal-
85:15
80:20
75:25
70:30
65:35
60:40
54:46
13.4 Ϯ 1.4
12.2 Ϯ 1.2
11.2 Ϯ 1.5
11.6 Ϯ 1.7
5.4 Ϯ 0.7
7.9 Ϯ 0.6
4.6 Ϯ 2.1
2.5 Ϯ 0.5
2.5 Ϯ 0.5
a Complete slow-scan XRD spectra collected between 2␪s of 20 and
75° were used for the data fitting.
Downloaded on 2014-05-31 to IP 128.148.252.35 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

A1272
Journal of The Electrochemical Society, 151 ͑8͒ A1269-A1278 ͑2004͒
Figure 3. XPS spectra for the as-prepared PtRu alloy catalyst of Pt:Ru ratio of 54:46 atom %: ͑a͒ the spectra for the Pt 4f core region and ͑b͒ the Ru 3p core
region, ͑ᮀ͒ the experimental data, ͑ ͒ convoluted peaks, and ͑—͒ the sum of the deconvoluted peaks.
Fig. 4a-d shows typical COads stripping voltamograms for Pt
powder and PtRu powders of 70:30 and 54:46 atom % composition,
respectively. The current scales in these CVs are normalized using
the corresponding, experimental QCOads values to better compare the
data. As discussed in previous work,21 the COads oxidation potential
can be linked to the PtRu surface concentration for the case of
alloys. With an increase in atomic Ru content ͑up to ca. 46%͒, the
COads stripping potential is shifted to more negative values in the
same manner as also seen in our data. The COads stripping voltam-
mograms observed in this work also show that they can be used to
gain some insight into the PtRu alloy catalyst surface within a cer-
tain error range ͓mainly due to variations in the potential measure-
ments ͑Ϯ10 mV͔͒. The shape and characteristics of the entire COads
voltammograms observed in this work for a catalyst powder of a
particular composition are in good agreement with literature data for
bulk alloys of the corresponding Pt:Ru surface ratio.7,20 Therefore,
the COads stripping data suggest that the surface composition of
these catalyst powders prepared in this work are the same as the
bulk and nominal composition, as also suggested by the XRD and
XPS results. Furthermore, the good agreement between the COads
stripping characteristics of the powder and bulk alloys validates the
use of these powders to obtain mechanistic CH3OH oxidation infor-
mation. It is also seen that for the Ru powder ͑e.g., Fig. 4b Ru
powder͒, very positive potentials are required to strip the adsorbed
CO completely from the catalyst surface and the COads stripping
peaks become broad.
ized by the QCOads value determined for this particular electrode.
The first cycle ͑thick line͒ represents the CO stripping curve, while
the second cycle ͑thin line͒ is equivalent to the background CV for
this catalyst in 0.5 M H2SO4 . For polycrystalline Pt, typical hydro-
gen adsorption/desorption peaks are observed in the background cy-
clic voltammetry ͑CV͒. The experimentally observed QCOads values
were found to be 80% of two times the H desorption charge (QHdes
)
for Pt powder electrodes. This suggests that under the conditions
used in this work 0.8 of a CO monolayer is adsorbed on the Pt
surface sites, i.e., the correct QCOads to Pt surface area conversion
factor is 336 ␮C cmϪ2 (ϭ0.8 ϫ 420 ␮C cmϪ2). The conversion of
the QCOads values to Pt surface areas in the presence of Ru is more
complicated unless the number of CO molecules that adsorb per Ru
and Pt sites is known. In recent work in our laboratory,21 it has been
shown that activation-controlled (COOH)2 oxidation currents can
be used to obtain electroactive Pt areas even for multicomponent
catalysts such as PtRu powders. It has also been shown that the
combination of the (COOH)2 technique and COads stripping volta-
mmetry can be used to estimate the number of CO molecules that
adsorb on Ru and Pt sites.21 The QCOads value and the Pt surface area
(APt) estimated using the (COOH)2 oxidation currents are related
using an experimentally determined conversion factor (fAPt), as fol-
lows
QCOads
420␮CcmϪ2
APt
ϭ
f
͓5͔
*
APt
In this work, COads stripping voltammograms were also recorded
for the catalysts before and after CH3OH oxidation studies. The
COads stripping voltamograms were used to calculate Pt surface
areas and to ensure that the catalyst surfaces are not altered as a
result of the CH3OH oxidation experiments.
As discussed in previous work, the value of the fAPt factor de-
pends on the catalyst powder; it is unity for Pt-only powders and
less than unity if CO adsorbs on other, e.g., Ru, catalyst sites.21 In
the case of PtRu alloys, the same ratio of CO molecules per Pt and
Ru sites are suggested to adsorb, and the experimentally determined
Catalyst preparation parameters and resulting catalyst
properties.—Influence of the NaBH4 addition rate on the degree of
PtRu alloy formation.—In this work, the rapid addition of high con-
centrations of NaBH4 solutions to the Pt and Ru precursor solutions
was found essential to incorporate the Ru into the Pt fcc lattice. The
influence of the rate and amount of NaBH4 addition is more pro-
nounced for the synthesis of the catalyst powders made of higher Ru
͑Ͼ30 atom %͒ ratio. The influence of the NaBH4 addition rate is
f
APt factor is equivalent to the atomic fraction of Pt surface sites. For
example, for a Pt:Ru surface ratio of 85:15 atom %, the fAPt factor is
0.85; for a Pt:Ru surface ratio of 70:30 atom %, the fAPt factor is
0.7,21 etc. These fAPt factors and Eq. 5 will be used in this work to
obtain the real catalytic activities for the PtRu catalysts toward the
CH3OH oxidation reaction, i.e., to obtain CH3OH oxidation currents
per Pt surface area.
Downloaded on 2014-05-31 to IP 128.148.252.35 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, 151 ͑8͒ A1269-A1278 ͑2004͒
A1273
Figure 4. COads stripping voltammograms recorded at 10 mV sϪ1 in 0.5 M H2SO4 solutions for ͑a͒ Pt powder, ͑b͒ Ru powder, ͑c͒ PtRu 70:30 atom % alloy,
and ͑d͒ PtRu 54:46 atom % alloy. ͑—͒ The first oxidation cycle, i.e., the anodic stripping of the COads , and ͑
͒ the second complete oxidation/reduction
cycle, which is equivalent to the background CV of the catalyst powder electrode in 0.5 M H2SO4 .
discussed in this section using Pt:Ru precursor ratios of 54:46 atom
%. The slow addition of NaBH4 ͑i.e., the dropwise addition of 75
mL of 0.2 M NaBH4 over a period of 30 min͒ was found to not
entirely reduce the Ru͑ϩIII͒-chloride precursor salt, while UV-
visible ͑UV-vis͒ spectroscopy did not show detectable amounts of
dissolved Pt. These differences can be assigned to different reduc-
tion rates of the Ru and Pt precursor salts, i.e., to a faster reduction
rate of PtCl4 than RuCl3 .
Values of aPt estimated using raw XRD data recorded for pow-
ders 1 ͑slow addition rate, 30 min for the addition of 75 mL of 0.2
M NaBH4) and 2 ͑more rapid addition rate, 5 min for the addition of
75 mL of 0.2 M NaBH4) showed that PtRu alloys of less than 46
atom % Ru were formed. aPt values of 0.387192 Ϯ 3 ϫ 10Ϫ5 and
0.387102 Ϯ 8 ϫ 10Ϫ5 nm and corresponding Pt:Ru alloy ratios of
ca. 62:38 and 61:39 atom % for powders 1 and 2, respectively, were
estimated. These aPt values also suggest lower Pt:Ru alloy ratios
than for the rapidly synthesized powder 3. XPS analyses were also
carried out for these PtRu catalyst powders of nominal 54:46 atom
% ratio. The XPS data suggest Pt:Ru atom % ratios of 12:88 for
powder 1, 28:72 for powder 2, and 55:45 for powder 3 ͑rapid addi-
tion rate and increased NaBH4 amount, less than 1 min for the
addition of 75 mL of 0.4 M NaBH4). These data indicate that Ru is
preferentially located on the catalyst surface for the powders pre-
pared using a smaller amount and slower addition rate of the NaBH4
reducing agent, and hence, the catalyst particle composition ͑surface
Figure 5. Typical XRD spectra for PtRu 54:46 atom % catalyst powders
subjected to heat-treatments for 1 h in an air oven at ͑a͒ 600, ͑b͒ 450, ͑c͒ 350,
͑d͒ 250, and ͑e͒ 100°C. The spectra are corrected using the position of the
internal Si powder standard at 28.44°. The XRD spectra were collected using
a step size of 0.06° and 40 s accusation time. The spectra were collected
between 2␪’s of 20 and 60°; however, a zoomed-in window of 30-45° is
shown in Fig. 5.
Downloaded on 2014-05-31 to IP 128.148.252.35 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

A1274
Journal of The Electrochemical Society, 151 ͑8͒ A1269-A1278 ͑2004͒
Figure 7. COads stripping voltammograms for PtRu 54:46 atom % catalyst
powders subjected to different heat-treatments. The first anodic cycle ͑after
CO adsorption at 0.15 V͒ is shown. The CVs were recorded at 10 mV sϪ1 in
Figure 6. Dependence of the lattice parameter of ͑ϩ͒ fcc Pt on the tempera-
ture used to treat a 54:46 atom % PtRu powder catalyst in air for 1 h. ͑᭺͒
The dependence of atom % Ru in the fcc Pt phase on the temperature. The
atom % Ru values were calculated using the lattice parameter values and the
theoretical Vegard’s law relationship shown in Fig. 2.
0.5 M H2SO4 solutions. ͑
͒ The COads stripping voltammogram for the
catalyst powder heated at 100°C, ͑—͒ the same for the catalyst powder
heated at 250°C, and ͑—͒ the stripping voltammogram for the catalyst pow-
der heated at 600°C. The currents were normalized by the COads charge to
better compare the data.
vs. bulk͒ of the ‘‘more slowly’’ synthesized PtRu powders 1 and 2 is
not uniform. XPS analysis for powders of nonuniform conditions
such as powders 1 and 2 likely yields only an estimate of the Pt:Ru
surface ratio.
in catalyst particle size. This was confirmed from particle size cal-
culations for the full XRD spectra using the Topas software. The 2␪
position of the Pt͑111͒ peak is seen to shift to smaller values, indi-
cating that the amount of Ru in the fcc Pt lattice is smaller, i.e., the
Pt:Ru alloy ratio decreases with increasing temperature. This is re-
flected more quantitatively in the lattice parameter values estimated
for the full spectra shown in Fig. 6. Figure 6 also shows the depen-
dence of the atom % of Ru in the fcc Pt phase, calculated from the
aPt lattice parameter and using the theoretical Vegard’s law relation-
ship shown in Fig. 2, showing a clear depletion of Ru in the PtRu
alloy with increasing temperature. Best fits of the XRD spectra for
the powders heat-treated at 350°C and higher temperatures were
only obtained using Pt, Ru, and RuO2 phases.
Heat-treatment of the prepared catalyst powders.—In this work,
care was taken to avoid the exposure of the PtRu catalysts to high
temperatures. Treatment at high temperatures of PtRu alloys typi-
cally results in Pt segregation on the catalyst surface.22 In this sec-
tion, the influence of heat-treatment on the prepared catalyst pow-
ders is discussed to ensure that suitable catalyst preparation
conditions are used. PtRu alloy catalysts of 54:46 atom % compo-
sition were heated in air for 1 h at either ͑a͒ 600, ͑b͒ 450, ͑c͒ 350,
͑d͒ 250, or ͑e͒ 100°C. Figure 5 shows an enlarged region of the
XRD spectra recorded for the heat-treated catalyst powders. The
XRD spectra clearly indicate the formation of a RuO2 phase for the
catalysts treated at 600°C. The Pt͑111͒ peak is seen to become nar-
rower with increasing temperature, indicating ͑an expected͒ increase
Changes in the surface properties of the PtRu catalyst induced
Figure 8. CVs ͑5th cycle͒ in 0.5 M CH3OH ϩ 0.5 M H2SO4 recorded at 10 mV sϪ1 between 0.09 and 0.8 V for Pt powder, PtRu alloy ͑54:46 atom %͒, and
1 day old Ru powder. The background CV for the Ru powder electrode recorded in the 0.5 M H2SO4 (CH3OH-free͒ solution is also shown. ͑a, left͒ The CVs
recorded at 20°C, and ͑b, right͒ the CVs recorded at 60°C. The current scales were normalized for the Pt ϩ Ru area.
Downloaded on 2014-05-31 to IP 128.148.252.35 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, 151 ͑8͒ A1269-A1278 ͑2004͒
A1275
Activation enthalpies and CH3OH oxidation activities: PtRu
alloys vs. Ru powder electrodes.—In this section the CH3OH oxi-
dation activities as a function of temperature for the PtRu alloys ͑of
54:46 and 70:30 atom %͒ and the freshly synthesized Ru metal
powder electrodes were obtained using cyclic voltammetry ͑CV͒.
Figures 8a and b show typical CVs for the CH3OH oxidation reac-
tion for Pt powder, PtRu alloys 54:46 atom %, and Ru powder
catalyst electrodes. In all cases, five cycles were sufficient to reach
current density-voltage ͑i-V͒ characteristics that were independent
of subsequent potential cycling. Therefore, the 5th CV cycles are
shown in Fig. 8a and b. All currents were normalized for the Pt
ϩ Ru surface area estimated using COads stripping voltammetry, as
discussed previously. Figure 8a shows the CH3OH oxidation char-
acteristics for 20°C, while Fig. 8b shows the situation for 60°C. A
background CV for the Ru powder catalyst in 0.5 M H2SO4 is also
shown in the two figures for comparison. It is seen that the freshly
prepared Ru powder shows a recognizable activity toward the
CH3OH oxidation reaction at both 20 and 60°C. However, its activ-
ity is clearly smaller than for the PtRu 54:46 atom % alloy powder.
The activity of the Ru powder toward the oxidation of CH3OH
appears high when compared to literature data.3 However, a freshly
synthesized ͑1 day old͒ Ru catalyst powder was used for the CV
data shown in Fig. 8a and b. The activity of the Ru powder toward
CH3OH oxidation reaction was observed to decrease with time, e.g.,
a 2 week old powder exhibited only ca. 20% of CH3OH oxidation
activity of the 1 day old powder, while a 6 month old Ru powder
showed no activity for the COads and CH3OH oxidation reactions.
This decay in activity may be due to the oxidation of the Ru surface
in air. The mechanism and final product͑s͒ of the CH3OH oxidation
using Ru is not known and may be different than for the Pt and PtRu
alloy electrodes.
Figure 9. Arrhenius plots, i.e., ln j vs. 1000/T graphs for the CH3OH oxi-
dation reaction using PtRu alloys of ͑ᮀ͒ 70:30 atom %, ͑᭝͒ 54:46 atom %,
and ͑छ͒ 1 day old Ru powder. The current density data were extracted at
0.68 V vs. RHE from slow-sweep CV data ͑see Fig. 8͒.
by the heat-treatment were also observed by COads voltammetry.
Figure 7 shows the COads stripping voltammograms ͑normalized for
the corresponding COads charge values͒ for the PtRu 54:46 atom %
catalysts heated at 100, 250, and 600°C. The COads stripping vol-
tammograms for the catalysts heated at 100°C is essentially the
same as for the same catalysts but dried at 80°C ͑Fig. 4c͒, as well as
room temperature, indicating that the catalyst surface is not affected
by the 100°C heat-treatment. However, differences in the COads
stripping voltammograms are seen for the catalyst powders heated at
250 and 600°C. Both the COads oxidation peak and oxidation onset
potential are shifted more positive with increasing temperature. This
indicates that the catalyst surface properties are altered by the heat-
treatment, resulting in conditions that are less favorable for the
COads oxidation reaction. It is most likely that the Pt: Ru site distri-
bution of the heat-treated catalysts ͑250°C and higher͒ is less ran-
dom than for the 100°C and lower temperature-treated catalysts, as
suggested by the XRD data that show a decrease in the amount of
Ru dissolved in the Pt fcc lattice with increasing temperature.
Figure 9 shows Arrhenius plots, i.e., plots of the logarithm of the
CH3OH current density ln(j) vs. the inverse of the temperature
͓
͔
(1/T) for two PtRu alloys, i.e., 54:46 ͑᭝͒ and 70:30 ͑ᮀ͒ atom %, as
well as for 1 day old Ru powders ͑छ͒. The CH3OH current density
values were extracted from steady-state ͑5th͒ CV data at 0.68 V, i.e.,
from CV data as shown in Fig. 8a and b. The experimentally ob-
served CH3OH oxidation current is seen to increase with increasing
temperature, as expected.3 For all tested temperature and potential
ranges, the PtRu 70:30 atom % alloy catalyst shows the highest
CH3OH oxidation activity, i.e., higher than the PtRu 54:46 atom %
Figure 10. CH3OH oxidation activity dependence on atom % Ru at ͑a, left͒ 0.3 and ͑b, right͒ 0.4 V vs. RHE. ͑ϫ͒ The current density values obtained from i-t
transients after 5 min in 0.5 M CH3OH and 0.5 M H2SO4 at 20°C. The current values are normalized by the Pt ϩ Ru surface area, as described in the text. ͑᭿͒
The theoretical current density values per Pt ϩ Ru surface area calculated assuming that CH3OH to CO2 oxidation requires three Pt and one adjunct Ru sites,
as discussed in the text and using Eq. 8.
Downloaded on 2014-05-31 to IP 128.148.252.35 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

A1276
Journal of The Electrochemical Society, 151 ͑8͒ A1269-A1278 ͑2004͒
alloy and the Ru powder. These data suggest that the optimal Pt:Ru
ratio for the CH3OH oxidation reaction is independent of tempera-
ture ͑between 20 and 60°C͒, consistent with Kabbabi et al.’s
conclusions4 and the raw data in Gasteiger et al.’s work.3 Activation
the cathodic current was found to result in the complete recovery of
the catalyst activity for the CH3OH oxidation reaction.
Figures 10a and b shows the CH3OH oxidation activities at
20°C as a function of the atom % of Ru at 0.3 and 0.4 V, re-
spectively. The crosses show the experimentally observed CH3OH
oxidation currents divided by the Pt ϩ Ru surface area, while the
squares show theoretical curves obtained as described below. The
experimental CH3OH current density values for the 65:35 and 54:46
atom % alloy are similar to previously reported values for bulk
metal alloys of similar composition and comparable experimental
conditions.3 This agreement shows that these unsupported catalyst
powders can be used for fundamental and reaction mechanistic
studies. It is seen that for both potentials the 70:30 atom % com-
position yields the highest CH3OH oxidation activity. The 70:30
atom % maximum and the shape of the curves can possibly be
explained by considering a simple statistical model that views the
alloys as fcc lattice made up of randomly distributed Pt and Ru sites.
The CH3OH oxidation reaction is assumed to involve three neigh-
boring Pt sites and one Ru site, while all other sites are assumed not
to be notably involved in the CH3OH oxidation reaction. Electro-
static interactions between adsorbed CH3OH molecules that may
lead to a decrease in CH3OH adsorption and possible alloying ef-
fects such as electronic influences of Ru on the Pt sites are ignored
in this model. Using this simple model the number of catalyst sites
involved in the CH3OH oxidation reaction per total surface sites,
i.e., the theoretical catalyst site utilization percentage ͑csu %͒ can be
calculated as a function of the Pt:Ru atom % ratio. The csu % is
defined as follows
enthalpy values (⌬Ho#) were also extracted from the slope of these
¯
Arrhenius plots at a particular potential ͑E͒ employing Eq. 623
o#
¯
ץ ln j
Ϫ⌬H
slope ϭ
ϭ
͓6͔
1
ͩ ͪ
ץ
R
T
E
͑Note that in the case of measurements in aqueous solutions, the
terms activation enthalpy and energy can be used interchangeably.23͒
In Eq. 6, T is temperature and R the gas constant ͑8.314 J KϪ1
molϪ1 24͒. In this work, activation enthalpy values of 30, 33, and
28 Ϯ 5 kJ molϪ1 were obtained at 0.68 V for the 54:46 atom %
alloy, the 70:30 atom % alloy, and the Ru powders, respectively. The
activation enthalpy values for the two PtRu alloys are lower than
previously reported for similar catalysts that were estimated, how-
ever, at lower overpotentials.3 In fact, activation enthalpies similar
to literature values of ca. 60 kJ molϪ1 were estimated in this work
for the 70:30 and 54:46 atom % PtRu alloys at 0.5 V. The CH3OH
oxidation activity for the Ru powder was too small at lower poten-
tials to allow the extraction of the activation enthalpy value. The
activation enthalpy values estimated at 0.68 V are similar for the
three powders. These data suggest that the activity for the CH3OH
oxidation reaction of the Ru does not increase more strongly with
no. of catalyst sites actively utilized in the CH3 OH oxidation reaction
csu % ϭ
͓7͔
no. of total catalyst surface sites
increasing temperature than for the PtRu alloy electrodes. A signifi-
cant contribution of Ru sites toward the CH3OH dehydrogenation
and oxidation reaction in PtRu alloys appears further questionable,
as current density values for Ru are clearly lower than for PtRu
alloys even at elevated temperatures. It should also be noted that
contrary to the case of the Pt and the PtRu alloys, the relationship
between the number of CO molecules that adsorb on Ru-only pow-
ders and its real Ru surface area is not known and the true Ru
surface area may actually be higher. This would result in smaller
CH3OH oxidation current density values on the Ru-only powders
than reported here.
In Eq. 7, Pt ϩ Ru, Pt, and Ru sites are used to calculate Pt ϩ Ru,
Pt, and Ru csu % numbers, respectively, shown in Table III as a
function of the PtRu alloy composition. It is seen that the 70:30
atom % alloy yields the maximum utilization of Pt ϩ Ru catalyst
sites ͑of 92%͒ consistent with the experimental data. The Pt ϩ Ru
csu % numbers shown in Table III were further used to calculate
theoretical CH3OH oxidation currents (JCH OH(theoretical)) as a func-
3
tion of the Pt:Ru atom % to better compare the catalyst site utiliza-
tion and the experimental CH3OH oxidation data, as shown in Eq. 8
CH3OH oxidation currents as a function of Pt:Ru atomic
concentration.—In this section, the influence of the Pt:Ru atomic
ratio on the CH3OH oxidation activity is investigated using a wide
range of catalyst compositions. CH3OH oxidation activities were
estimated from current-time ͑i-t͒ transients recorded at 0.3 and 0.4 V
Table III. Theoretical catalyst utilization percentage „csu %…
numbers as a function of the Pt:Ru atomic ratioa
atom % Ru
Pt csu %c
Ru csu %d
Pt ϩ Ru csu %b
9
15
20
25
30
35
40
45
43
56
78
83
92
84
70
57
43
56
78
83
94
86
83
79
100
100
100
83
82
83
after
5 min, i.e., when pseudo-steady-state conditions were
achieved. The amount of catalyst powders used for the measure-
ments was adjusted depending on the potential range and the PtRu
ratio studied. Pt ϩ Ru areas of 2-100 cm2 were generally used for
measurements at 0.3 V, while Pt ϩ Ru areas of less than 10 cm2
were used for the measurements at 0.4 V. The use of the higher
surface area electrodes at the lower potential allowed for sufficiently
high current measurements to achieve good quality data, while use
of the lower surface area electrodes at more positive potentials
avoided the production of large amounts of H2 at the counter elec-
trode. The latter is particularly important as an undivided cell is used
in this work. Between the recordings of the individual i-t transients,
a cathodic current of 5 ϫ 10Ϫ5 A was applied until the monitored
electrode potential reached a value of 0.025 V. The application of
50
30
a csu % numbers are calculated assuming that an assembly of 3 Pt and
1 adjunct Ru site are the prerequisite for the CH3OH to CO2 oxida-
tion reaction, as discussed in the text.
b Pt ϩ Ru csu % is the percentage of Pt ϩ Ru sites utilized per total
Pt ϩ Ru surface sites.
c Pt csu % is the percentage of Pt sites utilized per Pt surface sites.
d Ru csu % is the percentage of Ru sites utilized per Ru surface sites.
Downloaded on 2014-05-31 to IP 128.148.252.35 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, 151 ͑8͒ A1269-A1278 ͑2004͒
A1277
suggest the Pt:Ru surface ratio to be the same as the nominal Pt:Ru
ratio, thus suggesting that PtRu particles of uniform compositions
were formed. Furthermore, in situ electrochemical COads stripping
voltammograms recorded for the various Pt, Ru, and PtRu alloy
powders were consistent with previously reported voltammograms
for bulk PtRu alloys of corresponding Pt:Ru surface ratio, indicating
that catalyst powders of similar nominal, bulk, and surface ratios
were made. XPS data also suggest the Pt and Ru surface compo-
nents to mainly consist of Pt and Ru metal as well as lower, reduc-
ible Pt and Ru oxides. It is noteworthy that chemical reduction route
synthesis provides a simple and less expensive method for the
preparation of PtRu catalysts as compared to arc-melting and chemi-
cal vapor deposition techniques. Furthermore, the method described
in this work allows the preparation of catalyst powders at low tem-
peratures, thus minimizing undesirable segregation of Pt on the cata-
lyst surface. Many powders of a wide range of Pt:Ru atomic ratios
can be easily prepared at ‘‘low’’ cost and used for detailed studies.
Also, catalyst powder electrodes of variable surface areas can be
prepared, thus allowing the study of the CH3OH oxidation activity
over a broad potential range, and most importantly, including lower
potential ranges that are of interest to real fuel-cell applications.
Activation enthalpy values for the CH3OH oxidation reaction
were extracted for PtRu alloys and Ru powder electrodes over a
temperature range of 20-60°C. It was found that freshly prepared Ru
powders also exhibit an activity for the CH3OH oxidation reaction
even at low temperatures such as 20°C. However, the CH3OH oxi-
dation activity for the Ru powder is clearly smaller than for PtRu
alloys and decays with the age of the Ru powder, eventually becom-
ing completely inactive. The activation enthalpy values extracted at
0.68 V were essentially the same for the Ru and the PtRu alloy
powders. This indicates that the optimal PtRu alloy composition for
the CH3OH oxidation does not change with temperature. This con-
clusion is further supported by CH3OH oxidation activities obtained
from current-time transients at 20 and 60°C for PtRu alloys of 70:30
and 54:46 atom % ratio and is consistent with actual raw data
reported in the literature.3
Table IV. Average CH3OH oxidation current densities at 20 and
60°C estimated at 0.4 V vs. RHE.a
JCH
JCH
OH
3
OH
JCH
JCH
͑60°C͒
OH
OH
3
Pt:Ru ratio
atom %
3
Ratio:
at 20°C
at 60°C
b
20°C͒
͑
͑A cmϪ2
͒
͑A cmϪ2
͒
b
3
6.4 ϫ 10Ϫ6
3 ϫ 10Ϫ6
5.8 ϫ 10Ϫ5
2.6 ϫ 10Ϫ5
70:30
54:46
9
8.7
a The CH3OH activities are estimated from pseudo-steady state i-t tran-
sients recorded in 0.5 M CH3OH ϩ 0.5 M H2SO4 solution, as de-
scribed in the text;
b Current density values are normalized for total Pt ϩ Ru surface area.
JCH OH͑exp .70.30
atom %͒
3
*
Pt ϩ Ru csu %
JCH OH͑theoretical͒
ϭ
3
92%
͓8͔
Equation 8 uses the experimental CH3OH oxidation current
for the 70:30 atom % alloy (JCH OH(exp .70:30 atom %), namely, 4
3
ϫ 10Ϫ7 and 6.5 ϫ 10Ϫ6 A per cmϪ2 Pt ϩ Ru area, at 0.3 and 0.4
V, respectively, and the theoretical Pt ϩ Ru csu % value of 92% for
the 70:30 atom % PtRu alloy to calculate the theoretical CH3OH
current densities shown as squares in Fig. 10a and b. It is seen that
the theoretical and experimental CH3OH oxidation current densities
per total Pt ϩ Ru surface area are essentially the same ͑with a few
exceptions that may be due to experimental factors͒, thus supporting
the rather simple model used here. These results also suggest that
CH3OH oxidation current arises mainly from CH3OH molecules
that are adsorbed and oxidized on assemblies consisting of three
neighboring Pt and one Ru site, and that one Ru and three Pt sites
are involved in the oxidation of a CH3OH molecule within the po-
tential range studied here. Table III also shows the calculated Pt and
Ru csu % numbers, i.e., the percentage of Pt sites utilized per Pt
sites on the catalyst surface and the percentage of Ru sites utilized
per Ru sites on the catalyst surface, respectively. The data suggest
that a significant number of Pt surface sites is not involved in the
CH3OH to CO2 oxidation reaction for the PtRu alloys of lower Ru
composition ͑Ͻ20 atom % Ru͒, while the Ru sites are entirely uti-
lized. For the PtRu alloys of 30 atom % and higher values, the
utilization of both the Pt and Ru sites decreases with increasing Ru
content. However, all the theoretical CH3OH current density and
csu % values calculated in this work reflect trends and the true val-
ues may be different, as only a simple model has been used that
ignores a number of possibly relevant factors, as mentioned previ-
ously.
CH3OH oxidation activities were also obtained as a function of
Ru content for the PtRu alloy powders using many PtRu alloy pow-
ders of different Pt:Ru ratios at 0.3 and 0.4 V. For all tested poten-
tials, the 70:30 atom % PtRu alloy was found to yield the highest
CH3OH oxidation activity. Theoretical CH3OH oxidation activities
were calculated as a function of the Pt:Ru ratio using a simple model
that assumes that the CH3OH to CO2 oxidation reaction requires
three Pt sites and one neighboring Ru site. The theoretical and ex-
perimental data were found to show the same behavior, thus sup-
porting the model.
Acknowlegments
Table IV shows CH3OH oxidation currents obtained at 20 and
60°C for PtRu alloys of 70:30 and 54:46 atom % PtRu alloy pow-
ders. It is seen that the CH3OH oxidation activity is higher for the
70:30 atom % than for the 54:46 atom % alloy independent of tem-
perature. In fact, the CH3OH oxidation current increases for both
catalysts by a factor of ca. 9 with an increase in temperature from 20
to 60°C. This increase is consistent with literature data3 and the
activation enthalpy values reported in this work.
The authors thank P. L’Abbe ͑National Research Council
Canada͒ for making the electrochemical cells used in this work
and G. Pleizier ͑National Research Council Canada͒ for the XPS
analyses. Helpful discussions with Dr. E. Gileadi ͑Tel-Aviv Univer-
sity, Israel͒ are also greatly appreciated.
The National Research Council of Canada assisted in meeting the
publication costs of this article.
References
Conclusions
1. J. O’M. Bockris and H. Wroblowa, J. Electroanal. Chem. Interfacial Electrochem.,
7, 428 ͑1964͒.
2. M. Watanabe and S. Motoo, J. Electroanal. Chem. Interfacial Electrochem., 60,
267 ͑1975͒.
3. H. A. Gasteiger, N. Markovic, P. N. Ross, and E. J. Cairns, J. Electrochem. Soc.,
147, 1795 ͑1994͒.
4. A. Kabbabi, R. Faure, R. Durand, B. Beden, F. Hahn, J.-M. Leger, and C. Lamy, J.
Electroanal. Chem., 444, 41 ͑1998͒.
5. T. Iwasita, H. Hoster, A. John-Anacker, W. F. Lin, and W. Vielstich, Langmuir, 16,
522 ͑2000͒.
6. D. Chu and S. Gilman, J. Electrochem. Soc., 143, 1685 ͑1996͒.
7. H. A. Gasteiger, N. Markovic, P. N. Ross, and E. J. Cairns, Jr., J. Phys. Chem., 98,
617 ͑1994͒.
A wide range of PtRu powders of Pt:Ru ratio up to 54:46 atom %
have been synthesized using a simple chemical reduction method
involving low temperatures. Pt and Ru powder catalysts were also
synthesized using the same method. Lattice parameter estimation
for the fcc Pt shows a linear decrease in the lattice parameter value
with increasing Ru content. This is in agreement with Vegard’s
law and indicates the formation of PtRu alloys. The relationship of
the fcc Pt lattice parameter on the nominal atom % of Ru was found
to be essentially the same as predicted by ab initio calculations13
and previously reported data for bulk PtRu alloys.3 XPS studies
Downloaded on 2014-05-31 to IP 128.148.252.35 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

A1278
Journal of The Electrochemical Society, 151 ͑8͒ A1269-A1278 ͑2004͒
8. S. F. Huang, Y. Chi, C. S. Liu, A. J. Carty, K. Mast, C. Bock, B. MacDougall, S. M.
Peng, and G. H. Lee, Chem. Vap. Deposition, 9, 151 ͑2003͒.
9. B. Gurau, R. Viswanathan, R. Liu, T. J. Lafrenz, K. L. Ley, E. S. Smotkin, E.
Reddington, A. Sapienza, B. C. Chan, T. E. Mallouk, and S. Sarangapani, J. Phys.
Chem. B, 102, 9997 ͑1998͒.
10. A. J. Bard and L. R. Faulkner, Electrochemical Methods, Fundamentals and Ap-
plications, John Wiley & Sons, Inc., New York ͑1980͒.
11. X-Ray diffraction database, Diffraction Management System for Windows NT,
Scintag, Inc., Cupertino, CA ͑1997͒.
17. G. M. Bancroft, I. Adams, L. L. Coatsworth, C. D. Bennewitz, J. D. Brown, and
W. D. Westwood, Anal. Chem., 47, 586 ͑1975͒.
18. C. D. Wagner, Handbook of X-Ray Photoelectron Spectroscopy, Perkin-Elmer
Corporation, Physical Electronics Division, Eden Prairie, MN ͑1979͒.
19. A. S. Arico, P. Creti, P. L. Antonucci, J. Cho, H. Kim, and V. Antonucci, Electro-
chim. Acta, 43, 3719 ͑1998͒.
20. H. N. Dinh, X. Ren, F. H. Garzon, P. Zelenay, and S. Gottesfeld, J. Electroanal.
Chem., 491, 222 ͑2000͒.
12. B. E. Warren, X-Ray Diffraction, Dover Publications, New York ͑1990͒.
13. Y. Le Page, Unpublished results.
14. M. T. M. Koper, T. E. Shubina, and R. A. van Santen, J. Phys. Chem. B, 106, 686
͑2002͒.
15. Q. Ge, S. Desai, M. Neurock, and K. Kourtakis, J. Phys. Chem. B, 105, 9533
͑2001͒.
16. E. Antolini, L. Giorgi, F. Cardellini, and E. Passalacqua, J. Solid State Electro-
chem., 5, 131 ͑2001͒.
21. C. Bock and B. MacDougall, J. Electrochem. Soc., 150, E377 ͑2003͒.
22. P. Ross, in Electrocatalysis, J. Lipkoswki and P. R. Ross, Editors, Wiley-VCH, Inc.,
New York ͑1998͒.
23. E. Gileadi, Electrode Kinetics for Chemists, Chemical Engineers, and Material
Scientists, VCH Publishers, Inc., New York ͑1993͒.
24. H. Kuchling, Physik, Formeln und Gesetze, Buch- and Zeit-Verlagsgesellschaft
mbH Koeln ͑1982͒.
Downloaded on 2014-05-31 to IP 128.148.252.35 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).