Novel Pt-Ru nanoparticles formed by vapour deposition as efficient electrocatalyst for methanol oxidation: Part I. Preparation and physical characterization

Article abstract of DOI:10.1016/j.electacta.2004.12.005

Bimetallic Pt-Ru nanoparticles supported on carbon substrates have been prepared reproducibly by a simple method that utilizes commercially available metal-organic precursors at low temperature in vacuum. Particles morphology, composition and structure have been investigated using HRTEM, EDX, selected area electron diffraction (SAED) and powder XRD analysis. TEM shows that the obtained nanoparticles are homogeneously dispersed on the substrate surface and exhibit narrow size distribution, the average diameter being ca. 2 nm. Point resolved EDX analysis demonstrates co-presence of both Pt and Ru in each particle, thereby indicating that truly bimetallic nanoparticles have been obtained. Moreover, EDX performed on several areas of the sample evidences uniform particles composition. The latter can be controlled very easily and effectively by regulating the operation temperature during particles preparation. HRTEM imaging shows that the particles possess crystalline structure. Both SAED and XRD analyses indicate presence of nanoparticles exhibiting structure consistent with that of an f.c.c. Pt-Ru alloy. Besides the f.c.c. alloy, an additional crystalline phase might also be present as noticed by SAED. These nanoparticles display electrocatalytic activity with regard to methanol oxidation as evidenced by cyclic voltammetry (CV).

Full text of DOI:10.1016/j.electacta.2004.12.005

Electrochimica Acta 50 (2005) 3312–3319
Novel Pt–Ru nanoparticles formed by vapour deposition as
efficient electrocatalyst for methanol oxidation
Part I. Preparation and physical characterization
Pasupathi Sivakumar, Randa Ishak, Vincenzo Tricoli∗
Dipartimento di Ingegneria Chimica, Chimica Industriale e Scienza dei Materiali, Universita` di Pisa, 56126 Pisa, Italy
Received 30 July 2004; received in revised form 24 November 2004; accepted 4 December 2004
Available online 23 January 2005
Abstract
Bimetallic Pt–Ru nanoparticles supported on carbon substrates have been prepared reproducibly by a simple method that utilizes commer-
cially available metal-organic precursors at low temperature in vacuum. Particles morphology, composition and structure have been investigated
using HRTEM, EDX, selected area electron diffraction (SAED) and powder XRD analysis. TEM shows that the obtained nanoparticles are
homogeneously dispersed on the substrate surface and exhibit narrow size distribution, the average diameter being ca. 2 nm. Point resolved
EDX analysis demonstrates co-presence of both Pt and Ru in each particle, thereby indicating that truly bimetallic nanoparticles have been
obtained. Moreover, EDX performed on several areas of the sample evidences uniform particles composition. The latter can be controlled
very easily and effectively by regulating the operation temperature during particles preparation. HRTEM imaging shows that the particles
possess crystalline structure. Both SAED and XRD analyses indicate presence of nanoparticles exhibiting structure consistent with that of
an f.c.c. Pt–Ru alloy. Besides the f.c.c. alloy, an additional crystalline phase might also be present as noticed by SAED. These nanoparticles
display electrocatalytic activity with regard to methanol oxidation as evidenced by cyclic voltammetry (CV).
© 2004 Elsevier Ltd. All rights reserved.
Keywords: Pt–Ru nanoparticles; Vapour deposition; Electrocatalyst; Fuel cell; Methanol
1. Introduction
to Pt without Ru because of their markedly improved resis-
tance to carbon monoxide (CO) poisoning [7–9]. Much of
the current understanding of the mechanisms involved in ox-
idation of methanol or hydrogen in the presence of CO traces
is based on studies of Pt–Ru bulk alloy electrodes [10–12].
However, an adequate understanding of the compositional
and size dependencies of the electrocatalytic reaction rates
over supported nanometer-sized generally bimetallic parti-
cles remains to be realized, mostly owing to difficulties in
the preparation of such small particles with well defined size,
structure and composition [13].
The aim of this work is to accomplish a method to prepare
bimetallic Pt–Ru nanoparticles. We require that our method
be simple, reproducible and versatile. We require that our
nanoparticles fulfill the following conditions: (1) the parti-
cles should exhibit small size (of the order of 1 nm) with as
narrow as possible size distribution; (2) both metals should
Metallic and alloyed polymetallic nanoparticles with par-
ticles size approaching molecular dimensions are of great
fundamental and technological interest due to their catalytic
properties. The literature concerning preparation or studies
of these nanoscopic catalysts has become rather voluminous
especially during the past few years and a number of reviews
of this research topic have appeared [1–4]. Recently, increas-
ing interest in low temperature fuel cells has driven attention
to alloyed Pt–Ru nanoparticles as anode catalysts for hydro-
gen or methanol electrooxidation [5–7]. Pt–Ru alloys display
superior activity as fuel cells anode catalyst in comparison
∗
Corresponding author. Tel.: +39 050511246; fax: +39 050511266.
E-mail address: v.tricoli@ing.unipi.it (V. Tricoli).
0013-4686/$ – see front matter © 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2004.12.005

P. Sivakumar et al. / Electrochimica Acta 50 (2005) 3312–3319
3313
be alloyed in the particles with uniform “bulk” composition
2. Experimental
over the support surface; (3) the particles should be well dis-
persed on the substrate (should not form agglomerates); (4)
they should be dispersed on a substrate that facilitates their
characterization.
2.1. Materials and preparation methods
Platinum(II) acetylacetonate (Pt(acac)2) and ruthe-
nium(III) acetylacetonate (Ru(acac)3) were purchased from
Alfa Aesar GmbH, Germany. Hydrogen and nitrogen gases
(grade 4.5 and 4.6, respectively) were supplied by SOL SpA,
Italy. Carbon coated nickel TEM grids (EMS, Fort Wash-
ington, USA) or Vulcan® XC72R (Cabot) carbon powder,
were used as the nanoparticles substrate. Suitable amounts
of Pt(acac)2 and Ru(acac)3 were placed at the bottom of
a home-made vacuum flask along with the substrate. Di-
rect contact of the precursors with the substrate was care-
fully avoided by placing the carbon coated grids on top of a
clean graphite tablet (2 mm thickness). When carbon pow-
der was used as the substrate, the powder was placed in
a small watch-glass. The flask was evacuated (10−2 mbar),
sealed and transferred into a oven where it was heated at a
fixed temperature within the range 170–240 ◦C for half-an-
hour. During this period, the two precursors sublime with-
out decomposing and their vapours fill all empty space in-
side the flask. The latter was then taken out of the oven,
set to cool at room temperature and subsequently air vented.
During cooling stage, precursor vapours become solid and
deposit onto all the exposed surfaces inside the flask. The
precursor impregnated substrate was placed inside a quartz
tube and transferred into a tubular furnace, where it was
heated at 320 ◦C for 3 h under moderate hydrogen or ni-
trogen flow. The latter operation served to decompose the
metal-organic precursor molecules deposited on the substrate
surface (carbon coated TEM grids were found to be unaf-
fected).
We believe that Pt–Ru nanoparticles exhibiting the above
characteristics can be successfully optimized for use as effi-
cientelectrocatalystinmethanolfuelcells. Themostcommon
technique utilized in previous studies for the preparation of
Pt–Ru nanoparticles is reduction by gas phase hydrogen of
metal salt particles disposed on a support by impregnation
with the salt solution [14,15]. A variant of this procedure uti-
lizes hydrazine [16] or NaBH4 [17] to reduce salt in the liquid
phase. In our laboratory, such method allowed limited ability
to control the distribution of the particles size. Other methods
involve synthesis of colloidal precursors. Then, the colloidal
particles are collected from a suspension onto a surface and
subsequently decomposed to give Pt–Ru nanoparticles. In
some cases, these methods were reported to give good par-
ticle size monodispersity [18–20]. In a different approach,
a bimetallic molecular cluster precursor (Pt/Ru = 1:5) has
been prepared and disposed on a support from a tetrahydro-
furan solution. Upon decomposition, the precursor yielded
well defined Pt–Ru nanoparticles with fixed composition
[13,21]. Later on, preparation of supported Pt–Ru nanopar-
ticles from acetone- or CH2Cl2-dissolved bimetallic precur-
sors with 1:1 and 3:1 Pt/Ru atomic ratio was also reported
[22,23]. Miller and Dunn have utilized ruthenium(III) acety-
lacetonate as a metal-organic precursor to obtain nanometer
size monometallic Ru particles distributed homogeneously
throughout the porous structure of carbon aerogel [24]. The
precursor was first sublimed at moderate temperature and the
resulting vapours were let to impregnate the porous carbon.
After subsequent cooling, the precursor was decomposed at
higher temperature.
2.2. Physical characterization
The approach we propose in this work to accomplish
Pt–Ru nanoparticles having the above noted characteris-
tics is based on a suitable modification of the proce-
dure employed by Miller and Dunn. Our approach in-
volves simple operations and utilizes metal-organic precur-
sors that are commercially available. The obtained parti-
cles were characterized with respect to their morphology
using transmission electron microscopy (TEM). Particles
structure was assessed by high resolution TEM (HRTEM),
selected area electron diffraction (SAED) and powder X-
ray diffraction (XRD). Particles composition was inves-
tigated by point-resolved energy dispersive X-ray (EDX)
analysis. We also explored preliminarly in the present
work the electrocatalytic activity of our Pt–Ru nanopar-
ticles toward methanol oxidation using cyclic voltam-
metry. In the accompained paper, part II [25], we ex-
ploit the potential of this method to prepare supported
Pt–Ru nanoparticles with optimal electrocatalytic activity
for methanol oxidation. The latter is investigated exten-
sively in that work also in comparison with state of the art
catalysts.
After preparation all samples were preserved under moist
free environment before characterization. Particles size and
dispersion as well as particles nanostructure were investi-
gated using TEM, HRTEM and SAED analysis performed
on a JEOL 2010 microscope operating at 200 kV. The in-
strument was also equipped with an EDX analyzer (OX-
FORD Pentafet). Point resolved EDX analysis was carried
out on individual nanoparticles to ascertain co-presence of
Pt and Ru. However, single particle EDX analysis turned
out not to be satisfactory to quantify composition accurately,
because the output signal was not sufficiently strong. Reli-
able and reproducible average composition determinations
were achieved by carrying out EDX analysis on sample ar-
eas exposing an adequate number of nanoparticles. Overall
metal content (Pt + Ru) in Vulcan® XC72R was determined
by “bulk” EDX analysis. Mass gain determinations of the car-
bon black support after loading the metal precursors as well
as after thermal decomposition were also carried out. Pow-
der samples of Pt–Ru nanoparticles supported on Vulcan®
XC72R carbon black were also investigated by XRD analysis

3314
P. Sivakumar et al. / Electrochimica Acta 50 (2005) 3312–3319
performed on a Siemens instrument model 810 utilizing Cu
Kα radiation.
2.3. Electrocatalytic activity
Cyclic voltammetry (CV) in 0.5N H2SO4/1 M CH3OH
solution was performed to explore electrocatalytic activity
of our Pt–Ru nanoparticles towards methanol electrooxida-
tion. To this purpose Pt–Ru nanoparticles were prepared on
Vulcan® XC72R support at 170, 190, 220 and 240 ◦C. A fixed
amount of Pt–Ru/Vulcan® catalyst prepared at the various
temperatures was anchored onto a glassy carbon electrode.
The experimental procedure is described in detail in the ac-
compained paper, part II [25].
3. Results and discussion
3.1. Particles morphology
In Fig. 1, TEM micrographs are reported of nanoparticles
prepared on carbon coated nickel grids at various tempera-
tures between 170 and 240 ◦C. The images are representative
of the entire surface of the respective samples and show ho-
mogeneousdistributionofthemetalparticlesonthesubstrate.
The micrographs seem to indicate a higher particle coverage
on the support surface at increasing preparation temperature.
They also indicate that the preparation temperature has no
notable effect on particles size. Also shown in Fig. 1 is a
TEM image of nanoparticles prepared at 220 ◦C on Vulcan®
XC72R carbon black substrate. In this case, the apparent sur-
face coverage seems higher than for the carbon coated grid
substrate. That probably reflects the larger effective surface
area of the carbon black support. In all cases, it can be noted
that the particles size is in the nanometer range. Fig. 2 reports
the histograms of particles size distribution for samples pre-
pared at temperatures from 170 to 240 ◦C. Size measurements
of about 250 particles were collected for each sample. At all
temperatures, the histograms show a rather narrow particle
size distribution with the average diameter being ca. 2 nm.
With concern to preparation procedure, it should be pointed
out that, after sublimation, the precursors molecules are de-
posited as solid particles onto the carbon substrate by letting
the vacuum flask cool to room temperature spontaneously.
We have not investigated the influence of the cooling rate on
the size of the deposited precursor particles, i.e. on the ul-
timate Pt–Ru particles size. However, we deem that cooling
rate may prove a relevant parameter that can be regulated
to control particles size. We also prepared supported single
metal Pt or Ru nanoparticles by depositing and subsequently
decomposing vapours of single Pt(acac)2 or Ru(acac)3 pre-
cursor molecules (sublimation temperature 220 ◦C). Again,
TEM micrographs (not shown) of these single metal parti-
cles evidenced uniform coverage of substrate surface with
narrow particles size distribution and ca. 2 nm average di-
ameter. Finally, it is worth mentioning that no difference was
Fig. 1. TEM micrographs of Pt–Ru particles prepared on carbon coated
TEM grids at temperatures: (a) 170 ◦C; (b) 190 ◦C; (c) 220 ◦C; (d) 240 ◦C;
(e) prepared on Vulcan® XC72R carbon black substrate at T = 220 ◦C.
noted between nanoparticles prepared upon decomposing the
precursors in hydrogen or in nitrogen atmosphere, which in-
dicates 320 ◦C as an adequate temperature to decompose both
Pt(acac)2 and Ru(acac)3. This is also in agreement with ear-
lier studies concerning thermal stability of metal acetylacet-
onates [26,27].
3.2. Particles composition
Point resolved EDX analysis with spot size of 10 nm per-
formed on a single nanoparticle showed co-presence of both
platinum and ruthenium. The analysis was carried out on
several single particles at different zones of the substrate: no

P. Sivakumar et al. / Electrochimica Acta 50 (2005) 3312–3319
3315
large number of nanoparticles on the substrate. In fact, EDX
performed on a E-TEK (Somerset, USA) commercial cata-
lyst (Pt/Ru nanoparticles supported on carbon black with 1:1
Pt/Ru nominal atomic ratio) gave 50.2 at.% Pt and 49.8 at.%
Ru, thereby confirming reliability of the method. This anal-
ysis was performed on our samples with the particles being
supported on the Vulcan® XC72R substrate. As an example,
we report in Fig. 3 the EDX spectrum for a sample prepared
at 220 ◦C. Besides carbon, in addition to presence of Pt and
Ru, trace amounts of sulphur (<0.5 wt.%) are noticed. For
samples prepared at 170 ◦C, the average bulk particles com-
position was found to be 68:32 Pt/Ru atomic ratio. We found
that the percentage of ruthenium increases on increasing the
temperature at which the precursors are sublimed, reaching a
maximumat220 ◦C(ca. 50:50Pt/Ruatomicratio). Increasing
the temperature further does not alter particles composition.
Table 1 shows the composition of the particles prepared at
different temperatures, the value is an average of four to five
analysis on different zones of the sample with the difference
between each pair of values being around 1–2%. Values ob-
tained on different samples prepared according to the same
protocol were also highly reproducible. To understand the
effect of sublimation temperature on Pt–Ru particles compo-
sition we have examined the temperature dependence of the
vapour pressure at saturation for the two precursors: Pt(acac)2
and Ru(acac)3. To evaluate the volatility of such compounds
Fig. 2. Particle size distributions for nanoparticle prepared at temperatures:
(a) 170 ◦C; (b) 190 ◦C; (c) 220 ◦C; (d) 240 ◦C.
single metal particle was found; all the particles displayed co-
presence of the two metals. This indicates that truly bimetallic
Pt–Ru nanoparticles have been obtained. EDX analysis, al-
though unsuitable for accurate quantitative determination of
Pt/Ru ratio in single nanoparticle, proved effective and reli-
able to determine bulk average composition of a sufficiently
Fig. 3. EDX spectrum for a Pt–Ru/Vulcan® prepared at 220 ◦C.

3316
P. Sivakumar et al. / Electrochimica Acta 50 (2005) 3312–3319
Table 1
Particles composition vs. preparation temperature as determined by EDX analysis
Sublimation
temperature (◦C)
Composition/Pt:Ru
atomic ratio
Pt + Ru content
(by EDX, wt.%)
Mass gain after
deposition (wt.%)
Mass gain after
decomposition (wt.%)
170
190
220
240
68:32
61:39
50:50
52:48
2.5
7.3
8.8
11
7.1
19.3
24
3
9.1
11
Not measured
Not measured
at the temperatures of interest we have turned to the correla-
tion proposed by Morozova et al.
determined by EDX is similar to the mass gain value after
decomposition.
P = exp(α − β/T )
3.3. Particles structure
where P is the vapour pressure at saturation and T the ab-
solute temperature. α and β are correlation parameters that
were determined experimentally [26]. In Fig. 4, the parti-
cles composition (Pt/Ru atomic ratio) is plotted as a func-
tion of temperature along with relative volatility of the two
precursors (i.e. the ratio between their saturated vapour pres-
sures: PPOt/PROu). It can be noted that the two curves overlap
each other remarkably well. This confirms our expectation
that metal particles composition essentially mirrors the com-
position of the bulk mixed vapour phase constitued by the
precursor molecules.
Also reported in Table 1 is overall metal loading (Pt + Ru)
on the Vulcan® substrate at the various preparation tempera-
tures as determined by “bulk” EDX analysis along with mass
gain of the substrate upon precursor loading before as well
as after decomposition. Each EDX value reported is an aver-
age of six to seven data with the maximum difference being
around 20%. The mass gain values measured before and
after thermal decomposition treatment are consistent with
each other at all temperatures. In fact, the mass gain de-
crease measured upon thermally decomposing the deposited
Pt(acac)2 and Ru(acac)3 well enough corresponds to the cal-
culated mass loss due to liberation of the organic fragments.
It is furthermore noted in Table 1 that the Pt + Ru content as
3.3.1. Single metal Pt and Ru particles
Fig. 5a and b shows the HRTEM images of two selected
nanoparticles produced, respectively, from sole Pt(acac)2 and
sole Ru(acac)3 precursor. The presence of crystalline planes
is clearly noted in both cases. Further evidence of parti-
cle crystallinity was obtained by electron diffraction per-
formed on selected areas of each sample. Fig. 6a and b refers
to the SAED of nanoparticles obtained, respectively, from
sole Pt(acac)2 and sole Ru(acac)3 precursor. The character-
istic ring diffraction patterns, which are typical of crystalline
nanoparticles, are observed, although the rings are somewhat
diffuse. Rings 1a, 2a, 3a and 4a in Fig. 6a correspond to d-
˚
spacing of 2.26, 1.95, 1.35 and 1.15 A, respectively. These
values can unambiguously be assigned respectively to the
(1 1 1), (2 0 0), (2 2 0) and (3 1 1) planes of the face centered
cubic (f.c.c.) structure of metal Pt. We conclude that Pt(acac)2
precursor deposited after sublimation onto the substrate sur-
face is fully decomposed upon heating at 320 ◦C under ni-
trogen flow to yield metal Pt nanoparticles. We now turn our
Fig. 4. Particles composition as a function of preparation temperature. Also
shownistherelativevolatilityofthetwometal-organicprecursors(PPOt/PROu).
Fig. 5. HRTEMimagesof(a)Pt, (b)Ruand(c)Pt–Runanoparticlesprepared
at 220 ◦C on carbon coated TEM grids.

P. Sivakumar et al. / Electrochimica Acta 50 (2005) 3312–3319
3317
Fig. 7. Powder XRD patterns: (a) Pt–Ru nanoparticles prepared at 220 ◦C on
Vulcan® XC72R carbon black substrate; (b) commercial E-TEK nanoparti-
cle catalyst Pt–Ru/Vulcan® (20 wt.% metal content, 50:50 Pt:Ru).
lattice planes of a bimetallic Pt–Ru nanoparticle for a sample
prepared at 220 ◦C. Similar images were generally observed
for almost all particles disposed all over the sample surface.
Few particles not displaying structural planes recognizable at
HRTEM inspection were also encountered. Absence of lat-
tice fringes for such particles may be ascribed to either lack of
crystalline structure or, most likely, unfavourable orientation
of the particle. We performed SAED diffraction analysis to
investigate structure of the crystalline Pt–Ru nanoparticles.
The SAED pattern for these particles is shown in Fig. 6c.
Although the pattern is rather weak, presence can be noted of
several rings (or, in some cases, just a few spots) which cor-
respond to as many lattice planes reflections. Among those
we distinguish rings 1c, 2c, 3c and 4c which correspond to
Fig. 6. SAED patterns of (a) Pt, (b) Ru and (c) Pt–Ru nanoparticles prepared
at 220 ◦C on Vulcan® XC72R carbon black substrate.
consideration to the diffraction pattern of Fig. 6b. Diffrac-
tion rings 3b, 5b, 6b, 7b and 8b correspond respectively to
1
˚
d-spacing of 1.99, 1.34, 1.21, 1.13 and 0.84 A, respectively.
Such values in turn fit the reflections of the (1 0 1), (1 1 0),
(1 0 3), (2 0 1) and (1 1 4) lattice planes of hexagonal close-
packed (h.c.p.) metal Ru. Rings 1b and 2b correspond to d-
˚
d-spacing of 2.30, 1.99, 1.39 and 1.19 A, respectively. These
˚
spacing of 3.18 and 2.54 A, respectively, and cannot be as-
values fit well the (1 1 1), (2 0 0), (2 2 0) and (3 1 1) lattice
planes of the f.c.c. platinum structure. However, the pattern
displays several additional reflections that cannot be assigned
to f.c.c. platinum or h.c.p. ruthenium. Nor do such reflections
fit the structure of crystalline PtO2 or RuO2. Further evidence
of Pt–Ru nanoparticles exhibiting f.c.c. structure comes from
powder XRD analysis. Fig. 7a shows the XRD pattern of our
Pt–Ru particles supported on Vulcan® XC72R carbon black.
This is analogous to the pattern of a commercial E-TEK cat-
alyst made out of Pt–Ru nanoparticles supported on Vulcan®
XC72R with 20 wt.% metal and 50:50 at.% Pt:Ru (Fig. 7b).
We note two peaks at 2θ = 25◦ and 40◦ and also two less ev-
signed to the structure of h.c.p. Ru. However, they are in good
conformity with the reflections of the (1 1 0) and (1 0 1) lat-
tice planes of tetragonal ruthenium oxide RuO2. Finally, ring
˚
4b (d-spacing 1.66 A) could be assigned to either the (1 0 2)
lattice plane of metal Ru or, more likely, the (2 1 1) reflec-
tion of RuO2. These observations indicate that Ru(acac)3 is
in fact decomposed fully into metal Ru nanoparticles under
the above noted conditions. Such particles are then partially
oxidized to RuO2 upon exposure to air at room temperature.
On the other hand, Ru clusters are known to be sensitive to
oxygen even at mild conditions and to be oxidized easily in
air [28,29].
◦
◦
∼
ident reflections at 2θ 69 and 82 . Using the Bragg’s law,
=
we find the first reflection angle corresponding to d-spacing
of 3.57 A, which is associated with the Vulcan XC72R sub-
®
˚
3.3.2. Bimetallic Pt–Ru particles
Metal nanoparticles produced from mixed Pt(acac)2/
Ru(acac)3 precursors were also examined by HRTEM. The
image shown in Fig. 5c evidences fringes representing the
strate [30]. The remaining three peaks correspond, respec-
tively, to d-spacing of 2.25, 1.37 and 1.17 A. In turn, these
values are consistent with the (1 1 1), (2 2 0) and (3 1 1) re-
flections of f.c.c. platinum structure. In summary, HRTEM
imaging revealed that nanoparticles obtained from mixed
Pt(acac)2/Ru(acac)3 precursors generally possess crystalline
˚
1
Although rings 6b and 8b are hardly visible in Fig. 5b, they were clearly
seen in the original and indexed.

3318
P. Sivakumar et al. / Electrochimica Acta 50 (2005) 3312–3319
4. Concluding remarks
The present work has demonstrated a method to prepare
reproducibly bimetallic Pt–Ru nanoparticles supported on
carbon substrates. The preparation method involves simple
operations and utilizes metal-organic Pt- and Ru-precursors
that are commercially available and prepared from low cost
organic ligands [26]. The obtained nanoparticles are homo-
geneously dispersed on the substrate surface and exhibit aver-
age diameter of ca. 2 nm with narrow size distribution. Point-
resolved EDX analysis has evidenced co-presence of both
Pt and Ru in each particle. HRTEM imaging has shown that
these particles possess crystalline structure. SAED and pow-
der XRD analyses have demonstrated the particles to be truly
bimetallic with Pt and Ru atoms alloyed in an f.c.c. structure,
although presence of another crystalline phase (not identified
in this work) has also been noticed. Particles composition has
been found to be quite uniform over the surface of the sub-
strate. Importantly, this preparation method affords to control
particles composition simply by changing the temperature at
which precursors sublimation is carried out: we have been
able to control particles composition in the range 30–50 at.%
Ru.
Fig. 8. CV in 0.5N H2SO4/1 M CH3OH solution at 25 ◦C of supported
Pt–Ru/Vulcan® nanoparticles prepared at 190, 220 and 240 ◦C. The base
voltammogram was recorded for the 190 ◦C nanoparticles in sole 0.5N
H2SO4 solution. Shown current values refer to unit geometric area of the
electrode.
structure. Both SAED and powder XRD analyses show that
the particles structure is consistent with the f.c.c. structure of
metal platinum. On the other hand, EDX investigation proved
co-presence of both Pt and Ru in each particle. Thus, these
observations suggest that Pt and Ru form a solid solution in
such nanoparticles with Ru atoms replacing Pt atoms in an
f.c.c. structure. SAED analysis also evidences clearly pres-
ence of an additional crystalline phase whose lattice reflec-
tions are not ascribable to f.c.c. platinum or to h.c.p. ruthe-
nium. Nor do these reflections fit the structure of crystalline
PtO2 or RuO2. We suspect that such phase be made out of
mixedPt/Ruoxide. However, furtherinvestigationwoulddef-
initely be required for an unambiguous understanding of this
point.
It is useful to remark that the described preparation method
allows to increase easily nanoparticles loading on the car-
bon support by repeating the precursors deposition and de-
composition operations. Moreover, the method seems readily
amenable to preparation of tri- or polymetallic nanoparticles
as well.
We believe that carbon supported Pt–Ru nanoparticles like
those prepared in this work can be of considerable interest as
anode catalysts in low temperature fuel cells, owing to their
morphological, structural and compositional characteristics.
In fact, presence of Pt in the particle makes it active toward
H H or C H bond breaking; presence of Ru makes the cat-
alyst particle resistant to CO poisoning. Most importantly,
the very small particles size translates into a large surface
area, i.e. high reaction rate at the anode surface per unit cat-
alyst mass. Indeed, an exploratory CV study demonstrates
that the Pt–Ru nanoparticles prepared in this work do exhibit
clear catalytic activity toward methanol electrooxidation. An
extensive electrochemical investigation of supported Pt–Ru
nanoparticles with optimal electrocatalytic activity, as com-
pared to state of the art catalysts, is presented in the accom-
pained paper, part II [25]. Preparation and electrocatalytic
activity of ternary metal nanoparticles are also under study.
3.4. Electrocatalytic activity
Fig. 8 shows the voltammograms at 25 ◦C of our Pt–Ru
nanocatalysts prepared at three different temperatures. The
base voltammogram, in the absence of methanol, is also
reported as the reference. A clear onset of the anodic cur-
rent at 400 mV versus SHE, corresponding to methanol elec-
trooxidation, is noted for the catalysts prepared at 220 and
240 ◦C. The current is higher for the catalyst prepared at
240 ◦C, whereas low current is observed for the one prepared
at 190 ◦C. Similarly small anodic current was displayed by
the catalyst prepared at 170 ◦C (voltammogram not shown).
Thus, the higher is the preparation temperature of the cata-
lyst the greater is the current. This is largely due to a greater
amount of active phase (Pt–Ru) deposited on the substrate
at higher preparation temperature, as also indicated by TEM.
Another reason is probably ascribable to catalyst composition
(Pt/Ru ratio), which varies as the preparation temperature is
changed (Table 1).
Acknowledgements
We thank Prof. M. De Sanctis of the University of Pisa
for allowing use of TEM and for valuable discussions. We
are also grateful to Mr. C. Uliana of the University of Gen-
ova for his help with TEM and EDX. Mr. P. Narducci of the
University of Pisa is also gratefully acknowledged for addi-
tional support with EDX. Vulcan® XC72R carbon powder

P. Sivakumar et al. / Electrochimica Acta 50 (2005) 3312–3319
3319
was kindly provided by Carbocrom (Milano). This research
was financed by the Italian MURST.
[15] M. Watanabe, M. Uchida, S. Motoo, J. Electroanal. Chem. 229
(1987) 395.
[16] A.S. Arico`, P. Cret`ı, N. Giordano, V. Antonucci, P.L. Antonucci, A.
Chuvilin, J. Appl. Electrochem. 26 (1996) 959.
[17] K.L. Ley, R. Liu, C. Pu, Q. Fan, N. Leyarovska, C. Segre, E.S.
Smotkin, J. Electrochem. Soc. 144 (1997) 1543.
[18] T.J. Schmidt, M. Noeske, H.A. Gasteiger, R.J. Behm, P. Britz, W.
Brijoux, H. Bonnemann, Langmuir 13 (1997) 2591.
[19] C. Pan, F. Dassenoy, M.-J. Casanove, K. Philippot, C. Amiens, P.
Lecante, A. Mosset, B. Chaudret, J. Phys. Chem. B 103 (1999)
10098.
[20] T.J. Schmidt, M. Noeske, H.A. Gasteiger, R.J. Behm, P. Britz, H.
Bonnemann, J. Electrochem. Soc. 145 (1998) 925.
[21] C.W. Hills, M.S. Nashner, A.I. Frenkel, J.R. Shapley, R.G. Nuzzo,
Langmuir 15 (1999) 690.
[22] E.S. Steigerwalt, G.A. Deluga, D.E. Cliffel, C.M. Lukehart, J. Phys.
Chem. B 105 (2001) 8097.
[23] J.T. Moore, J.D. Corn, D. Chu, R. Jiang, D. Boxall, E.A. Kenik,
C.M. Lukehart, Chem. Mater. 15 (2003) 3320.
[24] J.M. Miller, B. Dunn, Langmuir 15 (1999) 799.
[25] P. Sivakumar, V. Tricoli, submitted for publication, 2004.
[26] N.B. Morozova, G.I. Zharkova, P.P. Semyannikov, S.V. Sysoev, I.K.
Igumenov, N.E. Fedotova, N.V. Gelfond, J. Phys. IV 11 (2001) Pr3-
609.
References
[1] A.T. Bell, Science 299 (2003) 1688.
[2] G.A. Somorjai, Y.G. Borodko, Catal. Lett. 76 (2001) 1.
[3] J.D. Aiken, R.G. Finke, J. Mol. Catal. A 145 (1999) 1.
[4] P.V. Kamat, D. Meisel, Curr. Opin. Colloid Interf. Sci. 7 (2002)
282.
[5] C.W. Hills, N.H. Mack, R.G. Nuzzo, J. Phys. Chem. B 107 (2003)
2636.
[6] V. Radmilovic, H.A. Gasteiger, P.N. Ross Jr., J. Catal. 154 (1995)
98.
[7] T.R. Ralph, M.P. Hogarth, Plat. Met. Rev. 46 (3) (2002)
117.
[8] J.M. Leger, C. Lamy, Ber. Bunsen-Ges. Phys. Chem. 94 (1990)
1021.
[9] F. Richarz, B. Wohlmann, U. Vogel, H. Hoffschulz, K. Wandelt,
Surf. Sci. 335 (1995) 361.
[10] H.A. Gasteiger, N. Markovic, P.N. Ross Jr., E.J. Cairns, J. Phys.
Chem. 97 (1993) 12020.
[11] A. Kabbabi, R. Faure, R. Durand, B. Beden, F. Hahn, J.M. Leger,
C. Lamy, J. Electroanal. Chem. 444 (1998) 41.
[12] H.A. Gasteiger, N. Markovic, P.N. Ross Jr., J. Phys. Chem. 99 (1995)
8290.
[13] M.S. Nashner, A.I. Frenkel, D.L. Adler, J.R. Shapley, R.G. Nuzzo,
J. Am. Chem. Soc. 119 (1997) 7760.
[27] A.F. Bykov, N.B. Morozova, I.K. Igumenov, S.V. Sysoev, J. Therm.
Anal. 46 (1996) 1551.
[28] J.O. Malm, J.O. Bovin, A. Petford-Long, D.J. Smith, G. Schmid, N.
Klein, Angew. Chem. Int. Ed. Engl. 27 (1988) 555.
[29] F. Dassenoy, W. Vogel, N. Alonso-Vante, J. Phys. Chem. B 106
(2002) 12152.
[14] A. Giroir-Fendler, D. Richard, P. Gallezot, Faraday Discuss. 92
(1991) 69.
[30] C. He, H.R. Kunz, J.M. Fenton, J. Electrochem. Soc. 144 (1997)
970.