A comparative study of Pt and Pt-Pd core-shell nanocatalysts

Article abstract of DOI:10.1016/j.electacta.2011.07.090

This comparative study characterizes two types of metallic and core-shell bimetallic nanoparticles prepared with our modified polyol method. These nanoparticles consist of Pt and Pt-Pd core-shell nanocatalysts exhibiting polyhedral morphologies. The controlled syntheses of Pt metallic nanoparticles in the 10-nm regime (4-8 nm) and Pt-Pd bimetallic core-shell nanoparticles in the 30-nm regime (15-25 nm) are presented. To realize our ultimate research goals for proton exchange membrane fuel cells (PEMFCs) and direct methanol fuel cells (DMFCs), we thoroughly investigate the dependence of the electrocatalytic properties of the nanoparticles on the structure, size and morphology. Significant differences in the electrocatalysis are also explained in experimental evidences of both Pt and Pt-Pd nanocatalysts. We suggested that the core-shell controlled morphologies and nanostructures of the Pd nanoshell as the Pd atomic monolayers will not only play an important role in producing inexpensive, novel Pt- and Pd-based nanocatalysts but also in designing more efficient Pt- and Pd-based nanocatalysts for practical use in DMFC technology. Our comparative results show that Pt-Pd nanocatalysts with Pd nanoshells exhibited much better electrocatalytic activity and stabilization compared to Pt nanocatalysts. Interestingly, we found that the size effect is not as strong as the nanostructuring effect on the catalytic properties of the researched nanoparticles. A nanostructure effect of the core-shell bimetallic nanoparticles was identified.

Full text of DOI:10.1016/j.electacta.2011.07.090

Electrochimica Acta 56 (2011) 9133–9143
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
A comparative study of Pt and Pt–Pd core–shell nanocatalysts
a,b,c,d,∗
d
c
a
a,∗∗
Nguyen Viet Long
, Michitaka Ohtaki , Tong Duy Hien , Jalem Randy , Masayuki Nogami
a
Department of Materials Science and Engineering, Nagoya Institute of Technology, Gokiso-cho, Showa-ku, Nagoya 466-8555, Japan
Posts and Telecommunications Institute of Technology, km 10 Nguyen Trai, Thanh Xuan, Ha Dong, Hanoi, Vietnam
Laboratory for Nanotechnology, Vietnam National University, Ho Chi Minh, Linh Trung, Thu Duc, Ho Chi Minh, Vietnam
b
c
d
Department of Molecular and Material Sciences, Interdisciplinary Graduate School of Engineering Sciences, Kyushu University, 6-1 Kasugakouen, Kasuga, Fukuoka 816-8580, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 29 April 2011
Received in revised form 15 July 2011
Accepted 20 July 2011
Available online 10 August 2011
This comparative study characterizes two types of metallic and core–shell bimetallic nanoparticles
prepared with our modified polyol method. These nanoparticles consist of Pt and Pt–Pd core–shell
nanocatalysts exhibiting polyhedral morphologies. The controlled syntheses of Pt metallic nanoparti-
cles in the 10-nm regime (4–8 nm) and Pt–Pd bimetallic core–shell nanoparticles in the 30-nm regime
(15–25 nm) are presented. To realize our ultimate research goals for proton exchange membrane fuel
cells (PEMFCs) and direct methanol fuel cells (DMFCs), we thoroughly investigate the dependence of
the electrocatalytic properties of the nanoparticles on the structure, size and morphology. Significant
differences in the electrocatalysis are also explained in experimental evidences of both Pt and Pt–Pd
nanocatalysts. We suggested that the core–shell controlled morphologies and nanostructures of the Pd
nanoshell as the Pd atomic monolayers will not only play an important role in producing inexpensive,
novel Pt- and Pd-based nanocatalysts but also in designing more efficient Pt- and Pd-based nanocatalysts
for practical use in DMFC technology. Our comparative results show that Pt–Pd nanocatalysts with Pd
nanoshells exhibited much better electrocatalytic activity and stabilization compared to Pt nanocata-
lysts. Interestingly, we found that the size effect is not as strong as the nanostructuring effect on the
catalytic properties of the researched nanoparticles. A nanostructure effect of the core–shell bimetallic
nanoparticles was identified.
Keywords:
Nanocatalysts
Core–shell bimetallic nanoparticles
Pt nanoparticles
Pt–Pd core–shell nanoparticles
Fast hydrogen adsorption, Direct methanol
fuel cells (DMFCs)
©
2011 Elsevier Ltd. All rights reserved.
1
. Introduction
FCs. Unfortunately, the cost of Pt is still high, which has hindered
the large-scale production and commercialization of these tech-
nologies. At present, Pt- or Pd-based nanocatalysts, combined with
various types of carbon supports and polymer membranes, are
employed in low and high temperature FC technologies, enabling
promising FC technologies to be competitive [8–10]. Among them,
DMFC technology has potential applications in portable devices,
including cell phones and portable computers, and automobiles.
For DMFC, methanol is one of the most widely used fuels because of
its low toxicity and large abundance. The technology for methanol
production has been well developed [1–3]. Today, Pt–Ru nanocat-
alysts that show good catalytic activity are successfully tested in
DMFCs; however, the high cost of their commercialization remains
a significant challenge [11,12,78]. Recently, an attractive method
of using ultrasound for the fabrication of nanocatalysts has been
tested in various FCs, showing the advantages of Pt–Ru nanocat-
alysts [13,78]. To achieve practical, cost-effective design, various
types of new Pt-based catalysts have been developed as bimetallic
alloys [11–13]. However, the successful synthesis of core–shell
morphologies in the sub-30-nm regime is relatively rare. In addi-
tion, one of the great challenges for researchers in the field is the
difficulty in fabricating the thin Pt shell as atomic monolayers on
the definite core [14,15]. The common strategies of controlling the
Pd–Pt bimetallic nanoparticles of the core–shell structure focus on
Traditionally, Pt- or Pd-based nanocatalysts are developed
with fuel cell (FC) technologies because they are highly active
electrocatalysts for oxygen reduction reactions (ORR), methanol
oxidation reactions (MOR) and ethanol oxidation reactions (EOR)
[
1–3]. These nanocatalysts are adopted in the cathodes and anodes
of fuel cell systems [1–6]. Their electrocatalysis was derived from
the surface-to-volume ratio and quantum-size effect. Additionally,
a synergistic effect was discovered to be the cause of the enhanced
catalytic activity [7]. At present, FC technologies cover a remarkably
wide variety of practical applications. Typically, they include pro-
ton exchange membrane fuel cells (PEMFCs) and direct methanol
fuel cells (DMFCs) [3–8]. In addition, alcohol fuels are utilized for
∗ Corresponding author at: Department of Materials Science and Engineering,
Nagoya Institute of Technology, Gokiso-cho, Showa-ku, Nagoya 466-8555, Japan;
Department of Molecular and Material Sciences, Interdisciplinary Graduate School
of Engineering Sciences, Kyushu University, 6-1 Kasugakouen, Kasuga, Fukuoka 816-
8
8
∗
580, Japan. Tel.: +81 090 9930 9504/81-(0)92 583 8835; fax: +81 092 583 8835/
1 (0)92 583 8835.
∗ Corresponding author.
E-mail addresses: nguyenviet long@yahoo.com, nguyenvietlong01@gmail.com
(
N.V. Long), nogami@nitech.ac.jp, mnogami@mtj.biglobe.ne.jp (M. Nogami).
0
013-4686/$ – see front matter © 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2011.07.090

9
134
N.V. Long et al. / Electrochimica Acta 56 (2011) 9133–9143
studying the nature of the nanoparticles. Therefore, the techniques
can be extended to cheap nanocores such as oxides, glasses, alloys,
and ceramics to design new functional materials, which are poten-
tially used for various FC systems [16–18]. In this paper, we report
a simple and effective method of preparing precious metallic and
bimetallic nanoparticles. Pt and Pt–Pd core–shell nanoparticles
were synthesized with controlled structure, size and morphology
by the reduction of H PtCl6 and Na PdCl4 in ethylene glycol (EG)
using AgNO3 as a structure-controlling agent. The methods of
transmission electron microscopy (TEM) and high resolution (HR)
TEM were adopted to characterize their morphology and size. The
results showed that the size and morphology of Pt nanoparticles
20 min prior to synthesis. The solution was refluxed for 2 h in an
Ar gas atmosphere and sealed to prevent air from entering the
flask. Next, 30 ␮L of 0.0625 M H PtCl was added to the flask at
2
6
◦
160 C followed by 60 ␮L of 0.375 M PVP so that their total vol-
umes, including 1.5 mL of H PtCl6 and 3 mL of PVP, were reacted
2
under stirring for 15 min. The volume ratio of PVP and H PtCl was
2
6
2:1 (60 ␮L/30 ␮L), and it was observed that the fast reduction of
H PtCl6 by EG occurred. The resultant mixture was stirred in the
2
2
2
◦
flask at 160 C for 15 min. As a result, a dark brown product con-
taining Pt nanoparticles was obtained. In Section 2, Pt nanoparticles
synthesized by the above reliable procedure were subsequently
used as the prepared cores for producing Pt–Pd core–shell nanopar-
ticles (15–25 nm). Here, 30 ␮L of 0.0625 M Na PdCl was added to
can be controlled with a small amount of AgNO . The prepared
3
2
4
◦
Pt nanoparticles of 4–8 nm exhibited a polyhedral morphology.
The Pt–Pd core–shell nanoparticles of 15–25 nm were controllably
synthesized and exhibited polyhedral morphology. A comparison
of their electrocatalysis was conducted to confirm the advantages
of the polyhedral morphology and core–shell configuration. The
significant catalytic enhancement and stabilization originating
from the nanoshell layer were identified. Our results thereby
revealed that a new strategy for creating nanostructured cata-
lysts with high hydrogen solubility and reactivity as well as fast,
sensitive, and stable hydrogen adsorption could be crucial for FCs.
the flask at 160 C, followed by 60 ␮L of 0.375 M PVP, so that their
total volumes, including 1.5 mL of Na PdCl4 and 3 mL of PVP, were
2
reacted under stirring. The volume ratio of PVP and Na PdCl4 was
2
2:1 (60 ␮L/30 ␮L). It was observed that the successive reduction
of Na PdCl4 by EG occurred under continuous stirring for 15 min.
2
Subsequently, the resultant mixture was kept in the flask under
stirring for some time (t ∼ 2–5 min) [19]. Then, the flask containing
the product solution was moved out of the boiling-oil bath. Through
this process, the only formation of Pt–Pd core–shell nanoparticles
was obtained in our final product.
2
2
2
. Experimental
2.2. Electrocatalyst preparation
.1. Synthesis of metallic and bimetallic nanoparticles
To obtain pure nanoparticles, the product containing the pre-
pared Pt and Pt–Pd nanoparticles were washed with a triple volume
of acetone, followed by centrifugation at 5000 rpm. Then, solid
products were obtained and re-dispersed in ethanol/hexane (vol-
ume ratio ∼1:3), and the mixtures were centrifuged at 3000 rpm.
The PVP-encapsulated Pt and Pt–Pd nanoparticles were washed
in ethanol/hexane several times. After washing, the nanoparticles
.1.1. Chemicals for synthesis
In our synthesis, chemicals from Aldrich (or Sigma–Aldrich)
were used in the experimental processes. These include
polyvinylpyrrolidone (PVP, a protective reagent) (FW: 55,000),
sodium tetrachloropalladate (II) hydrate (Na PdCl ) (ACS reagent),
2
4
−
1
and chloroplantinic acid hexahydrate (H PtCl ) (ACS reagent),
were dispersed in milli-Q water to achieve a density of ∼1 mg mL
2
6
which are the precursors for producing Pt nanoparticles and Pt–Pd
core–shell nanoparticles. Ethylene glycol (EG) was used as both
the solvent and the reducing agent. Silver nitrate (metal basis) was
used as a modifying agent. Solvents, including ethanol, acetone,
and hexane, were used for washing and cleaning. A Narnstead
nanopure H O purification system provided ionized, distilled, and
milli-Q water. All chemicals were of analytical grade and were
used without any further purification.
with the aid of an ICPS analyzer. For electrochemical measure-
ments, the working electrode was a glassy carbon rod (RA5, Tokai
Carbon Co., Ltd., diameter: 5.2 mm). First, the electrode surface was
cleaned and activated by using a polishing cloth (Buehler Textmet)
and an Al slurry (Aldrich, particle size ∼50 nm), followed by copious
washing with milli-Q water. This procedure was repeated until the
surface looked like a mirror. Then, a fixed weight of the prepared
catalyst loading was set onto the polished electrodes. The loaded
2
◦
◦
electrodes were dried for 3 h at 25 C and heated up to 450 C in air
◦
−1
2.1.2. Synthesis of Pt metallic nanoparticles
to remove the organic species (heating rate ∼1 C min , keeping
time ∼2 h). The slow heating rate was used to avoid the sintering
of the nanoparticles. After the electrode sintering heat treatment,
the electrodes were slowly cooled and subsequently exposed to a
To synthesize Pt nanoparticles (4–8 nm), 3 mL of EG, 1.5 mL of
.0625 M H PtCl , 3 mL of 0.375 M PVP, and 0.5 mL of 0.04 M AgNO
0
2
6
3
were used. The volumes of 4 mL of EG and 0.5 mL of 0.04 M AgNO3
were mixed in a flask and heated to 160 C. After the solution was
◦
◦
H2/N2 (20%, 80%) flow at 100 C for 3 h to reduce the existence of
mixed for 15 min, argon (Ar) gas was bubbled into the flask for
Pt–O and ensure a pristine catalyst surface. To improve mechanical
stability and durability, 10 ␮L of 5 wt% Nafion solution was added
onto the electrodes, and they were then dried overnight before
electrochemical measurements.
2
0 min prior to synthesis. The solution was refluxed for 2 h in an
Ar gas atmosphere. Next, 20 ␮L of 0.0625 M H PtCl was added to
2
6
◦
the flask at 160 C, sequentially followed by the injection of 40 ␮L
of 0.375 M PVP into the flask until 1.5 mL of 0.0625 M H PtCl6 and
2
3
mL of 0.375 M PVP were added to the flask and reacted for 15 min
2.3. Characterization
under stirring. The ratio of the solution volumes of PVP and H PtCl6
2
was 2:1 (40 ␮L/20 ␮L). The reduction of H PtCl by EG occurred. The
dark-brown mixture was stirred in the flask at 160 C for 15 min. As
a result, a dark-brown product of Pt nanoparticles was obtained.
2.3.1. Transmission electron microscopy
2
6
◦
To characterize the Pt nanoparticles of 4–8 nm and Pt–Pd
core–shell nanoparticles of 15–25 nm, copper grids containing the
Pt and Pt–Pd core–shell nanoparticles were maintained under vac-
uum by using a vacuum cabinet (JEOL-JEM-DSC10E Film Vacuum
Desiccator Cabinet) prior to TEM and HRTEM measurements. The
copper grids were treated again at a drying station (JEOL JDS-
230F Drying Station) using an ion cleaner that employed glow
discharge in vacuum to remove any contamination (Ion Cleaner
JIC-410 JEOL). The copper grids were maintained overnight under
high vacuum conditions in the transmission electron microscope
2.1.3. Synthesis of Pt–Pd core–shell bimetallic nanoparticles
To synthesize Pt–Pd core–shell nanoparticles, Pt nanoparticles
were made as the initial cores. In the synthesis of the Pt cores,
3
0
0
mL of EG, 1.5 mL of 0.0625 M H PtCl , 3 mL of 0.375 M PVP, and
2 6
.5 mL of 0.04 M AgNO3 were used. The volumes of 3 mL of EG and
◦
.5 mL of 0.04 M AgNO were mixed in a flask and heated to 160 C.
3
The solution was mixed for 15 min. Then, Ar gas was bubbled for

N.V. Long et al. / Electrochimica Acta 56 (2011) 9133–9143
9135
Fig. 1. TEM images of polyhedral Pt nanoparticles. Scale bars: (a)–(d) 20 nm.
(
JEOL JEM-2100F or JEM-2010) prior to the TEM and HRTEM mea-
the monolayer of hydrogen adsorbed onto the Pt–Pd nanocatalyst
surfaces.
surements. The TEM images were obtained using a transmission
electron microscope (JEOL JEM-2100F and JEM-2010) operated at
2
00 kV. Finally, DigitalMicrograph software (Gatan, Inc.) was used
2
.3.3. X-ray diffraction method
Starting from the products of the prepared Pt and Pt–Pd
in the TEM and HRTEM studies to acquire, visualize, analyze, and
process the digital image data of the Pt and Pt–Pd nanoparticles.
core–shell nanoparticles, 3 mL of an ethanolic solution of the pre-
pared Pt nanoparticles or Pt–Pd core–shell nanoparticles was set
onto the slides of special glass with a small area of approximately
2
.3.2. Electrochemical measurements
A typical setup of a three-electrode electrochemical system was
2
1
cm for the XRD method. The drops of the ethanolic solution of
the prepared Pt and Pt–Pd core–shell nanoparticles were poured
used in our cyclic voltammetry experiments at ambient room tem-
perature. The electrodes were connected to a potentiostat (SI 1287
Electrochemical Interface, Solartron). The cell was a 50 mL glass
◦
◦
onto the glass substrate and dried at 80 C and 100 C for 6 h prior
to use. These slides were treated with ethanol to remove any
impurities. X-ray diffraction patterns were recorded by a diffrac-
tometer (X’Pert-Phillips) operating at 45 kV/45 mA and using Cu K␣
radiation (0.54056 nm). The specific crystal nanostructures of the
prepared Pt and Pt–Pd core–shell nanoparticles were compared in
detail.
vial, which was treated with a mixture of H SO –HNO solution
2
4
3
and washed with milli-Q water. A leak-free AgCl/Ag/NaCl electrode
RE-1B, ALS) served as the reference, and the counter electrode
was a Pt coil (002234, ALS). All potentials were reported E vs.
Ag/AgCl)/V. The electrolyte solution was bubbled with N2 gas for
0 min before each measurement. The N2 blanket was maintained
throughout the course of potential sweeping. For the base voltam-
metry, the electrolyte was a solution of 0.5 M H SO (or 0.5 M H SO
(
(
3
3. Results and discussion
2
4
2
4
and 1 M CH OH). The potential window between E = −0.2 V and
3.1. Structure, size and morphology of metallic and bimetallic
nanoparticles
3
−1
E = 1.0 V with a sweep rate of 50 mV s was used. For the oxida-
tion of methanol, 1.0 M methanol in milli-Q water was added to
the electrolyte. The measurements were cycled until stable voltam-
mograms were achieved. The electrochemical surface areas (ECSA)
were estimated by considering the area under the curve in the
hydrogen desorption region of the forward scan with respect to
Fig. 1 shows the TEM images of the polyhedral Pt nanoparticles
with particle sizes of 5–8 nm that were produced by the modi-
◦
fied polyol method at 160 C for 15 min in the presence of PVP,
with the assistance of AgNO3 as the structure-controlling agent.

9
136
N.V. Long et al. / Electrochimica Acta 56 (2011) 9133–9143
These Pt nanoparticles exhibited truncated polyhedral morpholo-
gies at the corners. Most of the Pt nanoparticles exhibit the main
forms of cubic, octahedral, and tetrahedral morphologies. In addi-
tion, the truncated cubic, octahedral, and tetrahedral morphologies
were also observed. It is certain that the addition of AgNO3 influ-
enced the shaping of the specific morphology of Pt nanoparticles
associated with their short synthetic time. We estimate that this
occurred at approximately 10–15 min. The polyhedral morpholo-
gies also show very sharp corners, edges, and facets. In all cases, the
sharp and polyhedral Pt nanoparticles of the characteristic low and
high-index planes were considered for their practical applications
in catalysis [19–21]. They exhibited their various morphologies
of {1 1 1}, {1 0 0}, and {1 1 0} facets or other facets of various Pt
Pt core, the thickness of the Pd shell can be precisely manipu-
lated by the quantity of the Na PdCl4 precursor added. A classical
2
theory was used to explain the nucleation and growth of metal-
lic nanoparticles and was extended to study the nucleation and
growth of alloy and core–shell bimetallic nanoparticles. Gibbs free
energy and overall excess free energy were applied to understand
issues of the sizes and morphologies of the nanoparticles, and
synthesize uniform formations of the core and/or the shell. The
nucleation and growth modes used include Frank–van der Merwe
(FM), Volmer–Weber (VW), and Stranski–Krastanov (SK) modes.
These enable us to explain the formations of various nanostruc-
tures of Pt, Pt–based alloys and Pt–based core–shell nanoparticles
[4,19,27,28]. Thus, we need to control the various FM, VW, and SK
growth modes experimentally during the reductions of two metal-
lic salts. This strategy can lead to produce a homogeneous system
of core–shell bimetallic nanoparticles, which strongly depends on
controlling the dominant FM growth mode under certain experi-
mental conditions. Therefore, homogeneous and inhomogeneous
nucleation and growth was possibly achieved in the formation of
the nanoshells. The case of homogeneous nucleation and growth
of the shell by a layer-by-layer mechanism is a desired mode for a
controlled synthesis. The important features of Pt–Pd core–shell
nanoparticles consist of islands, terraces, and steps around the
edges and monatomic steps [4,19]. Here, we have observed the
uniformity of the directions of overgrowth of the Pd monolayer
on the Pt core via a layer-by-layer mechanism and the lattice
fringe for the Pd shell was estimated to be 0.246 nm. As shown
in Fig. 2, the atomic Pd–shell layers of the octahedral bimetallic
nanoparticles were observed in the thickness of 6 ± 1 monolayers
(2.64 nm) and 5 ± 1 monolayers (2.20 nm). In a core–shell struc-
ture, the Pd shell included a number of Pd atomic monolayers.
The shell layer displayed up to several dozen atomic monolay-
{h k l} planes. Therefore, a modified polyol with the assistance of
AgNO potentially offers viable ways to synthesize precious metal
3
nanoparticles under size and morphology control [20]. The char-
acterization of shape-dependent electrocatalytic activities of Pt
nanostructures has been investigated [21–23]. The polyhedral mor-
phology of low and high-index planes or various Pt {h k l} planes
exhibited a high electrocatalytic activity that offers an enhance-
ment in ORR. The highest stability of methanol electro-oxidation by
hydrogen adsorption and desorption was verified in the Pt nanopar-
ticles with polyhedral shapes [24]. The polyhedral Pt nanoparticles
exhibited good single-crystal structures that were controlled in
both size and morphology. In addition, the twinned structures of
the Pt nanoparticles were also observed. The results confirmed
the polyhedral Pt nanoparticles in the homogenous single-crystal
growth.
Fig. 2 shows the typical TEM and HRTEM images of the Pt–Pd
core–shell nanoparticles that were 10–25 nm in size, synthesized
by a polyol method in the presence of a trace amount of silver
nitrate (0.5 mL of 0.04 M AgNO ). The HRTEM images showed the
3
segregation of the Pt core and Pd shell without their alloying. The Pd
atomic monolayers in homogeneous nucleation and growth were
observed in the layer-by-layer mechanism. These monolayers pro-
tect the polyhedral morphology of the initial Pt cores. This provides
further evidence for the epitaxial shell growth mode. Fig. 2 also
shows the HRTEM images of the Pt–Pd core–shell nanoparticles,
and the vertically oriented overgrowth of Pd atomic monolayers
on the Pt core is examined. The thickness of the Pd atomic mono-
layers was estimated to be less than 4 nm, and the overgrowth
of these monolayers on the surface of the Pt core was due to the
layer-by-layer gradual growth mechanism [4,19].
The oriented overgrowth of Pd atomic monolayers on the sur-
face of the Pt nanocore showed a good lattice match. However,
we suggest that the re-crystallization of metallic and bimetallic
nanoparticles need to be studied at this stage. It is clear that Pt–Pd
core–shell nanoparticles typically exhibit cubic, octahedral, and
tetrahedral main core–shell morphologies. In this study, we suc-
cessfully produced Pt–Pd core–shell nanoparticles (15–25 nm). In
addition, we also observed that the Pd coatings (or the sharp and
polyhedral Pd shells) of the Pt–Pd core–shell nanoparticles not
only exhibit tetrahedral, cubic, and octahedral morphologies but
also possess highly symmetric crystal structures. The polyhedral
Pt cores were synthesized, which shaped the homogeneous nucle-
ation and growth of the final morphology of the polyhedral Pd
shells. In this mode, it can be suggested that the initial Pt nanocores
shaped the overgrowth of the nanoshells of the Pd atomic monolay-
ers. Here, the core–shell formation demonstrated the segregation of
the Pt cores and Pd shells without any alloying phenomenon. The
homogeneous and inhomogeneous nucleation and growth of the
shell at the core and shell interface of two different metals is due to
the epitaxial and/or non-epitaxial growth modes [4,19,25,26]. The
two Pt and Pd metals have the same crystal structure (fcc). There-
fore, the Pd shell could grow epitaxially on the Pt core because
of their similar crystal structures [4,19]. In comparison with the
ers, depending on the total volume and concentration of Na PdCl .
2 4
The Pt–Pd core–shell bimetallic nanoparticles showed the number
of Pd atomic monolayers in the 20-monolayer regime. Moreover,
Fig. 2 also shows a good octahedral Pt–Pd core–shell morphology;
however, these particles have truncated polyhedral corners. It is
clear that Pt–Pd core–shell nanoparticles also displayed arrange-
ments of highly symmetric atomic columns at their surfaces. The
nucleation and growth mechanism of Pt–Pd core–shell nanoparti-
cles were considered in the homogeneous nucleation and growth
of both the core and the shell. In addition, the observed morpholo-
gies of the Pt–Pd core–shell nanoparticles are strongly dependent
on their crystal orientation and the technique limitations of high
resolution TEM at approximately 2 nm. In the other hand, the Pt
core nanoparticles show the most characteristic morphologies,
including tetrahedral, cubic, and octahedral shapes. The forma-
tions of their truncated corners were also observed. However,
the simultaneous nucleation and growth of various polyhedral
nanoparticles need to be further studied. Moreover, the Pd mono-
layer shells were well grown with similar low-index planes of
{1 1 1}, {1 0 0}, and {1 1 0} or Pd {h k l}, which are the same as the
Pt cores with low-index planes of {1 1 1}, {1 0 0}, and {1 1 0} or Pt
{h k l}. It was suggested that small Pd nanoclusters were formed
and directly deposited onto the surfaces of the Pt cores. The single-
crystal Pd nanoshells were grown from these small Pd nanoclusters
that matched the surfaces of the single crystal Pt cores, exhibit-
ing sharp polyhedral morphology. The reduction of Na PdCl at
2
4
◦
160 C promotes not only homogeneous nucleation but also crys-
tal growth of the thin layer of Pd nanoshells on the surfaces of
the Pt nanocores [19]. In the nucleation and growth of the sin-
gle Pd nanoshell crystal, a very thin Pd layer was formed while
the nucleation and growth mechanisms of the Pd nanoshells retain
and protect the good morphology of the initial Pt nanocores. A
layer-by-layer mechanism was identified. The formation of the
Pt atomic monolayer is crucial to the controlled synthesis of the

N.V. Long et al. / Electrochimica Acta 56 (2011) 9133–9143
9137
Fig. 2. HRTEM images of Pt–Pd core–shell. The thin Pd shells protect polyhedral Pt cores. The nucleation and growth of Pd shells are controlled by a chemical synthesis. Scale
bars: (a)–(c) 20 nm. (d) 5 nm. (e) 5 nm. (f) 2 nm.
bimetallic core–shell nanoparticles. There is a very small lattice
mismatch at the transition between the crystalline orientation of
the Pt core and that of the Pd shell in the distinctly visible Pt–Pd
core–shell geometry observed inside Pt–Pd core–shell nanoparti-
cles (Fig. 2). Thus, high quality Pt–Pd core–shell nanoparticles were
achieved.
3.2. Electrocatalytic characterization
3.2.1. Electrochemical reactions
The electrode was swept from E = −0.2 to E = 1.0 V with respect
to the saturated electrode. In Fig. 3, there are seven specific regions
in the cyclic voltammogram exhibiting catalytic activity and

9
138
N.V. Long et al. / Electrochimica Acta 56 (2011) 9133–9143
planes. For Pt nanoparticles, the catalytic activity increased in the
order of {1 1 1} < {1 0 0} < {1 1 0} or {1 0 0} < {1 1 1} < {1 1 0} planes.
These mainly depend on the temperature and type of metal in
the nanoparticle-based catalyst [4,31–34]. In addition, the con-
firmation of H2 adsorption/desorption by high-index planes was
discussed [31–34]. However, researchers need further evidence
from both theoretical and experimental studies. The second region
showed important characteristics of charging and discharging,
which were identified by two double layers of QDL (charge) and
QDL (discharge) with respect to the standard hydrogen electrode
(SHE). The charging and discharging indicated by the double lay-
ers during the reverse sweep are the dominant contributors to
the current in a half-cell [8,35]. During the in situ CV measure-
ment across a FC, the second region might also contain a limiting
hydrogen oxidation current due to hydrogen crossover from the
anode chamber during the evaluation of the cathode catalyst [8,35].
Above E = 0.6 V, Pt nanocatalysts catalyzed water activation. In our
experimental observation, the Pt was covered with an oxide layer
at potentials close to E = 1.0 V. It was known that the third region
showed the formation of Pt–OH and the conversion of Pt–OH into
Pt(OH)2 between E = 0.8 V and E = 1.0 V. At the same time, the for-
mation of Pt–O and H O were seen in the fourth region. Beyond
2
E = 1.0 V, the evolution of oxygen gas began to dominate, and a
reverse sweep was immediately followed. At potentials close to
E = 0.45 V, the fifth region showed the formation of Pt–Pt and water.
Notably, the slow kinetics of Pt–O reduction was observed in the
fifth region. The pathways of Pt oxide reduction were different from
the first cycle to the twentieth cycle, and the stabilization of Pt oxide
reduction was achieved after the tenth cycle. During the reverse
sweep, hydrogen evolution potential was observed in the seventh
region, and hydrogen adsorption (Pt–Hads) was observed in the
sixth region. The hydrogen evolution potential was E = −0.2 V in the
seventh region. Hydrogen adsorption–desorption reactions using
electrodes in acidic media were discussed in detail [35,36]. The
hydrogen evolution reaction (HER) with the Pt catalyst followed
the reactions of Volmer, Tafel, and Heyrovsky [6]. The combi-
nations of these reactions were characterized as Volmer–Tafel
and Volmer–Heyrovsky mechanisms [6,35,36]. The processes that
occurred at the electrode surface exhibiting seven specific regions
in cyclic voltammograms follow Eqs. (A2)–(A8).
Fig. 3. Cyclic voltammograms of as-prepared Pt nanocatalysts in 0.5 M H2SO4 in the
ranges of E = −0.2 V to E = 1.0 V and E = −0.2 V to E = 0.2 V.
surface kinetics for the case of the Pt nanocatalysts of 4–8 nm. They
were characterized by the chemical activity occurring at the elec-
trode surface (Eqs. (A2)–(A8)). In the forward sweep, the first region
assigned to hydrogen desorption is crucial. Slow kinetics of hydro-
gen desorption was confirmed in the cell before the CV stabilization
was achieved from the first cycle to the twentieth cycle. We propose
to call this process the “slow hydrogen desorption” mechanism to
the single metallic nanocatalyst. Here, QA is the integrated anodic
charge in Coulombs representing the desorption of hydrogen (Eq.
Pt–Hads → Pt + H+ + e−
(A2)
(A3)
(A4)
(A5)
(A6)
(A7)
(A8)
QDL(Charge) ↔ QDL(Discharge)
Pt + H O → Pt–OH + H + e−
+
2
PtOH + H O → Pt(OH) + H + e−
+
(
A2)), and QC is the integrated cathodic charge representing the
2
2
adsorption of hydrogen (Eq. (A8)). The atomic hydrogen adsorption
charge density can be calculated in Fig. 3. To evaluate the catalytic
activity of the prepared nanocatalysts, the electrochemical active
surface area (ECSA) of the Pt nanocatalyst is calculated [8,29,30].
Pt–(OH) → PtO + H O
2
2
2PtO + 4H+ + 4e− → Pt–Pt + 2H O
Pt + H+ + e− → Pt–Hads
2
QH
ECSA =
,
(A1)
The seventh region is assigned to hydrogen evolution during
the reverse sweep. The hydrogen evolution reaction and hydro-
gen adsorption on the electrode surface of the Pt–Pd core–shell
nanoparticle catalyst showed much better chemical and catalytic
activity than the Pt nanocatalyst. The electrode was swept between
E = −0.2 V and E = 1.0 V with respect to the saturated electrode in
Fig. 4. Similarly, there were seven specific regions in the cyclic
voltammogram, exhibiting catalytic activity and surface kinetics in
the case of the Pt–Pd core–shell nanocatalysts (15–25 nm). Notably,
QA is the integrated anodic charge in Coulombs that represents the
desorption of hydrogen, and QC is the integrated cathodic charge
that represents the adsorption of hydrogen. The charges of QA and
QC showed the chemical activity of catalyst at the surfaces and
interfaces of the electrodes and electrolyte. For Pt–M (M = Ru, or
0
.21 × LPt
The regions after the background current subtraction were used
to accurately integrate the faradaic contribution. The Q /Q ratio
A
C
was characterized by the reversibility of the hydrogen adsorption
and desorption reactions on the electrode surface of the prepared
Pt nanocatalysts. The specific hydrogen adsorption/desorption
regions exhibited various facets at the surfaces of the Pt nanopar-
ticles of low- and high-index planes as well as the effects of
their various edges, corners, and terraces [31–34]. In our case,
hydrogen adsorption/desorption regions that correspond to the
low-index facets are difficult to distinguish in their separate
regions. In contrast, the differences were observed in both hydro-
gen desorption and oxygen adsorption for the low- and high-index

N.V. Long et al. / Electrochimica Acta 56 (2011) 9133–9143
9139
Fig. 4. Cyclic voltammograms of as-prepared Pt–Pd core–shell nanocatalysts in
.5 M H2SO4 in the ranges of E = −0.2 V to E = 1.0 V and E = −0.2 V to E = 0.2 V.
Fig. 5. Cyclic voltammograms of as-prepared Pt and Pt–Pd core–shell nanocatalysts
in 0.5 M H2SO4 in the range of E = 0.3 V to E = 1.0 V.
0
Rh, Sn, Pd,. . .), the two specific regions can be distinguished by
two anodic (desorption) and two cathodic (adsorption) peaks con-
sisting of H1A, H2A, H1C and H2C, respectively. These peaks are
interpreted by the strength of the bonded hydrogen atoms [37,38].
In the hydrogen desorption region, the peak at the more negative
potentials (H2A) is attributed to the hydrogen desorption from the
bulk of the metal in combination with the desorption of hydro-
gen adsorbed on the surface. The peak obtained at more positive
potentials (H1A) is attributed exclusively to the adsorbed hydrogen
after achieving good stabilization of the fast hydrogen-desorption
response and high sensitivity in our results after reaching the sta-
ble characterization after only the first cycle. Similarly, the H1C and
H2C peaks in the region were also clearly distinguished. Therefore,
the electrode with Pd–Hads and/or Pt–Pd–Hads surface monolay-
ers exhibited two types of weakly and strongly bonded hydrogen
atoms. The H2A and H2C peaks showed evidence of strongly bonded
hydrogen atoms within the sites. This is an important remark for
the case of Pt–Pd core–shell nanocatalysts with Pd nanoshells as
atomic monolayers in our study. This enables the realization of
robust and efficient Pt- or Pd-based core–shell nanocatalysts that
are extremely sensitive to the fast hydrogen desorption. The alter-
native method of improving the hydrogen reaction by core–shell
morphology can be realized. Similarly, QA was used to accurately
integrate the faradaic contribution after the background current
[
37,38]. In our results, for the case of Pt nanocatalysts, the H1A and
H2A peaks were not clearly distinguished after achieving the stabi-
lization of hydrogen desorption in the case of the Pt nanocatalyst.
Similarly, the H1C and H2C peaks in the region were also not clearly
distinguished. Therefore, Pt–Hads can be both weakly and strongly
bonded hydrogen atoms. Therefore, we need to create new active
sites on the surfaces of the nanocatalysts because they can give
the strongest hydrogen desorption and enhancement of hydrogen
desorption. Our proposal of improving Pt-based catalysts is to cre-
ate new, strongly bonded hydrogen atoms to increase the region
of hydrogen desorption. At the same time, ORR activity should be
necessarily maintained for the enhancement and stabilization of
ORR kinetics.
In Figs. 3–5, the ORR on the electrode of the prepared Pt
nanocatalyst is observed as being sluggish [39,40]. In contrast, the
fast enhancement of ORR on the electrode of the prepared Pt–Pd
core–shell nanocatalyst is clearly observed. Therefore, it is impor-
tant to use core–shell bimetallic nanocatalysts to increase the ORR
rate in the electrode catalyst. For the case of Pt–Pd core–shell
nanocatalysts, the H1A and H2A peaks were clearly distinguished
subtraction. In addition, the Q /QC ratio showed the reversibility of
A
the hydrogen adsorption and desorption reactions.
Figs. 4 and 5 showed the first region describing different pro-
cesses that occurred at the electrode surface. Hydrogen desorption
occurred at the anode containing Pt–Pd core–shell nanocatalysts.
These processes indicated the formation of a Pt–Pd–Hads layer at the
electrode. The hydrogen desorption was very good after only one
or two cycles. This shows the excellent hydrogen desorption of the
Pt–Pd core–shell configuration, and we call this process “fast hydro-
gen desorption” or “fast response of hydrogen desorption”. The
strong sensitivity and activity of hydrogen desorption is clear evi-
dence, and we hypothesize that the high sensitivity and activity are
due to the synergistic effect of the core–shell bimetallic nanopar-
ticles [7]. This possibility represents a progress in electrocatalysis

9
140
N.V. Long et al. / Electrochimica Acta 56 (2011) 9133–9143
0.0010
research. Thus, it can be concluded that Pt–Pd nanocatalysts accel-
erate the rate of the chemical reaction in hydrogen desorption.
It is obvious that the chemical nature of the nanostructure, mor-
phology and size effects of metallic and bimetallic nanoparticles in
the nanocatalysis were very different. In the second region, charg-
ing and discharging demonstrated good performance and stability.
This second region of Pt–Pd nanocatalyst (Fig. 4) from E = 0.1 V to
E = 0.4 V was much more stable than that of the Pt nanocatalyst
Pt-Pd core-shell nanocatalysts
Pt nanocatalysts
0
0
.0005
.0000
(
Fig. 3) from E = 0.15 V to E = 0.3 V. Therefore, the prepared Pt–Pd
core–shell nanocatalysts showed good stabilization of the rela-
tionship between charging and discharging. Above E = 0.6 V, Pt–Pd
core–shell nanocatalysts also catalyzed water activation. However,
the kinetics of their catalysis of water activation was very dif-
ferent from Pt nanocatalysts in Fig. 3. Certainly, the catalysis of
water activation of Pt–Pd core–shell nanocatalysts was much better
than that of Pt nanocatalysts. It was assumed that Pt–Pd core–shell
nanocatalysts are covered with an oxide layer at the maxima close
to E = 1.0 V. We assumed that these were in the form of Pd–O and/or
-0.0005
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
E vs (Ag/AgCl) / V
(
Pt–Pd)-O monolayers because the Pd nanoshell monolayers were
Fig. 6. Cyclic voltammogram of Pt nanocatalysts (a), and Pt–Pd core–shell nanocat-
alysts (b) on glassy carbon electrode in N2-bubbled 0.5 M H2SO4 electrolyte (scan
very thin (less than 20 atomic monolayers). In general, we assigned
them to be (Pt–Pd)–O monolayers. Between E = 0.8 V and E = 1.0 V,
the formation and variation of Pt–OH (or Pd), Pt–(OH)2 (or Pd),
and Pt–O (or Pd) with water were determined. In our results, the
third region showed the formation of Pd–OH and the conversion of
Pd–OH into Pd–(OH)2 and/or the formation of (Pt–Pd)–OH and the
−
1
rate: 50 mV s ).
The area within the curve in the relevant region can give QT and QDL
and can be obtained by taking the area under the same region but
with the upper and lower boundaries as horizontal lines passing on
a data point outset of the hydrogen desorption/adsorption waves.
The electrochemical surface area (ECSA) is calculated with Eq. (A1)
conversion of (Pt–Pd)–OH into (Pt–Pd)–(OH) . At the same time,
2
the formation of Pd–O and/or (Pt–Pd)–O and water were seen in
the fourth region. In contrast, a new phenomenon was observed
in the fourth region, in which the current density was very high
during the first cycle. The stability of positive current density was
achieved after cyclic voltammetry. At potentials close to E = 0.52 V,
the fifth region showed the formation of Pd–Pd and/or Pt–Pd–Pd
and water. Notably, the fast response and stable kinetics of the
Pd and/or Pt–Pd oxide reduction was observed (fifth region). The
pathways of Pt or Pt–Pd oxide reduction were different from the
first cyclic voltammogram to the twentieth cyclic voltammogram
−
2
[8,29,30]. The conversion factor of ∼0.21 (in mC cm ) was used for
a monolayer of hydrogen. LPt corresponds to the loading of the cata-
−
2
lyst on the glassy carbon surface (in mg cm ). We have presented
the calculated ECSA values of the two samples (Pt nanocatalysts
and Pt–Pd core–shell nanocatalysts). The results showed a sig-
nificant difference in ECSA values between the Pt nanocatalyst
and Pt–Pd core–shell nanocatalyst. The Pt–Pd core–shell nanocat-
2
−1
−1
alysts exhibited a high value for ECSA (27.7 m g ), while the
2
Pt-nanocatalysts had a much lower value (10.5 m g ). These high-
lighted the main contributing factors for the improvement and
fine-tuning of the nanoparticle surface. Therefore, Pt–Pd core–shell
nanocatalysts exhibiting efficient electrocatalysis are considered to
replace Pt–Ru electrocatalysts with low Pt content for formic acid
electro-oxidation [43,44]. Additionally, core–shell electrocatalysts
(
CV) in the values of current density and voltage; however, their
CV curve forms were similar in appearance. The stabilization of Pd
and/or Pt–Pd oxide reduction was achieved after only three cyclic
voltammograms, which showed better ORR characterization. Dur-
ing the reverse sweep back to the hydrogen evolution potentials
in the seventh region, the following surface reactions took place.
These showed hydrogen adsorption (Pt–Hads or Pt–Pd–Hads) in the
sixth region. The limitation of hydrogen evolution potentials was at
a more negative potential close to E = –0.2 V; however, the absolute
current density was negative. Through the importance of study-
ing electrochemical reactions in detail, the performance of new
Pt-based catalysts can be significantly improved.
with Pt shell monolayers on Pd Co cores significantly enhanced the
3
durability of the ORR activity [44]. In this regard, ternary Pt–Fe–Co
alloy electrocatalysts of Pt85Fe10Co5 formed by electrodeposition
showed mass activities that were 3.5 times higher than that of
pure Pt nanocatalysts [45]. In particular, the amount of Pt and Pd
used (two Pd atomic layers and a Pt atomic half-monolayer) is
very small; however, they exhibit very high activity for the electro-
oxidation of formic acid [46]. A small amount of Pt deposited on the
PdCo-Pd/C core–shell nanoparticles indicated that the core–shell
nanoparticles decorated by Pt displayed enhanced stability and
electrocatalytic activity for ORR activity [47]. Interestingly, a com-
bination of two types of Pd and Pt nanoparticles (Pd on cubic Pt
seeds) for investigation of electrocatalytic formic acid oxidation
showed much less self-poisoning and a lower activation energy rel-
ative to Pt nanocubes [48]. Indeed, the method uses a combination
of mixed metal Pt monolayer electrocatalysts to improve the ORR
kinetics relative to the pure Pt nanocatalyst. Its origin was the Pt
islands or small clusters or monolayers on the Pd nanoparticles (Pd
inslands or clusters or monolayers on Pt nanoparticles) [49]. There
is still some debate over the potentially stronger enhancement of
Pt alloy electrocatalysts for the ORR [50]. At present, we cannot
confirm their catalytic characterization and stabilization with cer-
tainty and cannot determine whether the prepared nanocatalysts
are better than the bimetallic catalysts (e.g., Pt–Pd, Pt–Cu, Pt–Ru)
as multi-component alloy nanocatalysts. In addition, the syntheses
3.2.2. Comparison of catalytic activity
Fig. 6 shows the cyclic voltammograms of the Pt nanocata-
lysts (wine color) and Pt–Pd nanocatalysts (line: magneta color)
−1
acquired at 50 mV s
in 0.5 M H SO . The voltammograms dis-
2 4
play the typical shape for base voltammetry. Beyond E = 0.4 V in
the forward sweep, the Pt–O formation is observed, and its corre-
sponding reduction is given by the negative peak in the backward
sweep. In the case of bimetallic nanoparticles, catalytic activity
is determined and enhanced by the electronic tuning on the sur-
face of the first metal core brought about by the adsorbed second
metal. Such a mechanism may be optimized by controlling the
shell thickness and the choice of the second metal as well as the
nanoparticle-shape dependence and stability [41,42]. The specific
charge transfer (QH) due to hydrogen adsorption and desorption is
given by: QH = (QT − QDL)/2, where QT denotes the total amount of
charge during hydrogen adsorption and desorption on Pt sites and
QDL relates to the charge due to the double layer capacitance [7,8].

N.V. Long et al. / Electrochimica Acta 56 (2011) 9133–9143
0.0025
9141
of the nanocatalysts show high complexity in confirming their var-
ious compositions, sizes, nanostructures, and characterizations. To
demonstrate the difference of catalytic activity between two types
of metallic and bimetallic systems during a typical catalyst-assisted
reaction, cyclic voltammetry in methanol electro-oxidation was
performed. To change the ORR property, Pt-based catalysts, such as
Pd–Pt and Pd–Pt–Ni electrocatalysts (binary and ternary composi-
tions), or catalysts with non-Pt compositions were tested with the
rotating disk electrode (RDE); however, there was no clear exper-
imental proof of higher activity in ORR compared to the binary
alloys [51]. In addition, the method of using RDE may possibly
lead to a high solubility of nanocatalysts in the electrolytes. To
Pt-Pd core-shell nanocatalysts
Pt nanocatalysts
0
0
.0020
.0015
0.0010
0
.0005
improve the ORR property, Pt–Pd nanocatalysts (PtxPd1 -based
0.0000
−x
nanocatalysts) supported on various carbon nanomaterials (carbon
nanotubes (CNTs) such as single-walled CNTs (SWCNTs) or multi-
walled CNTs (MWCNTs)) showed high surface area and stability
-
0.0005
0.0
0.2
0.4
0.6
0.8
1.0
[
52]. In addition, Pt–CO poisoning or CO-stripping voltammetry
E vs (Ag/AgCl) / V
causes a significant decrease in the overall efficiency of DMFCs.
Therefore, CO poisoning should be controlled at a low CO tolerance
Fig. 7. Cyclic voltammograms towards methanol electro-oxidation of Pt nanocata-
[
2,10]. Pt–M (M = Co, Ni, Mn, or Fe, etc.) or ternary Pt-base alloys
lyst and Pt–Pd nanocatalyst.
were prepared in an effort to find highly actively surfaces and sites.
Their usefulness in PEMs was reduced because of the dissolution
of Co, and Ni into the solution during the operation of the FCs.
This led to the conclusion that multi-composition nanocatalysts
are less desirable because of the high complexity in the prepara-
tion and application [53]. Moreover, quaternary Pt–Ru–Os–Ir alloy
nanocatalysts or multi-metallic nanocatalysts showed slightly bet-
ter electrocatalytic activity than Pt–Ru nanocatalysts; however, its
preparation procedure was much more complex [54,55]. Based on
the density functional theory in surface chemistry and catalysis,
Pt–Pd catalysts are great candidates in catalysis [56]. The main
reasons are the high oxygen reduction activity of Pt and Pd metal
catalysts because they are confirmed in their activity to be on the
top of the volcano curve among transition metals, such as Ir, Cu, and
Ag, for obtaining high ORR activity [56,57]. Among the noble met-
als, the four metals Pd, Pt, Rh, and Ir were confirmed to have high
In electrochemical measurements, the electrolyte solution is 0.5 M
−
1
H SO + 1.0 M CH OH, and a scan rate of 50 mV s was used. Fig. 7
2
4
3
shows a comparison of the cyclic voltammograms between the
Pt nanocatalysts and Pt–Pd core–shell nanocatalysts in the mix-
ture of 0.5 M H SO4 and 1 M CH OH. A stable voltammogram was
2
3
attained after 20 cycles of sweeping between 0 and 1 V. Two oxi-
dation peaks are noticeable; the first one is between E = 0.6 V and
E = 0.8 V in the forward sweep, and the other is between 0.4 and
0.5 V in the backward sweep. The two peaks are directly related to
methanol oxidation and the associated intermediates. Moreover,
the curve shape is in excellent agreement with other publications
[67–69]. The shoulder at E ∼ 0.8 V relates to the complete oxidation
of intermediates, causing the current to rise again at a higher poten-
tial. To explain clearly, the anodic current decreased significantly
when the voltage was raised beyond the peak current potential
because of the blocking effect by the adsorbed intermediate species
on the active sites. Afterward, the anodic current increased again
because the concentration of available active sites increased and
the methanol molecules were oxidized again. In the backward scan,
the second peak is related to the removal of the carbonaceous
products from the forward scan. The peak current density in the
adsorption energy characterization. Accordingly, the Pt M catalysts
3
(
M = Ni, Co, Fe, Ti) showed their catalytic activity vs. the d-band cen-
ter. According to DFT theory, the volcano-shaped dependency was
presented in the best electrocatalytic activity of Pt Ni, comparable
3
to that of Pt Co in this experiment [39]. At present, the appropriate
3
approach for improving the catalytic surface and reducing the load-
ing of the Pt-catalyst is to produce novel Pd or Pt based core–shell
bimetallic nanocatalysts using Pt monolayers (nanoshells) or Pd
monolayers (nanoshells), with a strong interest in Pt–Ru, Pt–Cu
bimetallic nanoparticles. The core–shell nanostructures of Pt-based
electrocatalysts have been introduced in various solutions [58]. This
is meaningful in the continuous efforts of the design of Pt or Pd
nanoshells with one or more atomic multi-monolayers [59]. The
innovative solutions of Pt monolayers on the definite cores in the
nanosized ranges have been highly evaluated because the cores can
be rich and diverse in materials such as metal, bimetal, alloy, and
glass, as well as multi-composition materials [60,61]. In addition,
the cost of Pd is around 20% of the cost of Pt per weight, leading
to a significant reduction of the Pt loading. Importantly, the ORR of
Pd metal is very close to that of Pt metal in their catalytic activity
forward scan (i ) is commonly associated with the catalytic activity
f
for methanol electro-oxidation. For the fabricated catalyst samples,
there is a significant difference in the magnitude of i as well as their
f
corresponding voltage values. The high value of i is observed for
f
−
3
−2
Pt–Pt core–shell nanocatalyst (j = 1.5 × 10 A cm ). This current
response is located more negatively (E = 0.63 V) in terms of E vs.
f
(Ag/AgCl)/V than the rest, indicating the most effective catalyst for
the oxidation of methanol molecules. Based on other reports, the
obtained results are comparable considering that no carbon sup-
port was used and some inaccessible catalyst sites could be present
during measurement [70]. In some recent works, Pt–Pd alloy elec-
trocatalysts or Pt–Pd alloy electrocatalysts supported on various
carbon nanomaterials have been developed for methanol toler-
ant oxygen reduction as well as for an improvement of catalytic
activity and stability in methanol electro-oxidation, especially for
the stability of the core–shell structure [71,72]. Moreover, Pd–Pt
alloyed structures such as solid-solution alloys possess enhanced
hydrogen-storage capacity. The structures of Pd–Pt alloys were
reconstructed from the core–shell configuration through hydro-
gen adsorption and desorption [73]. Therefore, we can use Pt–Pd
core–shell nanocatalysts with shells fabricated with Pt monolayers
or Pd monolayers. Pt–Pd core–shell nanocatalysts are less expen-
sive and can replace the present technology of Pt–Ru nanocatalysts
[
62]. At present, a great deal of research in the preparation of the
electrocatalysis of Pt–Pd alloy nanocatalysts or Pt–Pd alloy nanocat-
alysts with or without various supports has been reported [63,64].
Recently, Pt-based alloy nanocatalysts (Pt–M, M = Co, Fe, Ni, Pd)
with cubic and octahedral morphologies have been prepared. For
example, cubic and octahedral Pt Ni catalysts exhibited higher ORR
3
activity than that of Pt/C, showing one of the highest known values
of ORR activity [65]. A new hydrogen absorption site in the hetero-
interface between the Pd core and Pt shell of the Pd–Pt core–shell
nanoparticle has been studied to improve hydrogen storage [66].

9
142
N.V. Long et al. / Electrochimica Acta 56 (2011) 9133–9143
0
0
0
0
.0015
.0010
.0005
.0000
a 140
120
Pt-Pd core-shell nanocatalysts
Pt nanocatalysts
100
8
0
0
0
0
6
4
2
0
2
000
4000
t / s
6000
30
40
50
60
70
80
90
2 theta / º
Fig. 8. Chronoamperometry data of Pt and Pt–Pd nanoparticles. Electrolyte solution
of 0.5 M H2SO4 + 1.0 M CH3OH and polarization potential about E = 0.5 V.
b 280
240
in the development of DMFC applications [74,75]. At the same time,
the electrocatalytic activity and selectivity of the Pt–Pd core–shell
nanocatalysts must be intensively studied to realize the DMFCs
with respect to the membrane electrode assembly (MEA) tech-
nology [9,10]. At present, DMFCs can work appropriately between
2
00
160
120
80
40
0
◦
the temperatures of 30 and 110 C with various proton exchange
membranes, such as Nafion 117, as the electrolyte for methanol
oxidation at the electrode surface. Fig. 8 shows the chronoamper-
ometry results taken from the Pt and Pt–Pd nanocatalysts. The
polarization potential was E = 0.5 V and was held over a period
of 2 h. Based on this graph, the Pt–Pd core–shell nanocatalysts
(
t = 15 min) exhibited an excellent current response much higher
3
0
40
50
60
70
80
90
than that of Pt nanocatalysts throughout the measurement. As the
polarization progressed, intermediate species accumulated on the
accesible catalytic sites. In effect, it was expected that the current
response would decline over time because of two likely reasons. The
first suggests that the stripping of intermediates is usually slower
than the rate at which methanol molecules are oxidized, and the
second showed that nanoparticles usually undergo particle coars-
ening during work operation [76]. However, these can be improved
if a second metal is incorporated and an optimum configuration
is achieved. In the case of the two samples, a stable characteriza-
tion was observed for Pt–Pd core–shell nanocatalysts. They showed
2
theta / º
Fig. 9. XRD patterns of (a) polyhedral Pt nanoparticles, and (b) Pt–Pd core–shell
nanoparticles.
appearance and shape; however, their intensities and peak posi-
tions were different. The (2 0 0) peaks indicated the alignment of
Pt and Pt–Pd core–shell nanoparticles of a narrow shape distribu-
tion on the planar substrate. The XRD peaks of the Pt nanoparticles
were broad and comparable to those of the corresponding bulk Pt
material in a system of (1 1 1), (2 0 0), (3 1 1), (2 2 2), (4 0 0), (3 3 1),
−3
−2
with 30.01%
a very high initial current of j = 1.29 × 10 A cm
current left after 2 h of polarization. Meanwhile, the Pt catalyst
exhibited the highest retained current of 3.67%, although its ini-
(
4 2 0), (4 2 2), (3 3 3), (5 1 1) or more (h k l). By using one XRD peak,
the average size of the Pt nanocrystallites can be calculated by
the width of the reflection according to the Debye–Scherrer equa-
tion: D = 0.9ꢁ/(ˇ cos ꢀ), where ˇ is the full width at half maximum
FWHM) of the peak, ꢀ is the angle of diffraction, and ꢁ is the
wavelength of the X-ray radiation. Here, the (2 2 0) reflections of
Pt nanoparticles were used to calculate the crystallite size.
−4
−2
tial current of j = 4.33 × 10 A cm is much lower. Recently, Pt–Pd
nanoparticles and their nanoalloying catalysts on various commer-
cial carbon black and multi-walled carbon nanotubes (MWCNs)
have been prepared in DMFCs. Importantly, these nanocatalysts
could be easily recovered from their final reaction solution via
a conventional filtration [77]. In addition, Pt–Pd nanocatalysts
will become a suitable choice of replacing Pt–Ru nanocatalysts
and reducing the high cost of Pt loading [78]. Because of their
importance, Pt–Pd bimetallic nanoparticles prepared by the polyol
method were intensively investigated to show their high activity
for ORR [79–81]. Our next studies will emphasize the problems of
Pt-based multi-composition nanocomposite catalysts.
(
4. Conclusions
In this study, we controllably synthesized polyhedral Pt
nanoparticles and Pt–Pd core–shell nanoparticles via a modi-
fied polyol method using AgNO . The overgrowth of the thin Pd
3
Fig. 9 shows the two XRD patterns of the prepared Pt and Pt–Pd
core–shell nanoparticles. The results showed a crystalline Pt face
centered cubic (fcc) phase. The peaks were characterized by (1 1 1),
nanoshells on the Pt nanocores due to epitaxial growth mode was
observed. We predicted that Pt–Pd core–shell nanoparticles were
formed in both the FM and SK growth modes. In particular, the
core–shell nanoparticles with a thin Pd monolayer shell demon-
strated potential as an electrocatalyst with great stabilization and
reliability, far exceeding Pt nanoparticles with the same polyhedral
morphologies.
(
2 0 0), (2 2 0), (3 1 1), and (2 2 2) peaks corresponding to 2ꢀ values
◦
of ∼40.9, 46.4, 67.9, 81.8, and 86.2 , respectively [20]. Therefore,
the results are the agreement with the TEM data. Two XRD pat-
terns of Pt and Pt–Pd core–shell nanoparticles showed the same

N.V. Long et al. / Electrochimica Acta 56 (2011) 9133–9143
9143
Acknowledgements
[39] Y. Bing, H. Liu, L. Zhang, D. Ghosh, J. Zhang, Chem. Soc. Rev. 39 (2010) 2184.
[
[
[
40] Y. Qiao, C.M. Li, J. Mater. Chem. 21 (2011) 4027.
41] Y. Lu, W. Chen, Chem. Commun. 47 (2011) 2541.
42] A. Lebon, A. García-Fuente, A. Vega, F. Aguilera-Granja, Phys. Rev. B 83 (2011)
125427.
We thank the Institute of Ceramics Research and Education
(
ICRE) and the Nagoya Institute of Technology (NITECH) for sup-
[
43] E.A. Baranova, N. Miles, P.H.J. Mercier, Y.L. Page, B. Patarachao, Electrochim.
Acta 55 (2010) 8182.
porting our research program in science and nanotechnology in
Japan. We also thank the Kyushu University, the Novel Carbon
Resources Sciences, the Kyushu University Global GOE Program for
kind support and help.
[
44] J.X. Wang, H. Inada, L. Wu, Y. Zhu, Y. Choi, P. Liu, W. Zhou, R.R. Adzic, J. Am.
Chem. Soc. 131 (2009) 17298.
[45] S.J. Hwang, S.J. Yoo, S. Jang, T. Lim, S.A. Hong, S. Kim, J. Phys. Chem. C 115 (2011)
483.
2
[
46] P. Fang, S. Duan, X. Lin, J.R. Anema, J. Li, O. Buriez, Y. Ding, F. Fan, D. Wu, B. Ren,
Z.L. Wang, C. Amatore, Z. Tian, Chem. Sci. 2 (2011) 531.
47] D. Wang, H.L. Xin, Y. Yu, H. Wang, E. Rus, D.A. Muller, H.D. Abru n˜ a, J. Am. Chem.
References
[
[
[
[
[
[
1] A. Chen, P. Holt-Hindle, Chem. Rev. 110 (2010) 3767.
2] E. Antolini, Energy Environ. Sci. 2 (2009) 915.
3] A.A. Gewirth, M.S. Thorum, Inorg. Chem. 49 (2010) 3557.
4] Z. Peng, H. Yang, Nano Today 4 (2009) 143.
5] R.R. Adzic, J. Zhang, K. Sasaki, M.B. Vukmirovic, M. Shao, J.X. Wang, A.U. Nilekar,
M. Mavrikakis, J.A. Valerio, F. Uribe, Top. Catal. 46 (2007) 249.
6] N.M. Markovic, P.N. Ross, Surf. Sci. Rep. 45 (2002) 117.
7] B. Corain, G. Schmid, N. Toshima, Metal Nanoclusters in Catalysis and Materials
Science: The Issue of Size Control, Elsevier, Amsterdam, 2004.
8] S. Basu, Recent Trends in Fuel Cell Science and Technology, Anamaya Publishers,
New Delhi, India, 2007.
Soc. 132 (2010) 17664.
[48] H. Lee, S.E. Habas, G.A. Somorjai, P. Yang, J. Am. Chem. Soc. 130 (2008) 5406.
[49] J. Zhang, M.B. Vukmirovic, K. Sasaki, A.U. Nilekar, M. Mavrikakis, R.R. Adzic, J.
Am. Chem. Soc. 127 (2005) 12480.
[50] J.R. Croy, S. Mostafa, L. Hickman, H. Heinrich, B.R. Cuenya, Appl. Catal. A 350
(2008) 207.
[51] J. Zhao, A. Manthiram, Appl. Catal. B 101 (2011) 660.
[52] B. Wu, Y. Kuang, X. Zhang, J. Chen, Nano Today 6 (2011) 75.
[53] Y. Sung, J. Hwang, J.S. Chung, Int. J. Hydrogen Energy 36 (2011) 4007.
[54] B. Gurau, R. Viswanathan, R. Liu, T.J. Lafrenz, K.L. Ley, E.S. Smotkin, E. Redding-
ton, A. Sapienza, B.C. Chan, T.E. Mallouk, S. Sarangapani, J. Phys. Chem. B 102
(1998) 9997.
[55] C. Zhong, J. Luo, B. Fang, B.N. Wanjala, P.N. Njoki, R. Loukrakpam, J. Yin, Nano-
technology 21 (2010) 062001.
[56] J.K. Nørskov, J. Rossmeisl, A. Logadottir, L. Lindqvist, J. Kitchin, T. Bligaard, H.
Jónsson, J. Phys. Chem. B 108 (2004) 17887.
[57] J.K. Nørskova, F. Abild-Pedersen, F. Studt, T. Bligaard, Density Func-
tional Theory in Surface Chemistry and Catalysis, www.pnas.org/cgi/doi/
10.1073/pnas.1006652108, pp. 1–7.
[58] C. Kulp, X. Chen, A. Puschhof, S. Schwamborn, C. Somsen, W. Schuhmann, M.
Bron, Chem. Phys. Chem. 11 (2010) 2854.
[59] K. Sasaki, H. Naohara, Y. Cai, Y.M. Choi, P. Liu, M.B. Vukmirovic, J.X. Wang, R.R.
Adzic, Angew. Chem. Int. Ed. 49 (2010) 8602.
[60] Y. Xing, Y. Cai, M.B. Vukmirovic, W. Zhou, H. Karan, J.X. Wang, R.R. Adzic, J. Phys.
Chem. Lett. 1 (2010) 3238.
[61] A. Yin, Y. Zhang, C. Yan, J. Am. Chem. Soc. 133 (2011) 3816.
[62] B. Lim, M. Jiang, T. Yu, P.H.C. Camargo, Y. Xia, Nano Res. 3 (2010) 69.
[63] X. Zhou, L. Fan, Electrochim. Acta 55 (2010) 8111.
[64] W. He, M. Chen, Z. Zou, Z. Li, X. Zhang, S. Jin, D.J. You, C. Pak, H. Yang, Appl. Catal.
B 97 (2010) 347.
[
[
[
[
9] S. Bose, T. Kuila, T.X.H. Nguyen, N.H. Kim, K. Lau, J.H. Lee, Prog. Polym. Sci. 36
(
2011) 813.
[
[
10] E. Antolini, Appl. Catal. B 88 (2009) 1.
11] A.M. Zainoodin, S.K. Kamarudin, W.R.W. Daud, Int. J. Hydrogen Energy 35 (2010)
4
606.
[
[
[
[
12] S. Alayoglu, A.U. Nilekar, M. Mavrikakis, B. Eichhorn, Nat. Mater. 7 (2008) 333.
13] B.G. Pollet, Int. J. Hydrogen Energy 35 (2010) 11986.
14] H. Yang, Angew. Chem. Int. Ed. 50 (2011) 2.
15] K. Sasaki, J.X. Wang, H. Naohara, N. Marinkovic, K. More, H. Inada, R.R. Adzic,
Electrochim. Acta 55 (2010) 2645.
16] Y. Wang, N. Toshima, J. Phys. Chem. B 101 (1997) 5301.
17] D. Kaplan, M. Alon, L. Burstein, Y. Rosenberg, E. Peled, J. Power Sources 196
2011) 1078.
18] H. Wang, C. Xu, F. Cheng, M. Zhang, S. Wang, S.P. Jiang, Electrochem. Commun.
0 (2008) 1575.
19] N.V. Long, N.D. Chien, T. Hayakawa, H. Hirata, G. Lakshminarayana, M. Nogami,
Nanotechnology 21 (2010) 035605.
20] H. Song, F. Kim, S. Connor, G.A. Somorjai, P. Yang, J. Phys. Chem. B 109 (2005)
[
[
(
[
[
[
1
1
88.
[
[
[
[
[
[
[
21] N. Tian, Z.Y. Zhou, S.G. Sun, J. Phys. Chem. C 112 (2008) 19801.
22] M. Subhramannia, V.K. Pillai, J. Mater. Chem. 18 (2008) 5858.
23] K. Lee, M. Kim, H. Kim, J. Mater. Chem. 20 (2010) 3791.
24] C. Susut, T.D. Nguyen, G.B. Chapman, Y. Tong, Electrochim. Acta 53 (2008) 6135.
25] R. Ferrando, J. Jellinek, R.L. Johnston, Chem. Rev. 108 (2008) 845.
26] J. Zhang, Y. Tang, K. Lee, M. Ouyang, Science 327 (2010) 1634.
27] J. Zhang, F.H.B. Lima, M.H. Shao, K. Sasaki, J.X. Wang, J. Hanson, R.R. Adzic, J.
Phys. Chem. B 109 (2005) 22701.
[65] J. Wu, A. Gross, H. Yang, Nano Lett. 11 (2011) 798.
[66] M. Yamauchi, H. Kobayashi, H. Kitagawa, Chem. Phys. Chem. 10 (2009) 2566.
[67] S. Jingyu, H. Jianshu, C. Yanxia, Z. Xiaogang, Int. J. Electrochem. Sci. 2 (2007) 64.
[68] S. Wei, D. Wu, X. Shang, R. Fu, Energy Fuels 23 (2009) 908.
[69] K. Kakaei, M. Zhiani, H. Gharibi, J. Phys. Chem. C 114 (2010) 5233.
[70] R.G. Freitas, E.P. Antunes, E.C. Pereira, Electrochim. Acta 54 (2009) 1999.
[71] W. Wang, Q. Huang, J. Liu, Z. Zou, Z. Li, H. Yang, Electrochem. Commun. 10
(2008) 1396.
[72] P. Ercius, T. Duden, V.R. Radmilovic, Microsc. Microanal. 16 (2010) 1650.
[73] H. Kobayashi, M. Yamauchi, H. Kitagawa, Y. Kubota, K. Kato, M. Takata, J. Am.
Chem. Soc. 132 (2010) 5576.
[74] M. Carmo, A.R. Santos, J.G.R. Poco, M. Linardi, J. Power Sources 173 (2007)
860.
[
[
28] L. Carbone, P.D. Cozzoli, Nano Today 5 (2010) 449.
29] K.R. Cooper, V. Ramani, J.M. Fenton, H.R. Kunz, Experimental Methods and Data
Analyses for Polymer Electrolyte Fuel Cells, Scribner Associates Inc., 2005.
30] Z. Liu, J.Y. Lee, W. Chen, M. Han, L.M. Gan, Langmuir 20 (2004) 181.
31] R. Gomez, A. Fernandez–Vega, J.M. Feliu, A. Aldaz, J. Phys. Chem. 97 (1993) 4769.
32] N. Furuya, S. Koide, Surf. Sci. 220 (1989) 18.
[
[
[
[
[75] I. Avila–Garcia, C. Ramirez, J.M.H. Lopez, E.M.A. Estrada, J. Alloys Compd. 495
(2010) 462.
33] T.J. Schmidt, V. Stamenkovic, M. Arenz, N.M. Markovic, J.P.N. Ross, Electrochim.
Acta 47 (2002) 3765.
[76] A.M. Remona, K.L.N. Phani, J. Fuel Cell Sci. Technol. 8 (2011) 011001.
[77] X. Chen, H. Wang, J. He, Y. Cao, Z. Cui, M. Liang, J. Nanotechnol. 10 (2010) 3138.
[78] O.A. Petrii, J. Solid State Electrochem. 12 (2008) 609.
[79] B. Lim, J. Wang, P.H.C. Camargo, M. Jiang, M.J. Kim, Y. Xia, Nano Lett. 8 (2008)
2535.
[80] N.V. Long, T. Asaka, T. Matsubara, M. Nogami, Acta Mater 59 (2011) 2901.
[81] N.V. Long, T.D. Hien, T. Asaka, M. Ohtaki, M. Nogami, Int. J. Hydrogen Energy 36
(2011) 8478.
[34] Z.L. Wang, J. Phys. Chem. B 104 (2000) 1153.
[35] I.
Ávila–García,
M.
Plata–Torres,
M.A.
Domínguez–Crespo,
C.
Ramírez–Rodríguez, E.M. Arce–Estrada, J. Alloys Compd. 434 (2007) 764.
36] L. Luo, M. Futamata, Electrochem. Commun. 8 (2006) 231.
37] Y. Lee, A. Ko, S. Han, H. Kim, K. Park, Phys. Chem. Chem. Phys. 13 (2011) 5569.
38] J. Solla–Glullón, V. Montiel, A. Aldaz, J. Clavilier, Electrochem. Commun. 4 (2002)
[
[
[
7
16.