NANO

LETTERS

Formation of Rectangularly Shaped 2006

Vol. 6, No. 10
Pd/Au Bimetallic Nanorods: Evidence 2290-2294
for Competing Growth of the Pd Shell
between the {110} and {100} Side
Facets of Au Nanorods
Yanjuan Xiang,† Xiaochun Wu,*,‡ Dongfang Liu,† Xingyu Jiang,‡ Weiguo Chu,‡
Zhiyuan Li,† Yuan Ma,§ Weiya Zhou,† and Sishen Xie*,†

Beijing National Laboratory for Condensed Matter Physics, Institute of Physics,

Graduate School of the Chinese Academy of Sciences, Beijing 100080, P R China,
National Center for Nanoscience and Nanotechnology, Beijing 100080, P R China,
and Electronmicroscopy Laboratory, Peking UniVersity, Beijing 100871, P R China

Received July 25, 2006; Revised Manuscript Received August 22, 2006

ABSTRACT
In this letter, we report the competing growth of a Pd shell on the {110} and {100} facets of Au nanorods (Au NRs). This results in the
disappearance of unstable {110} facets and the formation of rectangularly shaped Pd/Au bimetallic nanorods that show only four stable
{100} side surfaces. The energy minimization to a more stable morphology is believed to be the driving force for the formation of the rectangular
shape of the Pd shell.

The surface structure of metal nanocrystals is very important
to their surface properties.1 For example, shape and crystal-
lographic facets are the major factors in determining the
catalytic and surface activities of metal nanocrystals. There-
fore, great efforts have been made to control their size and
shape. For face-centered cubic (fcc) metal nanocrystals,
different shapes have been formed. They are found to have
the {100} facets in the cubic nanocrystals,2 the {111} facets
in the tetrahedral nanocrystals,2 or a mixture of {100} and
{111} in the truncated octahedral and multiply twinned
nanocrystals.2 Because {100} and {111} are low in surface
energy, their formation is expected. It was found recently,
however, that Au nanorods (Au NRs), synthesized via an
electrochemical procedure,3 show unique surface structures.4
Wang et al. have investigated the crystal structure of the Au
NRs with aspect ratios of 2-7 using high-resolution TEM.
They found that the side facets of the as-prepared nanorods
are dominated by the {110} planes with small {100} facets,
whereas the ends of the nanorods are enclosed by {111} and
{100} planes.3 The appearance of the unstable {110} facets
is somewhat surprising and raises at least two questions:
* To whom correspondence should be addressed. E-mail: wuxc@
nanoctr.cn; ssxie@aphy.iphy.ac.cn.
† Graduate School of the Chinese Academy of Sciences.
‡ National Center for Nanoscience and Nanotechnology.
§ Peking University.
10.1021/nl061722c CCC: $33.50 © 2006 American Chemical Society
Published on Web 09/08/2006
First, why are {110} facets present? Second, what kind of
properties will such {110} facets show? For the first question,
one tentative explanation is ascribed to the stabilization by
surfactants:4b,5 because of the larger interatomic distance, the
bonding between the surface atoms of {110} facets are not
strong, thus allowing them to form stronger bonds with the
surfactant molecules. For the second question, such a
crystallographic shape and the appearance of unstable {110}
facets are suggested to induce unique surface properties. For
example, competing crystal growth on different facets has
been implied and explored. Until now, competing growth6
and etching7 between end facets and side facets have been
observed. In contrast, the competing growth on {110} and
{100} side facets has not been reported. Apart from
preparation by the electrochemical methods, Au NRs with
the same crystal structures have also been synthesized by a
seed-mediated method.8 Here, we demonstrate that using Au
NRs as templates a competing growth of Pd shell on these
two side facets takes place. This competitive growth leads
to the formation of rectangularly shaped Pd shell/Au core
bimetallic nanorods.
Initial Au NRs were synthesized using a seed-mediated
growth method.8c The aspect ratio of the Au NRs used here
is 3.2. The solution of the Au NRs was centrifuged twice at
12 000 rpm for 10 min. After that, the precipitate of the Au
NRs was redispersed in deionized water and diluted to the
same volume as the stock solution. A typical procedure for
the growth of Pd shells with different thicknesses is as
follows: five samples of the Au NR solutions (4 mL) were
each mixed with 2 mL of 0.1 M cetyltrimethylammomium
bromide (CTAB) aqueous solution containing 20 µL of 0.1
M ascorbic acid (AA). Different volumes (125, 225, 325,
425, and 550 µL) of 2 mM PdCl42- (0.035 g of PdCl2 was
dissolved in 2 mL of 0.2 M aqueous HCl solution and then
diluted into 100 mL with deionized water) were added,
respectively. The mixtures were then shaken vigorously and
placed in a 30 °C water bath. Within several hours, the color
of the solution changed from brown to dark gray, suggesting
the formation of Pd shell. After 14 h, the reaction was
stopped by centrifuging (12 000 rpm 10 min) the solution
twice. The precipitates were redispersed in deionized water
for UV-visible absorption spectra characterization (Perkin-
Elmer Lamdba 950). For transmission electron microscopy
(TEM, Tecnai F30) and scanning electron microscopy (SEM,
Hitachi S-5200) measurements, two more centrifugations
(12 000 rpm 10 min) were applied to remove excess amounts
of CTAB.
Figure 1 shows the change in morphology as Pd was
coated on the Au NRs. In comparison with the cylindrical
shapes of the Au NRs, all Pd-coated rods are rectangular
with sharp edges (see Figure 1b-f insets). Increasing the
Pd/Au molar ratio (mPd/mAu) causes the rods to become
“fatter”, indicating a decrease in aspect ratio (R). The
thickness of the Pd shell is controlled by adjusting the amount
of PdCl42- added. Energy-dispersive X-ray (EDX) analysis
verifies the existence of Pd (Figure 1g). The relationship
between R and mPd/mAu is plotted in Figure 2A. The average
length and diameter are obtained based on the measurement
of over 50 rods for each Pd shell. The aspect ratio decreases
gradually with increasing Pd/Au ratio, from 3.2 of pure gold
NRs to 2.4 for Pd-coated rods with a Pd/Au ratio of 2.75.
The decrease in aspect ratio can be understood from Figure
2B and C, which presents the relationship between the
increments in length (∆L) and in diameter (∆d) of the Figure 1. Formation of Au/Pd core/shell bimetallic nanorods by
bimetallic nanorods (compared with the length (LAu) and the chemically reducing Pd2+ ions onto the surface of Au NRs. SEM
diameter (dAu) of the Au NRs) and the Pd/Au molar ratio. images were taken before (a) and after adding (b) 125, (c) 225, (d)
Both ∆L and ∆d show good linearity, and the slopes of both 325, (e) 425, and (f) 550 µL of a 2 mM PdCl42- aqueous solution
to the reaction system. The insets are the corresponding STEM
lines are quite close (3.5 for ∆L and 3.3 for ∆d, respectively). images. The scale bars for both SEM and STEM are 90 nm. (g)
If we use this relationship, then R should be equal to 1.08 + EDX analysis for Pd-coated Au NRs with a Pd/Au ratio of 2.75.
(LAu - 1.08dAu)/(3.3mPd/mAu+ dAu), inversely proportional
to the Pd/Au ratio. This qualitatively agrees with that shown broadens with increasing Pd layer thickness. With a shell
in Figure 2A. By assuming a complete conversion of PdCl42- thickness of 1.2 nm, the LSPR feature from the Au NRs is
to Pd, we compared the calculated Pd/Au ratios with the still visible, albeit severely damped (broadening in peak
measured ratios (from EDX analysis, average from five width and reducing in intensity). Further increasing the
consecutive measurements). They show a linear relationship thickness of the Pd shell, the LSPR feature from Pd
with a slope of 0.82, indicating a quantitative but incomplete dominates, exhibiting a broad shoulder between 500 and 700
conversion of Pd ions to Pd (Figure 2D). nm. To gain more insight into the UV-visible extinction
Figure 3A shows the UV-visible absorption spectra of spectra exhibited by Pd/Au bimetallic nanorods, we per-
the as-prepared Au NRs and Pd-coated gold nanorods. The formed discrete dipole approximation (DDA) calculations
as-prepared Au NRs show a longitudinal surface plasma using the lengths and diameters given in Figure 2B and C.
resonance (LSPR) band at 730 nm with a full width at half- The calculated results are presented in Figure 3B. They
maximum (fwhm) of 153 nm and a weak transverse SPR reproduce the main features of the experimental spectra. For
band at 514 nm. The LSPR band gradually blue-shifts and the single Au nanorod, the calculated LSPR peak locates at
Nano Lett., Vol. 6, No. 10, 2006 2291
Figure 2. Changes in average aspect ratio (A), increments in length
(B) and diameter (C), and the measured Pd/Au molar ratios using
EDX (D) with the calculated Pd/Au molar ratios of 0.5, 1.2, 1.8,
2.2, and 2.8 by assuming a complete conversion of the added Pd2+
ions to Pd.
770 nm with a 33 nm fwhm. It shows a 40 nm red-shift in
comparison with the experimental value, and its width is
much narrower than the measured one. This indicates that
the prepared Au NRs have a distribution in aspect ratios, as
evidenced by the standard deviation of 12%. This distribution
is also responsible for the absence of fine structures in
measured spectra. In Figure 3B, at a thickness of 1.2 nm for
the Pd shell, the extinction spectrum shows a broadened peak
at 700 nm and a shoulder around 590 nm. Further increasing
the Pd shell thickness causes this peak to blue-shift and its
intensity decreases further. The shoulder also shows slight
blue-shifts, and its intensity shows a small increase. At a Pd
shell thickness of 4.8 nm, the original shoulder (ca. 590 nm)
becomes a broad peak at 560 nm, while the original peak
(700 nm) turns out to be a shoulder around 670 nm.
Therefore, we can roughly divide the extinction spectrum at
a wavelength range between 400 and 1000 nm into two
parts: one from the LSPR of the Au NR core and the other
from the LSPR of the Pd shell. The LSPR peak of the Au
core blue-shifts due to the decrease in overall aspect ratio
and due to the difference in dielectric constant between Au
and Pd. Its reduced intensity and broadened bandwidth come
from increasing damping from the Pd shell. This can be seen
more clearly in Figure S1A, where only the calculated LSPR
bands are shown. Similarly, the LSPR of the Pd shell also
shows a small blue-shift due to the decrease in aspect ratio.
By fixing the overall aspect ratio at 2.4, we calculated the
LSPR spectra of bimetallic nanorods by decreasing the length
of the Au core gradually while keeping its diameter
unchanged (Figure S1B). Because the molar fraction of Pd
has reached 71%, the extinction spectrum is dominated by
the Pd shell. Decreasing the Au core length causes further
increase of the molar fraction of Pd. This leads to a small
blue-shift of the LSPR band. Therefore, not only the overall
2292
Figure 3. (A) UV-visible absorption spectra of (a) the as-prepared
Au NRs dispersion and (b-f) Pd-coated gold nanorod dispersions
with the amount of Pd2+ ions given in Figure 1. Each spectrum is
normalized by its absorption at 350 nm. (B) Calculated extinction
spectra for single nanorod predicted by DDA calculation using the
lengths and diameters given in Figure 2B and C. Each spectrum is
normalized by its absorption at 300 nm.
aspect ratio but also the fraction of Pd in the bimetallic
nanorod influences the position of the LSPR band. At the
Au core length of zero, a pure Pd nanorod forms. It has a
LSPR position of 540 nm and a fwhm of 400 nm (much
broader than that of a pure gold nanorod). In contrast, a 50
nm Pd solid cube has a SPR position at 390 nm.2c Note that
the volume of the Pd nanorod here (58.4 nm long and 24.3
nm wide) is only 28% of the volume of the 50 nm Pd cube.
Therefore, in regard to pushing the SPR band of Pd to the
visible region, a nanorod structure is more effective. In
addition, by forming Pd shell/Au core bimetallic nanorods,
the LSPR band can be red-shifted further. The broad SPR
spectral feature of the Pd shell can be ascribed to the
dielectric function of Pd and has been confirmed both
experimentally and theoretically for Pd nanoboxes and
triangular and hexagonal nanoplates. Their UV-vis extinc-
tion spectra show a quite broad SPR peak and are in
agreement with corresponding DDA analysis.9
The low-magnification TEM image shown in Figure 4A
demonstrates a distinct variation in contrast between the
darker gold core and the lighter palladium shell, verifying
the formation of a Pd shell around the Au core. The
difference in morphology, especially at rod ends, can be also
seen clearly. HRTEM images (Figure 4B and C) provide
direct evidence of the crystal structure of the Pd shell. The
Pd shell grows epitaxially around the Au NR and is thus a
Nano Lett., Vol. 6, No. 10, 2006

Figure 4. (A) Low-resolution TEM image of single bimetallic
nanorods clearly revealing the formation of core-shell structure.
High-resolution TEM images of bimetallic nanorods when the
electron beam is aligned in the [001] (B) and [011] (C) directions,
showing the distribution of the atoms on the {001} and {111} faces,
respectively. The insets are the corresponding fast Fourier transform
images. (D) cross-section showing the morphology transition from
pure Au NRs to Pd/Au bimetallic nanorods.
single-crystal structure. When the electron beam is aligned
in the [001] direction, the distribution of the atoms on the
{001} faces of the Pd shell is imaged. The lattice distance
is 2 Å, corresponding to a lattice distance of {002} surfaces
of Pd. This is also confirmed by the corresponding fast
Fourier transform (FFT) analysis (in the inset): the FFT
pattern corresponds to only one set of diffraction spots:
{002}. In contrast, when the electron beam is aligned in the
[011] direction, both the (111) and (002) lattice planes of
the Pd shell are imaged (see the {002} and {111} diffraction
spots in the corresponding FFT pattern). Therefore, the Pd
shell grows along the [001] direction and all six surfaces
are terminated by {100} facets, resulting in the rectangularly
shaped Pd Shells.
Using Au NRs as templates, different metals such as Ag
and Pd have been employed to form shell structures around
them.6,10 Among these metals, the Ag shell has been studied
the most. For Au core/Ag shell structures, under certain
conditions, all of them show dumbbell shapes, although the
detailed morphologies of the dumbbell may be varied
(presumably because of the different reaction conditions used
in each procedure).6 A similar mechanism is employed to
explain the formation of dumbbell shapes: bilayers of the
cationic surfactants are assumed to stabilize the Au NRs and
make them water soluble.4 These surfactant molecules are
expected to form better close-packed structures on {110}
Nano Lett., Vol. 6, No. 10, 2006
surfaces than on {111} surfaces.4 Hence, the {111} facet is
more accessible to silver atoms than the {110} facet.
Therefore, deposition of more silver occurs preferentially at
the ends. Thus, the two ends of nanorods get thicker, and
the aspect ratio of the nanorods gets reduced. The same
mechanism is believed to also be responsible for the
dumbbell shapes observed in the overgrowth of Au on Au
NRs.10 Basically, this competing growth is more related to
the difference in packing density of CTAB bilayers on
different facets rather than the activities of the facets
themselves. Until now, competing growth between the side
facets {110} and {100} due to their different stabilities has
not been exhibited, albeit suggested. Here in our case, instead
of dumbbell structures, we observed a rectangular shape of
the Pd shell. We believe that it is due to the competing
growth between the {110} and {100} side facets. Because
the surface energy of the {110} surface is higher than that
of the {100} surface, a driving force exists to minimize the
overall surface energy by changing morphologies to more
stable ones. Thus, palladium grows faster along the {110}
planes than along the {100} planes. We observe an enlarge-
ment of the {100} surfaces in sacrificing of {110}. The final
consequence is the disappearance of all {110} facets. This
process leads to the formation of a rectangularly shaped Pd
shell terminated with only four {100} side surfaces (Figure
4D). We believe, therefore, that surface-energy minimization
is responsible for the growth of the Pd shell in our case.
Song et al. coated a Pd shell around gold nanorods
synthesized by an electrochemical procedure.10 In their case,
instead of forming a single-crystal Pd shell, polycrystalline
Pd nanoparticles with diameters of about 2-4 nm are formed
around the surface of the Au NRs. They believe that this is
due to lattice mismatch between Au and Pd. The lattice
constant of Pd (3.89 Å) is smaller than that of Au (4.08 Å).
A lattice mismatch of 4.6% may induce island nucleation/
growth of Pd nanoclusters on Au nanorod surfaces. Our
results indicate that by fine control of reaction conditions,
epitaxial growth can be also achieved. Our results also
indicate that although CTAB can stabilize the {110} surfaces
of Au, changes of interfacial environments can destabilize
this interaction. The stabilizing role of CTAB to the {110}
facets of Pd might be less than that of Au; thus, {110}
surfaces get reduced to minimize the surface energy.
However, the exact mechanism behind it needs further
investigation. With a fine control of growth conditions, we
can realize competing growth on the side facets. Because
the unstable {110} facets are coated with Pd and the outmost
surfaces of Au/Pd nanostructures are enclosed with stable
{100} surfaces, the core-shell NRs we demonstrated here
should be more stable than pure Au NRs.
The surfaces of Au NRs are coated with palladium by
reducing PdCl42- on the metallic surface. The thickness of
the Pd shell is tailored through controlled addition of PdCl42-
and tuned from 1.2 to 4.8 nm here, with a possibility to
further scale down or up. We showed that the changes in
the morphology of the resulting nanorods are due to the
competing growth on two side facets. We believe that the
energy minimization to a more stable morphology is the
2293
driving force for the formation of the rectangular shape of
the Pd shell. The appearance of an absorption shoulder
around 550 nm makes a strong EM enhancement in the
visible region feasible. In addition, because of the epitaxial
growth of the Pd shell, the single-crystal structure and well-
defined surfaces also make the study of another enhance-
ment factor (chemical enhancement) possible. These novel
Pd/Au bimetallic nanostructures may also serve as good
materials for SERS applications.
Acknowledgment. The work was supported by National
Natural Science Foundation of China (Grant No. 10334060)
and the “973” National Key Basic Research Program of
China (Grant No. 2005CB623602 and 2006CB705600). Z.L.
thanks the financial support of the National Natural Science
Foundation of China 10525419.
Supporting Information Available: Figures showing the
calculated LSPR extinction spectra of (a) a pure gold
nanorod, (b-f) Pd-coated gold nanorods, and the calcu-
lated LSPR extinction spectra of the bimetallic nanorods by
fixing the overall aspect ratio at 2.4 and by decreasing the
length of the Au core from 48.8, 38.8, 28.8, and 18.8 to 0
nm. The morphology of the bimetallic nanorod and the
calculated molar fraction of Pd in the corresponding nanorod
by using their sizes are shown in the right panel. This
material is available free of charge via the Internet at http://
pubs.acs.org.
2294
References
(1) (a) El-Sayed, M. A. Acc. Chem. Res. 2001, 34, 257. (b) Murphy, C.
J.; Sau, T. K.; Gole, A. M.; Orendorff, C. J.; Gao, J.: Gou, L.;
Hunyadi, S. E.; Li, T. J. Phys. Chem. B 2005, 109, 13857.
(2) (a) Wang, Z. L.; Ahmad, T. S.; El-Sayed, M. A. Surf. Sci. 1997,
380, 302. (b) Sau, T. K.; Murphy, C. J. J. Am. Chem. Soc. 2004,
126, 8648. (c) Xiong, Y. J.; Chen, J. Y.; Wiley, B.; Xia, Y. N. J.
Am. Chem. Soc. 2005, 127, 7332.
(3) Yu, Y. Y.; Chang, S. S.; Lee, C. L.; Wang, C. R. J. Phys. Chem. B
1997, 101, 6661.
(4) (a) Wang, Z. L.; Mohamed, M. B.; Link, S.; El-Sayed, M. A. Surf.
Sci. 1999, 440, L809. (b) Wang, Z. L.; Gao, R. P.; Nikoobakht, B.;
El-Sayed, M. A. J. Phys. Chem. B 2000, 104, 5417.
(5) (a) Nikoobakht, B.; El-Sayed, M. A. Langmuir 2001, 17, 6368. (b)
Gai, P. L.; Harmer, M. A. Nano. Lett 2002, 2, 771. (c) Johnson, C.
J.; Dujardin, E.; Davis, S. A.; Murphy, C. J.; Mann, S. J. Mater.
Chem. 2002, 12, 1765.
(6) (a) Ah, C. S.; Hong, S. D.; Jang, D. J. J. Phys. Chem. B 2001, 105,
7871. (b) Huang, C.; Yang. Z.; Chang. H. T. Langmuir 2004, 20,
6089. (c) Liu, M.; Guyot-Sionnest, P. J. Phys. Chem. B 2004, 108,
5882.
(7) Jana, N. R. Gearheart, L.; Obare, S. O.; Murphy, C. J. Langmuir
2002, 18, 922.
(8) (a) Jana, N. R.; Gearheart, L.; Murphy, C. J. AdV. Mater. 2001, 13,
1389. (b) Nikoobakht, B.; El-Sayed, M. A. Chem. Mater. 2003, 15,
1957. (c) Gou, L. F.; Murphy, C. J. Chem. Mater. 2005, 17, 3668.
(9) (a) Xiong, Y.; Wiley, B.; Chen, J.; Li, Z.-Y.; Yin, Y.; Xia, Y. Angew.
Chem., Int. Ed. 2005, 44, 7913. (b) Xiong, Y.; Chen, J.; Wiley, B.;
Xia, Y.; Yin, Y.; Li, Z.-Y. Nano Lett. 2005, 5, 1237. (c) Xiong, Y.;
McLellan, J. M.; Chen, J.; Wiley, B.; Yin, Y.; Li, Z.-Y.; Xia, Y. J.
Am. Chem. Soc. 2005, 127, 17118.
(10) Song, J. H.; Kim, F.; Kim, D.; Yang, P. D. Chem.sEur. J. 2005,
11, 910.
NL061722C
Nano Lett., Vol. 6, No. 10, 2006