Detailed structural examinations of covalently immobilized gold nanoparticles onto hydrogen-terminated silicon surfaces

Article abstract of DOI:10.1002/chem.200500455

The modification of flat semiconductor surfaces with nanoscale materials has been the subject of considerable interest. This paper provides detailed structural examinations of gold nanoparticles covalently immobilized onto hydrogen-terminated silicon surfaces by a convenient thermal hydrosilylation to form Si-C bonds. Gold nanoparticles stabilized by ω-alkene-1-thiols with different alkyl chain lengths (C₃, C₆, and C ₁₁), with average diameters of 2-3 nm and a narrow size distribution were used. The thermal hydrosilylation reactions of these nanoparticles with hydrogen-terminated Si(111) surfaces were carried out in toluene at various conditions under N₂. The obtained modified surfaces were observed by high-resolution scanning electron microscopy (HR-SEM). The obtained images indicate considerable changes in morphology with reaction time, reaction temperature, as well as the length of the stabilizing ω-alkene-1-thiol molecules. These surfaces are stable and can be stored under ambient conditions for several weeks without measurable decomposition. It was also found that the aggregation of immobilized particles on a silicon surface occurred at high temperature (> 100°C). Precise XPS measurements of modified surfaces were carried out by using a Au-S ligand-exchange technique. The spectrum clearly showed the existence of Si-C bonds. Cross-sectional HR-TEM images also directly indicate that the particles were covalently attached to the silicon surface through Si-C bonds.

Full text of DOI:10.1002/chem.200500455

DOI: 10.1002/chem.200500455
Detailed Structural Examinations of CovalentlyImmobilized Gold
Nanoparticles onto Hydrogen-Terminated Silicon Surfaces
[
a]
[b]
[a]
[c]
Yoshinori Yamanoi, Naoto Shirahata, Tetsu Yonezawa,* Nao Terasaki,
[
c]
[d]
[d]
[d]
Noritaka Yamamoto, Yoshitaka Matsui, Kazuyuki Nishio, Hideki Masuda,
Yuichi Ikuhara, and Hiroshi Nishihara*
[
e]
[a]
Abstract: The modification of flat semi-
conductor surfaces with nanoscale ma-
terials has been the subject of consider-
able interest. This paper provides de-
tailed structural examinations of gold
nanoparticles covalently immobilized
onto hydrogen-terminated silicon surfa-
ces by a convenient thermal hydrosilyl-
ation to form SiÀC bonds. Gold nano-
hydrogen-terminated Si(111) surfaces
were carried out in toluene at various
conditions under N2. The obtained
modified surfaces were observed by
high-resolution scanning electron mi-
croscopy (HR-SEM). The obtained
images indicate considerable changes
in morphology with reaction time, reac-
tion temperature, as well as the length
of the stabilizing w-alkene-1-thiol mol-
ecules. These surfaces are stable and
can be stored under ambient conditions
for several weeks without measurable
decomposition. It was also found that
the aggregation of immobilized parti-
cles on a silicon surface occurred at
high temperature (> 1008C). Precise
XPS measurements of modified surfa-
ces were carried out by using a Au–S
ligand-exchange technique. The spec-
trum clearly showed the existence of
SiÀC bonds. Cross-sectional HR-TEM
particles stabilized by w-alkene-1-thiols
with different alkyl chain lengths (C3,
C , and C ), with average diameters of
6
11
Keywords: gold · nanotechnology ·
scanning probe microscopy · silicon ·
transmission electron microscopy
2–3 nm and a narrow size distribution
images also directly indicate that the
particles were covalently attached to
the silicon surface through SiÀC bonds.
were used. The thermal hydrosilylation
reactions of these nanoparticles with
Introduction
[
a] Dr. Y. Yamanoi, Prof. Dr. T. Yonezawa, Prof. Dr. H. Nishihara
Department of Chemistry, School of Science
The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku
Tokyo 113-0033 (Japan)
Fax : (+81)3-5841-2356
E-mail: yonezawa@chem.s.u-tokyo.ac.jp
nisihara@chem.s.u-tokyo.ac.jp
The modification of semiconductor surfaces by the covalent
attachment of organic molecules has been an active topic of
investigation in the field of surface sciences. The most im-
[1]
portant widespread semiconductor material is silicon. Or-
ganic monolayers immobilized on silicon substrates are cru-
cial for microelectronics and sensors, as well as for funda-
mental studies. In the past, these monolayers were usually
prepared by the reaction of organosilicon derivatives (such
as alkylchlorosilanes, alkylalkoxysilanes, and alkylaminosi-
[
b] Dr. N. Shirahata
International Center for Young Scientists (ICYS)
National Institute for Materials Science (NMIS)
1
-1 Namiki, Tsukuba, Ibaraki 305-0044 (Japan)
[
c] Dr. N. Terasaki, Dr. N. Yamamoto
National Institute of Advanced Industrial Science
and Technology (AIST)
[2]
lanes) with hydroxylated silicon surfaces. However, this
modified surface has a problem in that a silica layer, con-
nected to the silanol groups via SiÀOÀSi bonds, forms on
1
-8-31 Midorigaoka, Ikeda, Osaka 563-8577 (Japan)
[
d] Dr. Y. Matsui, Dr. K. Nishio, Prof. Dr. H. Masuda
Department of Applied Chemistry, School of Engineering
Tokyo Metropolitan University
the surface. Besides the problems associated with the amor-
phous substrate and moisture sensitivity of the precursor
molecules, the silicon dioxide thin film essentially blocks the
electrical communication between the organic molecules
and the silicon surface. Another useful approach to the co-
valent attachment of organic molecules is the hydrosilyla-
tion reaction of 1-alkenes with a hydrogen-terminated sili-
1
-1 Minami-Osawa, Hachioji, Tokyo 192-0397 (Japan)
[
e] Prof. Dr. Y. Ikuhara
Institute of Engineering Innovation
School of Engineering, The University of Tokyo
2
-11-16 Yayoi, Bunkyo-ku, Tokyo 113-8656 (Japan)
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
[3]
con (HÀSi) surface. Both crystalline flat HÀSi(111) and
314
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 314 – 323

FULL PAPER
[
4]
HÀSi(100) surfaces, as well as hydrogen-terminated porous
Results and Discussion
[5]
silicon, have been used as substrates. In this case, organic
monolayers formed through a covalent SiÀC bond provide
Preparation and characterization of w-alkene-1-thiol-cov-
electronic coupling between organic functionalities and the
semiconductors without the interference of interfacial oxide
ered gold nanoparticles: The use of w-alkene-1-thiols (CH =
2
CH(CH ) SH), which undergoes reactions typical for both
2
n
[3–7]
thin film.
aliphatic thiols and olefins, is an excellent example of a
strategy for immobilization onto hydrogen-terminated sili-
con surfaces. The aliphatic thiol parts can act as stabilizers
Recently, much attention has centered on colloidal metal
nanoparticles because of their potential applications in areas
[8]
[8]
of electronics, photonics, and catalysis. In particular, thiol-
stabilized gold nanoparticles have become an important
model system in nanomaterial research due to their stability,
easy preparation and chemical versatility. Despite increasing
efforts in the past few years, it is still a great challenge to
fabricate well-controllable nanostructures by using colloidal
particles as the structural elements. A key to achieving this
goal will be the ability to assemble nanoparticles onto a de-
sired surface to enable efficient utilization of their unique
nanostructural properties. Much work has been performed
to design and produce position-controlled assemblies of
nanoparticles on solid surfaces, which are essential for nano-
of gold nanoparticles. The C=C bonds can also participate
in the hydrosilylation reaction on a hydrogen-terminated sil-
[3–6]
icon surface.
In this section, we report the preparation
and characterization of gold nanoparticles covered with w-
alkene-1-thiols.
We computed the energy-minimized preferred conforma-
tion of the stabilizing thiol forms and the lengths of the mol-
ecules using MM2 of Chem3D Pro V5.0. The total lengths
of 1a, b, and c (from the thiol proton to the terminal vinyl
proton) were calculated to be 0.56, 0.93, and 1.6 nm, respec-
tively (see Figure 1).
[9,10]
device studies.
Ideally, it would be desirable to immobi-
lize gold nanoparticles directly onto the surface by strong
covalent bonds. Considering the potential applications of hy-
drogen-terminated silicon(111) surfaces mentioned above,
nanoparticles could be immobilized onto a silicon surface
À1 [11]
through the highly stable SiÀC bond (347 kJmol ), if we
could prepare gold nanoparticles that are protected by w-
alkene-1-thiols and perform a hydrosilylation reaction. How-
ever, there have only been a few reported investigations of
nanometer-sized materials attached to a silicon surface
through strong SiÀC bonds, one of which is our short com-
Figure 1. Energy-minimized conformation of 1a, 1b, and 1c.
[12,13]
munication.
This full paper focuses on the detailed
Gold nanoparticles 2a were prepared by using a thiol sub-
stitution reaction between the free thiol 1a and 1-propane-
thiol-stabilized gold nanoparticles (Scheme 2). The nanopar-
ticles were prepared by a method based on the two-phase
solution synthesis reported by Brust et al., and were isolated
structural examinations of immobilized gold nanoparticles
on hydrogen-terminated silicon surfaces (Scheme 1) by
using HR-SEM at various reaction conditions, XPS mea-
surements and cross-sectional TEM observations.
[14–16]
and characterized prior to the substitution reaction.
On
the other hand, gold nanoparticles, 2b and 2c, capped with
[17]
thiol 1b or 1c, respectively, were directly synthesized by
[15,16]
the modified Brustꢁs two-phase method.
These gold
Scheme 1. Immobilization of w-alkene-1-thiol-functionalized gold nanoparticles onto a hydrogen-terminated silicon(111) surface.
Chem. Eur. J. 2006, 12, 314 – 323
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
315

T. Yonezawa, H. Nishihara et al.
Scheme 2. Synthetic route for the preparation of gold nanoparticles stabilized by w-alkene-1-thiols.
nanoparticles were isolated as a black powder and spectro-
scopically characterized. The UV-visible absorption spectra
of particle dispersions in toluene were measured. All spectra
exhibited peak absorbance wavelengths at around 520 nm,
which is a typical plasmon resonance band for gold nanopar-
ticles (diameter >2 nm), suggesting the formation of gold
[18]
nanoparticles (Figure 2).
Direct evidence concerning the size and morphology of
the nanoparticles were obtained by transmission electron
microscopy (TEM). Figure 3 shows the TEM images and
size histograms of gold nanoparticles 2a–c. These particles
are spherical and two-dimensionally dispersed. The average
diameters of 2a–c were 3.1Æ0.7, 2.3Æ0.7, and 2.1Æ0.4 nm,
respectively.
Figure 2. UV-visible absorption spectra of a toluene dispersion with 2a,
2b, and 2c.
1
H NMR spectra of these nanoparticles were measured to
assess the purity of the nanoparticles, especially in order to
reveal the saturation ratio of
the terminal C=C unsaturated
bonds of the stabilizing w-
alkene-1-thiols. Spectra ob-
tained in C D CD (2a) or
6
5
3
CDCl3 (2b and 2c) showed
broad peaks, and free w-alkene-
1-thiol (e.g., 2.50 ppm) was
almost completely absent, as
shown in Figure 4. The broad-
ening effect is derived from the
structural features of the mono-
layer on the gold surface. It is
well known that NMR line
broadening for proteins and
polymers is dominated by their
slow rotation in solution. Anal- Figure 3. TEM images and size histograms of gold nanoparticles a) 2a, b) 2b, and c) 2c.
316
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 314 – 323

Immobilization of Gold Nanoparticles
FULL PAPER
[19]
10 mL)
and then added to the hydrogen-terminated
silicon(111) surface at 508C for 24 h under a dry nitrogen at-
mosphere. The substrate was then rinsed with toluene, soni-
cated in toluene for 10 min, and dried with a stream of ni-
trogen. In order to avoid the oxidation of the modified sur-
faces, they were stored in a sealed nitrogen-filled container
until analyses were performed.
Figure 5 shows HR-SEM images of gold nanoparticles im-
mobilized onto a hydrogen-terminated silicon surface ob-
tained by thermal hydrosilylation at various conditions.
From Figure 5a, it can be seen that the immobilized colloids
have an average diameter of about 10–40 nm by using C -al-
3
kenethiol-stabilized gold nanoparticles (2a). This behavior
has previously been found for gold nanoparticles with short-
[20]
chain surfactants (ca. 0.5 nm). Traditional Ostwald ripen-
ing is a possible explanation for the changes in the size and
shape of the gold nanoparticles induced by immobilization
[21]
onto the silicon surface. Thus, gold nanoparticles initially
land on the silicon substrate in a randomly scattered way
and become firmly bound to the hydrogen-terminated sili-
con surface through SiÀC bonds. The randomly bound nano-
particles then act as seed or template particles to attract the
subsequently landing particles next to them. Thus, the subse-
quently landing gold nanoparticles first adhere to the al-
ready-anchored particles through metallic core fusion be-
tween them. The large particles with irregular sizes and
shapes were formed by the increased collision and aggrega-
tion of nucleation sites. According to the literature, this ag-
gregation takes place as a result of a reduction of the parti-
cle surface energy when short surface-stabilizing reagents
1
Figure 4. H NMR spectra of gold nanoparticles a) 2a (500 MHz,
C
6
D
5
CD
3 3
), b) 2b, and c) 2c (500 MHz, CDCl
). *: vinyl protons, *: ter-
minal methyl protons.
ogously, the aliphatic thiol-protected gold nanoparticles are
slowly rotating macromolecules. These observations suggest
that the hydrocarbon species derived from w-alkene-1-thiols
are densely attached to the particle surface through the
sulfur atoms. Furthermore, several alkenethiols on each par-
ticle were hydrogenated during the formation of functional-
ized gold nanoparticles. This ratio was determined by com-
paring the integrated areas for the vinyl protons of alkenyl
thiolates and the terminal methyl groups of alkyl thiolates.
The surface ratios of alkenyl thiolates/alkyl thiolates on par-
ticles 2a, 2b, and 2c were 49/51, 50/50, and 83/17 (mol/mol),
respectively.
[20]
are used. In contrast, no aggregation was observed in the
immobilization of 2b (Figure 5b), which indicates that nano-
particles were distributed on the silicon surface, forming
nearly a monolayer coverage with approximately 15000 par-
2
ticles per mm . We found an ordered monolayer region on
this modified silicon surface. The average interparticle dis-
tance was about 2.0 nm, close to twice the distance between
1b particles (0.83 nm). It was furthermore observed that the
coverage density on Si(111) decreased as the chain length of
stabilizers increased from C (2b) to C (2c) (Figure 5c).
6
11
Immobilization of w-alkene-1-thiol-stabilized gold nanopar-
ticles—Influence of the alkyl chain length on the immobi-
lized structure: The hydrogen-terminated silicon surface
reacts with alkenes under photochemical, thermal, or cata-
lytic conditions to generate an alkyl monolayer on the sur-
face. The formation of such an organic monolayer was
shown to be chemically robust. We have used the thermal
route to coat the hydrogen-terminated silicon surface with
w-alkene-1-thiol-stabilized gold nanoparticles. The surface
functionalization was carried out using the following proce-
dures. An atomically flat hydrogen-terminated Si(111) sur-
face was prepared by etching cleaned shards of silicon first
in 1% HF solution and then in 40% ammonium fluoride as
The low surface coverage using 2c should be ascribed to the
decrease in terminal C=C bond reactivity due to the strong
hydrophobic interaction among the stabilizer molecules. Ac-
cording to these results, the medium chain length thiol (C )-
6
stabilized gold nanoparticle (2b) is the most suitable for
nanostructuring by a thermal hydrosilylation reaction. Sub-
sequent TEM observations of particle dispersions after the
hydrosilylation reaction showed little improvement in size
distribution (Figure 6).
We also confirmed the stability of the modified silicon
surfaces. Characterization of the surface after long-term (2–
4 weeks) exposure to ambient laboratory conditions showed
no changes.
[11]
described in the literature.
This treatment removes the
SiO surface layer and etches the wafer to form an atomical-
ly flat surface. The w-alkene-1-thiol-stabilized gold nanopar-
ticles were first dissolved in degassed toluene (30 mg/
Immobilization of 5-hexene-1-thiol-stabilized gold nanopar-
ticles (2b)—Influence of reaction time and temperature:
One of the aims of the present detailed study is to create a
2
Chem. Eur. J. 2006, 12, 314 – 323
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
317

T. Yonezawa, H. Nishihara et al.
the particle immobilization de-
pended largely on reaction
time. Gold nanoparticles at-
tached in a random manner to
the silicon substrate with low
surface coverage at the initial
stage (Figure 5d). Longer reac-
tion time at 508C resulted in a
significant increase in the densi-
ty of the coverage. We fabricat-
ed isolated nanoparticle arrays
on silicon surfaces after approx-
imately 24 h of reaction time.
After 48 h, the surface was
densely covered with colloidal
gold particles with self-fusion in
a small partial area (Figure 5f).
The effect of reaction tem-
perature on the fixation of gold
nanoparticles (2b) onto a sili-
con surface was also explored
(
Figure 5g, b, and h). The hy-
drogen-terminated silicon sur-
face was immersed into a tolu-
ene dispersion of 2b (30 mg/
1
0 mL) and heated for 24 h at
Figure 5. High-resolution SEM images after various immobilization conditions in approximately 30 mg/10 mL
solutions of gold nanoparticles 2a–2c. a) 2a, 24 h, 508C, b) 2b, 24 h, 508C, c) 2c, 24 h, 508C, d) 2b, 2 h, 508C, temperatures ranging from 25
e) 2b, 6 h, 508C, f) 2b, 48 h, 508C, g) 2b, 24 h, 258C, h) 2b, 24 h, 1008C, and i) 2b, 24 h, 508C in the dark. to 1008C. Under these experi-
mental conditions, particle im-
mobilization onto the surface
began at around 508C (Fig-
ure 5b). Below the activation
temperature (258C), the 5-
hexene-1-thiol-protected
gold
nanoparticles remained almost
unreacted (Figure 5g). At high
temperature (1008C), the parti-
cles were aggregated and irregu-
larly shaped, as expected. It is
argued that the capping ligand
is ripped off the surface of the
nanoparticle during heat treat-
ment (>1008C) and that small
spherical particles coalesce to
[22]
form bigger particles.
The
Figure 6. TEM images and size distributions of gold nanoparticles a) 2a, b) 2b, and c) 2c after heating at 508C
for 24 h in toluene.
TEM observation of these re-
fluxed gold nanoparticles in dis-
persion showed
a small im-
one-particle layer without aggregation. A systematic study
of the surface immobilization reaction on Si(111) is de-
scribed below. To investigate the immobilization process of
gold nanoparticles onto a silicon surface, we carried out the
immobilization reaction at 508C for different periods of
time. Gold nanoparticles of the type 2b, as described in the
previous section, was used as a model substrate. Figure 5d,
e, b and f shows the HR-SEM images of the reacted sample
after heating for 2, 6, 24, and 48 h, respectively. Obviously,
provement in the size distribution (3.4Æ0.8 nm, Figure 7).
However, the immobilized particles were bigger than ap-
proximately 10 nm. Ostwald ripening on the silicon surface
may also explain this mechanism, as described in the previ-
ous section. Recently, the photochemical attachment of 1-al-
kenes or 1-alkynes onto hydrogen-terminated silicon surfa-
[23]
ces using a visible light source was reported.
However,
the HR-SEM image of the 2b-immobilized silicon surface
was almost the same in the dark for 24 h at 508C as it was
318
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 314 – 323

Immobilization of Gold Nanoparticles
FULL PAPER
sion of particles from the surface was expected. However,
HR-SEM analysis of the surface showed little change. The
reaction of 1-undecanethiol (C11 alkanethiol)-coated gold
nanoparticles with a hydrogen-terminated silicon surface at
5
08C for 24 h provided no immobilized particles, which was
[25]
confirmed by HR-SEM observation. This result indicates
that the C=C bond in w-alkene-1-thiol on the particle is in-
dispensable to immobilize nanoparticles onto hydrogen-ter-
minated silicon through the SiÀC bond.
Figure 7. TEM image and size distribution of 2b nanoparticles after heat-
ing at 1008C for 24 h in toluene.
To observe the formation of SiÀC bonds directly, X-ray
photoelectron spectroscopy (XPS) was selected. XPS is a
very useful tool for the study of covalently attached mono-
in visible light for 24 h at 508C. In this case, the immobiliza-
tion process took place exclusively under thermal initiation.
Judging from these observations, silicon surfaces with
well-spaced nanoparticles were apparent only up to the re-
action time of 24 h at 508C and 30 mg/10 mL toluene solu-
tion: longer reaction time (>24 h) and higher temperature
[26]
layers on silicon. In the initial work, XPS of a particle-im-
mobilized silicon surface was employed; however, it was
found that the XPS signal from SiÀC was too poor to be re-
solved from the C(1s) region. A ligand-exchange reaction of
the 2c-immobilized surface with a large excess amount of 1-
hexadecanethiol (C H SH) was then applied to remove all
(
>508C) led to the merging of the gold nanoparticles.
1
6
33
gold nanoparticles without cleavage the SiÀC bonds on the
Investigation of the immobilizing structure: To establish that
the particles were indeed covalently attached and not just
adsorbed to the surface, the modified surfaces were subject-
ed to ultrasonication in toluene for 1 h. Since this rigorous
treatment is capable of inducing bond cleavage, some ero-
silicon surface. The HR-SEM measurement of the silicon
surface indicated successful removal of gold nanoparticles
(Figure 8a and b). The XPS results for this system are also
shown in Figure 8c and d. The freshly prepared hydrogen-
terminated Si(111) surface shows no SiÀC bond and only a
[24]
Figure 8. a) SEM images comparing 2c immobilized silicon surface (n=9), b) the silicon surface after removal of 2c with 1-hexadecanethiol, c) a carbon
1s) narrow XPS scan for the hydrogen-terminated silicon surface and d) the modified silicon surface after the cleavage of gold nanoparticle 2c using a
(
large excess amount of 1-hexadecanethiol.
Chem. Eur. J. 2006, 12, 314 – 323
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
319

T. Yonezawa, H. Nishihara et al.
small CÀC signal, which is presumably due to physically ad-
sorbed hydrocarbons (Figure 8c). As shown in Figure 8d, the
C(1s) core-level spectrum of the surface can be fitted with
two peaks having binding energies at approximately 284 and
temperature, and reaction time are important factors influ-
encing the modified silicon surfaces. Using HR-SEM, XPS
with removal of particles by an Au–S ligand-exchange
method, and cross-sectional TEM analysis, we confirmed
that the gold nanoparticles were directly attached to the sili-
con surface through robust SiÀC bonds. Our total structural
2
85 eV, attributable to the CÀSi and CÀC species, respec-
tively. This spectrum shows the carbon 1s signal from the
silicon(111) surface, modified by the attachment of 11-unde-
cene-1-thiol. From this result, the w-alkene-1-thiol-stabilized
gold nanoparticles were thought to be attached to the hy-
examinations of the immobilization of nanomaterials onto
silicon surfaces should become a standard combination of
analytical methods. This surface-modification method is ge-
neric in the sense that it should be easy to extend to a broad
range of nanosized compounds that expose unsaturated
carbon–carbon bonds. Gold nanoparticles are expected to
behave as quantum dots, and we believe that our approach
for successful and reproducible fabrication of nanostructures
will bring the manufacture of nanoparticle-based devices a
step closer to realization.
[27]
drogen-terminated silicon surface through SiÀC bonds. As
a control, the Si(111)-H surface was exposed to neat 1-hexa-
decanethiol (10 mL) and 10-undecene-1-thiol in toluene
(
0.3 mg/10 mL) at room temperature for 7 d. No SiÀC bond
[28]
signal was detected by XPS measurement in either case.
Apart from conventional XPS, gold nanoparticles immo-
bilized onto silicon surfaces were also directly observed with
[29]
cross-sectional TEM. Figure 9 shows cross-sectional TEM
images of 2b immobilized onto a silicon surface. The over-
view image in Figure 9a demonstrates that a uniform gold
nanoparticle submonolayer is formed on the Si(111) surface.
The high-resolution cross-sectional TEM image is shown in
Figure 9b. This image clearly shows 2b gold nanoparticles in
the near surface layer, surrounded by 5-hexene-1-thiol (1b).
Thus, the distance between the particles and the surface in
Figure 9b was estimated to be 0.95 nm, in fairly good agree-
Experimental Section
Materials and chemicals: All experiments were carried out in a dry nitro-
gen atmosphere. All starting materials were purchased from Aldrich or
other reagent manufacturers and used without further purification. All
2
organic solvents were distilled from CaH and stored over MS 4 Š. De-
ionized water (resistivity higher than 18.2MWcm) was purified with a
Millipore Milli-Q water system. N-type silicon wafers with (111) orienta-
tion were obtained from Chiyoda Koeki Co. Ltd. and used as a substrate
after suitable chemical etching.
[30]
ment with the expected distance (0.90 nm).
General methods: NMR spectroscopy:
1
H NMR studies of nanoparticles and
their precursors were recorded on a
Bruker DRX-500 spectrometer at
room temperature. Chemical shifts in
ppm were referenced to tetramethylsi-
lane (0.00 ppm) as an internal stan-
dard. FT-IR spectroscopy: FT-IR spec-
tra were recorded on a Jasco FT/IR-
6
20v spectrometer. Mass spectrometry:
MS (EI) measurements were recorded
on a Shimadzu GCMS-QP5000A. Ele-
mental analysis: Elemental analysis of
Figure 9. a) Cross-sectional TEM image of 2b gold nanoparticles immobilized onto a silicon surface. b) Calcu-
lated and experimentally determined height in the formation of 2b nanoparticles immobilized onto a silicon
surface.
the products was performed on
a
Yanaco MT-6 CHN recorder at the El-
emental Analysis Center of the Uni-
versity of Tokyo. UV-visible spectros-
These observations provide strong evidence that these
particles are directly immobilized onto a hydrogen-terminat-
ed silicon surface through SiÀC bonds.
copy: UV-visible spectra were recorded using a Hewlett-Packard 8453
UV/Vis spectrometer at 208C in the visible region to find the location
and intensity of the surface plasmon resonance peak for the particles.
Structural examinations: TEM: The TEM images were recorded at
2
00 kV by using a Hitachi HF-2000. The TEM samples of gold nanoparti-
Conclusion
cles were prepared at room temperature by depositing a droplet of a tol-
uene solution of particles onto a carbon film supported on a copper grid.
The micrographs were then analyzed by using Scion image analysis soft-
ware. More than 500 particles were used in the statistical analyses to de-
termine the size distribution, which indicated the mean diameter and its
standard deviation. HR-SEM: The SEM images were recorded at 30 kV
by using a Hitachi S-5000 or S-5200. The preparation of specimens is de-
scribed below. Cross-sectional HR-TEM: The cross-sectional TEM
images were taken at 300 kV using a Topcon. The samples for cross-sec-
tional TEM images were prepared by a conventional method. XPS:
Sample surfaces were characterized by using X-ray photoelectron spec-
troscopy (XPS; ULVAC-PHI, Quantum-2000 system) with monochromat-
ic AlKa (hn=1486.6 eV) radiation. The monochromatic Al anode was op-
erated at 15 kV and 20 W. Static point acquisitions were collected with a
Herein we presented detailed structural observations of co-
valently immobilized gold nanoparticles on hydrogen-termi-
nated silicon surfaces by a thermal hydrosilylation reaction
under an inert atmosphere. A series of w-alkene-1-thiol-pro-
tected gold nanoparticles was successfully prepared using a
method based on the two-phase solution synthesis and spec-
troscopically characterized. The modified silicon surface was
easily obtained by thermal hydrosilylation of the hydrogen-
terminated silicon(111) surface with these particles in tolu-
ene. The effect of the length of the stabilizer thiols, reaction
320
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 314 – 323

Immobilization of Gold Nanoparticles
FULL PAPER
2
1
5
00 mm analysis area. Spectra were recorded at constant pass energy of
500 MHz, selected peaks): d = 5.70–5.61 (brs, 1H), 5.04–4.88 (brs, 2H);
IR (KBr): n˜ = 3074, 2952, 2908, 1631, 1412, 912 cm ; UV/Vis (toluene):
À1
.85 eV, with a step size of 0.025 eV. The core-level signals were obtained
at a photoelectron takeoff angle of 458 (with respect to the sample sur-
l
max = 517 nm; elemental analysis calcd (%) for C342
H570Au459S114 (Au459-
face). During operation of the spectrometer, the pressure in the analysis
chamber was less than 4”10 Torr. All the binding energies were nor-
malized to that of the Si 2p peak appearing at 99.3 eV, which was refer-
enced to a p-type Si wafer. The resultant XPS profiles were analyzed by
using a Gaussian–Lorentzian fit program.
(C S)114): C 4.16, H 0.58; found C 5.10, H 0.85.
3 5
H
À10
Gold nanoparticles 2b and 2c: Gold nanoparticles 2b and 2c were also
[15]
synthesized as described by Brust et al. Hydrogen tetrachloroaurate(iii)
tetrahydrate (1.00 g, 2.4 mmol) in deionized water (75 mL) was added to
a vigorously stirred solution of tetra-n-octylammonium bromide (6.02 g,
11 mmol) in toluene (250 mL). The yellow aqueous solution became
clear immediately and the toluene phase turned brown. The organic
phase was transferred into a 1 L round-bottom flask, and w-alkene-1-
thiol (2.4 mmol) was added to the solution. Sodium borohydride (0.25 g,
Synthesis
w-Alkene-1-thiol 1b and 1c: 5-Hexene-1-thiol (1b) and 10-undecene-1-
thiol (1c) were each prepared from corresponding alkenyl bromide sub-
[
31]
strate according to the well-known thiourea route. A typical synthesis
was carried out as follows. Thiourea (7.61 g, 100 mmol) and alkenyl bro-
mide (20 mmol) were added to ethanol (300 mL) in a two-neck flask
equipped with a reflux condenser. The reaction mixture was heated to
reflux for 24 h. The reaction was allowed to cool to room temperature,
and then 2m NaOH aqueous solution (50 mL) was added. The solution
was heated to reflux for 24 h. The reaction was monitored by silica gel
thin-layer chromatography (TLC). After cooling to room temperature,
the solution was treated with 1m HCl to pH 7. The organic layer was sep-
6
.6 mmol) in deionized water (50 mL) was slowly added to a vigorously
stirred solution. The resulting solution was stirred for 3 h at room tem-
perature. The organic phase was separated and dried over sodium sulfate,
and the solvent was removed under reduced pressure without exceeding
a temperature of 508C. The dark residue was suspended in ethanol (1 L)
and kept in a refrigerator at À108C overnight. The precipitate was col-
lected by filtration on a membrane filter and washed thoroughly with
ethanol. This process was repeated to ensure the removal of any free w-
alkene-1-thiol in the product. The complete removal of unabsorbed w-
arated, and the aqueous layer was extracted with CH
2 2
Cl three times.
1
The combined extracts were dried over Na SO . The solvent was evapo-
2
4
alkene-1-thiol from the nanoparticles was ascertained by H NMR. The
purified gold nanoparticles were kept in powder form and redispersed
rated under slightly reduced pressure, and the residue was dissolved in
hexane. The solution was passed through a short column packed with
silica gel (hexane) to give the corresponding w-alkene-1-thiol in moder-
3 2 2
into organic solvents (e.g., toluene, CHCl , CH Cl , and THF). Finally,
1
the nanoparticles were characterized by H NMR, FT-IR, UV/Vis, ele-
mental analysis, and TEM.
ate to good yield.
1
1
Compound 1b: colorless oil (0.79 g, 34%); H NMR (CDCl
3
, 500 MHz):
Compound 2b: 445 mg, 78%; H NMR gave very broad peaks. The ratio
1
d=5.80 (ddt, J=17.2, 10.3, 6.7 Hz, 1H), 5.02 (ddt, J=17.1, 2.0, 1.6 Hz,
H), 4.97 (ddt, J=10.2, 2.0, 1.1 Hz, 1H), 2.54 (q, J=7.4 Hz, 2H), 2.07 (q,
J=7.2 Hz, 2H), 1.63 (quin, J=7.4 Hz, 2H), 1.49 (quin, J=7.5 Hz, 2H),
of alkyl thiolates to alkenyl thiolates on particles was 50:50. H NMR
(CDCl , 500 MHz, selected peaks): d = 5.90–5.69 (brs, 1H), 5.03–4.95
3
(brs, 2H), 3.34–3.31 (m, 2H), 1.71–1.66 (m, 2H), 1.39–1.37 (m, 2H),
1
À1
1
9
.34 (t, J=7.9 Hz, 1H); IR (neat): n˜ = 3076, 2928, 2855, 1641, 991,
0.84–0.81 (m, 2H); IR (KBr): n˜ = 3074, 2918, 2848, 1637, 1414, 906 cm
UV/Vis (toluene): lmax
;
À1
+
11 cm ; EI-MS: m/z: 115 [MÀH] .
=
519 nm; elemental analysis calcd (%) for
1
C
552
H
1012Au309
S
92 (Au309(C
6
H
11S)92): C 9.28, H 1.39; found C 8.91, H 1.53.
Compound 1c: colorless oil (3.09 g, 83%); H NMR (CDCl
3
, 500 MHz):
[
17]
1
d = 5.81 (ddt, J=17.1, 10.3, 6.7 Hz, 1H), 4.99 (dd, J=17.1, 1.7 Hz, 1H),
Compound 2c:
520 mg, 80%; H NMR gave very broad peaks. The
4
2
3
.93 (d, J=10.0 Hz, 1H), 2.52 (q, J=7.4 Hz, 2H), 2.04 (q, J=7.1 Hz,
ratio of alkyl thiolates to alkenyl thiolates on particles was 17:83.
H NMR (CDCl , 500 MHz, selected peaks): d = 5.80 (brs, 1H), 4.94
3
1
H), 1.61 (quin, J=7.3 Hz, 2H), 1.39–0.92 (m, 13H); IR (neat): n˜
076, 2920, 2850, 1462, 910, 723 cm ; EI-MS: m/z: 185 [MÀH] .
=
À1
+
(brs, 2H), 2.03–0.87 (m, 2H); IR (KBr): n˜ = 3074, 2920, 2848, 1412,
À1
9
08 cm ; UV/Vis (toluene): lmax = 515 nm; elemental analysis calcd (%)
Gold nanoparticles 2a: Gold nanoparticles 2a were prepared by the
ligand-exchange reaction of allyl mercaptan with 1-propanethiol-protect-
ed gold nanoparticle. The 1-propanethiol-stabilized gold nanoparticles
were synthesized by using a modified version of a method in the litera-
ture. Hydrogen tetrachloroaurate(iii) tetrahydrate (6.00 g, 14.6 mmol)
in deionized water (200 mL) was added to a vigorously stirred solution of
tetra-n-octylammonium bromide (22.4 g, 41.0 mmol) in toluene (600 mL).
The yellow aqueous solution immediately became clear and the toluene
phase turned brown. The organic phase was transferred into a 1 L round-
bottom flask, and 1-propanethiol (1.5 mL, 16.8 mmol) was added to the
solution. Sodium borohydride (3.18 g, 84.0 mmol) in deionized water
for C583
3.08.
H1113Au140S53 (Au140(C11H21S)53): C 18.72, H 3.00; found C 18.83, H
[
13]
Preparation of silicon substrate: A typical experimental procedure for
the immobilization of gold nanoparticles onto a hydrogen-terminated sili-
con surface is as follows. The hydrogen-terminated silicon(111) surface
[
15]
[11]
was prepared by chemical etching as previously reported. Silicon(111)
wafers were cut into squares (10 mm”10 mm). The silicon substrates
were cleaned ultrasonically in toluene for 10 min. The substrates were
etched in 1% HF aqueous solution at 708C for 1 min (in a Teflon vessel)
4
and further etched in 30% NH F aqueous solution at room temperature
(
200 mL) was slowly added to a vigorously stirred solution. The resulting
for 5 min (in a Teflon vessel) to remove surface oxide and to etch the sur-
face anisotropically. The fresh hydrogen-terminated silicon surface was
transferred under nitrogen into a Schlenk tube containing gold nanoparti-
cles in toluene (30 mg/10 mL) and kept at 508C for 24 h. After cooling to
room temperature, the sample was removed from the solution, rinsed
with toluene, and sonicated in toluene for 10 min. Finally, the modified
silicon substrate was rinsed briefly with toluene and dried in a stream of
argon gas. CAUTION! HF is very dangerous, particularly in contact with
skin, and should be handled with extreme care.
solution was stirred for 3 h at room temperature. The solvent was re-
moved under reduced pressure without exceeding a temperature of 508C.
The dark residue was suspended in ethanol (1 L) and kept in a refrigera-
tor at À108C overnight. The precipitate was collected by filtration on a
membrane filter and washed thoroughly with ethanol. The purified gold
nanoparticles were isolated as a black powder (3.48 g). Further surface
functionalization was achieved by a ligand-exchange reaction with 1-pro-
panethiol-protected gold nanoparticle and allyl mercaptan. 1-Propane-
thiol-protected gold nanoparticles (529 mg) were dissolved in CH
2 2
Cl
Detachment of gold nanoparticles from 2c-immobilized silicon surfaces
bya Au–S ligand-exchange method : The freshly prepared hydrogen-ter-
minated silicon(111) surface was immersed into the toluene dispersion of
gold nanoparticle 2c (30 mg/10 mL) and heated to 1008C for 4 d. After
immobilization, the silicon wafer was removed from the solution, rinsed
thoroughly with toluene, cleaned sonically in toluene for 10 min, and
dried under a flow of nitrogen. The cleavage of the gold nanoparticles
from this specimen was accomplished in a nitrogen atmosphere by the
addition of 1-hexadecanethiol (10 mL). The reaction was carried out at
room temperature for 7 d. Subsequently, the silicon piece was removed
from the solution, rinsed thoroughly with toluene, sonicated in toluene
(
10 mL). To this solution was added allyl mercaptan (0.5 mL), and the re-
action mixture was stirred at room temperature for 2 h. The solvent was
evaporated under reduced pressure. The dark residue was suspended in
ethanol (1 L) and kept overnight in a refrigerator at À108C. The precipi-
tate was collected by filtration on a membrane filter and washed thor-
oughly with ethanol. Gold nanoparticles 2a were obtained as a black
1
powder (504 mg). The nanoparticles were characterized by H NMR, FT-
IR, UV-vis, elemental analysis, and TEM.
[
16] 1
Compound 2a:
H NMR gave very broad peaks. The ratio of alkyl thi-
1
olates to alkenyl thiolates on the particles was 51:49. H NMR (C
6
5 3
D CD ,
Chem. Eur. J. 2006, 12, 314 – 323
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
321

T. Yonezawa, H. Nishihara et al.
for 10 min, and dried under a flow of nitrogen. The specimen was then
promptly subjected to HR-SEM and XPS measurement.
[8] For recent reviews, see: a) Nanoparticles (Ed.: G. Schmid), Wiley,
Weinheim, 2004; b) J. C. Love, L. A. Estroff, J. K. Kriebel, R. G.
Nuzzo, G. M. Whitesides, Chem. Rev. 2005, 105, 1103–1169; c) B. L.
Cushing, V. L. Kolesnichenko, C. J. OꢁConnor, Chem. Rev. 2004, 104,
3
3
893–3946; d) M. C. Daniel, D. Astruc, Chem. Rev. 2004, 104, 293–
46; e) G. Schmid, B. Corain, Eur. J. Inorg. Chem. 2003, 3081–3098;
f) N. A. Kotov, “Layer-by-layer Assembly of Nanoparticles and Col-
loids: Intermolecular Interactions, Structure, and Material Perspec-
tives” in Multilayer Thin Films (Eds.: G. Decher, J. B. Schlenoff),
Wiley, Weinheim, 2003, pp. 207–269; g) T. Yonezawa, N. Toshima,
“Polymer-stabilized Metal Nanoparticles” in Advanced Functional
Molecules and Polymers (Ed.: H. S. Nalwa), Gordon and Breach,
London (UK), 2001, pp. 65–86; h) A. C. Templeton, W. P. Wuelfing,
R. W. Murray, Acc. Chem. Res. 2000, 33, 27–36.
Acknowledgements
The authors thank Hitachi Science Systems and Hitachi High Technolo-
gies for their experimental assistance with the HR-SEM observation.
This work was financially supported in part by a Grant-in-Aid for Scien-
tific Research from the Ministry of Education, Culture, Sports, Science,
and Technology, Japan. T.Y. also thanks the Mitsubishi Chemical Corpo-
ration Fund for financial support.
[
9] For example, see: a) E. W. L. Chan, D.-C. Lee, M.-K. Ng, G. Wu,
K. Y. C. Lee, L. Yu, J. Am. Chem. Soc. 2002, 124, 12238–12243;
b) E. W. L. Chan, L. Yu, Langmuir 2002, 18, 311–313; c) H. X. He,
H. Zhang, Q. G. Li, T. Zhu, S. F. Y. Li, Z. F. Liu, Langmuir 2000, 16,
3846–3851; d) A. C. Templeton, F. P. Zamborini, W. P. Wuelfing,
R. W. Murray, Langmuir 2000, 16, 6682–6688; e) J. A. Harnisch,
A. D. Pris, M. D. Porter, J. Am. Chem. Soc. 2001, 123, 5829–5830;
f) T. A. Taton, R. C. Mucic, C. A. Mirkin, R. L. Letsinger, J. Am.
Chem. Soc. 2000, 122, 6305–6306; g) T. Okamoto, I. Yamaguchi, J.
Phys. Chem. B 2003, 107, 10321–10324; h) S. Chen, R. Pei, T. Zhao,
D. J. Dyer, J. Phys. Chem. B 2002, 106, 1903–1908; i) H. Tanaka, M.
Mitsuishi, T. Miyashita, Langmuir 2003, 19, 3103–3105; j) Q. Li, J.
Zheng, Z. Liu, Langmuir 2003, 19, 166–171; k) R. D. Tilley, S. Saito,
Langmuir 2003, 19, 5115–5120; l) S. Chen, R. W. Murray, J. Phys.
Chem. B 1999, 103, 9996–10000; m) Y. Joseph, I. Besnard, M. Rose-
nberger, B. Guse, H.-G. Nothofer, J. M. Wessels, U. Wild, A. Knop-
Gericke, D. Su, R. Schlçgl, A. Yasuda, T. Vossmeyer, J. Phys. Chem.
B 2003, 107, 7406–7413; n) J. Zheng, Z. Chen, Z. Liu, Langmuir
[
1] For recent reviews, see: a) D. Y. Petrovykh, F. J. Himpsel, “Self-as-
sembled Nanostructures at Silicon Surface” in Encyclopedia of
Nanoscience and Nanotechnology, Vol. 9 (Ed.: H. S. Nalwa), Ameri-
can Scientific Publishers, Stevenson Ranch, California, 2004,
pp. 497–528; b) C. H. Choi, M. S. Gordon, “Chemistry on Silicon
Surface” in Chemistry of Organic Silicon Compounds, Vol. 3 (Eds.:
Z. Rappoport, Y. Apeloig), Wiley, 2001, pp. 821–851.
[
2] For recent reviews, see: a) P. K. Jal, S. Patel, B. K. Mishra, Talanta
2
2
004, 62, 1005–1028; b) F. Schreiber, Prog. Surf. Sci. 2000, 65, 151–
56; c) A. Ulman, Chem. Rev. 1996, 96, 1533–1554.
[
3] For recent reviews, see: a) J. M. Buriak, Chem. Rev. 2002, 102,
271–1308; b) D. D. M. Wayner, R. A. Wolkow, J. Chem. Soc.
1
Perkin Trans. 2 2002, 23–34; c) N. Shirahata, A. Hozumi, T. Yoneza-
wa, Chem. Rec. 2005, 5, 145–159.
[
4] For representative examples, see: a) M. R. Linford, C. E. D. Chidsey,
J. Am. Chem. Soc. 1993, 115, 12631–12632; b) M. R. Linford, P.
Fenter, P. M. Eisenberger, C. E. D. Chidsey, J. Am. Chem. Soc. 1995,
2
000, 16, 9673–9676.
1
17, 3145–3155; c) C. H. de Villeneuve, J. Pinson, M. C. Bernard, P.
[
10] Electrophoretic deposition, see: a) M. Yamada, T. Tadera, K. Kubo,
H. Nishihara, Langmuir 2001, 17, 2363–2370; b) M. Yamada, H.
Nishihara, Langmuir 2003, 19, 8050–8056.
Allongue, J. Phys. Chem. B 1997, 101, 2415–2420; d) A. B. Sieval,
A. L. Demirel, J. W. M. Nissink, M. R. Linford, J. H. van der Mass,
W. H. de Jeu, H. Zuilhof, E. J. R. Sudhçlter, Langmuir 1998, 14,
[
[
[
11] R. T. Sanderson,
Chemical Bonds and Bond Energy, Academic
1
759–1768; e) R. Boukherroub, S. Morin, F. Bensebaa, D. D. M.
Press, New York, 1976.
12] Preliminary communication: Y. Yamanoi, T. Yonezawa, N. Shiraha-
ta, H. Nishihara, Langmuir 2004, 20, 1054–1056.
Wayner, Langmuir 1999, 15, 3831–3835; f) A. B. Sieval, R. Linke,
H. Zuilhof, E. J. R. Sudhçlter, Adv. Mater. 2000, 12, 1457–1460.
5] For representative examples, see: a) J. M. Buriak, Adv. Mater. 1999,
[
13] a) Miller et al. reported the preparation of the covalently bonded
1
1, 265–267; b) H. C. Choi, J. M. Buriak, Chem. Mater. 2000, 12,
C
1
60 monolayers on silicon surface, see: W. Feng, B. Miller, Langmuir
999, 15, 3152–3156; b) Quite recently, Altavilla et al. reported the
monolayers of w-alk-1-ene-silane functionalized magnetite (Fe
nanoparticles covalently linked to a hydrogen-terminated silicon-
111) surface by a thermal hydrosilylation reaction. See: C. Altavil-
la, E. Ciliberto, D. Gatteschi, C. Sangregorio, Adv. Mater. 2005, 17,
084–1087.
2
151–2156; c) J. M. Buriak, M. P. Stewart, T. W. Geders, M. J. Allen,
H. C. Choi, J. Smith, D. Raftery, L. T. Canham, J. Am. Chem. Soc.
999, 121, 11491–11502; d) J. M. Schmeltzer, L. A. Porter, M. P.
3
4
O )
1
Stewart, J. M. Buriak, Langmuir 2002, 18, 2971–2974; e) R. Bou-
kherroub, A. Petit, A. Loupy, J.-N. Chazalviel, F. Ozanam, J. Phys.
Chem. B 2003, 107, 13459–13462; f) J. E. Bateman, R. D. Eagling,
D. R. Worrall, B. R. Horrocks, A. Houlton, Angew. Chem. 1998, 110,
(
1
2
829–2831; Angew. Chem. Int. Ed. 1998, 37, 2683–2685; g) M. P.
[14] For examples of ligand-exchange reactions on monolayer-protected
gold nanoparticles, see: a) M. J. Hostetler, A. C. Templeton, R. W.
Murray, Langmuir 1999, 15, 3782–3789; b) R. S. Ingram, M. J. Hos-
tetler, R. W. Murray, J. Am. Chem. Soc. 1997, 119, 9175–9178;
c) M. J. Hostetler, S. J. Green, J. J. Stokes, R. W. Murray, J. Am.
Chem. Soc. 1996, 118, 4212–4213.
[15] M. Brust, M. Walker, D. Bethell, D. J. Schiffrin, R. Whyman, J.
Chem. Soc. Chem. Commun. 1994, 801–802.
[16] One referee asked for an explanation about these experiments. Al-
though gold nanoparticles 2a were prepared according to Brustꢁs
Stewart, J. M. Buriak, Angew. Chem. 1998, 110, 3447–3450; Angew.
Chem. Int. Ed. 1998, 37, 3257–3260.
[
6] For representative recent examples of stable SiÀC linked monolay-
ers with hydrogen-terminated silicon surfaces, see: a) T. Yamada, T.
Inoue, K. Yamada, N. Takano, T. Osaka, H. Harada, K. Nishiyama,
I. Taniguchi, J. Am. Chem. Soc. 2003, 125, 8039–8042; b) B. Fabre,
D. D. M. Wayner, Langmuir 2003, 19, 7145–7146; c) P. T. Hurley,
A. E. Ribbe, J. M. Buriak, J. Am. Chem. Soc. 2003, 125, 11334–
1
8
1339; d) W. H. Yu, E. T. Kang, K. G. Neoh, Langmuir 2004, 20,
294–8300; e) B. J. Eves, Q.-Y. Sun, G. P. Lopinski, H. Zuilhof, J.
two-phase method from HAuCl
report, the reproducibility of the method was poor. Accordingly, we
prepared 2a by thiolate substitution reaction in CH Cl . From this
4
and allyl mercaptan in our previous
Am. Chem. Soc. 2004, 126, 14318–14319; f) D. Xu, E. T. Kang,
K. G. Neoh, A. A. O. Tay, Langmuir 2004, 20, 3324–3332; g) C. M.
Yam, J. Cho, C. Cai, Langmuir 2004, 20, 1228–1233.
2
2
result, we noticed that significant particle aggregation was observed
when the fraction of allyl mercaptan was increased (more than ap-
proximately 60%). On the other hand, high reproducibility was
found for the synthesis of gold nanoparticle 2b and 2c by the use of
Brustꢁs two-phase method. Therefore, these particles were not pre-
pared by ligand-exchange reaction.
[
7] For SiÀC bond formation by using cycloaddition of organic mole-
cules with semiconductor surfaces, see: a) F. Tao, G. Q. Xu, Acc.
Chem. Res. 2004, 37, 882–893; b) S. F. Bent, J. Phys. Chem. B 2002,
1
06, 2830–2842; c) R. J. Hamers, S. K. Coulter, M. D. Ellison, J. S.
Hovis, D. F. Padowitz, M. P. Schwartz, C. M. Greenlief, J. N. Russell,
Jr., Acc. Chem. Res. 2000, 33, 617–624.
322
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 314 – 323

Immobilization of Gold Nanoparticles
FULL PAPER
[
17] McCarley et al. reported the preparation of 10-undecene-1-thiol-
protected gold nanoparticles. J. S. Peanasky, R. L. McCarley, Lang-
muir 1998, 14, 113–123.
[26] a) J. Terry, M. R. Linford, C. Wigren, R. Cao, P. Pianetta, C. E. D.
Chidsey, Appl. Phys. Lett. 1997, 71, 1056–1058; b) N. Shirahata, T.
Yonezawa, Y. Miura, K. Kobayashi, K. Koumoto, Langmuir 2003,
19, 9107–9109; c) N. Shirahata, T. Yonezawa, W.-S. Seo, K. Koumo-
to, Langmuir 2004, 20, 1517–1520; d) A. Bansal, X. Li, S. I. Yi,
W. H. Weinberg, N. S. Lewis, J. Phys. Chem. B 2001, 105, 10266–
10277; e) N. Y. Kim, P. E. Laibinis, J. Am. Chem. Soc. 1998, 120,
4516–4517; f) R. L. Cicero, M. R. Linford, C. E. D. Chidsey, Lang-
muir 2000, 16, 5688–5695; g) R. Boukherroub, S. Morin, F. Bense-
baa, D. D. M. Wayner, Langmuir 1999, 15, 3831–3835; h) H. B. Liu,
R. J. Hamers, Surf. Sci. 1998, 416, 354–362.
[
18] a) M. M. Alvarez, J. T. Khoury, T. G. Schaaff, M. N. Shafigullin, I.
Vezmar, R. L. Whetten, J. Phys. Chem. B 1997, 101, 3706–3712;
b) T. G. Schaaff, M. N. Shafigullin, J. T. Khoury, I. Vezmar, R. L.
Whetten, W. C. Cullen, P. N. First, C. J. Gutiꢂrrez-Wing, Ascensio,
M. J. Jose-Yacamꢃn, J. Phys. Chem. B 1997, 101, 7885–7891.
19] Gold nanoparticles 2a–2c were added in great excess (more than
[
[
1
000 times the required quantity).
20] For recent examples, see: a) P. Buffat, J.-P. Borel, Phys. Rev. A 1976,
1
1
3, 2287–2298; b) L. J. Lewis, P. Jensen, J.-L. Barrat, Phys. Rev. B
997, 56, 2248–2257; c) C. J. Kiely, J. Fink, M. Brust, D. J. Schiffrin,
[27] We also carried out a ligand-exchange reaction of the 2a or 2b-im-
mobilized silicon surface with 1-hexadecanethiol, followed by XPS
measurement of the modified silicon surface. The signal of SiÀC
Nature 1998, 396, 444–446; d) T. Yonezawa, S. Onoue, N. Kimizuka,
Chem. Lett. 2002, 1172–1173.
could not be observed.
[
[
[
21] A. Stella, P. Cheyssac, R. Kofman, “Fundamental Properties and
Developments of Metallic and Semiconducting Nanoparticles” in
Science and Technology of Thin Films (Eds.: F. C. Matacotta, G. Ot-
taviani), World Scientific, River Edge, NJ, 1996, pp. 517–587.
22] a) M. M. Maye, W. Zheng, F. L. Leibowitz, N. K. Ly, C.-J. Zhong,
Langmuir 2000, 16, 490–497; b) M. M. Maye, C.-J. Zhong, J. Mater.
Chem. 2000, 10, 1895–1901; c) T. Teranishi, S. Hasegawa, T. Shimi-
zu, M. Miyake, Adv. Mater. 2001, 13, 1699–1701.
23] a) Q.-Y. Sun, L. C. P. M. de Smet, B. van Lagen, M. Giesbers, P. C.
Thꢄne, J. van Engelenburg, F. A. de Wolf, H. Zuilhof, E. J. R.
Sudhçlter, J. Am. Chem. Soc. 2005, 127, 2514–2523; b) Q.-Y. Sun,
L. C. P. M. de Smet, B. van Lagen, A. Wright, H. Zuilhof, E. J. R.
Sudhçlter, Angew. Chem. 2004, 116, 1376–1379; Angew. Chem. Int.
Ed. 2004, 43, 1352–1355; c) M. P. Stewart, J. M. Buriak, J. Am.
Chem. Soc. 2001, 123, 7821–7830.
[28] The reactions of 1-hexadecanethiol (neat) and 1-undecenethiol in
toluene (3 mg/10 mL) with a hydrogen-terminated silicon(111) sur-
face were carried out at room temperature for 7 d. The signal of SiÀ
C bonds was not detected by XPS measurement in both cases. See
Supporting Information for the XPS.
[29] For cross-sectional TEM observations of gold nanoparticles on sur-
faces, see: a) K. Sun, S. Zhu, R. Fromknecht, G. Linker, L. M.
Wang, Mater. Lett. 2004, 58, 547–550; b) S. H. Cho, S. Lee, D. Y. Ku,
T. S. Lee, B. Cheong, W. M. Kim, K. S. Lee, Thin Solid Films 2004,
447–448, 68–73.
[30] The calculated distance between particle 2b and the silicon surface
was obtained by
a semi-empirical calculation (MOPAC, PM3,
closed-shell-restricted) in Chem3D Pro.
[31] A. J. Speziale, Org. Synth. 1963, Vol. V, 401.
[
24] The use of ultrasound in the simple removal of contaminants attach-
ed as surface coatings is well known, see: T. J. Mason, J. P. Lorimer,
Sonochemistry, Wiley, Chichester (UK), 1988.
Received: April 22, 2005
Revised: June 25, 2005
[
25] See Supporting Information for the SEM image.
Published online: October 6, 2005
Chem. Eur. J. 2006, 12, 314 – 323
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
323