Dumbbells, trikes and quads: Organic-inorganic HYBRID NANOARCHITECTURES BASED on "cLICKED" GOLD NANOPARTICLES

Article abstract of DOI:10.1002/smll.201300839

The controlled assembly of gold nanoparticles in terms of the spatial arrangement and number of particles is essential for many future applications like electronic devices, sensors and labeling. Here an approach is presented to build up oligomers of mono functionalized gold nanoparticles by the use of 1,3-bipolar azide alkyne cycloaddition click chemistry. The gold nanoparticles of 1.3 nm diameter are stabilized by one dendritic thioether ligand comprising an alkyne function. Together with di-, tri- and tetra-azide linker molecules the gold nanoparticle can be covalently coupled by a wet chemical protocol. The reaction is tracked with IR and UV-vis spectroscopy and the yielded organic-inorganic hybrid structures are analyzed by transmission electron microscopy. To evaluate the success of this click chemistry reaction statistical analysis of the formed oligomers is performed. The geometric and spatial arrangements of the found oligomers match perfectly the calculated values for the used linker molecules. Dimers, trimers and tetramers could be identified after the reaction with the corresponding linker molecule. The results of this model reaction suggest that the used click chemistry protocol is working well with mono functionalized gold nanoparticles. Functionalized gold nanoparticles are interlinked by a click chemistry protocol into defined oligomers. The spatial and geometric arrangement is directed by the size and numbers of functional moieties of the linker molecule. Electron transmission microscopy reveal the formation of dimers, trimers and tetramers.

Full text of DOI:10.1002/smll.201300839

Nanoarchitectures
Dumbbells, Trikes and Quads: Organic–Inorganic
Hybrid Nanoarchitectures Based on “Clicked” Gold
Nanoparticles
Fabian Sander, Ulrike Fluch, Jens Peter Hermes, and Marcel Mayor*
T he controlled assembly of gold nanoparticles in terms of the spatial arrangement
and number of particles is essential for many future applications like electronic
devices, sensors and labeling. Here an approach is presented to build up oligomers
of mono functionalized gold nanoparticles by the use of 1,3-bipolar azide alkyne
cycloaddition click chemistry. The gold nanoparticles of 1.3 nm diameter are
stabilized by one dendritic thioether ligand comprising an alkyne function. Together
with di-, tri- and tetra-azide linker molecules the gold nanoparticle can be covalently
coupled by a wet chemical protocol. The reaction is tracked with IR and UV–vis
spectroscopy and the yielded organic-inorganic hybrid structures are analyzed by
transmission electron microscopy. To evaluate the success of this click chemistry
reaction statistical analysis of the formed oligomers is performed. The geometric and
spatial arrangements of the found oligomers match perfectly the calculated values
for the used linker molecules. Dimers, trimers and tetramers could be identified
after the reaction with the corresponding linker molecule. The results of this model
reaction suggest that the used click chemistry protocol is working well with mono
functionalized gold nanoparticles.
review articles highlight the importance of this research
1
. Introduction
[
10–13]
field.
Various concepts were developed in order to con-
Gold nanoparticles (Au NPs) are intensively studied as easy
accessible model compounds and functional units in various
applications. Au NPs are utilized for example in catalysis,
sensor application, nano-electronic devices
ling.
nect Au NPs into more complex structures by exploiting
[14]
different binding concepts. The use of π–π interactions,
[
1,2]
[15–17]
[18]
hydrogen bonding,
charge transfer
or coordinative
were investigated. The use of host-guest interac-
[
3]
[4,5]
and labe-
[19]
bonds
[6–9]
For most of these applications a controlled assembly
[20–23]
[24–33]
tions
and DNA recognition subunits
were also
of Au NPs into defined architectures is essential and several
used to build up well defined assemblies of nanoparticles.
Covalent bonds are among the most stable ways to assemble
Au NPs into complex structures. A widely used tool to
interlinke Au NPs or to decorate their surface is the Cu(I)
F. Sander, U. Fluch, Dr. J. P. Hermes,
Prof. Marcel Mayor
University of Basel
[34]
catalyzed “Sharpless click reaction”,
as the mild room
temperature version of the Huisgen 1,3-dipolar cycloaddi-
tion between azide and an alkyne forming a 1,2,3-triazole.
The covalent interlinking of Au NPs has been used for the
optical detection of copper by color change of Au NP solu-
Department of Chemistry
St. Johannsring 19
CH-4056, Basel, Switzerland
E-mail: marcel.mayor@unibas.ch
Prof. M. Mayor
[35–38]
tions or by precipitation after forming huge networks.
Karlsruhe Institute of Technology (KIT)
Institute of Nanotechnology (INT)
P. O. Box 3640, D-76021, Karlsruhe, Germany
The detection of proteins has been achieved by using func-
tional Au NPs, which create large assemblies resulting in an
[
39,40]
optical response.
Au NPs were also assembled by click
[
41]
DOI: 10.1002/smll.201300839
chemistry to form photo responsive networks,
liquid
small 2012,
DOI: 10.1002/smll.201300839
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
wileyonlinelibrary.com
1

full papers
F. Sander et al.
using click chemistry. In particular our
attempts to assemble dimers, trimers and
tetramers are reported.
2
. Concept and Strategy
The dendritic thioether ligand G2
Scheme 1) efficiently stabilizes Au NPs
(
of a size of about 1.3 nm. The Au NPs
are formed by reduction of choloro auric
acid in present of the ligand G2. While
this reduction the growing Au NPs are
enwrapped by the dendritic ligand that sta-
bilizes and traps the Au NPs with an almost
monodisperse size distribution around
1.3 nm that is given by the size of ligand’s
cavity. One G2 dendrimer bearing a tri-
isopropylsilyl (TIPS) protected acetylene
moiety is capable to stabilize an entire Au
NP (Au/G2) and for that reason provides
full control over the chemical reactivity
of the organic shell surrounding the Au/
Figure 1. 3D sketch of the click reaction and the resulting oligomers. G2 stabilized Au NPs
(
(
left) are clicked together with benzyl azides (middle) into dimers, trimers and tetramers
right).
[55,56]
G2.
The acetylene moiety is attached
crystals,[42] oligomers,
metal nanoparticles
[43]
chains,[31] combined networks of to an oligo phenylethylene (OPE) rod in order to have the
and robust nanostructures of dif- functionality in the NP’s periphery. A pyridine moiety was
[
44,45]
[
46]
ferent shapes of Au NPs. Furthermore the assembly of Au chosen as anchor of the OPE to the NP surface because it
NPs via click chemistry on functionalized metal and polymer is known to provide a fixed perpendicular arrangement of
[
47–51]
[54]
surfaces was performed.
the OPE spacer on the Au NP surface. This rigid arrange-
These approaches were usually controlled by the max- ment increases the spatial control over the NPs forming the
imum number of reactants fitting around one NP or were organic/inorganic hybrid structure. Click chemistry seems
performed in a statistical fashion. Our goal is to obtain well- particularly suited to covalently assemble the monofunctional
defined oligomers of Au NPs in means of spatial arrangement Au NPs Au/G2 in well-defined structures. The mild condi-
and number of NPs within each assembly. To obtain such tions of this well-known azide-alkyne click chemistry reaction
structures the ideal Au NP would be characterized by three make it perfect candidate for this purpose. All reagents used
key features: good chemical stability, narrow size-distribution to catalyze this 1,3-bipolar cycloaddition are easily removed
and well controlled number and spatial arrangement of func- after the reaction by aqueous workup. In order to perform
tional groups exposed at its surface. In 2008 our research click chemistry azide functionalities have to be installed at
group presented the formation of stable Au NPs stabilized the periphery of the interlinking structure. By the number of
by octadentate thioether ligands.[52] It was found that two the installed azide moieties, the size of the interlinking struc-
ligands are coating the NP’s surface, leading to bifunctional ture and their geometric arrangement full control is gained
Au NPs upon introduction of a functional moiety into the over the number of linked particles and their spatial arrange-
[
53]
ligands backbone. These functional NPs carrying two ter- ment. Here benzyl azides are investigated due to the easy
minal protected acetylenes were coupled to form organic/ synthetic accessability of this functional group and its known
inorganic superstructures by applying a wet-chemical oxida- reactivity in click reactions. On the one hand rotational
[
54]
tive acetylene homocoupling protocol.
In order to create freedom around the benzylic carbon-carbon bond is expected
monofunctionalized NPs, larger dendritic ligands were to provide an increased reactivity because the huge particles
designed. Exactly one second generation dendrimer con- can evade each other and do not cause sterical hindrance.
sisting of 20 thioether moieties was able to stabilize an entire One the other hand the distance between the particles can
[
55]
Au NP leading to monofunctionalized Au NPs, which were differ more if they can form varying arrangements around
used to form dumbbell structures by oxidative acetylene the linker molecules. However, the difference in the distance
[56]
coupling.
These alkyne-monofunctionalize Au NPs can should be small enough that a statistical evaluation by trans-
be considered as artificial molecules and in order to further mission electron microscopy (TEM) imaging is still possible.
explore the range of applicable reactions, we became inter- Different numbers of benzylic azides evenly arranged on a
ested in the copper catalyzed mild click reaction between central benzene ring are expected to direct the assembly of
alkynes and azides.
Au NPs into unique geometric arrangements, which can be
Here we present the assembly of organic/inorganic hybrid observed by TEM investigations. Figure 1 sketches the con-
architectures by interlinking mono-alkyne functionalized Au cept of controlling the spatial assembly of the interlinked NPs
NPs with small oligoazide functionalized linker structures by the azide substitution pattern of the linker.
2
www.small-journal.com
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2012,
DOI: 10.1002/smll.201300839

Organic–Inorganic Nanoarchitectures Based on “Clicked” Gold Nanoparticles
Scheme 1. Synthesis of ligand G2 using literature known fragments G1[SH] and 1.
Brust et al.[61] Hydrochloroauric acid was solved in H O and
then transferred into the organic phase by adding tetraoc-
3
. Synthesis
2
3
.1. Synthesis of the ligand
tylammonium bromide (TOAB) as phase transfer agent in
CH Cl . The dendrimer G2 solved in CH Cl was added to
2
2
2
2
this mixture and stirred for 15 min. Subsequently a solution
The G2 dendrimer, which is used to stabilize the Au NPs, was
synthesized from literature known dendron [G1.SH] and the
functional OPE rod (1, see Scheme 1).
To yield G2 two equivalents of [G1.SH] were reacted with
one equivalent of the OPE rod (1) in tetrahydrofurane (THF)
with sodium hydride (NaH) as base. After aqueous workup
and purification with silica column chromatography and
finally recycling gel permeation chromatography (GPC) the
dendritic ligand was obtained as colorless oil in good yields.
of sodium borohydride in H O was added as reducing agent
2
[56]
leading to an immediate color change to dark brown indi-
cating the formation of Au NPs. This mixture was stirred addi-
tional 30 min. After aqueous workup and removal of ligand
excess by gel permeation chromatography (GPC) the size of
the Au NPs was determined by UV–vis and TEM. The UV–
vis was measured in CH Cl and showed a very weak plasmon
2
2
band at 520 nm, indicating Au NPs with a size below 2 nm.The
features of the OPE rod were observed at 280 nm (Figure 2).
A diluted sample was transferred on a carbon coated copper
grid (CCCG) to determine the particle size by TEM images.
Small Au NPs with a narrow dispersity were observed on
the TEM pictures from which the size was measured by an
3
.2. Synthesis of the Multi-Triazide Functionalized Linkers
The azide functionalized linker molecules 2–4 were synthesized
by a classical S 2 reaction following a reported protocol. The
N
benzylbromide derivative of interest was treated with an excess
of sodium azide (approximately 25 equivalent) as 0.5 M solution
[57]
in dimethylsulfoxide (DMSO) (Scheme 2). After stirring for
8 h, the reactions were quenched and washed five times with
1
water to remove remaining DMSO. The crude products were
purified by column chromatography. Those compounds were
obtained in quantitative yields and were stable at ambient con-
ditions for several days. Starting from 1,4-bis(bromomethylene)
benzene
5
the 1,4-bis(azidomethylene)benzene
2
was
obtained as colorless oil. The 1,3,5-tris(azidomethylene)ben-
zene 3 was synthesized from 1,3,5-tris(bromomethylene)ben-
zene 6 as colorless oil. The 1,2,4,5-tetrakis(azidomethylene)
benzene 4 was isolated as colorless solid starting from
1
,2,4,5-tetrakis(bromomethylene)benzene 7. The literature
1
13
known azides 2–4 were characterized by H- and C NMR,
mass spectrometry (MS) and match the reported data.
[58–60]
3
.3. Synthesis of the Au NPs
The Au NPs were synthesized in presence of the dendrimer
G2 in a H O/CH Cl solvent mixture following a protocol by Scheme 2. Synthesis of azide linker molecules 2–4.
2
2
2
small 2012,
DOI: 10.1002/smll.201300839
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com 3

full papers
F. Sander et al.
automatic investigation method using the software ImageJ.[62]
As recently published, the yielded Au NPs show narrow size
[56]
distribution with a mean diameter of 1.3 ± 0.4 nm. Similar
diameters were found for related ligand-stabilized NPs sys-
tems based on unfunctionalized thioether dendrimers or
[52–54,63]
linear thioether ligands.
The 1/1 ratio of stabilizing
ligand per Au NPs was corroborated by thermogravimetry. As
only one ligand is coating the entire Au NP, only one func-
tional group is exposed per Au NP. It is noteworthy that sim-
ilar monofunctionalized NPs formed exclusively dimers upon
oxidative coupling conditions, further supporting the hypoth-
[56,64]
esis of monofunctionalized Au NPs.
3.4. Interlinking of the Monofunctionalized Au NPs
The oxidative homocoupling of the acetylene functionalized
particles in recent publications was conducted with Cu(I)
Figure 2. UV–vis spectra to monitor the click reaction.
Scheme 3. Synthesis of nano-scale objects by wet-chemical assembly of mono-ethynyl functionalized nano-particles and the azide linkers 2–4
under “click-chemistry” conditions. Synthesis of a “dumbbell” dimer A), a trimer B) and a tetramer C); (the particle coating dendritic ligand is only
sketched as formula to increase clarity).
4
www.small-journal.com
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2012,
DOI: 10.1002/smll.201300839

Organic–Inorganic Nanoarchitectures Based on “Clicked” Gold Nanoparticles
as catalyst in stoichiometric amounts. In order to favor the
cycloaddition and avoid homocoupling as competing reac-
tion, a protocol was used with a low concentration of Cu(I)
as active catalytic species which was formed in situ by reduc-
tion of Cu(II) by sodium ascorbate. As test reaction the Au
NPs were exposed to all reagents in CH Cl and CH Cl /H O
2
2
2
2
2
mixtures in order to investigate the stability of the NPs under
the reaction conditions. No precipitation or changes in the
UV–vis spectra were observed during this treatment. Also
subsequent TEM investigations showed a similar size distri-
bution as the initial sample, pointing at the stability of the
particles under the reaction conditions of interest.
The synthesis of NP oligomers by interlinking the
ethynyl-exposing NPs with the oligoazides 2–4 is displayed in
Scheme 3. First the Au NPs were dissolved in CH Cl and the
2
2
TIPS protection group of the ethynyl function was removed
by fluoride ions coming from subsequently added tetrabu-
tylammonium fluoride (TBAF). The reaction was quenched
after 1 h and washed with water. After these deprotected
AuNPs exposing a free ethynyl group were dried and redis-
solved in CH Cl , an equivalent of triazides was provided
Figure 3. Normalized ATR-IR Spectra to monitor the decrease of azide
groups (2100 cm ).
2
2
−1
by adding the required amounts of the multi-triazide linker
dissolved CH Cl . Thus 1/2 equivalent 2, 1/3 equivalent 3
and 1/4 equivalent 4 were added respectively. Subsequently
2
2
latter two reflect dimensional details of the hybrid architec-
ture. Thus TEM analyses were ideally suited to investigate
the nature and the ratios of the formed hybrid oligomers. In
all three samples of the reactions with the respective linker
molecules the NP sizes maintained pointing at their stability
in the applied reaction conditions. The yield of the obtained
NP oligomers was counted on randomly recorded TEM pic-
tures of very diluted solutions. High dilution was chosen to
minimize wrong hits eventually emerging from agglomer-
ating hybrid objects and/or NPs, which might result in sim-
ilar inter NP spacing as the desired hybrid objects. On these
diluted micrographs all particles where counted and the
amount of oligomeric architectures was determined manu-
ally. A set of NPs was considered as a hybrid architecture if
their lateral separation was within the distances dmax given
by the following considerations. In order to determine dmax
the maximal expansion of the interlinking molecular struc-
ture was analyzed. These structures combine the functional
units of the NPs fused to the corresponding linker mole-
cule and were created with the software SPARTAN V4.1.1
and shown as simplified drawing in Figure 4. The modeled
structures are called ms2 , ms3 and ms4 . To determine their
maximal extensions dmax the OPE structures were rotated
around the benzylic C–C bond until the maximum possible
distance between two terminal N atoms of the structure was
found (Figure 4).
a 1 M CuSO solution in followed by 25 mol% sodium ascor-
4
bate as 1 M solution in water were added to the vigorously
stirred reaction mixture. For the reactions of all linker mol-
ecules a similar analytical procedure was applied. First a
sample of the CH Cl phase was taken and investigated by
2
2
UV–vis spectroscopy to monitor the Au NPs size (Figure 2).
No increase in intensity or shift of the plasmon band was
observed pointing at the structural integrity of the Au NPs
under the applied reaction conditions. In contrast to that
the signal of the OPE structure around 280 nm displayed a
pronounced increase in intensity together with a slight shift
to longer wavelength. These spectral changes point at an
increased delocalization of the π-system, as it is expected by
the formation of the triazole heterocycles as products of the
“
click-chemistry”.
The click reaction was further monitored by attenuated
total reflectance infrared spectroscopy (ATR-IR) where
mainly the consumption of the azide linker was observed
−1
(
Figure 3). The triazide band at 2100 cm decreased to a
very small remaining trace showing the almost complete
transformation of the triazide group during the course of the
“
click” reaction.
Finally the reaction was stopped by allowing the two
phases to separate and the dark brown reddish colored
CH Cl phase was dried over MgSO .
2
2
4
In order to consider potential imprecision of the TEM
images, the obtained maximum distance between both ter-
minal nitrogen’s was increased by 0.4 nm, which is the size
of almost two pixels. So two particles were considered as
4
. TEM Analysis
The solved coupled Au NPs were further diluted by approxi- connected if their distance dmax was within the calculated
mately a factor of 1000. This diluted solution was spotted on value dmax 3.9 nm for ms2, dmax = 3.8 nm for ms3 and
=
a CCCG in order to investigate structural features of the dmax = 3.9 nm for ms4. For the statistical evaluation all dis-
reaction products by TEM. Of particular interest are sizes, tances of Au NPs fitting this selections rule were counted.
distances and spatial arrangements of Au NPs. While the Also the ratio of different oligomers was investigated for
first parameter mainly allows analyzing the NPs’ stability, the each sample.
small 2012,
DOI: 10.1002/smll.201300839
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com 5

full papers
F. Sander et al.
Figure 4. Schematic drawing of model structures ms2, ms3 and ms4 to determine maximum possible distance dmax between two NP
4.1. Dimer Structures
of single structures points at particles lacking an exposed
and reactive acetylene group. As NMR studies document the
completeness of the deprotection reaction, potential hypoth-
eses are either sterical shielding by the large NP or com-
peting degradation of the acetylene during the course of the
coupling reaction.
The linker molecule 2 carrying two benzylic azide functions
assembles NP with a maximum interparticle distance of dmax
=
3.9 nm. To yield as many desired dimers as possible 0.5 eq.
of 2 were reacted with 1 eq. Au NPs. On TEM images 249
particles were counted. 94 NPs were found as single struc-
tures, 128 NPs assembled as dimers, 12 NPs as trimers and
The interparticle distance d measured of the oligomers
display mean value of 2.9 nm calculated by a Gaussian fit
(
Figure 5B). All distances of particles matching the selection
1
6 NPs within higher oligomers (Figure 5A). Trimers and
rule mentioned above are taken in account.
tetramers can occur because of aggregation of the Au NPs.
The found trimers can be the aggregation product of three
single Au NPs or more likely because of the high dilution
one dimer and a monomer. The tetramers are also aggregates
of either a dimer with two monomers on each side or two
dimers. Over all about 51% of the counted particles show the
expected formation of dimers (Figure 5C). The high amount
4
.2. Trimer Structures
The evaluation of the click reaction with linker molecule 3
was performed following the same procedure. To yield the
Figure 5. A) Distribution of the formed oligomers, B) Distance distribution between the oligomer forming particles with Gaussian fit (red curve),
C) Representative TEM pictures of formed dimer with linker molecule, more pictures are displayed in the supporting information (Figures S3, S4),
D) Zoom with overlayed 3D sketch of proposed structure.
6
www.small-journal.com
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2012,
DOI: 10.1002/smll.201300839

Organic–Inorganic Nanoarchitectures Based on “Clicked” Gold Nanoparticles
Figure 6. A) Distribution of the formed oligomers, B) Distance distribution between the oligomer forming particles comprising a Gaussian fit (red
curve), C) Representative TEM pictures of formed trimers with linker molecule 3, additional TEM records are displayed in the supporting information
(
Figures S5, S6), D) Zoom with overlayed 3D sketch of the proposed structure.
desired trimers equimolar ratios of azide groups in linker 4.3. Tetramer Structures
molecule 3 and Au NPs were mixed together under click
conditions. The calculated maximum distance dmax = 3.8 nm Subsequently the click reaction of an equimolar amount
of the clicked triangular structure ms3 is just slightly shorter deprotected NPs with linker molecule 4 was examined. The
than the distance of the clicked assembly formed by ms2 . maximum length dmax of ms4 was calculated to be 3.9 nm,
For this arrangement the third NP would form an isos- which is the same as the stretched conformation of the inter-
celes triangle if the other two are in a stretched arrange- linking structure ms2 . For interlinked substructure ms4 a rec-
ment. Therefore smaller distances are expected as three tangular shape with two different sides (Figure 8) is expected.
NPs should form an equilateral triangle for less sterical Just the four sides were measured for the evaluation of
repulsion. A total of 264 NPs were evaluated, of whom the interparticle distance; the diagonals were not taken in
1
04 were present as monomers, 22 arranged in dimers, 117 account. This procedure was applied because also dimers
connected to trimers (Figure 6A) and 21 lying together as and trimers where measured. In the case of the trimers it
higher oligomers. The found dimers are probably linkers cannot be distinguished whether a edge or diagonal distance
that only reacted with two Au NP, which seems the more is measured. To measure a equal representative amount of
likely scenario than two aggregated Au NPs, considering distances all four edges of a tetramer structure were con-
the high dilution deposition protocol. The found tetramers sidered without the diagonals. For a dimer one distance is
are supposed to be aggregates of trimers with a single Au measured, for a trimer three distances and four distances for
NP. Again a reasonable amount of monomers could be a tetramer. The diluted TEM images of the clicked samples
observed. Because of the low ratio of dimers a passivation displayed nanoparticles present as monomers, dimers, trimers
of the linking structure during the reaction must occur. If and tetramers without a clear peak for the desired oligomer
only sterical reasons would cause this distribution a higher as in the first experiments (Figure 7A). Again about 30% of
amount of dimers would be expected. The coupling reaction the Au NPs were present as monomer structures what is in
is relatively fast as no further increase in the yield of oli- analogy with the previous reactions. One difference in this
gomers is observed after 2 h. This observation further favors case is the decreasing number of found oligomers that is a
the hypothesis of competing side reactions over solely ster- hint for additional restraining effects. Our hypothesis is that
ical arguments.
this might be due to the limited space around the linker mol-
A clear trend can be observed as about 44% of the nano- ecule 4. This is leading to a higher yield of undesired dimers
particles were found in triangular assemblies (Figure 6C). and trimers instead of tetramers. The first two NPs probably
The distance distribution measured from TEM is broader react in the same manner as for linker molecule 2 because the
than the first sample and shows a smaller mean interpar- additional benzylic azides are not very sterical demanding.
ticle distance of 2.4 nm (Figure 6B). The broadening can be As soon as two NP are attached to linker 4 both remaining
explained by the higher flexibility of three possible rotations azides are already shielded by the present NPs hindering the
around the benzylic C–C bond leading to a larger variety of next reaction step to the trimer. The reaction of the trimer to
NPs spacing.
the tetramer is even more hindered, as the linker molecule
small 2012,
DOI: 10.1002/smll.201300839
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com 7

full papers
F. Sander et al.
Figure 7. A) Distribution of the formed oligomers, B) Distance distribution between the oligomer forming particles with a Gaussian fit (red curve),
C) Representative TEM pictures of formed tetramers with linker molecule 4, more TEM pictures are displayed in the supporting information
(
Figures S7–S9), D) Zoom with overlayed 3D sketch of proposed structure.
is rather small compared to the particles. Therefore, the for- 4.4. Discussion of the Measured Distances
mation of tetramers is not favored and an equal distribution
of oligomers was obtained. Of 299 NPs that were evaluated
for the investigation 131 of them were present as monomers.
All found interparticle distances are all considerably shorter
than the calculated maximum possible spacing. This behavior
was expected because the maximum distances dmax were cal-
culated to define a selection rule in order to determine parti-
cles as coupled to each other. In reality the spacing between
the particles must be shorter because the interlinking struc-
ture will arrange in a more folded geometry (Figure 8).
9
8 NPs were present as dimers, 93 as trimers and 64 NPs were
reacted to tetramers (Figure 7A). Extended duration of the
click reaction could not improve the ratio of the oligomers
towards more tetramers. The mean distance for all oligomers
stated as connected to each other is 2.4 nm (Figure 7B).
Figure 8. Energy minimized model (MMFF94) structures ms2, ms3, and ms4 in stretched out geometry representing the expected maximum
spacing between Au NPs interlinked by the molecules 2, 3, and 4 respectively.
8
www.small-journal.com
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2012,
DOI: 10.1002/smll.201300839

Organic–Inorganic Nanoarchitectures Based on “Clicked” Gold Nanoparticles
dcalc with the measured distances d fully
supports the hypothesis of formed trimers
and their flat relaxed arrangement as equi-
lateral triangle on the CCCG of the TEM.
In similarity to the trimers above, a
flat arrangement is also assumed for the
tetramer structures on the CCCG. For the
rectangular structures ms4 the two dif-
ferent sides l2A and l2B must be taken into
account. The calculated inter particle dis-
tances dA and dB are calculated from the
values l1A, l2A, l1B, l2B and r (Figure 8) with
formula 1 in the same manner as for the
triangle structures (Figure 9). The meas-
ured mean distance of 2.4 nm matches
calculated mean distance 2.4 nm of dA
=
Figure 9. Sketch to visualize calculations for Au NP inter particle distances. Left side dimer
structure, middle trimer structure and right tetramer structure. All lengths are calculated by
using molecular mechanics (MMFF94) from ms2, ms3, and ms4 (see Figure 8).
2
.2 and d = 2.7 for a flat assembly per-
B
fectly (Table 2). The fact that the meas-
ured distances of the reaction to tetramers
matches the calculated values supports
An approximate value to evaluate the experimental found the hypothesis that also the found dimers and trimers arise
distances was calculated for each click reaction. The dimer from not completely reacted linker molecules, because they
structure ms2 is able to rotate around both benzyl C – C assemble in similar distances to each other. These results
bounds what leads to various stretched and folded conforma- reflect the flat arrangement of the tetramer structures as well
tion of interlinked particles. The minimum distance for the as the hypothesis of the high amount of dimers and trimers
folded conformation is calculated to be l1 fold = 1.3 nm for the caused by sterical repulsion of the Au NPs during the click
stretched conformation a maximum value of l1 stre = 3.2 nm reaction. All values for the calculation are listed in Table 1.
is calculated. The mean diameter of all found dimers show a
nice distribution from 1.3 nm to dmax = 3.9 nm with its mean tances are summarized for all synthesized Au NP oligomers.
value at 2.9 nm. Because of the huge size and the resulting
In summary, the observed nanoparticle oligomers
In Table 2 the experimental found and calculated dis-
repulsion of the Au NPs a stretched assembly is more prob- reflecting the numbers and spatial orientation of the func-
ably. That is reflected in the found distance d = 2.9 nm which tional groups of the oligoazide precursors, the interparticle
is at the upper end of the possible calculated value matching distance analysis in the obtained organic/inorganic hybrid
nearly calculated distance for a stretched assembly.
In contrast to the above discussed stretched dimers, the
measured interparticle distances in trimers and tetramers is
Table 1. All calculated lengths for l , l , l2A, l2B by 3D model using
molecular mechanics (MMFF94) and measured radius of Au NPs by
TEM.
1
2
no longer directly reflecting the interlinking structure and its
correlation must thus be corrected. For the trimer structure a
flat arrangement like an equilateral triangle is expected. This
arrangement would provide a maximum of space for each
coupled Au NP. From a MM2 relaxed planar fixed model of
structure ms3 the distance dcalc for an equilateral triangle is
calculated with equation (1).
Dimer [nm]
Trimer [nm]
Tetramer [nm]
l1
3.2
x
1.6
2.6
x
1.6
x
l2
l2A
l2B
r
x
2.1
2.7
l
2
· (l
2
+ r)
x
x
d
calc
=
− 2r
(1)
l
2
0.6
The value l1 is measured form the coupled structures
center ms3 to the nitrogen atom of the clicked OPE rod of Table 2. Maximum distance for particles count as oligomer, average
experimental found distance, calculated distance value form energy
minimized model (MMFF94).
the ligand G2 and l2 from one nitrogen atom to the next
neighbored in this structure (Figure 8). With the radius r of
the Au NPs the distance d can be calculated by applying the
Distance [nm]
Dimer
3.9
Trimer
3.8
Tetramer
3.9
intercept theorem as displayed in Figure 9 using equation
1). The found mean distance d of 2.4 nm matches the calcu-
Maximum Distance dmax
(
lated value dcalc of 2.4 nm perfectly. Of course the distances Found Distance d
2.9
2.4
2.4
can vary when the structure is rotated around the benzylic
C–C bounds from a minimum value of 1.3 nm to a maximum
value of 3.1 nm that is well reflected in the distance distribu-
2.4a)
Calculated Distance dcalc
3.2
2.4
a)
For tetramers two different distances of NP neighbors (2.1 nm) and NP on oppo-
site sites (2.7 nm) can be expected. Mean value of both distances calculated for
tion (Figure 6B). The perfect matching of the calculated value the tetramer.
small 2012,
DOI: 10.1002/smll.201300839
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com 9

full papers
F. Sander et al.
architectures, and the disappearance of the azide bands in
the IR spectra are corroborating the successful interlinking
of the ethynyl functionalized particles by “click chemistry”.
The UV–vis spectra further support the formation of triazole
linkers and corroborate the stability and homogeneity of the
ligand coated gold nano-particles under the applied reaction
conditions.
[6] I. Willner, B. Willner, Nano Lett. 2010, 10, 3805–3815.
[
[
7] R. D. Powell, J. F. Hainfeld, Micron 2011, 42, 163–174.
8] J. I. Cutler, E. Auyeung, C. A. Mirkin, J. Am. Chem. Soc. 2012, 134,
376–1391.
9] L. Dykman, N. Khlebtsov, Chem. Soc. Rev. 2012, 41, 2256.
1
[
[
10] F. Westerlund, T. Bjørnholm, Curr. Opin. Colloid Interface Sci.
009, 14, 126–134.
[11] K. J. M. Bishop, C. E. Wilmer, S. Soh, B. A. Grzybowski, Small
009, 5, 1600–1630.
12] C. L. Choi, A. P. Alivisatos, Annu. Rev. Phys. Chem. 2010, 61,
69–389.
13] M. Grzelczak, J. Vermant, E. M. Furst, L. M. Liz-Marzán, ACS Nano
010, 4, 3591–3605.
2
2
[
[
3
5
. Conclusion
2
The formation of Au NPs into defined organic/inorganic
hybrid nanoarchitectures in means of quantity and spatial
arrangement utilizing a wet click chemistry protocol is suc-
cessfully demonstrated with mono acetylene functional-
ized Au NPs. A series of azide linker molecules was used to
[
[
14] K. Naka, H. Itoh, Y. Chujo, Langmuir 2003, 19, 5496–5501.
15] C. R. van den Brom, P. Rudolf, T. T. M. Palstra, B. Hessen, Chem.
Commun. 2007, 4922.
[
16] H. Ozawa, M. Kawao, H. Tanaka, T. Ogawa, Langmuir 2007, 23,
6365–6371.
build up dimer, trimer and tetramer structures reflecting the [17] C.-P. Chak, S. Xuan, P. M. Mendes, J. C. Yu, C. H. K. Cheng, K. C.-F.
Leung, ACS Nano 2009, 3, 2129–2138.
18] K. Naka, H. Itoh, Y. Chujo, Langmuir 2003, 19, 5496–5501.
19] H. Ozawa, M. Kawao, H. Tanaka, T. Ogawa, Langmuir 2007, 23,
linker’s geometry. Our current attempts are geared towards
[
[
optimizing the reactive sites for the Huisgen 1,3-dipolar
cycloaddition in order to improve the efficiency of the cou-
pling reaction. We are searching also for ligand structures
providing larger monofunctionalized metallic particles dis-
playing plasmon resonance bands which are of interest for a
6
365–6371.
[
20] R K. la jn , M .A . Olson , .P J. Wesson , L. Fa ng , A. Coskun , A. Tr abolsi ,
S. Soh, J. F. Stoddart, B. A. Grzybowski, Nat. Chem. 2009, 1,
733–738.
variety of optical experiments and devises. Finally strategies [21] M. A. Olson, A. Coskun, R. Klajn, L. Fang, S. K. Dey, K. P. Browne,
B. A. Grzybowski, J. F. Stoddart, Nano Lett. 2009, 9, 3185–3190.
to improve the stability of the particles are very appealing as
the diversity of applicable reactions scales with the stability
of the reacting species.
[
22] M. A. Olson, A. Coskun, R. Klajn, L. Fang, S. K. Dey, K. P. Browne,
B. A. Grzybowski, J. F. Stoddart, Nano Lett. 2009, 9, 3185–3190.
23] Q. Zeng, R. Marthi, A. McNally, C. Dickinson, T. E. Keyes, R. J. For-
ster, Langmuir 2010, 26, 1325–1333.
[
The proof of concept to use Au NPs as artificial molecules
engaged in click reactions is appealing for numerous applica- [24] A. P. Alivisatos, K. P. Johnsson, X. Peng, T. E. Wilson, C. J. Loweth,
tions ranging from labels to functional units in electronic and
optical devises. The future success of the approach however
will depend strongly on the nature, variety, and quality of
available particles.
M. P. Bruchez, P. G. Schultz, Nature 1996, 382, 609–611.
25] C. A. Mirkin, R. L. Letsinger, R. C. Mucic, J. J. Storhoff, Nature
[
[
[
[
[
[
[
[
[
[
1
996, 382, 607–609.
26] Z. Deng, Y. Tian, S.-H. Lee, A. E. Ribbe, C. Mao, Angew. Chem., Int.
Ed. 2005, 44, 3582–3585.
27] F. A. Aldaye, H. F. Sleiman, Angew. Chem., Int. Ed. 2006, 45,
2
204–2209.
28] F. Huo, A. K. R. Lytton-Jean, C. A. Mirkin, Adv. Mater. 2006, 18,
304–2306.
29] S. E. Stanca, R. Eritja, D. Fitzmaurice, Faraday Discuss. 2006,
31, 155.
2
Supporting Information
Supporting Information is available from the Wiley Online Library
1
30] J. H. Lee, D. P. Wernette, M. V. Yigit, J. Liu, Z. Wang, Y. Lu, Angew.
Chem., Int. Ed. 2007, 46, 9006–9010.
31] M. Fischler, A. Sologubenko, J. Mayer, G. Clever, G. Burley,
J. Gierlich, T. Carell, U. Simon, Chem. Commun. 2008, 169.
32] M. H. S. Shyr, D. P. Wernette, P. Wiltzius, Y. Lu, P. V. Braun, J. Am.
Chem. Soc. 2008, 130, 8234–8240.
33] M. H. S. Shyr, D. P. Wernette, P. Wiltzius, Y. Lu, P. V. Braun, J. Am.
Chem. Soc. 2008, 130, 8234–8240.
34] V. V. Rostovtsev, L. G. Green, V. V. Fokin, K. B. Sharpless, Angew.
Chem. Int. Ed. 2002, 41, 2596–2599.
or from the author.
Acknowledgements
We are grateful to the University of Basel and the Karlsruhe Insti-
tute of Technology for ongoing support. Financial support by the
EU through the project FUNMOL (no. 213382 of the call FP7-NMP-
[
35] . YZ hou ,S.Wang , K.Zhang , X. J ia ng ,A ngew. Chem., Int. Ed. 2008,
2
007-SMALL-1), the Swiss National Science Foundation (SNF), the
4
7, 7454–7456.
Swiss Nanoscience Institute (SNI) and the National Research Pro-
gram 62 “smart materials” is gratefully acknowledged.
[
[
36] X. Xu, W. L. Daniel, W. Wei, C. A. Mirkin, Small 2010, 6, 623–626.
37] C. Hua, W. H. Zhang, A. De, S. Ciampi, D. Gloria, G. Liu,
J. B. Harper, J. J. Gooding, Analyst 2012, 137, 82–86.
[
[
[
[
38] Z. Lin, S. Gao, J. Lin, W. Lin, S. Qiu, L. Guo, B. Qiu, G. Chen, Anal.
Methods 2012, 4, 612–615.
39] M.-X. Zhang, B.-H. Huang, X.-Y. Sun, D.-W. Pang, Langmuir 2010,
[
[
1] A. Corma, H. Garcia, Chem. Soc. Rev. 2008, 37, 2096–2126.
2] C. Della Pina, E. Falletta, L. Prati, M. Rossi, Chem. Soc. Rev. 2008,
2
6, 10171–10176.
3
7, 2077–2095.
40] K. Zhu, Y. Zhang, S. He, W. Chen, J. Shen, Z. Wang, X. Jiang, Anal.
Chem. 2012, 84, 4267–4270.
41] A. Kimoto, K. Iwasaki, J. Abe, Photochem. Photobiol. Sci. 2010, 9,
[
[
3] X. Zhang, Q. Guo, D. Cui, Sensors 2009, 9, 1033–1053.
4] M. Homberger, U. Simon, Phil. Trans. R. Soc. A 2010, 368,
1
405–1453.
1
52–156.
[
5] G. Schmid, Chem. Soc. Rev. 2008, 37, 1909–1930.
1
0
www.small-journal.com
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2012,
DOI: 10.1002/smll.201300839

Organic–Inorganic Nanoarchitectures Based on “Clicked” Gold Nanoparticles
[
[
[
[
[
[
[
[
[
[
[
[
[
42] S. Mischler, S. Guerra, R. Deschenaux, Chem. Commun. 2012, 48, [55] J. P. Hermes, F. Sander, T. Peterle, R. Urbani, T. Pfohl,
2
183–2185.
43] M. Homberger, S. Schmid, J. Timper, U. Simon, J. Cluster Sci.
012, 23, 1049–1059.
44] W. Maneeprakorn, M. A. Malik, P. O’Brien, J. Am. Chem. Soc.
010, 132, 1780–1781.
D. Thompson, M. Mayor, Chem. Eur. J. 2011, 17, 13473–
13481.
[56] J. P. Hermes, F. Sander, U. Fluch, T. Peterle, D. Thompson,
R. Urbani, T. Pfohl, M. Mayor, J. Am. Chem. Soc. 2012, 134,
14674–14677.
2
2
45] E. Locatelli, G. Ori, M. Fournelle, R. Lemor, M. Montorsi, F. Comes, [57] S. G. Alvarez, M. T. Alvarez, Synthesis 1997, 1997,
Chem. Eur. J. 2011, 17, 9052–9056. 413–414.
46] M. Xie, L. Ding, Z. You, D. Gao, G. Yang, H. Han, J. Mater. Chem. [58] J. R. Thomas, X. Liu, P. J. Hergenrother, J. Am. Chem. Soc. 2005,
012, 22, 14108–14118. 127, 12434–12435.
47] T. Zhang, Z. Zheng, X. Ding, Y. Peng, Macromol. Rapid Commun. [59] Y. Song, E. K. Kohlmeir, T. J. Meade, J. Am. Chem. Soc. 2008, 130,
008, 29, 1716–1720. 6662–6663.
48] H. Gehan, L. Fillaud, M. M. Chehimi, J. Aubard, A. Hohenau, [60] H. E. Montenegro, P. Ramírez-López, M. C. de la Torre, M. Asenjo,
N. Felidj, C. Mangeney, ACS Nano 2010, 4, 6491–6500. M. A. Sierra, Chem. Eur. J. 2010, 16, 3798–3814.
49] H. Gehan, L. Fillaud, N. Felidj, J. Aubard, P. Lang, M. M. Chehimi, [61] M. Brust, M. Walker, D. Bethell, D. J. Schiffrin, R. Whyman, Chem.
C. Mangeney, Langmuir 2010, 26, 3975–3980. Commun. 1994, 801–802.
50] S. A. Krovi, D. Smith, S. T. Nguyen, Chem. Commun. 2010, 46, [62] M. D. Abramoff, P. J. Magalhães, S. J. Ram, Biophotonics Int.
277–5279. 2004, 11, 36–42.
51] M. Guerrouache, S. Mahouche-Chergui, M. M. Chehimi, [63] J. P. Hermes, F. Sander, T. Peterle, M. Mayor, Chimia 2011, 65,
B. Carbonnier, Chem. Commun. 2012, 48, 7486–7488. 219–222.
52] T. Peterle, A. Leifert, J. Timper, A. Sologubenko, U. Simon, [64] D. Thompson, J. P. Hermes, A. J. Quinn, M. Mayor, ACS Nano
2
2
5
M. Mayor, Chem. Commun. 2008, 3438–3440.
2012, 6, 3007–3017.
53] T. Peterle, P. Ringler, M. Mayor, Adv. Funct. Mater. 2009, 19,
3
497–3506.
Received: March 15, 2013
Revised: May 24, 2013
Published online:
54] J. P. Hermes, F. Sander, T. Peterle, C. Cioffi, P. Ringler, T. Pfohl,
M. Mayor, Small 2011, 7, 920–929.
small 2012,
DOI: 10.1002/smll.201300839
© 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com 11