Core - Shell and hollow nanocrystal formation via small molecule surface photodissociation; Ag@Ag2Se as an example

Article abstract of DOI:10.1021/jp0616011

Metallic Ag nanoparticles have been converted to Ag₂Se nanoparticles at ambient temperature and open atmosphere by UV photodissociation of adsorbed CSe₂ on the Ag core surface. The photolysis could be prevented at any stage yielding Ag@Ag₂Se core - shell structures of different thickness. Depending on the initial Ag nanoparticle size, either hollow or filled nanocrystals of Ag₂Se could be prepared. The Kirkendall effect has been proposed to account for the formation of hollow nanoparticles. A coated-sphere Drude model has been used to explain the redshift of the Ag plasmon band as a function of the Ag ₂Se shell thickness as well as to provide the first estimates of the wavelength-dependent dielectric function of Ag₂Se. This photochemical method might be especially promising for carrying out a direct room-temperature phototransformation of metallic into semiconductor nanostructures already assembled on surface templates.

Full text of DOI:10.1021/jp0616011

15812
J. Phys. Chem. B 2006, 110, 15812-15816
Core-Shell and Hollow Nanocrystal Formation via Small Molecule Surface
Photodissociation; Ag@Ag2Se as an Example
Hua Tan, Shuping Li, and Wai Yip Fan*
Department of Chemistry, National UniVersity of Singapore, 3 Science DriVe 3, Singapore 117543
ReceiVed: March 15, 2006; In Final Form: June 15, 2006
Metallic Ag nanoparticles have been converted to Ag2Se nanoparticles at ambient temperature and open
atmosphere by UV photodissociation of adsorbed CSe2 on the Ag core surface. The photolysis could be
prevented at any stage yielding Ag@Ag2Se core-shell structures of different thickness. Depending on the
initial Ag nanoparticle size, either hollow or filled nanocrystals of Ag2Se could be prepared. The Kirkendall
effect has been proposed to account for the formation of hollow nanoparticles. A coated-sphere Drude model
has been used to explain the redshift of the Ag plasmon band as a function of the Ag2Se shell thickness as
well as to provide the first estimates of the wavelength-dependent dielectric function of Ag2Se. This
photochemical method might be especially promising for carrying out a direct room-temperature phototrans-
formation of metallic into semiconductor nanostructures already assembled on surface templates.
Introduction
diffraction (ED), energy-dispersive X-ray analysis (EDX), and
Fourier transform infrared (FTIR) spectroscopy.
The differences between the optical and electronic properties
of nanoparticles and their corresponding bulk solids have
rendered them invaluable for applications in catalysis, photonics,
optoelectronics, and biological processes.1-3 The syntheses of
metallic nanoparticles of controlled size and shape are being
pursued actively so that their properties could be tuned and
tailored for specific uses. Core-shell structures of alloys,
metal-insulator and metal-semiconductor types, have also
generated much interest by offering even more variations
in tunability of the physical properties,4-7 eventually leading
to novel optical and semiconductor devices. Although nano-
structures are usually made using chemical reduction methods
at elevated temperatures, better control of core-shell and
nanoshell formation is certainly needed if their properties are
to be studied systematically and their potential uses are to be
exploited fully.
The use of light to control the properties of nanostructures
has rarely been explored. The Mirkin group has recently shown
that silver nanosphere to nanoprism formation could be induced
by light.8,9 We have also demonstrated that visible light
photodissociation of small molecules such as iodine could be
used to prepare Ag@AgI core-shell nanoparticles of different
shell sizes.10 We focus on extending this technique to silver
selenide (Ag2Se) nanoparticles which have been investigated
via many other methods.11-13 The photochemical method has
since been found to be very versatile, and in this work, we report
the controlled formation of different thicknesses of Ag2Se shell
surrounding a Ag core by irradiating a solution of dissolved
carbon diselenide, CSe2, and Ag colloids under open atmosphere
at ambient temperatures. Depending on the initial size of the
Ag core, a progression from core-shell structures to either fully
formed Ag2Se nanoparticles or hollow Ag2Se nanoshells has
been observed and their physical properties have been character-
ized by UV-vis absorption spectroscopy, transmission electron
microscopy (TEM), powder X-ray diffraction (XRD), electron
Experimental Section
A silver colloid solution was prepared by reducing AgNO3
with sodium citrate according to the protocol in the literature;14
in this work, 18 mg AgNO3 was dissolved in 100 mL of H2O
and was brought to boiling. A 2-mL solution of 1% sodium
citrate was added, and the solution was boiled for 1 h. Silver
selenocyanate (AgNCSe) was synthesized from an equimolar
reaction of potassium selenocyanate (KNCSe) with AgNO3
in water. The obtained precipitate was washed with water
several times and was dried under vacuum overnight. CSe2 vapor
could be generated from the vacuum thermal decomposition
(250-300 °C) of silver selenocyanate (AgNCSe). The presence
of CSe2 vapor was verified using FTIR absorption spectroscopy
(ν1 ) 1300 cm-1, FTIR spectrum15), and with the aid of a
nitrogen gas flow, the vapor was bubbled into ethanol sol-
vent which turned slightly yellow. For the preparation of
Ag-Ag2Se core-shell and Ag2Se hollow nanoparticles, 1 mL
of the 0.35 M CSe2 ethanolic solution was mixed with the 4
mL 10-3 M Ag colloidal solution in a quartz cell under open
atmospheric conditions and then was placed under irradiation
of a Hanovia 254-nm lamp (20 W). After a certain period of
time, a UV-vis absorption spectrometer (Shimadzu UV-2550)
was used to record the change in the plasmon peak of Ag
colloid. The reaction can be stopped at any time by turning off
the UV lamp, and the final products, such as core-shell
nanoparticles or hollow nanoparticles, can be obtained by
controlling the molar ratio of Ag and CSe2. An aliquot of the
solution was removed for TEM, ED, EDX, and XRD measure-
ments. The TEM images (JEOL-2010 or JEOL-3010) were
taken of direct sampling of the solution on carbon-coated copper
grids at certain intervals of the irradiation. The ED and EDX
spectra were taken during TEM (JEOL-3010) imaging process.
Powder X-ray diffraction (XRD) patterns were recorded from
a large quantity of colloids supported on a glass slide using
SIEMENS D5005 diffractometer (CuKa λ ) 0.15418 nm) at a
scanning rate of 0.02° s-1 for 2θ in the range from 20° to 70°.
* Author to whom correspondence should be addressed. Fax: 6567791691;
e-mail: chmfanwy@nus.edu.sg.
10.1021/jp0616011 CCC: $33.50 © 2006 American Chemical Society
Published on Web 07/22/2006

Core-Shell and Hollow Nanocrystal Formation
J. Phys. Chem. B, Vol. 110, No. 32, 2006 15813
Figure 1. UV-vis absorption spectra showing the Ag plasmon bands
at various intermediate stages during the progression from Ag to core-
shell Ag@Ag2Se and finally to Ag2Se nanoparticles. (a) Ag colloids
prepared from 10-3 M AgNO3, (b) Ag colloids in excess CSe2, (c)
after 10 min, (d) after 30 min, (e) after 60 min, and (f) after 90 min of
irradiation.
We have used two methods to estimate the size of the
nanoparticles: first, the Debye-Scherrer formula using the data
from the XRD measurements and second and more commonly,
the actual diameter measurement of each nanoparticle from a
sample of more than 200 nanoparticles in the TEM images. The
detection of OCSe vapor was accomplished by carrying out the
irradiation in a three-neck round-bottom flask equipped with a
quartz window on one neck for passage of the UV light and on
the middle neck, a cylindrical glass cell of 15-cm length and
3-cm diameter equipped with CaF2 windows at each end. An
FTIR spectrum was recorded of the vapor phase across 1000-
4000 cm-1 at a resolution of 1 cm-1. For any part of the
experiment that required the use of lasers, a frequency-doubled
10-ns pulsed Nd:YAG laser (Continuum Surelite III-10, 532-
nm wavelength) was used to provide radiation energy of 10-
100 mJ/pulse.
Figure 2. TEM images showing the production of Ag, Ag@Ag2Se,
and Ag2Se at various stages of irradiation, corresponding to positions
a, c, e, and f of Figure 1. The left panel shows the distribution of
nanoparticles over a wide area (≈1 × 1 µm) and the right panel contains
enlarged images of single nanoparticles from each stage showing the
different structures. (a) Ag, (c) Ag@Ag2Se, (e) Ag@Ag2Se with hollow
interior beginning to form, and (f) hollow Ag2Se nanoparticles.
the low-resolution TEM image of the particles in Figure 2
corresponding to position c began to show evidence of core-
shell structures. An image contrast between the core and the
shell could be clearly seen, and a high-resolution TEM image
(JEOL-3010) taken of the core showed the existence of poly-
crystalline silver nanoparticles characterized by a lattice spacing
of 2.0 Å (200 plane) while patterns taken mainly of the shell
possessed orthorhombic polycrystalline Ag2Se attributes with
lattice spacings of 2.58 Å (121 plane) and 2.67 Å (112 plane).16
Results and Discussion
An ethanolic solution of CSe2 was mixed with a Ag colloidal
solution prepared earlier by reducing AgNO3 with sodium citrate
in water. The progress of the reaction during certain intervals
of irradiation by a 20 W Hanovia 254-nm lamp was followed
by UV-vis absorption spectroscopy. The UV-vis spectrum
showed the presence of the surface plasmon band of Ag
nanoparticles followed by a slight 10-nm redshift upon addition
of CSe2 (see Figure 1). Upon photolysis, the plasmon band was
broadened and redshifted further. In addition, an isosbestic point
at 370 nm was recorded. If excess CSe2 was used, a broad
absorption peak which tailed off at around 800 nm was
produced. The reaction could also be monitored easily by
following the color changes from yellow, red, blue, and finally
brown over a period of 1-2 h for typical concentrations of 10-3
M AgNO3 in excess CSe2.
The transformation to core-shell structures could be seen
when the TEM images taken of the nanoparticles formed under
conditions corresponding to positions a-f of Figure 1 were
examined (Figure 2). Initially, near-spherical silver nanoparticles
stabilized by citrate ions were observed with a mean size of 50
( 10 nm. Smaller mean sizes (10 ( 5 nm and 30 ( 10 nm)
have also been prepared by varying the relative concentration
of the citrate ions and AgNO3 in solution. Upon irradiation,
EDX analysis of the nanoparticles revealed two peaks
corresponding to Ag and Se only (Figure 3). As expected, the
Ag:Se ratio was >2 for core-shell structures since the Ag core
still remained to be fully converted. In the first stage of the
EDX spectrum (Figure 3a), only the Ag signal was observed.
As the irradiation got underway, some Ag2Se began to form
and in the EDX of Figure 3c the Ag-to-Se ratio had become
4:1. Upon further progress in the reaction, the EDX spectrum
showed a smaller Ag:Se ratio of 3:1 in Figure 3e and finally
the stoichiometric ratio was approached more closely during
the final stages of photolysis where only pure Ag2Se nanostruc-
tures were observed (Figure 3f). XRD patterns16 of the nano-
particles confirmed the formation of phase-pure orthorhombic
Ag2Se with a concomitant decrease of cubic Ag signals, in
agreement with the EDX analysis. All the XRD peaks could be
assigned to either Ag2Se or Ag.
We have also obtained electron diffraction patterns for the
pure Ag and pure Ag2Se nanoparticle with a hollow interior
(Figure 4). For each case, the patterns could be assigned only
to one species in polycrystalline form, confirming the full
conversion of Ag to Ag2Se upon irradiation. However, for
intermediate cases where both Ag and Ag2Se are present, the

15814 J. Phys. Chem. B, Vol. 110, No. 32, 2006
Tan et al.
Figure 3. EDX (top panel) and XRD (bottom panel) plots showing the evolution of Ag to Ag2Se nanoparticles at various stages of irradiation,
corresponding to positions a, c, e, and f of Figure 1. The Cu signals of the EDX plots belong to the carbon-coated copper grid used for nanoparticle
deposition. The triangular and square labels in the XRD graphs refer to cubic Ag and orthorhombic Ag2Se signals, respectively.
electron diffraction patterns are much more complicated and
difficult to analyze because of overlapping patterns.
measurements also verified the full conversion of Ag to Ag2Se
nanoparticles. The production of a contrast at the center of the
nanoparticle indicated a small electron density which im-
mediately suggested a hollow interior.
The formation of Ag2Se could be prevented at any stage of
irradiation, hence, the thickness of the Ag2Se shell enveloping
the Ag core could be controlled. The shell grew thicker at the
expense of the core with increasing irradiation period until the
nanoparticle became fully Ag2Se in chemical composition. For
small particles (10 ( 5 nm diameter), only a fully filled Ag2Se
was seen. However, evidence of Ag2Se hollow nanoshells was
seen for larger nanoparticles of mean diameter 50 ( 10 nm as
provided by images of Figure 2f. Magnification of the TEM
image showed that the core of the nanoparticle was again of a
different contrast to the shell. Electron diffraction carried out
on these particles revealed only rings belonging to Ag2Se; no
evidence of the Ag core was detected. EDX and XRD
The formation of Ag2Se in its various forms was initiated by
the addition of CSe2 into the Ag colloidal solution. An
immediate 10-nm redshift in the UV-vis spectrum indicates
the onset of a little amount of silver nanoparticulate aggregation.
Partial displacement of the citrate ion surfactants by neutral CSe2
could have caused a reduction in the intercolloidal electrostatic
repulsion.17,18 The formation of Ag2Se shells most probably took
place at the Ag surface upon photolysis of adsorbed CSe2. Since
only UV (<300 nm) rather than visible light (>360 nm)
irradiation could transform Ag to Ag2Se, it implied that
dissociation of one of the CdSe bonds of CSe2 had occurred.19

Core-Shell and Hollow Nanocrystal Formation
J. Phys. Chem. B, Vol. 110, No. 32, 2006 15815
from Ag nanoparticle induced by visible laser photolysis. While
such an effect has been detected for benzyl phenyl sulfide (BPS)
adsorbed on Ag where the S-benzyl bond was broken upon
electron attachment and subsequent dissociation,22,23 we have
failed to observe the occurrence of a similar process for CSe2
since both broad-band visible radiation and 532-nm pulsed laser
photolysis could not produce Ag2Se.
We have not observed any selenium or carbon precipitation
upon irradiation of the Ag colloid/CSe2 solution although the
CSe transient species, which could give rise to these deposits,
must have been produced together with Se atoms upon CSe2
photodissociation. As the photolysis has been performed under
open atmosphere, it is possible for CSe to react with O2 to form
carbonyl selenide (OCSe) gas. Using FTIR spectroscopy, the
intense ν1 band (2020-2050 cm-1 24
belonging to OCSe was
)
indeed observed in the vapor phase above the solution saturated
with O2 in a sealed reaction vessel. The release of OCSe into
the gas phase also has an added advantage of reducing unwanted
side products in the solution.
Further insights into the reaction pathway could be inferred
from the evolution of the UV-vis absorption spectrum. The
spectrum showed an isosbestic point at 370 nm which only
appeared after irradiation of the Ag colloid/CSe2 mixture had
been initiated. The presence of the isosbestic point showed that
a clean conversion of the reactants into a single product (Ag)n@-
(Ag2Se)m had occurred, where n became smaller and m became
larger with increasing photolysis time until full conversion to
Ag2Se.
Ag + CSe2(ads) f (Ag)n@(Ag2Se)m f Ag2Se
However, the UV-vis spectrum of Ag2Se at the final stage
(Figure 1f) did not cross the isosbestic point since the equilib-
rium would be destroyed at the point where complete conversion
had taken place. Upon closer inspection, spectrum 1a and b also
did not cross the point as no core-shell structures had been
made yet.
Initially, it appears unusual that a redshift has occurred upon
irradiation because the gradual decrease in the silver core size
ought to have shifted the plasmon band to the blue with a
concomitant decrease in intensity. The band usually vanishes
when a diameter of 1-2 nm is attained for Ag nanoparticles.
Instead, the redshift could be understood by the changes in the
dielectric function of Ag caused by the coated Ag2Se shell. The
plasmon bandshift was estimated by using the coated sphere
Drude model treatment25
Figure 4. Electron diffraction patterns for (a) polycrystalline Ag
nanoparticles obtained before irradiation and (b) polycrystalline Ag2-
Se hollow nanoparticle formed after irradiation. The crystal planes are
shown in the diagrams.
ωsurf ) ωbulk/(1 - ꢀ1)1/2
(1)
The presence of a Se atom directly above the silver surface
would easily undergo a redox reaction with Ag to form Ag2Se.
This is not uncommon since alkali halide salts could be made
from gas-phase reactions of alkali with halogen atoms.20,21 We
have also considered the formation of Ag2Se from a direct
bimolecular process between the Ag colloids and the Se atoms.
However, this pathway has been ruled out because of the low
concentration of either species. A simple calculation showed
that there were ≈5.2 × 106 atoms in a 50-nm diameter spherical
Ag nanoparticle, and assuming that a 10-3 M AgNO3 solution
has been fully reduced to Ag atoms, the concentration of the
colloids would only be 1.9 × 10-10 M. The expected low
concentration of reactive Se atoms in the solution upon CSe2
dissociation also did not favor a nonsurface mediated bimo-
lecular reaction.
ꢀ1 ) -2ꢀ2[ꢀ2(1 - f) + (2 + f)ꢀm]/[ꢀ2(2f + 1) + 2(1 - f)ꢀm]
(2)
where ꢀ1, ꢀ2, and ꢀm are the dielectric functions of Ag core,
Ag2Se, and water/ethanol solvent (ꢀm ≈ 1.4);26 f is the volume
fraction of Ag occupied in the nanoparticle, ωsurf is the surface
plasmon band frequency, and ωbulk is the bulk plasmon
frequency. The volume fraction (f) occupied by Ag and the
dielectric function (ꢀ2) of the coated material, Ag2Se, play a
major role in determining its band shift.
According to the Drude theory, surface plasmon resonance
for silver nanoparticles is achieved when ꢀ1 ) -2ꢀmedium. In
the absence of the Ag2Se shell, ꢀmedium is ꢀm. A simple inspection
of eq 1 showed that this frequency would be lower for more
negative ꢀ1 values. From eq 2, as long as ꢀ2 is larger than ꢀm,
the values for ꢀ1 are always smaller than -2 whenever f is less
Another possible mechanism leading to Ag2Se formation
involves the dissociation of CSe2 upon initial electron transfer

15816 J. Phys. Chem. B, Vol. 110, No. 32, 2006
Tan et al.
than 1. Surprisingly, wavelength-dependent dielectric functions
for bulk Ag2Se are to the authors’ knowledge, not available in
the literature. However, a survey of dielectric functions of solid-
state materials showed a wide variation of values from ≈1.5
for NaF to more than 10 for Ge throughout the UV-vis range.26
If the dielectric function of Ag2Se falls within this range, which
itself is not an unreasonable estimate, a redshift of the plasmon
peak would be observed. This is certainly in agreement with
the experimental data. However, a quantitative estimate of the
dielectric function will need to await a more sophisticated
treatment which is beyond the scope of the work here.
Acknowledgment. The project was supported by the Agency
of Science, Technology, and Research (ASTAR) under Grant
No.143-000-198-305. H. Tan and S. Li thank the National
University of Singapore for a research scholarship. We thank
B. Liu for help in the acquisition of TEM images.
Supporting Information Available: The supplementary
information contains the gas-phase FTIR spectrum of the
detected product OCSe, TEM images of large 100 nm and small
10 nm Ag and Ag2Se nanoaprticles, and high-resolution TEM
images showing the lattice spacings of the various nanoparticles
during the conversion stages. This material is available free of
charge via the Internet at http://pubs.acs.org.
We have found that Ag nanoparticles of 10 ( 5 nm diameter
have increased slightly to 12 ( 5 nm upon full transformation
to Ag2Se. The increase was due to the differences in the densities
of the two structures assuming the nanoparticle densities were
the same as those of the bulk material (bulk FAg 10.5 g/cm3;
References and Notes
(1) Heath, J. R. Acc. Chem. Res. 1999, 32, 388.
(2) Volokitin, Y.; Sinzig, J.; Dejongh, L. J.; Schmid, G.; Vargaftik,
M. N.; Moiseevi, I. I. Nature 1996, 384, 621.
F
Ag Se 8.2 g/cm3).26 Calculations showed that a full conversion
2
to Ag2Se would increase the volume of the spherical nanopar-
ticle by a factor of 1.75 which is equivalent to ≈20% increase
in the radius. This fitted very well with our size determination
of the initial Ag and final Ag2Se nanoparticles. Because of the
hollow interior, Ag2Se nanoparticles which have been formed
from large Ag nanoparticles were slightly larger; for example,
Ag nanoparticles of 50 ( 10 nm diameter were converted to
Ag2Se of 70 ( 10 nm diameter. For such cases, the hollow
interior diameter of 40-50 nm was expected to be formed on
the basis of similar calculations as above. An inspection of the
TEM images indeed showed good agreement with the prediction
in which a mean diameter of 40 ( 5 nm (taken as the distance
between the two dark sections within each nanoparticle) could
be determined for the hollow interior.
(3) EI-Sayed, M. A. Acc. Chem. Res. 2001, 34, 257.
(4) Park, J. I.; Kim, M. G.; Jun, Y. W.; Lee, J. S.; Lee, W. R.; Cheon,
J. W. J. Am. Chem. Soc. 2004, 126, 9072.
(5) Lei, Y.; Chim, W. K. J. Am. Chem. Soc. 2005, 127, 1487.
(6) Kim, M.; Sohn, K.; Na, H. B.; Hyeon, T. Nano Lett. 2002, 2, 1383.
(7) Averitt, R. D.; Sarka, R. D.; Halas, N. J. Phys. ReV. Lett. 1997, 78,
4217.
(8) Jin, R. C.; Cao, Y. C.; Hao, E. C.; Metraux, G. S.; Schatz, G. C.;
Mirkin, C. A. Nature 2003, 425, 487.
(9) Jin, R. C.; Cao, Y. W.; Mirkin, C. A.; Kelly, K. L.; Schatz, G. C.;
Zheng, J. G. Science 2001, 294, 1901.
(10) Tan, H.; Fan, W. Y. Chem. Phys. Lett. 2005, 406, 289.
(11) Gates, B.; Wu, Y.; Yin, Y.; Yang, P.; Xia, Y. J. Am. Chem. Soc.
2001, 123, 11500.
(12) Gates, B.; Mayers, B.; Wu, Y.; Sun, Y.; Cattle, B.; Yang, P.; Xia,
Y. AdV. Funct. Mater. 2002, 12, 679.
(13) Jiang, Z.; Xie, Z.; Zhang, X.; Huang, R.; Zheng, L. Chem. Phys.
Lett. 2003, 378, 313.
The production of hollow nanocrystals when large silver
colloids were used has been attributed to the Kirkendall effect.
The effect is related, for example, to the different diffusion rates
of the Ag and Se species which formed a diffusion couple such
that either particle with hollow interiors or core-shell particles
with cracks in their shells would be obtained. Y.Yin et al. first
demonstrated the Kirkendall effect in nanoparticles by observing
the formation of platinum cobalt oxide yolk-shell structures.27
Hollow Pb@PbS structures caused by the faster diffusion of
Pb compared to S atom have since been made as well.28
However, Pb@Ag core-shell particles with cracked shells have
been observed instead owing to the faster diffusion of Ag
compared to Pb.28,29 In our case, Ag atoms must have diffused
faster outward from the core than Se atoms diffusing inward
toward the core. The net effect was that all the Ag would move
away from the center and get converted into Ag2Se albeit leaving
a hollow core behind. However, we have found that the diffusion
time differences are not significantly large enough such that
for small nanoparticles of 10-nm diameter, hollow nanoshells
were not formed; filled Ag2Se nanoparticles were observed
instead.
(14) Lee, P. C.; Meisel, D. J. Phys. Chem. 1982, 86, 3391.
(15) Buerger, H.; Willner, H. J. Mol. Spectrosc. 1988, 128, 221.
(16) JCPDF #241041 (database of X-ray diffraction patterns).
(17) Kim, S. J.; Kim, T. G.; Ah, C. S.; Kim, K.; Jang, D. J. Phys. Chem.
B 2004, 108, 880.
(18) Feilchenfeld, H.; Siiman, O. J. Phys. Chem. 1986, 90, 4590.
(19) Callear, A. B.; Tyerman, W. J. R. Trans. Faraday Soc. 1965, 61,
2395.
(20) Plane, J. M. C.; Husain, D. J. Phys. Chem. 1986, 90, 501.
(21) Husain, D.; Plane, J. M. C.; Xiang, C. C. J. Chem. Soc., Faraday
Trans. 2 1985, 81, 1675.
(22) Kim, K. L.; Lee, S. J.; Kim, K. J. Phys. Chem. B 2004, 108, 9216.
(23) Lee, I.; Han, S. W.; Kim, C. H.; Kim, T. G.; Joo, S. W.; Jang, D.;
Kim, K. Langmuir 2000, 16, 9963.
(24) Sueoka, K.; Hamada, Y.; Uehara, H. J. Mol. Spectrosc. 1988, 127,
370.
(25) Bohren, C. F.; Huffman, D. R. Absorption and Scattering of Light
by Small Particles; John Wiley: 1983.
(26) CRC Handbook of Chemistry and Physics, 85th ed.; CRC Press:
Cleveland, OH, 2004.
(27) Yin, Y.; Rioux, R. M.; Erdonmez, C. K.; Hughes, S.; Somorjai, G.
A.; Alivisatos, A. P. Science 2004, 304, 711.
(28) Wang, Y.; Cai, L.; Xia, Y. AdV. Mater. 2005, 17, 473.
(29) Prakash, S. G.; Ravi, R.; Chhabra, R. P. Chem. Phys. 2004, 302,
149.