Catalytic characterization of hollow silver/palladium nanoparticles synthesized by a displacement reaction

Article abstract of DOI:10.1016/j.electacta.2009.04.056

Hollow Ag/Pd nanoparticles have been successfully prepared by a galvanic displacement reaction, in which a small amount of Pd(NO₃)₂ is allowed to react with previously synthesized Ag nanoparticles that act as templates. The resulting

Full text of DOI:10.1016/j.electacta.2009.04.056

Electrochimica Acta 54 (2009) 5544–5547
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Catalytic characterization of hollow silver/palladium nanoparticles
synthesized by a displacement reaction
∗
Chien-Liang Lee , Chun-Ming Tseng, Rong-Bing Wu, Chen-Chung Wu, Shu-Ciao Syu
Department of Chemical and Materials Engineering, National Kaohsiung University of Applied Science, Kaohsiung 807, Taiwan, ROC
a r t i c l e i n f o
a b s t r a c t
Article history:
Hollow Ag/Pd nanoparticles have been successfully prepared by a galvanic displacement reaction, in
which a small amount of Pd(NO3)2 is allowed to react with previously synthesized Ag nanoparticles that
act as templates. The resulting hollow Ag/Pd (Ag/Pdhollow) nanoparticles are found to be icosahedral and
decahedral in structure. The kinetics of electroless copper deposition (ECD) catalyzed by these bimetallic
Received 19 January 2009
Received in revised form 16 April 2009
Accepted 22 April 2009
Available online 3 May 2009
(
Ag/Pdhollow) nanoparticles are analyzed using an electrochemical quartz crystal microbalance (EQCM).
The results reveal that these Ag/Pdhollow nanoparticles have better catalytic activities than monometallic
Ag and Pd nanoparticles. Furthermore, the catalytic activities of these hollow nanoparticles in the ECD
bath can be controlled by tuning their alloy ratios in a suitable manner.
Keywords:
Activation
Printed circuit board
Surface plasmon resonance
Platinum
©
2009 Elsevier Ltd. All rights reserved.
Catalysis
1. Introduction
enhanced catalytic activity [1]. In the present study, we synthe-
sized hollow Ag/Pd (Ag/Pdhollow) nanoparticles via a displacement
Nanoparticles fabricated from Ag-based alloys find extensive
applications in catalysis [1], electrocatalysis [2], and optics because
their surface plasmon resonance (SPR) bands appear in the visible
region [1–4]. One reason to determine these properties is the struc-
ture of the nanoparticles. Hollow nanoparticles fabricated from
Ag-based alloys have attracted attention due to their catalytic [5,6]
and electrocatalytic activities [2,7], and the SPR characteristics of
hollow nanoparticles [2,6,8,9] differ significantly from those of solid
nanoparticles. Hollow Ag/Pd nanoshells have been found to show
unique catalytic activities when used as catalysts for electroless
copper deposition (ECD) [5,10]. ECD has gained considerable signif-
icance in the field of nanocircuits [11]. The mechanism underlying
ECD involves activation, which is a catalytic reaction triggered by
active colloids on the surface of substrates dipped into the elec-
troless bath [12]. The active catalyst acts as an electron carrier
for the transfer of electrons from the reducing agent to the cop-
per ions. Hence, the structure and composition of the catalyst can
alter the deposition rate and reaction kinetics [13]. In a previous
study, the authors prepared spherical Pd nanoparticles and used
them as activators for ECD [12–14]. Additionally, for use in the elec-
troless reaction, they synthesized solid nanocatalysts with Pd-rich
shells from a Ag/Pd alloy by adding Ag particles to the microstruc-
tures of Pd nanoparticles. The nanocatalysts thus prepared showed
reaction. In this reaction, a small amount of Pd(NO ) was allowed
3 2
to galvanically react with the Ag nanoparticles. The slight deposi-
tion of Pd ions on the surface of the Ag nanoparticles resulted in a
closure of pores and the formation of hollow nanoparticles. Then,
we used the synthesized Ag/Pdhollow nanoparticles as ECD activa-
tors. We carried out electrochemical quartz crystal microbalance
(EQCM) measurements to compare the kinetics of ECD catalyzed by
Ag/Pdhollow nanoparticles with different alloy ratios (the Pd content
remained constant) and those catalyzed by solid Pd nanoparticles.
2. Experimental
Ag/Pdhollow nanoparticles of different compositions were pre-
pared in the following manner. Initially, silver nitrate (AgNO ,
3
−
5
∼5 × 10 mol) was slowly dissolved in 3 mL of a neutral aqueous
solution of 0.1 M sodium n-dodecyl sulfate (SDS). Approximately
25 ␮L of 0.1 M NaBH was added dropwise to the abovementioned
4
solution to reduce the Ag ions to metallic Ag. The solution tem-
◦
perature was maintained at 30 C. After 10 min, Ag nanoparticles
were obtained. Then, approximately 2.562 mg of Pd(NO ) ·H O was
3
2
2
+
2
slowly dissolved in 1 mL of 0.1 M HNO3 to obtain a Pd solution,
which was used for the subsequent galvanic displacement reaction.
−
7
Approximately 50 ␮L of this solution (5.56 × 10 mol) was added
to 3 mL of a well-stirred solution of Ag nanotemplates at a fixed
◦
temperature of 30 C. After 70 min, the dispersed nanoparticles,
∗ Corresponding author. Tel.: +886 7 3814526 5131; fax: +886 7 3830674.
E-mail address: cl lee@url.com.tw (C.-L. Lee).
in which the Ag/Pd molar ratio was 90:1, were collected. Ag45Pd1
nanoparticles were prepared in a similar manner; however, in this
0
013-4686/$ – see front matter © 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2009.04.056

C.-L. Lee et al. / Electrochimica Acta 54 (2009) 5544–5547
5545
case, the concentration of AgNO used for the preparation of the Ag
3
−5
nanotemplates was reduced to 2.5 × 10 mol. In order to carry out
a fair comparison of the optical or catalytic properties, Ag and Pd
−
5
nanoparticles were produced from the reduction of 5.0 × 10 mol
−7
of Ag ions and 5.56 × 10 mol of Pd ions, respectively, in 0.1 M SDS
solution (3 mL). In addition, a Pd/Sn colloidal solution with the same
Pd concentration as that used in the nanoparticles was available for
comparison.
The prepared nanoparticle solutions were placed on a carbon-
coated copper grid and allowed to dry naturally. Then, the
characteristic shapes and sizes of the nanoparticles were observed
under a transmission electron microscope (TEM; JEOL JEM-3000F)
and an energy-dispersive X-ray (EDX) spectroscope. The SPR spectra
of the nanoparticles were measured using a UV–vis spectropho-
tometer (Agilent 8453).
The synthesized hollow Ag90Pd and Ag Pd nanoparticles, the
1
45
1
solid Ag and solid Pd nanoparticles, and the Pd/Sn colloids were
then used as activators for ECD. In order to reduce the interac-
tion of free SDS with the obtained nanoparticles during the ECD
reaction, 1 mL of the nanoparticle solution was precipitated by
high-speed centrifugation (15,000 rpm) and redispersed in 1 mL of
H O. This process helped eliminate the free SDS. The ECD bath for
2
the EQCM measurements was prepared from 0.44 M formaldehyde,
0
.1 M EDTA, and 0.05 M CuSO . The pH of the ECD bath was adjusted
4
◦
to 12.3 using NaOH powder. The ECD baths were maintained at 30 C
and bubbled with N2 for 15 min prior to the measurements. The
working electrode used for the EQCM (Autolab PGATAT30 and Seiko
QCA922) experiment was prepared by coating 2 ␮L of the nanopar-
ticle solution or the Pd/Sn colloid solution in a uniform layer over
2
0
.159 cm of the Au surface of the QCM substrate. Au was sputtered
on both sides of a Ti film (thickness: 100 Å) on the QCM substrate
Seiko EG&G QA20-A9M-Au), which was connected to an oscilla-
tor manufactured in-house. The reference electrode (Hg/Hg Cl )
(
2
2
was separated from the compartment that housed the electrolyte
solution by a Luggin capillary filled with saturated KCl solution.
3. Results and discussion
Fig. 1A shows TEM images of the microstructures of the Ag90Pd1
nanoparticles prepared with the abovementioned galvanic dis-
placement method. Some of the hollow nanoparticles were formed
Fig. 1. TEM images of hollow Ag/Pd nanoparticles prepared by the displacement
−7
reaction. (A) TEM image of Ag90Pd . (B) TEM image of Ag Pd1.
after 5.56 × 10 mol of Pd(NO3)2 reacted with the Ag nanoparti-
1
45
−
5
cles obtained from the reduction of 5 × 10 mol of AgNO . The
3
0
half-cell reactions and standard reduction potentials (E ) for this
displacement reaction are shown below:
the displacement reaction. Ag nanoparticles synthesized by the dis-
placement reaction using HAuCl as the displacing salt are found to
4
Anodic reaction : Ag → Ag+ + e−
have similar morphologies [9].
EDX analyses of the synthesized hollow nanoparticles provided
information on their architectures, and this information is shown
in Table 1. The weight percentages of Pd estimated from the EDX
analyses of the Ag45Pd and Ag Pd nanoparticles are 15.43% and
E0 = −0.7996 V
Cathodic reaction : Pd2+ + 2e− → Pd
1
90
1
1
0.50%, respectively. The weight percentages of Ag in Ag45Pd and
1
E0 = 0.951 V
Ag90Pd1 are 84.57% and 89.5%, respectively. These results indicate
that the nanoparticles prepared by the method described in this
paper consist of both Ag and Pd. This confirms the formation of
Ag/Pdhollow nanoparticles.
The optical properties of the prepared hollow nanoparticles
were determined from their corresponding SPR spectra. Fig. 3
compares the SPR extinction spectra of the cage-like bimetal-
lic nanoparticles, their corresponding Ag nanotemplates, and Pd
nanoparticles. A sharp peak at 405 nm was detected in the spec-
trum of the Ag nanotemplates. However, the SPR peak of the Pd
nanoparticles could be detected only in the UV region. Comparison
Hence, the potential corresponding to the net reaction is
0.1514 V. The equlibrium constant (K) calculated using the for-
5
−1
−1
mula Ereaction = (RT/nF) ln K is 1.09 × 10 . Here, R = 8.314 J K mol
,
T = 303 K, n = 2, and F is Faraday’s constant. From the value of K and
the TEM results, it can be concluded that the displacement reac-
tion readily results in Ag/Pdhollow nanoparticles, even when a low
concentration of Pd(NO3)2 is used. Fig. 1B shows TEM images of
the microstructures of the hollow Ag Pd nanoparticles obtained
45
1
−5
using 2.5 × 10 mol of AgNO . It is evident that these Ag/Pd
3
hollow
nanoparticles contain pores at the central location. HR-TEM obser-
vations reveal that hollow icosahedral (Fig. 2A) and decahedral
of the spectra of the Ag90Pd and Ag nanoparticles revealed a red-
1
shift of the bimetallic SPR peak to 479 nm. Broadening of this peak
was observed after Pd was deposited on the Ag templates. How-
(
Fig. 2B) nanoparticles with porous {1 1 1} faces are formed from

5
546
C.-L. Lee et al. / Electrochimica Acta 54 (2009) 5544–5547
Table 1
Comparative EDX and EQCM results of hollow Ag45Pd1 and Ag90Pd1 nanoparticles, solid Ag and Pd nanoparticles, and Pd/Sn colloids.
Synthesis condition
EDX
EQCM measurement
Ag/Pd (g)
Deposition rate (␮g/cm2 s)
+
(Pd2+) M
Activity (␮g/␮g cm2 s)
(
Ag ) M
Pd (at%)
Ag (at%)
−
3
−4
−4
−6
−3
Hollow Ag45Pd1 nanoparticles
Hollow Ag90Pd1 nanoparticles
Ag nanoparticles
Pd nanoparticles
Pd/Sn colloids
8.197 × 10
1.823 × 10
1.823 × 10
15.43
10.5
84.57
89.5
6.67 × 10
6.44 × 10
1.88 × 10
6.33 × 10
2.36 × 10
0.04
0.018
0
0
0
5 × 10
−2
−6
−6
−6
−6
−3
1.64 × 10
2.795 × 10
8.197 × 10−
3
0
0
0
1.823 × 10−
1.823 × 10−
4
4
ever, no such broadening of the SPR peak was observed in the case
of solid Ag/Pd nanoparticles prepared via the chemical reduction
method [1]. The optical change observed in the present study was
possibly due to the formation of the hollow nanoparticles, as con-
firmed by the TEM images shown in Fig. 1A. It is noteworthy that
extent than those of the Ag90Pd1 nanoparticles. Thus, it would be
reasonable to conclude that the concentration of the Ag/Pdhollow
nanoparticles formed increases with the concentration ratio of Pd
to Ag.
The prepared Ag/Pdhollow nanoparticles were tested for their
effectiveness as activators for ECD. Fig. 4A shows the comparative
results of EQCM analyses carried out after in situ measurements of
the catalytic activities, while Fig. 4B shows the open circuit poten-
tials of the Ag/Pdhollow, Ag, and Pd nanoparticles, as well as the
Pd/Sn colloids in the ECD bath. The deposition rates shown in Fig. 4A
were recalculated from the changes in the electrode frequency
using Sauerbrey’s equation [12]. The results of these calculations are
summarized in Table 1. Ag nanoparticles obtained from the reduc-
the SPR peaks of the Ag45Pd nanoparticles broadened to a greater
1
−
3
tion of an 8.197 × 10 M solution of Ag ions and Pd nanoparticles
−
4
and Pd/Sn colloids obtained from the reduction of a 1.823 × 10
M
solution of Pd ions were found to be inactive for ECD. The aver-
age ECD deposition rates observed during the preparation of the
hollow Ag90/Pd and Ag45/Pd1 nanoparticles were approximately
1
2
2
0
.018 ␮g/cm s and 0.04 ␮g/cm s, respectively. These values indi-
cate that the Ag/Pdhollow nanoparticles prepared by this method
could be successfully employed as catalysts for ECD. The catalytic
activity of the nanocatalysts was calculated according to the follow-
ing relation:
deposition rate
weight of the alloy catalyst
catalytic activity =
.
The catalytic activities of the Ag Pd and Ag Pd nanoparticles
9
0
1
45
1
−
3
2
−3
2
were 2.795 × 10 ␮g/␮g cm s and 5 × 10 ␮g/␮g cm s, respec-
tively. Thus, it is evident that the electroless deposition rate and the
catalytic activity are enhanced by the more hollow nanoparticles.
As can be seen in Fig. 4B, a steady-state reaction potential (SRP)
is observed after the ECD reaction is triggered by the nanocat-
alysts. The SRPs measured for the inactive Ag nanoparticles, Pd
nanoparticles, and Pd/Sn colloids were −0.355 V, −0.409 V, and
Fig. 2. HR-TEM images of hollow Ag/Pd nanoparticles prepared by the displacement
Fig. 3. Comparative SPR spectra of hollow Ag90Pd1 and Ag45Pd1 and solid Ag and Pd
reaction. (A) Hollow icosahedral nanoparticle. (B) Hollow decahedral nanoparticle.
nanoparticles.

C.-L. Lee et al. / Electrochimica Acta 54 (2009) 5544–5547
5547
4. Conclusions
Hollow Ag/Pd nanoparticles with different alloy ratios were suc-
cessfully prepared by a galvanic displacement reaction, in which a
small amount of Pd(NO3)2 was allowed to react slowly with pre-
viously synthesized Ag nanoparticles that acted as templates. The
resulting hollow Ag/Pd nanoparticles were icosahedral and deca-
hedral in microstructure and showed better catalytic activities for
ECD than did the Ag nanoparticles, Pd nanoparticles, and Pd/Sn
colloids. In addition, the catalytic activities of these Ag/Pd nanopar-
ticles in the ECD bath could be controlled by suitably tuning their
alloy ratios. These Ag/Pdhollow nanoparticles thus constitute a novel
class of catalysts for ECD.
Acknowledgments
The authors would like to thank the National Science Coun-
cil of the Republic of China, Taiwan for financially supporting this
research under Contract No. NSC 97-2221-E-151-028.
References
[1] C.L. Lee, Y.C. Huang, L.C. Kuo, Electrochem. Commun. 8 (2006) 1021.
[2] C.L. Lee, C.M. Tseng, J. Phys. Chem. C 112 (2008) 13342.
[3] T.Y. Olson, A.M. Schwartzberg, C.A. Orme, C.E. Talley, B. O’Connell, J.Z. Zhang, J.
Phys. Chem. C 112 (2008) 6319.
[
[
4] W.H. Wang, G.Y. Cao, J. Nanopart. Res. 9 (2007) 1153.
5] C.L. Lee, C.M. Tseng, S.C. Wu, R.B. Wu, K.R. Yang, Electrochim. Acta 53 (2008)
5905.
[
6] C.M. Cobley, D.J. Campbell, Y.N. Xia, Adv. Mater. 20 (2008) 748.
Fig. 4. Comparative EQCM curves of ECD kinetics catalyzed with hollow Ag90Pd1
and Ag45Pd1 nanoparticles, solid Ag and Pd nanoparticles, and Pd/Sn colloids. (A)
Deposition rate. (B) ECD reaction potential.
[7] H.P. Liang, H.M. Zhang, J.S. Hu, Y.G. Guo, L.J. Wan, C.L. Bai, Angew. Chem. Int. Ed.
43 (2004) 1540.
[8] J.N. Gao, X.L. Ren, D. Chen, F.Q. Tang, J. Ren, Scripta Mater. 57 (2007) 687.
[9] X.M. Lu, H.Y. Tuan, J.Y. Chen, Z.Y. Li, B.A. Korgel, Y.N. Xia, J. Am. Chem. Soc. 129
(
2007) 1733.
[
10] C.L. Lee, C.M. Tseng, S.C. Wu, R.B. Wu, Electrochem. Solid-State Lett. 11 (2008)
−
0.450 V, respectively. The SRPs of the Ag/Pdhollow nanoparticles,
D27.
which could catalyze ECD, were more negative than those of the
Ag and Pd nanoparticles, which were unable to catalyze ECD. The
SRPs of the hollow Ag90/Pd1 and Ag45/Pd1 nanoparticles reached
approximately −0.680 V and −0.750 V, respectively. Thus, it may
be concluded that the ECD rate increases as the SRP becomes more
negative.
[
11] H. Akahoshi, M. Kawamoto, T. Itabashi, O. Miura, A. Takahashi, S. Kobayashi,
M. Miyazaki, T. Mutoh, M. Wajima, T. Ishimaru, IEEE Trans. Compon. Packag.
Technol. Part A 18 (1995) 127.
[
[
12] C.L. Lee, Y.C. Huang, L.C. Kuo, J. Solid-State Electrochem. 11 (2007) 639.
13] C.L. Lee, Y.C. Huang, C.C. Wan, Y.Y. Wang, Y.J. Ju, L.C. Kuo, J.C. Oung, J. Electrochem.
Soc. 152 (2005) C520.
[
14] C.L. Lee, Y.C. Huang, Electrochem. Solid-State Lett. 9 (2006) C196.