Seed-mediated growth of palladium nanocrystals on indium tin oxide surfaces and their applicability as modified electrodes

Article abstract of DOI:10.1021/jp0635157

Palladium nanoparticles (PdNPs) were successfully attached and grown on an indium tin oxide (ITO) surface using a seed-mediated growth method, i.e., via a simple two-step immersion of the ITO substrate into the seed and growth solutions. After the growth treatment for 24 h, PdNPs grew up to 60-80 nm, exhibiting crystal-like appearances and accompanying the formation of short rodlike nanocrystals as a minor product. Thus prepared PdNPs tend to stick each other, so that the dense gathering of PdNPs was observed on the ITO surfaces. Due to the dense attachment, the PdNPs directly attached to the ITO (PdNP/ITO) electrode had a significantly lowered charge-transfer resistivity compared with that of a bareITO, and the redox reaction of [Fe(CN)₆] ^3-/[Fe(CN)₆]^4- was observed as reversible in 0.1 M phosphate buffer solution. The electrocatalytic property of PdNPs was confirmed for the reduction of oxygen. In addition, some typical responses were observed in 0.5 M H₂SO₄ with the PdNP/ITO electrode, reflecting both the characteristics of NPs and the thin layer in nanoscale. The present preparation method of PdNP-attached surfaces would be promising for catalytic applications as well as electrochemical uses.

Full text of DOI:10.1021/jp0635157

20362
J. Phys. Chem. B 2006, 110, 20362-20368
Seed-Mediated Growth of Palladium Nanocrystals on Indium Tin Oxide Surfaces and
Their Applicability as Modified Electrodes
†
,‡
†
Gang Chang, Munetaka Oyama,* and Kazuyuki Hirao
Department of Material Chemistry, Graduate School of Engineering, and DiVision of Research InitiatiVes,
International InnoVation Center, Kyoto UniVersity, Nishikyo-ku, Kyoto 615-8520, Japan
ReceiVed: June 7, 2006; In Final Form: July 27, 2006
Palladium nanoparticles (PdNPs) were successfully attached and grown on an indium tin oxide (ITO) surface
using a seed-mediated growth method, i.e., via a simple two-step immersion of the ITO substrate into the
seed and growth solutions. After the growth treatment for 24 h, PdNPs grew up to 60-80 nm, exhibiting
crystal-like appearances and accompanying the formation of short rodlike nanocrystals as a minor product.
Thus prepared PdNPs tend to stick each other, so that the dense gathering of PdNPs was observed on the ITO
surfaces. Due to the dense attachment, the PdNPs directly attached to the ITO (PdNP/ITO) electrode had a
significantly lowered charge-transfer resistivity compared with that of a bare ITO, and the redox reaction of
3
-
4-
[
Fe(CN)
property of PdNPs was confirmed for the reduction of oxygen. In addition, some typical responses were
observed in 0.5 M H SO with the PdNP/ITO electrode, reflecting both the characteristics of NPs and the
6
] /[Fe(CN)
6
]
was observed as reversible in 0.1 M phosphate buffer solution. The electrocatalytic
2
4
thin layer in nanoscale. The present preparation method of PdNP-attached surfaces would be promising for
catalytic applications as well as electrochemical uses.
cal characteristics were observed.9
-11
The advantageous char-
Introduction
acteristic of thus formed AuNPs directly attached to the ITO
AuNP/ITO) surface or the AgNP/ITO surface is that the
Along with the increasing interests in the size- and shape-
dependent properties of metal nanoparticles (NPs) [e.g., see the
review on gold NPs (AuNPs)], modification of the electrode
(
1
electron-transfer resistivities are lower than those of the metal
NP-modified electrodes prepared using linker molecules.1 In
addition, the weakly capped surface of AuNPs is suitable for
0,11
surfaces with metal NPs has been extensively carried out
nowadays. While the electrochemical reductive deposition of
noble metal ions is a typical approach to prepare metal NPs on
1
2,13
the further modification, e.g. hemeproteins.
conducting surfaces,2-4 the use of linker molecules, which have
In the present work, as a series of our trials to fabricate metal-
NP-attached electrodes, the attachment and growth of palladium
NPs (PdNPs) on ITO surfaces are investigated using the same
5
bonding groups for both metal NPs and the substrate, have
been a popular method in recent years to attach metal NPs on
electrode surfaces via pure chemical treatments. Assuming the
attached metal NPs as a nanoelectrode ensemble, the fine-tuning
is possible by controlling the conditions for the chemical
seed-mediated growth method.
PdNPs are useful for catalytic purposes,1
7-19
so that active
researches have been devoted to the preparation methods of
20-23
6
,7
PdNPs in solution phase.
In addition, shape-controlled wet
preperations.
synthesis of unique PdNPs, i.e., nanocubes and nanoplates, has
As a novel approach to prepare metal NP-attached electrodes
2
4,25
been reported.
However, in contrast to the synthesis in
with wet chemical treatments, our group is proposing a seed-
mediated growth method for surface modification,8-13 which
solution, research works using the PdNP-modified electrodes
are not numerous. The electrochemical deposition was utilized
is an alteration of the original seed-mediated method for pre-
2
3
paring metal nanorods in solution.1
4,15
for the preparations of PdNPs on glassy carbon and gold elec-
trodes by Wang and Dong’s group. Also, Dong and co-workers
proposed a simple preparation method of multilayer polymer
films containing PdNPs on a 4-aminobenzoic acid-modified GC
In actual, the nanoseed
particles are attached on the surface by just immersing a piece
of conducting substrate, such as indium tin oxide (ITO), into a
seed solution containing metal NPs of ca. 4 nm. Then, by
immersing the nanoseed-particle-attached substrate into the
growth solution, which contained metal ions, ascorbic acid as
reductant, and cetyltrimethylammonium bromide (CTAB), the
crystal growth of metal NPs on the substrate’s surface is
possible. While the second process is chemical reductive growth,
the first process is physisorption of small nanoseed particles,
26
electrode. G o´ mez et al. reported the surface-enhanced Raman
spectroscopic results on nanostructured electrode prepared by
depositing PdNPs on either gold or platinum electrodes.27
However, compared with the cases of AuNPs and AgNPs, in
general, the attachment method of PdNPs on the electrode
surfaces are very limited, which is presumably because linker
molecules suitable for PdNPs are not established.
16
which is known as a precursor method in the field of catalysis.
This method was successfully applied to the attachment of
Despite of this scarceness of PdNP-modified electrodes, the
Pd nanostructured films are electrochemically interesting objec-
tives because they can be used for the detection of combustible
8
9
AuNPs and AgNPs on the ITO surfaces, and the electrochemi-
*
Corresponding author. Tel: +81-75-383-3074. Fax: +81-75-383-3031.
28
29
gases and nitrous oxide, surface enhanced Raman measure-
E-mail: oyama@iic.kyoto-u.ac.jp.
3
0
†
‡
ments, and elucidating the hydrogen adsorption and absorption
Department of Material Chemistry.
Division of Research Initiatives.
31
processes. Thus, if we can form the nanostructured ensembles
1
0.1021/jp0635157 CCC: $33.50 © 2006 American Chemical Society
Published on Web 09/12/2006

Seed-Mediated Growth of Palladium Nanocrystals
J. Phys. Chem. B, Vol. 110, No. 41, 2006 20363
of PdNPs without using strong binders or linker molecules, the
potential applicability of the formed Pd nanomaterials would
be significant including the uses as catalysts.
The first aim of the present paper is to show what kind of
nanostructured particles are formed in the case of Pd using the
seed-mediated growth method. While this method brought the
successes in attaching and growing AuNPs and AgNPs on
ITO,8 it was not valid for the case of platinum NPs (PtNPs),
so that we reported a different approach for fabricating the PtNP/
ITO electrode.32 However, in the case of Pd, the seed-mediated
growth has found to be applicable to allow the crystal growth
of Pd. Because the grown nanostructure of Pd was different
,9
8,9
from those of AuNPs and AgNPs, the structural differences
are described on the basis of the SEM images.
The second aim of the present work is to investigate the
properties of Pd nanocrystals grown on the ITO surfaces through
the electrochemical measurements. By using a thus prepared
PdNP/ITO electrode as the working electrode, the electrochemi-
cal characteristics and applicability are explored.
Experimental Section
Materials. Cetyltrimethylammonium bromide (CTAB) and
K2PdCl6 (99.998%) were purchased from Aldrich Co., Ltd.
NaH2PO4 and Na2HPO4 from Wako Pure Chemicals, Ltd. were
used to prepare phosphate buffer solutions. Other reagents were
obtained from Wako Pure Chemicals. Glass plates on which
ITO film was coated were purchased from CBC Ings Ltd. (sheet
resistance ca. 50 Ω sqr., size 18 mm × 18 mm). A piece of the
ITO glass was cut into four pieces and used after washing in
acetone, ethanol, and pure water with sonication and drying with
nitrogen gas. A conventional Pd bulk electrode (Pd disk; the
diameter, 0.16 cm) was purchased from BAS Inc. It was
polished to be mirrorlike, finished by a diamond suspension
(250 nm), and washed carefully by the same method as for ITO.
The reference electrode used was Ag/AgCl in 3 M NaCl
purchased from BAS Inc. In all the procedures and the prepa-
ration of solution, we used pure water prepared using Millipore
Autopure WR600A (Millipore, Ltd., resistivity > 18 MΩ cm).
Apparatus. The surface morphology of the PdNPs attached
on the ITO surfaces was observed with a field-emission scanning
electron microscopy (FE-SEM, JSM-7400F, JEOL). The electro-
chemical measurements were performed using an EG&G
M263A potentiostat/galvanostat (Princeton Applied Research
Figure 1. Typical FE-SEM images of Pd nanocrystals attached and
grown on the ITO surfaces observed after (A) immersing only into the
seed solution for 2 h and immersing into the seed solution for 2 h and
then into the growth solution for (B) 4 h and (C) 12 h.
Fabrication of PdNP/ITO Electrodes. To evaluate the
electrochemical properties of the PdNPs directly attached to ITO
(PdNP/ITO), a handmade PdNP/ITO working electrode was
(
PAR)) and a 5210 lock-in amplifier (PAR) controlled by a
fabricated as follows. At first, a piece of ITO on which PdNPs
were attached was connected to a strip of copper adhesive tape
computer with M270 and M398 programs (PAR). All the
electrochemical measurements were carried out at room tem-
perature, ca. 25 °C.
Procedures. Typical procedures of seed-mediated growth
method are as follows. First, the seed solution of ca. 4 nm Pd
colloid was prepared by adding 0.5 mL of ice-cold solution of
(
3M copper tape 1181, 50 mm × 6 mm) and then molded into
a piece of tape (3M filament tape 898, 50 mm × 18 mm) that
was made to have a 0.2-cm-diameter hole. For the electrochemi-
cal measurements, the exposed area of the diameter of 0.20 cm
was the working surface to contact with the sample solution.
0
.01 M NaBH4 into 18 mL of aqueous solution containing 0.25
mM K2PdCl6 and 0.25 mM trisodium citrate. A piece of the
ITO substrate was then immersed into the seed solution and
left for 2 h in order to attach the Pd nanoseed particles on the
surface. Then, ITO substrate with Pd nanoseeds attached was
immersed in the growth solution and left for 4, 12, or 24 h.
The growth solution was prepared by mixing the aqueous
solutions of 0.5 mL of 10 mM K2PdCl6, 18 mL of 0.10 M
CTAB solution, 0.1 mL of 1.0 M NaOH solution, and 0.1 mL
of 0.1 M ascorbic acid. Thus ITO fabricated with Pd nano-
crystals attached was removed from the growth solution, rinsed
with pure water several times, and then dried with nitrogen gas
for the FE-SEM characterizations and the electrochemical
measurements. All the wet chemical treatments were carried
out at room temperature, ca. 25 °C.
Results and Discussion
FE-SEM Observation of PdNP/ITO Surfaces. First, we
would like to show how the surface morphology changed in
the seed-mediated growth treatments, in particular, depending
on the immersion time into the growth solution. A typical FE-
SEM image observed just after immersing into the seed solution
is shown in Figure 1A. In this image, bright dots correspond to
the Pd nanoseed particles attached on the ITO surface, while
the background structure is that of ITO crystals. Although the
nanoseed particles were difficult to recognize in the case of
AuNPs and AgNPs on the ITO surfaces due to the resolution
8
,9
of the FE-SEM, some of the Pd nanoseed particles were
observable, as shown in this image.

20364 J. Phys. Chem. B, Vol. 110, No. 41, 2006
Chang et al.
Figure 3. UV-visible absorbance spectrum of the PtNP/ITO fabricated
with the growth time of 24 h.
To verify this hypothesis, we tried the growth of Pd after the
seeding of Au and consequently confirmed that the same
nanocrystal growth of Pd occurred. Thus, not depending on the
seeding process, PdNPs are expected to grow with increased
time in the growth solution.
The appearances of PdNPs were quite crystal-like after being
well grown, as shown in Figures 1C and 2A. So, we observed
the XRD spectra for the surface of PdNP/ITO. As a result, the
peaks of (111), (200), and (220) surfaces were observed in the
XRD pattern, showing the formation of typical face-centered
cubic (fcc) crystals. Therefore, the present treatment is consid-
ered as a practical method to gather Pd nanocrystals on the
supporting material with high density. As pointed out for the
growth of AuNPs with and without linker molecules, such a
formation of less capped nanocrystals is peculiar to our seed-
8
mediated growth.
In addition, utilizing the transparent characteristics of the ITO
substrate, the UV-visible absorption spectrum was observed
for the PdNP/ITO prepared via 24 h of growth. As the result,
a broad absorption peak was observed with the maximum wave-
length of 420 nm, as shown in Figure 3. This value would be
reasonable for the PdNPs of 60-80 nm, judging from the
absorption maxium reported for the 50-nm Pd nanocubes. Thus,
the optical characteristics of NPs are inferred to be reserved
despite the gathering of PdNPs.
Because the zoomed-out images (Figure 2B,C) shows that a
kind of porous surface composed of Pd nanocrystals was formed
on the ITO surface, the modified nanomaterials might be
effective for the electrochemical measurements due to the large
surface area or porous characteristics. Thus, in the following
sections, we observe some electrochemical responses using the
PdNP/ITO as the working electrode. While the Pd bulk working
electrode is used to compare the electrochemical responses, the
surface of the bulk metal is flat enough in comparison with the
PdNPs attached surfaces.
Figure 2. Typical FE-SEM images of Pd nanocrystals attached and
grown on the ITO surfaces observed after immersing into the seed
solution for 2 h and then into the growth solution for 24 h. The images
are shown in three different scales.
Parts B and C of Figure 1 show typical FE-SEM images
recorded after treating the Pd-nanoseed-particle-attached ITO
substrates in the growth solution for 4 and 12 h, respectively.
As apparent from the figures, the crystal growth of PdNPs from
Pd nanoseeds on the ITO crystals was observed with the increase
in the immersion time into the growth solution. While the size
of grown Pd nanocrystals was around 15 nm after 4 h of growth,
the size after 12 h was roughly 40 nm. In both cases, some
short nanorods were observed to form as a minor product.
Figure 2 shows typical FE-SEM images recorded after
treatment in the growth solution for 24 h. As shown in the
expanded image (Figure 2A), Pd nanocrystals grew up to 60-
Electrochemical Impedance of PdNP/ITO Electrodes
4-
3-
8
0 nm. The most characteristic feature of the formed nano-
Measured for the [Fe(CN)6] /[Fe(CN)6] Couple. To evalu-
ate the effect of Pd nanocrystals on the charge-transfer resistiv-
ity, we measured the electrochemical impedance of the PdNP/
ITO electrodes that were prepared with different growth periods.
Electrochemical impedance spectra were recorded in 0.1 M
phosphate buffer solution (pH 7.0) containing 0.5 mM
structure of Pd is that Pd nanocrystals tend to stick to each other.
In addition, the stuck gathering was not only on the surface but
three-dimensional, i.e., some nanocrystals were observed to grow
above the basal nanocrystals. As recognized when we compare
with the images in Figure 1B,C, the degree of the gathering
increased with the growth time. The nanostructure of Pd grown
on ITO was quite different from the cases of Au and Ag, because
3
-
4-
[Fe(CN)6] and 0.5 mM [Fe(CN)6]. As shown in Figure 4,
the Rct values were observed to decrease with the increase in
the growth period, reflecting the correlation of the Rct values
with the degree of the crystal growth of PdNPs in the FE-SEM
images (Figures 1 and 2). The Rct values were determined to
8
,9
AuNPs and AgNPs tend to form the dispersed states. Hence,
it is inferred that the identity of metal changed such gathering
characteristics in the growth process, even though the same
capping reagent, CTAB, was uses in all the growth procedures.
2
be 0.41 kΩ cm for the bare ITO electrode and 0.31, 0.23, and

Seed-Mediated Growth of Palladium Nanocrystals
J. Phys. Chem. B, Vol. 110, No. 41, 2006 20365
Figure 4. Electrochemical impedance spectra recorded at 0.22 V in
0
.1 M phosphate buffer solution (pH 7.0) containing 0.5 mM
3
-
4-
[Fe(CN)
6
]
and 0.5 mM [Fe(CN)
6
]
with use of bare ITO electrode
(a) and PdNP/ITO electrode prepared via (b) 4 h, (c) 12h, and (d) 24 h
of growth.
.14 kΩ cm for the PdNP/ITO electrodes prepared with a
2
0
growth time of 4, 12, and 24 h, respectively. For a reference,
2
that of the Pd bulk electrode was ca. 17 kΩ cm .
It is interesting that, while the Rct value of the Pd bulk
electrode was quite large, the increase of PdNPs on ITO was
found to reduce the Rct values. The decrease in Rct with the
attachment of NPs is similar to the cases of AuNPs and
]4 in (A) 1.0
-
Figure 5. Cyclic voltammograms of 1.0 mM [Fe(CN)
6
M KCl solution and (B) 0.1 M phosphate buffer solution (pH 7.0) with
a) a bare ITO electrode, (b) a bulk Pd electrode, and (c) PdNP/ITO
(
-1
electrode prepared via 24 h of growth. Scan rate, 50 mV s . Geometric
AgNPs.9
-11
The resistivity of Pd metal (10.8 × 10 Ω cm) is
-6
2
electrode areas: (a, c) 0.031 and (b) 0.020 cm . The current axis was
-
6
greater than that of Au and Ag metals (2.4 × 10 and 1.62 ×
normalized by dividing the surface areas.
-
6
1
0
Ω cm). However, the results of the impedance measure-
ment showed that the charge transfer resistivity was significantly
lowered by the crystal growth of PdNPs, and that the lowered
degree was comparable to the cases of AuNPs and AgNPs after
phosphate buffer solution. Hence, it is considered that the PdNP/
ITO has some additional effects to exhibit the reversible
response of Figure 5B, curve c.
9-11
4-
3-
24 h of growth.
This lowered Rct value would be a reflection
In the redox reaction of [Fe(CN)6] /[Fe(CN)6] , the re-
sponses should be governed by the diffusion processes, and the
diffusion layer is expected to be far more than the thickness of
the layer of PdNPs. Thus, the improvement characteristics of
the PdNP/ITO electrode cannot be explained with changes in
the increased surface of Pd in PdNPs. As shown by the imped-
ance results, the Rct values were 0.14, 0.41, and ca. 17 kΩ cm2
for the PdNP/ITO, base ITO, and Pd bulk electrodes, respec-
tively. Hence, by attaching on the surface of ITO, PdNPs are
inferred to change the nature of the interfaces to reduce the
charge-transfer resistivity. In addition, the porous surface of
PdNPs may reduce the adsorption of phosphate ions.
of the dense and more than monolayer attachment of Pd
nanocrystals, as shown in Figure 2, in which the state of Pd is
quite different from that in the Pd bulk electrode.
In addition, we carried out the electrochemical impedance
measurement also in 1.0 M KCl solution containing 0.5 mM
3
-
4-
[
Fe(CN)6] and 0.5 mM [Fe(CN)6]. The resulted plots
showed straight lines for all the PdNP/ITO electrodes, implying
the low charge-transfer resistivity in 1.0 M KCl. For a com-
2
parison, the Rct values were ca. 0.03 and 0.28 kΩ cm for the
bare ITO and Pd bulk electrodes, respectively.
Cyclic Voltammograms of the [Fe(CN)6] /[Fe(CN)6]
4-
3-
Couple on PdNP/ITO Electrodes. To investigate the appli-
cability of the PdNP/ITO electrode for typical redox systems,
cyclic voltammograms of [Fe(CN)6]4 were observed in 1.0 M
KCl and 0.1 M phosphate buffer solution. Hereafter, the PdNP/
ITO electrode used was that prepared after 24 h of growth. The
recorded voltammograms are shown in Figure 5 together with
those obtained with the bare ITO and Pd bulk electrodes.
In 1.0 M KCl solution, the redox behaviors were essentially
same for all the electrodes, except the currents observed at the
At any rate, as the most noticeable feature here, it was found
that the PdNP/ITO was superior to the Pd bulk electrode in
terms of the reversibility. If we assume only the surface of Pd
is working as electrode on the PdNP/ITO, such a difference is
difficult to consider. Thus, it is inferred that a unique interface
could be fabricated by attaching PdNPs on the ITO surface.
This result is similar to the case of AgNP/ITO electrode for the
-
4
-
3- 9
redox of [Fe(CN)6] /[Fe(CN)6] .
Electrocatalytic Property of PdNP/ITO Electrode for the
Reduction of Oxygen. Because the electrocatalytic property
of PdNPs for oxygen is well-known and examined in previous
4-
3-
sides of the redox peaks of the [Fe(CN)6] /[Fe(CN)6] couple,
as shown in Figure 5A. This result indicates that the difference
was not significant in 1.0 M KCl.
3
,26
works,
we observed the voltammetric responses with the
In contrast, an interesting difference was observed in 0.1 M
phosphate buffer, as shown in Figure 5B. That is, reversible
redox peaks were best observed with the PdNP/ITO electrode
present PdNP/ITO electrode. Figure 6 shows the cyclic voltam-
mograms recorded with the PdNP/ITO, Pd bulk, and bare ITO
electrodes in the air-saturated 0.1 M KCl solution, together with
that recorded with the PdNP/ITO electrode in the N2-saturated
0.1 M KCl solution.
As shown in this figure, an apparent catalytic reduction
response was observed for the PdNP/ITO and Pd bulk electrodes
having almost the same peak potential, though the current
magnitude was different. Compared with the Pd bulk electrode,
the peak current was ca. 1.8 times larger with the PdNP/ITO
(curves c), while the peaks were quasireversible with the bare
ITO electrodes (curves a) or less appeared with the Pd bulk
electrodes (curves b). Judging from the differences in the Rct
values between 1.0 M KCl and 0.1 M phosphate buffer solution
mentioned in the previous section and the probability of the
absorption of phosphate ion, etc., the quasireversible responses
in Figure 5B, curves a and b, may be reasonable in 0.1 M

20366 J. Phys. Chem. B, Vol. 110, No. 41, 2006
Chang et al.
Figure 6. Cyclic voltammograms recorded with the PdNP/ITO
electrode prepared via 24 h of growth for (a) air-saturated and (b) N
2
saturated 0.1 M KCl solutions and with (c) a Pd bulk electrode and (d)
a bare ITO electrode for air-saturated 0.1 M KCl solution. Scan rate:
-
1
2
5
0 mV s . Geometric electrode area: (a, b, d) 0.031 cm and (c) 0.020
2
cm . The current axis was normalized by dividing the surface areas.
Figure 8. Cyclic voltammograms obtained with (A) the PdNP/ITO
electrode and (B) the Pd bulk electrode in the solution containing 0.5 M
H
2
SO
4
. In part A, the potential was reversed at various potentials and
-
1
in part B at -0.28 V. Scan rate: 50 mV s . Geometric electrode area:
2
2
(A) 0.031 cm and (B) 0.020 cm . The current axis was normalized by
dividing the surface areas.
ITO electrodes were roughly similar, though a slight change of
the wave-shape and the magnitude of the reduction current was
observed (which are discussed later).
In contrast, in the negative-going scan for the potential region
of hydrogen adsorption and absorption, remarkable differences
were observed concerning the onset and the magnitude of the
reduction current (Figure 8). While the increase of the reduction
current started from ca. -0.2 V and the current magnitude was
limited even at -0.28 V as in Figure 8A, the onset of the reduc-
tion with the Pd bulk electrode was observed from ca. 0 V, and
the reduction current increased monotonically as in Figure 8B.
A similar observation for the reductive current with the
polycrystalline Pd electrode has been reported in comparison
with that of 200-nm Pd film, and the reason for the difference
was attributed to the different volume of Pd to absorb hydro-
Figure 7. Cyclic voltammograms obtained with (A) the PdNP/ITO
electrode and (B) the Pd bulk electrode in the solution containing 0.5 M
H
2
SO
4
. In part A, the potential was reversed at various potentials. Scan
-
1
2
31
31
rate: 50 mV s . Geometric electrode area: (A) 0.031 cm and (B)
gen. Compared with the results on the 200-nm Pd film, the
start of hydrogen adsorption/absorption was rather smaller and
the level-off current region lasted to the negative potential
region. Thus, it is expected that the present response with the
PdNP/ITO would be further evidence of the limited volume of
PdNPs, which should be inferred from the SEM images in
Figure 2. The present result in Figure 8 clearly indicated that
the PdNPs had characteristics similar to those of the Pd thin
film concerning very limited absorption of hydrogen into the
2
0
.020 cm . The current axis was normalized by dividing the surface
areas.
electrode. For this reason, it is expected that the total surface
area of Pd in the present Pd/ITO electrode is larger than that of
the Pd bulk electrode, as inferred from the FE-SEM image of
Figure 2C. Thus, depending on the target electrochemical
reactions, it was found that PdNPs could take part in the
electrode reactions effectively as electrocatalysts utilizing the
larger surface area.
Hydrogen Adsorption and Absorption on PdNP/ITO
Electrode. As recently carried out for Pd thin films of 10 mono-
layers and 100 µm,31 cyclic voltammetric measurements should
give useful information about the hydrogen adsorption and
absorption. Thus, the cyclic voltammograms in 0.5 M H2SO4
were recorded using the PdNP/ITO electrode.
31
Pd metal. Thus, it is expected that the PdNP/ITO electrode
would be an interesting alternative to the thin Pd film, though
the limited volume is similar but the composition is different,
i.e., smooth film and porous characteristics.
As a minor feature in the voltammeric responses with the
PdNP/ITO electrode in Figure 8A, neither a couple of the
absorption peak nor separated peaks for R- and â-phases, which
3
1
were observed on the Pd thin films, were recorded. This is
presumably because of the iR drop in the PdNP/ITO electrode,
because the shape of the present voltammograms is similar to
the result for the Pd thin film on reticulated vitreous carbon.34
In addition, we can extract some more features of the PdNP/
ITO as an electrode from the voltammetric results in Figure 7.
Figure 7 shows the cyclic voltammograms recorded in the
anodic scans by changing the reversed potentials. In this
potential region, the oxidative formation of PdO and PdO2 and
their reduction responses are observed.33 As shown in this figure,
the voltammeric responses observed with the Pd bulk and PdNP/

Seed-Mediated Growth of Palladium Nanocrystals
J. Phys. Chem. B, Vol. 110, No. 41, 2006 20367
The results in Figure 7A show that the reduction waves of PdO
increased, accompanying a slight negative potential shift as the
reversed potential became positive. This tendency is similar to
the results with the Pd bulk electrode33 and the Pd films.31
However, as a marked difference from the previous results, the
magnitude of the negative shift is smaller than those reported
with the Pd bulk electrode33 and the Pd films. The reason for
the negative shift of the reduction wave of PdO was explained
with the ever-increasing stability of the PdO layer as the positive
increase of the reversed potential.33 Therefore, in the present
PdNP/ITO electrode, it is expected that the proceeding formation
of the Pd oxides is different from those of the bulk surface and
film states. Because the surface oxidation of Pd should proceed
easily and evenly for the PdNPs judging from the limited very
small volumes of PdNPs contacting with the conducting ITO
support, the changes in the formed states of Pd oxides on PdNPs
should be less sensitive to the reversed potential in comparison
with the bulk surface or film. Thus, the reductive responses in
Figure 7A would be a reflection of the characteristics of PdNPs.
In addition, compared with the response recorded with the
Pd bulk electrode (Figure 7B), two features can be noted. The
first is the difference in the magnitude of the reduction peaks
of PdO. If the thin surfaces of all PdNPs are responsible for
the reduction peak, the magnitude of the reduction current should
have become larger in Figure 7A on the basis of the large surface
area per volume of PdNPs in the FE-SEM images (Figure 2).
However, actually, the peaks in Figure 7A recorded with the
PdNP/ITO electrode were smaller than that in Figure 7B
recorded with the Pd bulk electrode. Thus, it is inferred that all
the surfaces of PdNPs not necessarily took part in the oxidation
and the reduction, or the surface layer of the Pd bulk electrode
responsible for the oxide formation is thicker than the scale of
PdNPs. The second feature is a few humps observed in the
positive-going scan only with the PdNP/ITO electrode (Figure
promote the electron-transfer reactions. Furthermore, PdNPs
themselves have the electrocatalytic functions for peculiar reac-
tions, as shown for the reduction of oxygen.
The cyclic voltammetric measurements in 0.5 M H2SO4
showed some characteristics due to the limited volume and the
dispersed state of PdNPs. While Pd thin films have been utilized
in some applications, the present modification of PdNPs may
be the alternatives, judging from some advantages of the
proposed method, e.g., (1) the large surface area per volume,
(2) the not strongly capped or covered surfaces, and (3) the
ease of the preparation, i.e., just a simple two step immersion,
of the present PdNP/ITO.
Because the proposed seed-mediated growth was applicable
to the modification of the GC surface, the approach is expected
to be utilized in the surface modification for catalytic uses as
well as for electrochemical measurements.
31
Acknowledgment. This work was supported by the Kyoto
Nanotechnology Cluster Project, a Grant for Regional Science
and Technology Promotion, and by the computational materials
science unit in Kyoto University, both from the Ministry of
Education, Culture, Sports, Science and Technology (MEXT),
Japan. G.C. thanks the Monbukagaku-sho scholarship as a doctor
course student at Kyoto University.
References and Notes
(
1) Daniel, M.-C.; Astruc, D. Chem. ReV. 2004, 104, 293.
(2) Li, F.; Zhang, B.; Dong, S.; Wang, E. Electrochim. Acta 1997, 42,
563.
3) Shen, Y.; Bi, L.; Li, B.; Dong, S. New J. Chem. 2003, 27, 938.
4) El-Deab, M. S.; Sotomura, T.; Ohsaka, T. J. Electrochem. Soc. 2005,
152, C1.
(5) Freeman, R. G.; Grabar, K. C.; Allison, K. J.; Bright, R. M.; Davis
2
(
(
J. A.; Guthrie, A. P.; Hommer, M. B.; Jackson, M. A.; Smith, P. C.; Walter,
D. G.; Natan, M. J. Science 1995, 267, 1629.
(6) Cheng, W.; Dong, S.; Wang, E. Anal. Chem. 2002, 74, 3599.
7
A). Because some characteristic oxidation peaks have been
(7) Cheng, W.; Dong, S.; Wang, E. Langmuir 2002, 18, 9952.
(8) Kambayashi, M.; Zhang, J.; Oyama, M. Cryst. Growth Des. 2005,
35
reported for the well-defined surfaces of Pd, the appearance
might be due to the crystallographic surfaces of PdNPs.
In the above results, some characteristics of PdNPs such as
nanosized particles and thin layers were explicitly presented.
While the thin layer modification with Pd materials has been
mainly performed by forming the films, the present PdNPs’
attachment would be an alternative method to modify the
conducting surface with very small amount of Pd.
5
, 81.
9) Chang, G.; Oyama, M.; Hirao, K. J. Phys. Chem. B 2005, 109,
1204.
(10) Zhang, J.; Kambayashi, M.; Oyama, M. Electrochem. Commun.
004, 6, 683.
11) Zhang, J.; Kambayashi, M.; Oyama, M. Electroanalysis 2005, 17,
08.
(12) Zhang, J.; Oyama, M. Electrochim. Acta, 2004, 50, 85.
(
2
(
4
(
(
13) Zhang, J.; Oyama, M. J. Electroanal. Chem., 2005, 577, 273.
14) Jana, N. R.; Gerheart, L.; Murphy, C. J. J. Phys. Chem. B 2001,
1
05, 4065.
Conclusions
(15) Busbee, B. D.; Obare, S.; Murphy, C. J. AdV. Mater. 2003, 15,
414.
In the present work, Pd nanocrystals were successfully
attached and grown on the ITO surfaces using the seed-mediated
growth method. While some short nanorods were formed as a
minor product, the PdNPs having crystal-like appearance, whose
size was 60-80 nm, were densely modified on the ITO surface
after 24 h of growth. Because the PdNPs tend to stick to each
other in the growth treatment, which is particular to Pd and not
observed for Au or Ag, the ITO surfaces could be modified
with a kind of porous structure composed of Pd nanocrystals.
The electrochemical impedance measurement of the PdNP/
ITO electrode showed a significant decrease of the charge-
transfer resistivity, which is comparable to the cases of AuNPs
(16) B o¨ nnemann, H.; Braun, G.; Brijoux, W.; Brinkmann, R.; Schulze
Tilling, A.; Seevogel, K.; Siepen, K. J. Organomet. Chem. 1996, 520, 143.
17) Li, Y.; Hong, X. M.; Collard, D. M.; El-Sayed, M. A. Org. Lett.
000, 2, 2385.
18) Nishihata, Y.; Mizuki, J.; Akao, T.; Tanaka, H.; Uenishi, M.;
Kimura, M.; Okamoto, T.; Hamada, M. Nature 2002, 418, 164
19) Kidambi, S.; Dai, J.; Li, J.; Bruening, M. L. J. Am. Chem. Soc.
004, 126, 2658.
20) Teranishi, T.; Miyake, M. Chem. Mater. 1998, 10, 594.
(
2
(
(
2
2
(
(21) Narayanan, R.; El-Sayed, M. A. J. Phys. Chem. B 2004, 108, 8572.
(22) Son, S. U.; Jang, Y.; Yoon K. Y.; Kang, E.; Hyeon, T. Nano Lett.
004, 4, 1147.
(23) Xiong, Y.; McLellan, J. M.; Chen, J.; Yin, Y.; Li, Z. Y.; Xia, Y.
J. Am. Chem. Soc. 2005, 127, 17118.
(24) Xiong, Y.; Chen, J.; Wiley: B.; Xia, Y.; Li, Z.-Y. Nano Lett. 2005,
5, 1237.
9-11
and AgNPs.
Despite the high resistivity of Pd in comparison
(
25) Guliotti, L. A.; Feldheim, D. L.; Eaton, B. E. Science 2004, 304,
50.
(26) Liu, J.; Cheng, L.; Song, Y.; Liu, B.; Dong, S. Langmuir 2001,
with Au or Ag, the dense modification by Pd nanocrystals is
inferred to improve the charge transferring situations at the
interface. The same effect was observed for the electrochemical
8
1
7, 6947.
4-
3-
(27) G o´ mez, R.; P e´ rez, J. M.; Solla-Gull o´ n, J.; Montiel, V.; Aldaz, A.
reaction of the [Fe(CN)6] /[Fe(CN)6] couple in 0.1 M phos-
phate buffer. Thus, the PdNP/ITO electrode can be regarded as
a unique modified electrode in which the NPs at the interface
J. Phys. Chem. B 2004, 108, 9943.
28) Bartlett, P. N.; Guerin, S. Anal. Chem. 2003, 75, 126.
(29) Wang, B.; Li, X. Anal. Chem. 1998, 70, 2187.
(

20368 J. Phys. Chem. B, Vol. 110, No. 41, 2006
Chang et al.
(
30) Park, S.; Yang, P.; Corredor, P.; Weaver, M. J. J. Am. Chem. Soc.
002, 124, 2428.
31) Gabrielli, C.; Grand, P. P.; Lasia, A.; Perrot, H. J. Electrochem.
Soc. 2004, 151, A1937.
32) Chang, G.; Oyama, M.; Hirao, K. J. Phys. Chem. B 2006, 110,
860.
(33) Dall’Antonia, L. H.; Tremiliosi-Filho, G.; Jerkiewicz, G. J. Electro-
anal. Chem. 2001, 502, 72.
(34) Czerwinski, A.; Marassi, R.; Zamponi, S. J. Electroanal. Chem.
1991, 316, 211.
(35) Hoshi, N.; Kuroda, M.; Hori, Y. J. Electroanal. Chem. 2002, 521,
155.
2
(
(
1