Synthesis, stability, and surface plasmonic properties of rhodium multipods, and their use as substrates for surface-enhanced raman scattering

Article abstract of DOI:10.1002/anie.200503174

(Graph Presented) Multipodal Rh nanocrystals (see HRTEM image of a tripod) prepared through a polyol synthesis display surface plasmon resonance bands in the visible region. Surface-enhanced Raman scattering from 4-mercaptopyridine adsorbed on Rh multipods is 19-times stronger than that from Rh nanocubes.

Full text of DOI:10.1002/anie.200503174

Communications
Rhodium Nanoparticles
DOI: 10.1002/anie.200503174
Synthesis, Stability, and Surface Plasmonic
Properties of Rhodium Multipods, and Their Use
as Substrates for Surface-Enhanced Raman
Scattering**
Nobuyuki Zettsu, Joseph M. McLellan, Benjamin Wiley,
Yadong Yin, Zhi-Yuan Li, and Younan Xia*
Controlling the shape of nanocrystals is critical to modern
materials chemistry because the intrinsic properties of most
[
1]
nanomaterials are strongly dependent on their shape.
Tripod and tetrapod shapes have recently attracted much
attention owing to their potential use as building blocks in the
fabrication of complex, multiple-terminal devices through
[
2]
self-assembly. Multipods made of type II–VI semiconduc-
[
3]
[4]
[5]
tors, such as CdSe, CdS, and CdTe , have been known for
several years. A key parameter for growing multipods of these
solid materials is the energy difference between the zinc blend
and wurtzite structures at certain temperatures, as one
structure may be preferred during nucleation and the other
[
5]
during growth. As most metals only exist in one crystal
structure (often in the face-centered-cubic lattice), it is more
[
6]
challenging to grow branched nanocrystals from metals; it
[*] Dr. N. Zettsu, J. M. McLellan, Prof. Y. Xia
Department of Chemistry
University of Washington
Seattle, WA 98195-1700 (USA)
Fax: (+1)206-685-8665
E-mail: xia@chem.washington.edu
B. Wiley
Department of ChemicalEngineering
University of Washington
Seattle, WA 98195 (USA)
Dr. Y. Yin
The Molecular Foundry
Lawrence Berkeley National Laboratory
Berkeley, CA 94720 (USA)
Prof. Z.-Y. Li
Laboratory of OpticalPhysics
Institute of Physics
Chinese Academy of Sciences
Beijing 100080 (P.R. China)
[**] This work was supported in part by a grant from the NSF (DMR-
0451788) and a fellowship from the David and Lucile Packard
Foundation. Y.X. is a Camille Dreyfus Teacher Scholar (2002). N.Z.
was partially supported by a research fellowship from the Japan
Society for the Promotion of Science (JSPS). B.W. thanks the Center
for Nanotechnology at the UW for a fellowship. Z.-Y.L. was
supported by the NationalKey Basic Research SpecialFoundation
of China (Grant No. 2004CB719804). This work used the Nanotech
User Facility (NTUF), a member of the National Nanotechnology
Infrastructure Network (NNIN) funded by the NSF. We thank the
Molecular Foundry at the Lawrence Berkeley National Laboratory for
the HRTEM analysis.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
1
288
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 1288 –1292

Angewandte
Chemie
will be very difficult at least, if not impossible, to achieve
branched growth by means of structural transformation
between the nucleation and growth steps. Nevertheless, a
[
7]
[8]
[9]
number of noble metals including Au, Pt, and Rh have
recently been demonstrated to form branched nanocrystals.
The use of a capping agent such as cetyltriethylammonium
bromide (CTAB) or poly(vinyl pyrrolidone) (PVP) is imper-
ative, however, and this is probably just one of the prereq-
uisites for achieving a branched morphology. The growth
mechanism is yet to be completely understood for these
systems.
Herein, we report the synthesis of Rh multipods as well as
their stability, surface plasmonic properties, and use as
substrates for surface-enhanced Raman scattering (SERS).
Toshima and co-workers first reported their observation of
[
9a]
branched Rh nanocrystals in the late 1970s. Most recently,
Tilley and co-workers reported the polyol synthesis of Rh
multipods and proposed a mechanism based on homogenous
nucleation and seeded growth to account for the anisotropic
[
9b]
growth. However, neither of these reports described any
appealing properties associated with the Rh multipods. We
also note that in both cases, the arms of the branched
nanostructures were rather short, which could possibly be
[
10]
related to the involvement of water in the reaction media.
Here, we demonstrate that Rh multipods can be synthesized
Figure 1. TEM images of Rh nanocrystals synthesized in air at 1408C
in high yields when anhydrous Na RhCl6 (rather than
3
for different periods of time: a) 5 min; b) 10 min; c) 10 min (HRTEM
image); d) 90 min; e) 6 h; and f) 48 h. The inset in (f) shows a TEM
image of the sample at t=120 h.
RhCl ·nH O) is used as a precursor of Rh. We also report
3
2
that the branched morphology can only prevail when oxygen
is excluded from the reaction system. Otherwise, oxidative
etching and Ostwald ripening would transform the branched
nanocrystals into spheres in an effort to minimize the total
surface energy.
in the range 358 < 2q < 808 which correspond to the (111),
(200), and (220) diffractions from face-centered-cubic Rh
(JCPDS #05-0685).
The branched Rh nanocrystals were synthesized through a
polyol process. In a typical procedure, Na RhCl and PVP
3
6
were separately dissolved in ethylene glycol and the two
solutions were injected simultaneously into ethylene glycol
As the reaction continued (Figure 1d and e), the overall
length-to-width aspect ratio of the arm decreased as com-
pared to the sample shown in Figure 1b. The branched Rh
nanocrystals slowly evolved into rice-shaped particles at t =
48 h and eventually became spheres at t = 120 h. The overall
size of the nanocrystal also decreased in this process. The
transformation from the branched shape into a spherical one
could be interpreted in terms of oxidative etching and
Ostwald ripening. Like other face-centered-cubic metals,
the surface energy of Rh decreases in the order of (110),
(
preheated to 1408C) by a syringe pump. The color of the
solution rapidly turned from rose pink to dark brown within
3
ꢀ
1
min, indicating the formation of Rh nanocrystals as RhCl6
was reduced by ethylene glycol. The as-obtained Rh nano-
crystals were characterized using transmission electron micro-
scopy (TEM). Figure 1 shows TEM images taken from a
series of samples that were prepared by increasing the
reaction time from 5 min to 48 h. In the initial stage of the
reaction (Figure 1a and b), the particles exhibited a well-
defined, branched morphology with three (tripod) or four
[
11]
(100), and (111). In an effort to minimize the total surface
energy, the arm was slowly etched from the tip and the
resultant Rh ions were reduced and deposited on the central
portion of the nanocrystal to increase the percentage of
(111) facets on the surface. Comparison of the intensity ratios
of XRD peaks indicates that the transformation of shape
resulted in a decrease in the proportion of higher energy
(110) facets on the nanocrystals (see Supporting Informa-
(
tetrapod) arms. Among them, tripods were in the majority,
comprising approximately 70% of the total particle popula-
tion. The arms were roughly 11 nm long at t = 10 min, but the
apparent arm size varied slightly as a result of the difference
in their orientation on the TEM grid. High-resolution TEM
(
HRTEM) studies clearly showed that the same fringe pattern
[
12]
continuously extended throughout each branched nanocrys-
tal, suggesting the formation of a single crystal (Figure 1c).
The interplanar distance was measured to be 0.22 nm. This
value could be assigned to the spacing between {111} planes,
implying that the arms grew in the h111i directions. Powder
X-ray diffraction (XRD) patterns (see Supporting Informa-
tion) of the as-prepared Rh multipods exhibited three peaks
tion). As demonstrated for both Ag and Pd, the oxygen in
air plays a key role in the morphological evolution. Contin-
uous bubbling of Ar during the synthesis greatly improved the
stability of the branched Rh nanocrystals. Figure 2shows
TEM images of Rh nanocrystals that were obtained under Ar
protection at t = 10 min and 48 h, respectively. From these
images, it is apparent that the branched morphology could be
Angew. Chem. Int. Ed. 2006, 45, 1288 –1292
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1289

Communications
predict the contributions of scattering and absorption to the
extinction spectra as well as the peak positions. The calcu-
lations indicate that a spherical particle has an SPR peak in
the deep UV region while the tripod displays a peak around
3
80 nm. Although the tripod was assumed to have a planar
structure to save computational time, the DDA calculation
nearly replicated the observed UV/Vis absorption spectrum
of the branched Rh nanocrystals. The experimental spectrum
displayed a somewhat broader resonance peak, which is most
likely caused by the variation in both shape and size of the Rh
nanocrystals in the real sample. In the presence of air, the
SPR peak shifted toward shorter wavelengths as the reaction
proceeded (Figure 3a). This shift can be attributed to the
transformation from the branched shape into a spherical one.
The spectral changes indicate that oxidative etching had
already started at about 10 minutes into the reaction. In
contrast, the position of the SPR peak remained unchanged
over time (see Figure 3b) when the reaction was carried out
under Ar. The peak gradually became more intense with the
reaction time. This observation implies that protection by Ar
not only promoted anisotropic growth but also helped to
preserve the less stable, branched morphology.
Figure 2. TEM images of Rh nanocrystals synthesized under Ar at
408C for a) 10 min and b) 48 h.
1
preserved for a long period of time when oxidative etching
was excluded from the synthesis.
The UV/Vis absorption spectrum of the branched Rh
nanocrystals displayed an attractive surface plasmon reso-
nance (SPR) feature (Figure 3a); a weak and broad peak
We also systematically varied the reaction conditions to
find the optimal parameters for generating branched Rh
nanocrystals. In general, both the reaction temperature and
the concentration of the precursor played important roles in
determining the morphology of the product. As compared to
the sample prepared at 1408C, a decrease in the temperature
(e.g., to 1108C, Figure 4a) led to the formation of branched
structures with higher aspect ratios (with arms ca. 2nm wide
by ca. 10–11 nm long). In contrast, relatively isotropic,
cubelike nanocrystals were obtained at a higher temperature
(1808C, Figure 4b). When the reaction conditions were fixed
at 1408C and 10 min, a decrease in the concentration of
Na RhCl6 by five times (to 5 mm, Figure 4c) resulted in
3
embryonic multipods that featured largely faceted rather than
Figure 3. UV/Vis spectra of the Rh nanocrystals as a function of
reaction time. The reaction was performed under a) air and b) contin-
uous bubbling of Ar gas. Note that the surface plasmon resonance
peak of the Rh multipods was shifted toward shorter wavelengths with
increasing reaction time when air was present in the synthesis.
around 380 nm appeared in the initial stage of the reaction. It
is well documented that both thin films and spherical
nanoparticles of Rh do not exhibit any extinction peaks in
the visible region as a result of the very small imaginary part
[
13]
of the dielectric constant.
feature, we carried out discrete dipole approximation
DDA) calculations for an individual Rh spherical particle
To understand this optical
(
Figure 4. TEM images showing the effects of temperature and the
concentration of the precursor on the formation of branched Rh
nanocrystals: a) 25 mm Na RhCl at 1108C; b) 25 mm Na RhCl at
of 10 nm diameter and a tripod nanocrystal featuring an outer
triangle of 19 nm, a thickness of 5 nm, and an inner angle of
3
6
3
6
[
14]
1
208 (see Supporting Information).
In general, if the
1
808C; c) 5 mm Na RhCl at 1408C; and d) 50 mm Na RhCl at 1408C.
3
6
3
6
geometric parameters are known, DDA calculations can
All samples were collected at t=10 min into the reaction.
1
290 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 1288 –1292

Angewandte
Chemie
spherical nanostructures. When the concentration of
Na RhCl was doubled (to 50 mm, Figure 4d), the branched
3
6
morphology was also observed. Interestingly, the majority of
these nanostructures were tetrapods rather than tripods.
As the thermodynamically favored particle shape of a
face-centered-cubic metal is a truncated octahedron, the
branched morphology must be a kinetically controlled
product. The finding that anisotropic growth prevailed at
relatively lower temperatures is consistent with the results
that were recently obtained for Rh multipods synthesized
[
9b]
through a different polyol process.
Chen et al. proposed
that the shape of a particle of gold (or another face-centered-
cubic metal) is mainly determined by the competitive growth
[
7a]
Figure 5. SERS spectra (cps=counts per second) of 4-mercaptopyr-
idine on films of Rh nanocrystals: a) multipods synthesized under Ar
see Figure 2b); b) multipods synthesized in air (see Figure 1a); and
along the h111i and h100i directions.
As the reaction
temperature was increased, the rates of growth along these
two directions were decreased to generate cubic nanocrystals.
The shape of Rh nanocrystals was also strongly dependent on
the concentration of the precursors. Although branched seeds
were formed at all concentrations of Na RhCl examined in
(
c) 9-nm wide cubes (see Figure 4b).
sharp corners or edges are especially active SERS substrates,
3
6
2
the present work (5, 25, and 50 mm), TEM studies revealed no
additional growth of the crystals and a drastic transformation
of their shape with time in the case of 5 mm Na RhCl . These
as the local value of j E j at the tips can be more than
[
19]
500 times that of the applied field. While the SERS results
shown here are purely qualitative, the high-quality spectra
obtained with the Rh multipods are encouraging and may
warrant further exploration.
3
6
observations indicate that most of the precursor was con-
sumed in the nucleation stage. In comparison, there was
precursor left in the solution after nucleation when higher
In summary, we have demonstrated that Rh multipods can
be prepared in large quantities by reducing Na RhCl with
concentrations (e.g., 25 and 50 mm) of Na RhCl6 were
3
3
6
involved. As a result, the arms of each branched seed could
continue to grow. As long as the difference in growth rate
along the h111i and h100i directions was maintained, the
branched nanocrystals could grow into larger dimensions
without a change in their morphology.
ethylene glycol in the presence of PVP. These highly branched
nanocrystals are not stable in air at elevated temperatures;
they are slowly converted into spherical particles when the
synthesis was allowed to proceed for long periods of time. We
believe the formation of multipods is a kinetically controlled
process, and thus parameters such as precursor concentration
and temperature both play important roles in controlling the
morphology of the product. When the synthesis was protected
with continuous bubbling of Ar, both the anisotropy in growth
and the stability of the product could be enhanced. The
optical properties of the tripods were experimentally and
theoretically investigated by UV/Vis spectroscopy and DDA
calculations, with both methods indicating the presence of an
SPR peak around 380 nm for the tripods. These Rh multipods
display unprecedented SERS activities, which may find use in
applications such as the in situ monitoring of catalytic
The interesting plasmonic properties of these multipods
may make them attractive for use as SERS substrates. One of
the primary enhancement mechanisms in SERS is a result of
electromagnetic enhancement arising from the extremely
[
15]
high local fields due to surface plasmon resonance (SPR).
Previous explorations of SERS on Group VIIIB metals such
[
16]
as Rh have suffered from fairly weak enhancement.
However, these studies have mainly focused on Rh-coated
electrodes or Rh particles less than 10 nm in size, none of
which exhibited any SPR features in the visible region.
Figure 5 compares the SERS activities of thin films cast from
three types of Rh nanocrystals: a) multipods synthesized
under Ar, b) multipods synthesized in air, and c) nanocubes.
The nanocubes of 9 nm in edge length exhibited a SPR peak
at 250 nm, while the multipods showed SPR peaks at 360 nm
and 380 nm for those synthesized in air and under Ar,
respectively. We selected 4-mercaptopyridine (4-MP) as a
probe molecule because it has a large scattering cross-
[
20]
reactions.
Experimental Section
In a typical synthesis, ethylene glycol (EG; J. T. Baker; 2mL) was
placed in a three-necked flask equipped with a reflux condenser and a
magnetic stirrer bar embedded in PYREX glass and heated in air at
the designated temperature (e.g., 110, 140, or 1808C) for 1.5 h. A
temperature controller (ACE, 12125–14) equipped with a probe
(ChemGlass, 104 TC105) and mantle heater (Electrothermal,
HM0050-HS1) was used to maintain the temperature. In separate
[
17]
[18]
section and it has been well studied. Using films that
had been modified with 4-MP, it was found that both of the
multipod samples gave well-defined, relatively strong SERS
signals while the signal from the cubes was quite weak. The
multipods prepared under Ar had the highest activity, with an
enhancement that was 19 times stronger than that for the
nanocubes. These results suggest a trend of stronger SERS
signal with a red shift in the SPR peak. It is also possible that
the sharp tips of the multipods provided additional enhance-
ment. Recent studies have shown that nanoparticles with
ꢀ
2
vials, Na RhCl (34.6 mg, 9.0 ” 10 mmol, Aldrich) and poly(vinyl
3
6
ꢀ1
pyrrolidone) (PVP; M = 55000, Aldrich; 51.3 mg, 4.5 ” 10 mmol in
r
terms of the repeating unit) were each dissolved in EG (0.8 mL) at
room temperature. The molar ratio of Na RhCl and the repeating
3
6
unit of PVP was controlled at 1:5. These two solutions in EG were
injected simultaneously into the heated EG using a syringe pump at a
ꢀ1
rate of 0.2mLmin . All of the precursor was reduced to Rh
Angew. Chem. Int. Ed. 2006, 45, 1288 –1292
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1291

Communications
multipods, as indicated by UV/Vis spectroscopy. The reaction was
oxidative etching and thus greatly modify the shape of a
nanocrystal.
[11] Q. Jiang, H. M. Lu, M. Zhao, J. Phys. Condens. Matter 2004, 16,
521 – 530.
[12] a) B. Wiley, T. Herricks, Y. Sun, Y. Xia, Nano Lett. 2004, 4, 1733 –
1739; b) Y. Xiong, J. Chen, B. Wiley, Y. Xia, S. Aloni, Y. Yin, J.
Am. Chem. Soc. 2005, 127, 7332– 7333.
continued up to 120 h. A set of samples was taken over the course of
each synthesis using a glass pipette. For synthesis performed under
Ar, all other experimental parameters were kept the same except for
the continuous bubbling of Ar gas. The product was collected by
centrifugation and washed with acetone and water several times to
remove EG and excess PVP.
The as-obtained samples were characterized by transmission
electron microscopy (TEM; JEOL 1200EX II, operated at 80 kV),
powder X-ray diffraction (XRD; Philips 1820 diffractometer, l =
[13] Y. Wang, S. D. Russell, R. L. Shimabukuro, J. Appl. Phys. 2005,
97, 023708.
[14] a) B. T. Drani, P. J. Flatau, J. Opt. Soc. Am. A 1994, 11, 1491 –
1499; b) R. Jin, Y. Cao, C. A. Markin, K. L. Kelly, G. Schatz, J.-
G. Zheng, Science 2001, 294, 1901 – 1903; c) Y. Xiong, J. Chen, B.
Wiley, Y. Xia, Y. Yin, Z.-Y. Li, Nano Lett. 2005, 5, 1237 – 1242.
[15] N. Halas, MRS Bull. 2005, 30, 362– 367.
[16] Z.-Q. Tian, B. Ren, D.-Yin. Wu, J. Phys. Chem. B 2002, 106,
9463 – 9482.
[17] W. B. Cai, B. Ren, X. Q. Li, C. X. She, F. M. Liu, X. W. Cai, Z. Q.
Tian, Surf. Sci. 1998, 406, 9 – 2 2 .
[18] Z. Wang, S. Pan, T. D. Krauss, H. Du, L. J. Rothberg, Proc. Natl.
Acad. Sci. USA 2003, 100, 8638 – 8643.
[19] K. L. Kelly, E. Coronado, L. L. Zhao, G. C. Schatz, J. Phys.
Chem. B 2003, 107, 668.
[20] R. Gomez, J. Perez, J. Solla-Gullon, V. Montiel, A. Aldaz, J.
Phys. Chem. B 2004, 108, 9943 – 9949.
0
.154180 nm), and UV/Vis spectroscopy (Hewlett Packard 8452A
diode-array spectrometer). TEM samples were prepared by placing a
drop of the aqueous suspension on a carbon-coated copper grid (Ted
Pella, Redding, CA), and drying under ambient conditions. The
substrate for surface-enhanced Raman scattering (SERS) was
prepared by drying a 3-mL aliquot of the final product on a 25-nm
thick Al film supported on a Si wafer. The sample was then incubated
in an aqueous solution of 4-mercaptopyridine (4 mm) for 1 h, rinsed
with deionized water, and dried in air. SERS spectra were obtained
with laser excitation (3 mW at the sample) at 785 nm (Renishaw
HPNIR785 laser) and a spot size of approcimately 1.6 mm using a
Leica DM IRBE optical microscope equipped with a Renishaw inVia
Raman spectrometer and a thermoelectrically cooled CCD detector.
Received: September 7, 2005
Published online: January 17, 2006
Keywords: nanostructures · Raman spectroscopy · rhodium ·
.
surface plasmon resonance
[
1] a) Y. Xia, P. Yang, Y. Sun, Y. Wu, B. Mayers, B. Gates, Y. Yin, F.
Kim, H. Yan, Adv. Mater. 2003, 15, 353 – 389; b) Y. Xia, N. J.
Halas, MRS Bull. 2005, 30, 338 – 348; c) Y. Sun, Y. Xia, Science
2002, 298, 2176 – 2179; d) E. C. Scher, L. Manna, A. P. Alivisatos,
Philos. Trans. R. Soc. London Ser. A 2003, 361, 241 – 257; e) P.
Yang, MRS Bull. 2005, 30, 85 – 91; f) C. J. Murphy, Science 2002,
298, 2139 – 2141.
[
2] a) D. J. Milliron, S. M. Hughes, Y. Cui, L. Manna, J. Li, L.-W.
Wang, A. P. Alivisatos, Nature 2004, 430, 190 – 195; b) D. L.
Wang, C. M. Lieber, Nat. Mater. 2003, 2, 355 – 356.
[
3] a) L. Manna, E. C. Scher, A. P. Alivisatos, J. Am. Chem. Soc.
2000, 122, 12700 – 12706; b) X. Peng, Adv. Mater. 2003, 15, 459 –
463; c) Z. A. Peng, X. Peng, J. Am. Chem. Soc. 2002, 124, 3343 –
3353.
[
[
4] a) Y. W. Jun, S. M. Lee, N. J. Kang, J. Cheon, J. Am. Chem. Soc.
001, 123, 5150 – 5151; b) M. Chen, J. Lu, S. Zhang, Y. Xie, Y.
2
Xiong, Y. Qian, X. Liu, J. Mater. Chem. 2002, 12, 748 – 753.
5] a) L. Manna, D. J. Milliron, A. Meisel, E. C. Scher, A. P.
Alivisatos, Nat. Mater. 2003, 2, 382– 385; b) C. Y. Yeh, Z. W.
Lu, S. Froyen, A. Zunger, Phys. Rev. B 1992, 46, 10086 – 10097.
6] C. Kittel, Introduction to Solid State Physics, 7th ed., Wiley, New
York, 1996.
[
[
7] a) S. Chen, Z. L. Wang, J. Ballato, S. H. Foulger, D. Carroll, J.
Am. Chem. Soc. 2003, 125, 16186 – 16187; b) E. Hao, R. C.
Bailey, G. C. Schatz, J. T. Hupp, S. Li, Nano Lett. 2004, 4, 32 7 –
3
30.
8] a) J. Chen, T. Herricks, Y. Xia, Angew. Chem. 2005, 117, 2645 –
648; Angew. Chem. Int. Ed. 2005, 44, 2589 – 2592; b) T.
[
2
Herricks, J. Chen, Y. Xia, Nano Lett. 2004, 4, 2367 – 2371; c) X.
Teng, H. Yang, Nano Lett. 2005, 5, 885 – 891.
[
9] a) H. Hirai, Y. Nakao, N. Toshima, J. Macromol. Sci. Chem. A
1978, 12, 1117 – 1141; b) J. D. Hoefelmeyer, K. Niesz, G. A.
Somorjai, D. Tilley, Nano Lett. 2005, 5, 435 – 438.
[
10] We were unable to synthesize well-defined Rh multipods when
RhCl ·nH O was used as a precursor. Water seems to enhance
3
2
1
292
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 1288 –1292