Discotic liquid crystal-functionalized gold nanorods: 2- and 3D self-assembly and macroscopic alignment as well as increased charge carrier mobility in hexagonal columnar liquid crystal hosts affected by molecular packing and π-π interactions

Article abstract of DOI:10.1002/adfm.201401844

Gold nanorods functionalized with triphenylene-based discotic liquid crystal (LC) motifs show striking self-assembly behavior both on transmission electron microscopy (TEM) grids as well as in the bulk enforced by the π-π-stacking of triphenylene groups of adjacent nanorods. TEM images confirm that these discotic LC nanorods form ribbons of parallel-stacked nanorods several hundred nanometer long. The pursued silane conjugation approach to decorate the nanorods allows for the preparation of dispersions of the nanorods in the hexagonal columnar phases of parent discotic LCs, where the nanorods can be macroscopically aligned with almost 80% efficiency by a simple shearing protocol. Doping the parent host materials with about 1% by weight of the discotic LC-capped nanorods also reduces the lattice parameter and the intracolumnar packing, which gives rise to enhanced charge carrier mobility in these hosts as determined by time-of-flight measurements.

Full text of DOI:10.1002/adfm.201401844

www.afm-journal.de
www.MaterialsViews.com
Discotic Liquid Crystal-Functionalized Gold Nanorods:
2- and 3D Self-Assembly and Macroscopic Alignment as
well as Increased Charge Carrier Mobility in Hexagonal
Columnar Liquid Crystal Hosts Affected by Molecular
Packing and π–π Interactions
Xiang Feng, Lydia Sosa-Vargas, S. Umadevi, Taizo Mori, Yo Shimizu, and Torsten Hegmann*
significant interest over the past few
Gold nanorods functionalized with triphenylene-based discotic liquid crystal
years for the creation of metamaterials,[2]
(LC) motifs show striking self-assembly behavior both on transmission
for optical applications, such as high
efficiency polarizers[3] and display appli-
cations,[4] as well as for applications in
surface-enhanced Raman spectroscopy.[5]
However, the assembly and manipulation
of anisotropic building blocks into com-
plex superstructures is still a challenging
task in comparison to their corresponding
electron microscopy (TEM) grids as well as in the bulk enforced by the π–π-
stacking of triphenylene groups of adjacent nanorods. TEM images confirm
that these discotic LC nanorods form ribbons of parallel-stacked nanorods
several hundred nanometer long. The pursued silane conjugation approach
to decorate the nanorods allows for the preparation of dispersions of the
nanorods in the hexagonal columnar phases of parent discotic LCs, where
the nanorods can be macroscopically aligned with almost 80% efficiency by a
simple shearing protocol. Doping the parent host materials with about 1% by
weight of the discotic LC-capped nanorods also reduces the lattice parameter
and the intracolumnar packing, which gives rise to enhanced charge carrier
mobility in these hosts as determined by time-of-flight measurements.
quasi-spherical
counterpart
nano-
particles.[6] The key-limiting factor most
often described is the irreversible aggre-
gation of the anisometric nanomaterial
in a given fluid or polymer host medium.
It is well understood that the interactions
between nanostructures plays a critical
role during the self-assembly process.[7]
Ionic additives[8] or cationic surfactants[9] have been employed
to tune the attraction between nanoparticles in colloidal suspen-
sions and improve the ability of inducing nanorods assemblies.
In an earlier paper, we reported a fascinating large-area self-
assembly of nematic liquid crystal (LC)-functionalized gold
nanorods (GNRs) forming nematic- and smectic-like patterns
on substrates, and nematic-phase behavior in the bulk.[10] The
interaction between the LC molecules attached to the surface
of the GNRs was established as the driving force for the self-
assembly of such GNRs. The attached LC ligands tended to
pack together next to each other when the GNRs dispersion
became more and more concentrated during solvent evapora-
tion, and attractive interactions between the capping LC ligands
facilitated the formation of large-area nematic and smectic
superstructures of the GNRs. To extent this concept to higher
ordered LC phases that would facilitate long-range order, align-
ment, and even enhanced photoconductivity, we here present
the synthesis, characterization, as well as the surface and
bulk self-assembly of GNRs functionalized with discotic tri-
phenylene-based LC (DLC) molecules (Figure 1a). Strong π–π
interactions between the triphenylene cores should aid the self-
assembly as well as integration and alignment in structurally
identical triphenylene hosts. Yamada et al.[11] reported a 2D hex-
agonal superstructure of such DLC-functionalized gold nano-
particles on the substrate of transmission electron microscopy
1. Introduction
Approaches of manipulating anisotropic nanoparticles, in order
to arrange the particles into ordered assemblies,[1] generated
X. Feng, Dr. S. Umadevi,[+] Prof. T. Hegmann
Department of Chemistry
University of Manitoba
Winnipeg, Manitoba R3T 2N2, Canada
E-mail: thegmann@kent.edu
Dr. L. Sosa-Vargas, Prof. Y. Shimizu
Nanotechnology Research Institute
National Institute of Advanced Industrial
Science and Technology (AIST)
Ikeda, Osaka, Japan
Dr. T. Mori, Prof. T. Hegmann
Liquid Crystal Institute and Chemical Physics
Interdisciplinary Program
Kent State University
Kent, OH 44242, USA
Prof. T. Hegmann
Department of Chemistry & Biochemistry
Kent State University
Kent, OH 44242-0001, USA
[+]Present address: Department of Industrial Chemistry, Alagappa
University, Karaikudi-630 003, India
DOI: 10.1002/adfm.201401844
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201401844
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
1

www.afm-journal.de
www.MaterialsViews.com
alignment in the superlattices of the parent DLC host in planar
alignment cells and the effect of the GNR inclusions on the
charge carrier mobility of the DLC host.
2. Experimental Section
General Methods: All chemicals were obtained from com-
mercial sources (HAuCl4·3H2O, AgNO3, ascorbic acid, NaBH4,
cetyltrimethylammonium bromide (CTAB), B-bromocatecho-
lborane, Karstedt’s catalyst, octadecyltrimethoxysilane (ODS)
are all purchased from Sigma-Aldrich) and used without fur-
ther purification. High-purity deionized (DI) water (Millipore,
18 MΩ) and Aldrich purification grade solvents purified via a
PureSolv solvent purification system (Innovative Technology,
Inc.) were used for the synthesis and purification of the pre-
sented materials. All glassware used for synthesis and sample
handling was cleaned with aqua regia and thoroughly rinsed
with DI water. A highly pure sample of the LC H4TP was avail-
able in our laboratories.
UV–Vis–NIR absorption spectra were recorded using an
Agilent Cary 5000 UV–Vis–NIR spectrophotometer using
a quartz cell (1 cm path length). TEM images were recorded
on a Hitachi H 7000 microscope. The samples were prepared
by drop casting a 5 µL dispersion of the GNR solution onto
carbon-coated copper grids (400 meshes) and dried overnight
under ambient conditions. Polarized optical photoimages were
taken using an Olympus BX51-P polarizing optical micro-
scope equipped with a Linkam LS 350 heating/cooling stage.
Small-angle X-ray diffraction (SAXD) data were collected using
a Rigaku 3-pinhole camera (S-MAX 3000) equipped with a
Rigaku MicroMax+002 microfocus sealed tube (Cu Kα radia-
tion at λ = 1.54 Å) and Confocal Max-Flux optics operating at
40 W. The system has a 3 m, fully evacuated camera length,
and is equipped with 200 mm multiwire 2D detector for data
collection as well as an option to use image plates at various
distances away from the sample. The data reduction was per-
formed using Rigaku’s SAXGUI data processing software.
Mechanical shearing was used to obtain planar alignment
of the H6TP/GNRs composite in the Colho phase (85 °C). The
polarized UV–Vis–NIR spectrophotometry was carried at the
same temperature using the same Linkam heating stage in the
beam path of the spectrophotometer.
Figure 1. a) Structure and phase-transition temperatures of the DLC
silane precursor 3 and silane 4 as well as a 2D schematic representation
of the DLC-functionalized GNRs (DLC-GNR). b) UV–Vis–NIR spectra of
the DLC silane 4 and the DLC-GNRs.
(TEM) grids after solvent evaporation. The authors concluded
that the packing status of the functional LC ligands in the dis-
persion influenced the self-assembly of the nanoparticles on
TEM grids. Kumar and co-workers[12] presented data on such
DLC nanocomposites, which show that DLC-functionalized
gold nanoparticles[12a] and modified carbon nanotubes[12b]
enhance the electrical conductivity of the DLC hosts to a very
high level compared with the pure DLC. The authors also sug-
gested that the inclusion of shape-anisotropic nanotubes into
the DLC matrix leads to a homogeneous dispersion of the
nanotubes in the DLC superlattice, and predicted that align-
ment of 1D nanostructures can be achieved by employing well-
established LC-alignment technologies.
The charge carrier mobility measurements were carried out
using the time-of-flight (TOF) technique by injecting a sample
of the DLC-GNR-doped H4TP (1 and 2 wt%), DLC-GNR-doped
H6TP (1 wt%), and undoped, high-purity H4TP into ITO-
coated, sandwich-type cells (cell gap: 2–3 µm, measured by
interferometry). The H4TP used for the TOF experiments was
purified by successive recrystallizations in spectroscopic-grade
toluene and treatment with a chelating solution (aqueous eth-
ylenediaminetetraacetic acid) to remove ionic impurities. The
samples were injected by capillary action in the isotropic phase.
The TOF measurements were carried out on slow cooling from
the isotropic liquid to the crystalline phase. Carrier transport
was measured using the conventional TOF setup and a N2
laser (λ = 337 nm) as the excitation source, whilst recording
the photocurrents with an oscilloscope. Temperature varia-
tion was obtained with a hot-stage and temperature controller.
Herein, we describe for the first time GNRs functionalized
with DLC molecules via a silane conjugation approach,[13] the
self-assembly in the bulk and on TEM grids as well as their
©
2
wileyonlinelibrary.com
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201401844

www.afm-journal.de
www.MaterialsViews.com
Scheme 1. Synthesis route to triphenylene LC-decorated GNRs following a silane-conjugation approach.
Homeotropic alignment of the samples was monitored by
polarized optical microscopy and obtained by slow heating and
cooling from the isotropic liquid phase.[14]
catalyst (0.1 M in xylene) was added subsequently. The reaction
was stirred at room temperature for 24 h. The monosilane com-
pound 4 was purified by filtration through a 0.45 µm PTFE
syringe microfilter before silane conjugation to the (3-mercap-
topropyl)trimethoxysilane (MPS)-coated GNRs. Further details
can be found in the Supporting Information.
Synthesis: The monofunctionalized triphenylene derivatives
for the conjugation to the GNRs was synthesized according to
a previously published procedure,[15] and the hydrosilylation
reaction was following the same conditions described in our
earlier work.[10] The synthesis of the monoalkene-functional-
ized hexaalkoxytriphenylene (Scheme 1) was performed as fol-
lows:[15,16] A solution of H6TP, 1, (0.67 g, 0.5 × 10–3 mol) in
anhydrous CH2Cl2 (10 mL) was cooled to 0 °C before 1.2 eq.
(0.12 g, 0.6 × 10–3 mol) B-bromocatecholborane (dissolved in
1 mL CH2Cl2) was added dropwise. The mixture was kept stir-
ring at room temperature for 72 h. Thereafter, it was poured
on ice water (50 mL), and the water layer was extracted with
CH2Cl2 (two times 20 mL). The organic layer was separated
by column chromatography (hexanes/ethyl acetate = 30/1).
The purified monohydroxy triphenylene compound 2 (0.18 g,
0.24 × 10–3 mol) was dissolved in 10 mL THF, and KOH
(0.0132 g, 0.24 × 10–3 mol) was added under stirring. The
reaction was refluxed for 1 h before 11-bromo-1-undecene
(0.174 mL, 0.8 × 10–3 mol) was injected. The system was kept
under reflux for 24 h, and was then poured into 20 mL 1 M HCl,
and extracted with CH2Cl2 (twice 20 mL). The monoalkene
compound 3 (Cr 30.5 Col 57.5 Iso) was purified by recrystalliza-
tion from n-hexane. Compound 3 (0.09 g, 0.1 × 10–3 mol) was
then dissolved in 2 mL of dry toluene, and under stirring 64 µL
of trimethoxysilane (0.5 × 10–3 mol) and 20 µL of Karstedt’s
Next, the GNRs were synthesized following the modified
single-step growth method using CTAB as initial capping
agent, as described earlier.[10,17] Synthesis of CTAB-GNRs: a
250 mL water solution containing the following compounds
was prepared: [HAuCl4·3H2O] = 1.0 × 10–3 M, [CTAB] = 0.20 M,
[AgNO3] = 0.20 × 10–3 M, [ascorbic acid] = 2.0 × 10–3 M. Then
1.87 mL of 0.1 × 10–3 M ice-cold NaBH4 aqueous solution was
added into the mixture under stirring. After 2 min, the mixture
was kept for 24 h at 30 °C allowing the growth of the GNRs.
The resulting GNRs were washed by repeated agitation with
100 mL DI water and centrifugation at 16 000 rpm for 20 min.
The functionalization of the CTAB-GNRs was achieved by
a two-step silane hydrolysis–condensation reaction reported
previously:[10,17,18] 3 mL of as-synthesized CTAB-GNRs were
washed with 3 mL CHCl3, and then 120 × 10–6L of 3-mercapto-
propyltrimethoxysilane, MPS, (10 × 10–3 M in ethanol, calculated
to provide about 30 molecules nm–2) was added into the CTAB-
GNRs in water under vigorous stirring for 30 min. Thereafter,
3 mL of a CHCl3 solution (10 × 10–3 M) of the DLC silane 4 or
ODS was added, followed by 30 × 10–6 L of base (1.0 M NaOH).
The two-phase system was vortexed for several minutes until
the water phase became colorless and the organic phase turned
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201401844
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3

www.afm-journal.de
www.MaterialsViews.com
to red-brown, indicating the functionalization to LC-functional-
ized, hydrophobic GNRs. The organic layer was separated and
the GNRs was isolated by centrifugation (12 000 rpm, 20 min)
and washed with toluene.
3. Results and Discussion
Figure 1a shows the schematic representation of the as-syn-
thesized DLC-GNRs and Figure 1b shows the UV–Vis–NIR
absorbance spectrum of the DLC-GNRs in chloroform. The
absorption bands at 520 and 820 nm belong to the transversal
and longitudinal surface plasmon resonance (SPR) absorbances
of the DLC-GNRs, respectively, indicating that the aspect ratio
of the GNRs is about 3–4 on average.[19] Moreover, the absorb-
ance bands at 250–290 nm belong to the valence electron π–π*
transition of the triphenylene core of the DLC molecules. The
peak of this band shifted from 285 nm for the free DLC mol-
ecules (blue spectrum in Figure 1b) to 260 nm for the GNRs
dispersion (red spectrum in Figure 1b). This slight blueshift
can be explained by the local field enhancement near the sur-
face of the GNRs,[20] and was used as an evidence for the DLC
molecules attachment to the surface of the GNRs. The 2D-sim-
plified cartoon of the DLC-GNRs in Figure 1a also highlights
that during the silane conjugation a small amount of CTAB
surfactant is retained on the surface of the GNRs as established
in previous work using X-ray photoelectron spectroscopy.[10,17]
The self-assembly behavior of the neat DLC-GNRs on sub-
strates was first investigated by drop casting dispersions of the
GNRs in toluene on carbon-coated TEM copper grids after slow
evaporation of the solvent. Figure 2 shows some representa-
tive TEM images of the DLC-GNR samples dried from GNRs
dispersion in toluene (about 6 mg mL–1). Instead of an even,
yet random dispersion on the grid during the solvent evapora-
tion, the DLC-functionalized GNRs tend to gather together in a
side-by-side fashion to form what is best described as ribbons
of nanorods.
Figure 2. Transmission electron microscopy images of the DLC-GNRs
dried on TEM grids from dispersion in toluene.
also matches with the molecular length of the alkyl chains of
neighboring GNRs overlapping with one other. The key differ-
ence is that instead of the large-area assembly of DLC-GNRs
presented in Figures 2 and 3, the ODS-GNRs are showing only
short-range self-assembly (Figure S1, Supporting Information).
It is well known that the self-assembly of anisotropic colloidal
nanostructures is to a large extent determined by the type
and nature of the surface modification, in other words, a tai-
lored interaction between the rod-like nanoparticles. The DLC
molecules themselves, owing to their intrinsic packing in
columnarstacksformingahexagonalcolumnarphaseinacertain
The self-assembly potential of the DLC-GNR supported
by these images suggests that there are attractive interac-
tions between the GNRs driving the side-by-side long-range
assembly. In this case, the major driving forces are π–π interac-
tions between the triphenylene moieties of the DLC molecules
in addition to natural side-by-side assembly of anisometric rod-
like entities. As sown in Figure 3, these assemblies exist over
very large areas with several µm2 dimensions.
To obtain a monolayer of GNRs on the TEM grid that allows
us to measure the distance between neighboring GNRs, we
also tuned the concentration of the GNRs dispersion. The
image in Figure 4a and the corresponding cross section pro-
file (Figure 4b) shows that the gap between DLC-GNRs is
about 7 nm on average (peak widths were measured as the
full width at half maximum), which matches perfectly with
the length of the stacking molecules from neighboring GNRs
with the triphenylene core overlapping each other (see inset,
Figure 4a). For comparison, ODS-coated GNRs were also syn-
thesized, and a representative TEM image of the ODS-GNRs
drop cast and dried from toluene is shown in Figure 4c. From
the corresponding cross sectional profile (Figure 4d), the gap
between the ODS-GNRs was measured to be 3–4 nm, which
Figure 3. TEM image showing the large-area self-assembly of the DLC-GNRs.
Several GNR ribbons extend over several hundred nanometers.
©
4
wileyonlinelibrary.com
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201401844

www.afm-journal.de
www.MaterialsViews.com
Figure 4. a) TEM image of the DLC-GNRs, b) cross section profile of (a), c) TEM image of the ODS-GNRs, d) cross section profile of (c). Insets show
the arrangement of the GNRs in each case.
temperature interval due to strong π–π stacking, drive here the
self-assembly of the GNRs.
from the TEM image shown in Figure 2b using Image J®).
Considering the average diameter of the GNRs (11 1 nm), the
average gap between DLC-GNRs is approximately 7 nm, which
agrees perfectly with the number measured by analyzing TEM
images, as well as the bimolecular length of the DLCs with
overlapping triphenylene cores.
While the large-area self-assembly of the DLC-GNRs on TEM
grids is static, the essential π–π interactions should also facili-
tate alignment of the GNRs in a parent triphenylene host. To
study this, we prepared mixtures of the DLC-GNRs in two triph-
enylene-based hexagonal columnar LC hosts, H6TP forming a
Colho (ordered hexagonal columnar) and H4TP forming a Colhp
(hexagonal columnar plastic) phase, with the concentrations of
the DLC-GNRs at 1 or 2 wt% (much higher
To further demonstrate the π–π stacking assembly model
inside the multiple layers of the DLC-GNRs, a highly con-
densed dispersion of these GNRs was prepared and drop casted
on a Kapton® substrate and dried under ambient atmosphere
before performing small-angle X-ray scattering (SAXS) experi-
ments. Figure 5a shows the 2D scattering pattern from the
X-ray detector. The plot of intensity versus scattering vector q
(Figure 5b) shows a broad peak at q = 0.035, corresponding to
a d spacing of about 18 nm (d = 2π/q, in Å), which corresponds
to the fast Fourier transform obtained from the TEM images
(for example fast Fourier transform (FFT) in Figure 5b obtained
than reported for many other nanorods in
LCs).[21] This concentration will determine
the numerical ratio between the H6TP
(or H4TP) molecules and the DLC-GNRs.
Assuming approximately 7 × 106 free H6TP
molecules per GNR (at 1 wt% and 10 mg
H6TP) in a given mixture, the ratio between
free H6TP molecules and anchored discotic
mesogens on the GNR surface is about
1160:1, considering there are approximately
6000 DLC silane molecules attached to the
surface of each DLC-GNR (with 10% CTAB)
following calculations published by Gelbart
and co-workers[22] The DLC-GNRs easily
disperse in the parent DLC hosts (H6TP)
without showing any discernable aggregation
under the polarized light optical microscope
(even at 100× magnification). The thermal
behavior observed by differential scanning
calorimetry (DSC) measurements (Sup-
porting Information Figure S2 for H6TP)
Figure 5. a) 2D SAXS pattern, b) Fourier transform of the TEM image shown in Figure 2b, and
c) azimuthally averaged intensity data of the scattering vector (q in Å−1) from (a).
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201401844
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
5

www.afm-journal.de
www.MaterialsViews.com
macroscopically align the DLC-GNRs, a crit-
ical aspect for the use of anisometric metal
nanorods in potential applications as optical-
phase retarders or wave plates. To study
this in a rather simple approach, the 1 wt%
DLC-GNRs-doped H6TP sample was placed
between two plain glass slides, and planar
alignment of the Colho phase of H6TP was
achieved by shearing the cover slide in one
direction at a temperature of 85 °C. POM
images of the alignment obtained for this
cell are shown in Figure 8. Under crossed
Figure 6. POM photomicrographs (crossed polarizers) taken on cooling at 1 °C min–1 of: a)
neat H6TP at 99 °C (left) and 88 °C (right), b) H6TP doped with 1 wt% DLC-GNRs at 98 °C (left)
and 85 °C (right), and c) H6TP doped with 2 wt% DLC-GNRs at 96 °C (left) and 85 °C (right).
shows that the LC-phase transition temperatures are practically
identical for the pure hosts and the 1 and 2 wt%-doped samples
during both heating and cooling runs (Supporting Information
Table S1 for H4TP).
polarizers, when the shearing direction was parallel to the polar-
izer (director n of the aligned H6TP molecules, yellow arrow),
the aligned cells gave the weakest birefringence (Figure 8a).
The birefringence reached the maximum when the shearing
direction was 45° with respect to both polarizers (Figure 8b)
indicating the planar alignment of H6TP in these nontreated
cells. In order to investigate the orientation (alignment) of the
OCnH2n+1
OCnH2n+1
H2n+1CnO
H2n+1CnO
OCnH2n+1
OCnH2n+1
n=4 H4TP; Cr 85.6 Colhp 144 Iso
n=6 H6TP; Cr 67.6 Colho 99.0 Iso
At first, temperature-controlled polarized
light optical microscopy (POM) was per-
formed, as this technique always provides the
first clues if a particular nanomaterial tends
to aggregate in a LC-host phase. For the given
DLC-GNRs no such tendency was observed
and aggregates that may eventually form at the
higher concentration are not visible at mag-
nifications up to 100×. For H6TP, the texture
observed by POM for neat H6TP appear virtu-
ally unaltered after the addition and dispersion
of the DLC-GNRs. Figure 6a–c shows the typ-
ical dendritic (right at the transition from Iso
to Colho) and mosaic-type textures (deeper into
the Colho phase) textures for neat H6TP and
H6TP doped with 1 and 2 wt% DLC-GNRs.
Similar overall trends were also observed
for H4TP during POM analysis as shown in
Figure 7. For H4TP, however, highly bire-
fringent (and rather colorful) textures were
observed on cooling in a very narrow tem-
perature interval (≈1 °C–2 °C) immediately
at the transition from the isotropic liquid
to the Colhp phase. Similar textural features
have previously been described for tilted,
supramolecular chiral (or racemic) columnar
phases, and may provide the first hint for a
potentially induced tilt after inclusion of the
GNRs into the hexagonal lattice of the two
Figure 7. POM photoimages taken on cooling at 1 °C min–1 (left: uncrossed polarizers; right:
crossed polarizers): a) H4TP doped with 1 wt% DLC-GNRs and b) H4TP doped with 2 wt%
DLC-GNRs at various temperatures. Within a very narrow temperature interval on cooling,
highly birefringent and colorful textures are seen immediately below the isotropic liquid phase.
hosts.
First, however, we wanted to test if the
more fluid Colho of H6TP could be used to
©
6
wileyonlinelibrary.com
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201401844

www.afm-journal.de
www.MaterialsViews.com
A|| − A⊥
A|| + 2A⊥
S =
where Aʈ and A⊥ is the absorbance measured with the polarizer
either parallel or perpendicular to the director n, respectively.
The resulting order parameter S (dichroic ratio) was calculated
to be 0.38. Considering that the typical order parameter of
planar alignment of columnar phase of discotic LCs achieved
by mechanical shearing[24] is in the vicinity of 0.5, the DLC-
GNRs inserted into the planar alignment columnar phase are
aligned along the director n with an over 75 % efficiency.
With POM and UV–Vis–NIR experiments suggesting that
the DLC-GNRs are reasonably well dispersed in the hexagonal
columnar LC phases of both H4TP and H6TP, we sought to
determine if the suspended GNRs have any effect on the organ-
ization and structure of the host phase by performing medium
and wide-angle X-ray diffraction experiments. While we have
not been able to pinpoint and carefully study the potential high-
temperature modification exhibiting the bright, highly birefrin-
gent textures for the 1 and 2 wt% DLC-GNR in H4TP mixtures
(too narrow temperature interval), the obtained X-ray diffraction
data did reveal some rather unique trends when both hosts are
doped with the DLC-GNRs. In both cases, we selected the same
temperature well below the Iso/Colh phase-transition temper-
ature (since DSC and POM revealed no thermal effect of the
addition of the DLC-GNRs in the two hosts) on cooling (80 °C
for H6TP and 125 °C for H4TP) and collected X-ray diffraction
data, which are shown in Figure 9 and summarized in Table 1.
First, the diffraction pattern of both pure H6TP and H4TP
perfectly match data published previously.[25] Second, the
packing of H4TP appears to be more significantly impacted by
the dispersion of the DLC-GNRs than H6TP at least judging
from the plot of the scattering intensity versus q, with some of
the (311)/(320) diffraction peaks disappearing and a completely
altered intensity profile. The calculated data for the d spacing
and the lattice parameter ahex (from d100), however, indicate that
there is only a minor decrease in ahex for H4TP, but a signifi-
cant decrease for H6TP both doped with 1 wt% DLC-GNRs, but
not for the 2 wt%-doped sample. While one would expect an
increase in d spacing (and lattice parameter)[26] upon insertion
of GNRs with a much larger diameter, the lattice parameter for
the 1 wt% DLC-GNRs in H6TP sample decreases by about 8%,
which at this point could only be explained by a more efficient
interdigitation of the alkyl chains surrounding the triphenylene
cores of H6TP. Here, the insertion of the GNRs might lead
to packing mismatch between DLC molecules attached to the
GNR surface and DLC molecules of the host. At the volume
fraction of 1 wt% of DLC-GNRs in H6TP, such packing mis-
match could lead to packing frustration that might be com-
pensated by a larger degree of interdigitation. Such process,
however, would be more pronounced with H6TP than H4TP
because of the longer hydrocarbon chains (see schematics in
Figures 10 and 11). Such interdigitation would be less signifi-
cant for H4TP because of the shorter chain length, which is
supported by the less significant decrease in d (and ahex) for
H4TP doped with 1 wt% DLC-GNR (see Table 1). For H4TP this
packing frustration reaches its maximum at 2 wt% DLC-GNRs
indicated by a decrease of ahex to 1.93 nm (a 6% decrease).
Figure 8. a,b) Polarized optical images of the planar alignment cell of
H6TP doped with 1 wt% of DLC-GNRs, c) polarized UV–Vis–NIR spectra
of the DLC-GNRs doped in H6TP (planar-aligned cell) at different polar-
izer rotation angles.
DLC-GNRs doped in the superstructure of H6TP, the planar-
aligned cell was studied using linearly polarized UV–Vis–NIR
spectrophotometry. From the spectra of the polarized absorp-
tion experiments (Figure 8c), the absorbance band at around
800 nm corresponding to the longitudinal SPR of the GNRs
is tunable by rotating the planar-aligned thin Colho fi lm. The
intensity of this band decreased when the angle between polar-
izer and the director n is rotated from 0° to 90° and reaches the
minimum intensity at 90°, while the polarization of the beam
is perpendicular to the transversal direction of the DLC-GNRs.
This is in principle also true for the transversal SPR band cen-
tered around 535 nm (zero absorbance at 0° and 10°, maximum
absorbance at 90°; 45° making the logical exception), which
indicates that the majority of the DLC-GNRs in the aligned cell
follows the director n with their long axes a shown in the insert
of Figure 8c.
The polarized UV–Vis–NIR spectrum at 90° shows as
expected the weakest absorbance at 800 nm, but not zero, which
means that the alignment of H6TP is not perfect (as can actu-
ally be seen in Figure 8a), but potentially also that a portion of
the GNRs is not perfectly aligned along the director. As one can
see from Figure 8a, some areas still have weak birefringence,
and appear as bright domains. The order parameter S of the
DLC-GNRs can be calculated using the following equation[23]
:
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201401844
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
7

www.afm-journal.de
www.MaterialsViews.com
Figure 9. X-ray diffraction pattern and azimuthally averaged intensity data of the scattering vector (q in Å−1) from the 2D pattern shown as insets.
The intracolumnar spacing however remains virtually iden-
tical. For H6TP, both the intracolumnar distance between the
molten alkyl chains and the core–core distance simultaneously
decreases, most considerably for the 1wt%-doped H6TP, with
a core–core distance decrease from 0.379 to 0.367 nm. How-
ever, once the concentration of the DLC-GNRs is doubled to
2 wt%, most likely an onset of aggregation of the GNRs mini-
mizes these packing effects and the lattice relaxes back to the
values of the nondoped H6TP host phase. Another possibility
to explain a decrease of both ahex and core–core distance would
be a molecular tilt of the molecules within the columns (23° to
be exact), but this can be ruled out as the resulting phase would
have a different symmetry. Such molecular tilt would give
rise to an elliptical cross section of the columns leading to an
assignment as a rectangular columnar phase (Colr). In neither
case do the powder XRD pattern support such phase symmetry
and assignment. However, the insertion of the GNRs, at least
at lower volume fractions induces a tighter packing of the host
molecules both inter- and intracolumnar at 1 wt% for H6TP,
which should give rise to a higher charge carrier mobility in
these phases, which we tested next using transient photocur-
rent measurements using the TOF technique.[14] For H4TP
exclusively the intercolumnar packing decrease at 2 wt% DLC-
GNRs, which should in principle not affect the charge carrier
mobility unless other electronic processes or ionic contribu-
tions are active.
cells (cell gap: 2–3 µm), and the most reliable data sets were
acquired for the ultrapure H4TP synthesized and purified pre-
viously by Shimizu and co-workers.[25] The use of thin cells
for TOF measurements shows a direct effect on the align-
ment of the DLC-GNR composites. While we were unable to
obtain complete homeotropic alignment in cells with a ≈15 µm
cell gap, it occurred spontaneously for the 1 wt% DLC-GNR/
H4TP composite in the ≈3 µm thin cell, and remained present
throughout the whole temperature range analyzed on cooling
(from 141 to 20 °C). The 2 wt% composite also shows homeo-
tropic regions albeit significantly smaller (see Figure S3, Sup-
porting Information). The presence of homeotropic regions in
the DLC-GNRs composites contributes to the enhancement of
the hole mobility as it reflects an increased uniaxial ordering
along the columnar axis that is absent in the pure H4TP
sample.
Figures 12 and 13 show the obtained transient photocur-
rent decay curves for both hole and electron transports for
H4TP doped with 1 and 2 wt% DLC-GNRs and H6TP doped
with 1 wt% DLC-GNRs (hole mobility only). H4TP and H6TP
are known to show ambipolar character for the charge trans-
port in electronic processes,[27] and in comparison to those of
the pure, undoped compounds, the GNR-doped systems tend
to show noisy and less qualified decay curves, although an elec-
tric field-independent character of the mobility is still observed
(see Figure S4, Supporting Information). This may indicate
that local disordering of the discotic molecules is induced by
the addition of the surface modified DLC-GNRs. H4TP and
These measurements require extremely pure samples and
ideally excellent homeotropic alignment in thin, ITO-coated
©
8
wileyonlinelibrary.com
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201401844

www.afm-journal.de
www.MaterialsViews.com
Table 1. Measured scattering vectors (q in Å−1) with respective Miller indices, and calculated d spacing as well as lattice parameters of the Colh phase
in nm. The intracolumnar distance refers to the distance between molten hydrocarbon chains, the core–core distance to the distance between the
triphenylene cores within the columnar aggregates.
Stacking periodicities
[nm]
(hkl) [Å−1
]
Sample
H6TP
T
[°C]
d
[nm]
ahex
[nm]
Intracolumnar
Core–Core
q
80
1.953
2.252
0.495
0.379
q(100) = 0.3218
q(110) = 0.5447
q(200) = 0.656
q = 1.27
q = 1.66
80
80
1.803
1.945
1.775
2.082
2.246
2.05
0.477
0.491
0.479
0.367
0.379
0.358
H6TP + 1 wt%
DLC-GNRs
q(100) = 0.3485
q(110) = 0.614
q = 1.316
q = 1.714
H6TP + 2 wt%
DLC-GNRs
q(100) = 0.3231
q(110) = 0.608
q = 1.28
q = 1.658
H4TP
125
q(100) = 0.3539
q(110) = 0.645
q(200) = 0.7148
q(210) = 0.992
q(220)(211) = 1.311
q(311)(320) = 1.728
q(002) = 1.7557
q(100) = 0.3673
125
1.711
1.975
0.475
0.358
H4TP + 1 wt%
DLC-GNRs
q(110) = 0.6307
q(200) = 0.7415
q(220)(211) = 1.324
q(002) = 1.7534
q(100) = 0.3759
125
1.672
1.93
0.476
0.363
H4TP + 2 wt%
DLC-GNRs
q(110) = 0.6629
q(200) = 0.7176
q(210) = 0.95
q(220)(211) = 1.319
q(311)(320) = 1.729
q(002) = 1.7564
H6TP both exhibit hexagonal columnar (Colh) phases, however
with a slight difference in the molecular ordering. The shorter
chain homologue, H4TP shows a 3D-plastic columnar phase
with a 2D hexagonal arrangement of columns (Colhp phase),[28]
whilst H6TP displays a Colho mesophase with some degree of
ordering along the columnar axis (see X-ray diffraction data in
Figure 9). Their hole and electron mobilities are recorded on
the order of 10−2 and 10−4 cm2 V−1 s−1 for the Colhp and the
Colho phase, respectively, and both show largely tempera-
ture-independent character.[25] The temperature dependencies
of the charge carrier mobility for the two composites of H4TP
are depicted in Figure 14. All essentially show a temperature-
independent nature for both holes and electrons except for a
sudden jump at about 115 °C for the electron mobility. There
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201401844
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
9

www.afm-journal.de
www.MaterialsViews.com
and provide the maximum mobility. For the
1wt% DLC-GNR composite with H6TP, the
mobility behavior is similar to those of the
composites with H4TP. Here, the measured
hole mobility measured for the composite is
slightly higher than the one of pure H6TP
(≈3.2 × 10−4 cm2 V−1 s−1 for the com-
posite vs 2.0
× f
10−4 cm2 V−1 s−1
or neat H6TP over the Colho mesophase
range.[25] The carrier mobilities obtained
for undoped H4TP as well as the compos-
ites with DLC-GNR are summarized in
Table 2, and the data for H6TP are collected
in the Supporting Information (POM and
hole mobility, see Figure S5, Supporting
Information).
Kumar et al. reported on an enhancement
of the conductivity with the inclusion of GNRs
(1 wt% for H5TP) from 10−9 to 10−6 Sm−1,
which was even higher than in the isotropic
liquid phase (10−7 Sm−1).[26,29] A related
enhancement of the conductivity was also
reported for composites with gold nano-
particles dispersed in H6TP.[30] The conduc-
tivity is a function of the number of efficient
charges in the system and their mobility.
Enhancement of the conductivity is thus
the result of an increased number of car-
riers mobile in the system and/or improved
Figure 10. Schematic of the Colho phase of H6TP and H6TP doped with 1 wt% DLC-GNRs
showing the decrease in lattice parameter as well as intracolumnar packing.
is no evidence of a phase transition, although an optical texture
change indicates a potential phase or structure modification
within these domains around this temperature (see Figure 14b).
This change in the optical textures is best described by a slight
widening of the domain boundaries (Figure 7), although
DSC measurements show no first order phase transition in
this temperature range. Interestingly, this sudden change in
mobility is not seen for the hole mobility, which drops for the
2 wt% mixture, likely due to an onset of aggregation of the
DLC-GNRs in H4TP, but otherwise increases below the Cr/
Colhp phase transition (rather than decrease for neat H4TP).
Overall, the mobilities of both holes and electrons are compa-
rable to those of pure H4TP in the LC phase, and the mixture
of 1 wt% DLC-GNRs in H4TP shows a slightly higher value
of the mobility. This may indicate that the optimal concentra-
tion of the DLC-GNRs would exist around this concentration
mobility of the charge carriers. Hence, in both composites of
1 and 2 wt% DLC-GNRs in H4TP, the enhanced conductivity
reported is best explained by an increase in the number of effi-
cient charges. If we assume that the charge conduction in this
system is only arising from electronic processes, the doping
of GNRs would have to elevate the number of charges. On the
other hand, conductivity phenomena essentially include ionic
transport processes. In particular, soft states of matter such as
LC phases sometimes provide a good ionic charge transport
path.[31] Considering that the mobilities measured using the
TOF technique are rather high in comparison to those of ionic
mobility, and that the temperature dependence obeys Arrhe-
nius formalism, the enhanced conductivity is probably derived
from charge transport in ionic processes.
4. Conclusions
In continuation of our earlier experiments on the reconfigur-
able self-assembly of nematic LC-capped GNRs,[10,17] we here
demonstrate that GNRs functionalized with DLC π−-π stacking
motifs also foster long-range, bulk self-assembly and permit
alignment in a parent DLC host. The as-synthesized DLC-
GNRs can be homogeneously dispersed in organic solvents,
such as toluene and chloroform, at fairly high concentrations.
These dispersions, dried on TEM grids, flat substrates, or in
the bulk show large-area self-assembly of the as observed by
TEM and SAXS. However, true liquid crystalline behavior
could not be observed. The proposed stacking assembly model
is supported by distance measurements using TEM image
Figure 11. Schematic of the H6TP (or H4TP) doped with 1 wt%
DLC-GNRs (3D and top 2D view) showing potential packing inefficiencies
around a single-DLC-capped GNR that are potentially compensated as
determined by XRD.
©
10 wileyonlinelibrary.com
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201401844

www.afm-journal.de
www.MaterialsViews.com
Figure 12. Bias dependence of the hole mobility of: a) pure H4TP, b) 1 wt% DLC-GNRs in H4TP, c) 2 wt% DLC-GNRs in H4TP (all three at 140 °C in
the Colhp phase), and d) 1 wt% DLC-GNRs in H6TP (at 90 °C in the Colho phase).
Figure 13. Bias dependence of the electron mobility of: a) pure H4TP, b) 1 wt% DLC-GNRs in H4TP, c) 2 wt% DLC-GNRs in H4TP (at 140 °C in the
Colhp phase).
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201401844
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
11

www.afm-journal.de
www.MaterialsViews.com
Figure 14. Temperature dependence of hole and electron mobilities for: a) 1 wt% and b) 2 wt% DLC-GNR in H4TP at 10 kV cm−1.
analysis and SAXS showing identical distances between side-
by-side self-assembled GNRs based on π–π molecular stacks
between neighboring GNRs. In combination with our earlier
data regarding the distances between nematic LC-function-
alized GNRs,[10] the interparticle distance, i.e., the coupling
effect,[32] can be fine tuned by the LC-packing mode, which
shows promise for potential application as metamaterials or
field-enhanced spectroscopy units in electric and optical nan-
odevices.[11,33] Furthermore, we have demonstrated that the
DLC-functionalized GNRs are compatible with the parent
DLC hosts. The enhanced thermal stability provided by the
silane conjugation approach allows us to disperse these DLC-
capped GNRs without the risk of thiol desorption from the
GNR surface eliminating the risk of free thiols contributing
both positively or negatively to the observed effects. Homo-
geneous dispersion and uniform alignment of the DLC-func-
tionalized GNRs have been achieved in a planar-aligned H6TP
DLC host over areas as large as several cm2, employing a very
simple and well-established alignment technique for DLCs
without any applied electric or magnetic field. These compos-
ites show rather unique structural features particularly at the
lower concentration of 1% by weight. SAXD data reveal that
the lattice parameter decreases from 2.25 to 2.08 nm for the
host featuring longer hydrocarbon chains (H6TP) at 1 wt% of
dispersed GNRs and relaxes back to 2.25 nm when the con-
centration of the GNRs is doubled, likely indicating the onset
of some GNR aggregation. This effect is more pronounced for
the less viscous H6TP forming the Colho phase in compar-
ison to the more viscous H4TP forming a plastic Colhp phase.
Simultaneously, the intracolumnar distances also decrease at
the lower concentration of GNRs especially in H6TP, which
gives rise to an increase in the charge carrier mobility for both
electrons and holes. These results indicate that the inclusion
of the DLC-GNRs into the hexagonal columnar LC affects the
charge transport efficiency only slightly at the lower concentra-
tion range of dispersed GNRs. However, the interdigitation of
columnar semiconductors and functionalized metal nanorods
as conducting active parts and electrodes, respectively, may
provide a simple yet effective strategy for organic photovoltaics.
Table 2. Carrier mobility for H4TP and H4TP doped with 1 and 2 wt%
DLC-GNRs.
Charge carrier mobility [10−2 cm2 Vs–1
Hole Electron
Colhp
]
Sample
Colhp
1.7
Cr
Cr
neat H4TP
1.4
6.7
3.1
2.0
5.2
2.4
4.9
8.2
4.1
H4TP (1 wt% DLC-GNRs)
H4TP (2 wt% DLC-GNRs)
3.9
0.7
©
12 wileyonlinelibrary.com
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201401844

www.afm-journal.de
www.MaterialsViews.com
[7] a) Y. J. Min, M. Akbulut, K. Kristiansen, Y. Golan, J. Israelachvili,
Nat. Mater. 2008, 7, 527; b) K. J. M. Bishop, C. E. Wilmer, S. Soh,
B. A. Grzybowski, Small 2009, 5, 1600.
[8] T. S. Sreeprasad, T. Pradeep, Langmuir 2011, 27, 3381.
[9] A. Guerrero-Martinez, J. Perez-Juste, E. Carbo-Argibay, G. Tardajos,
L. M. Liz-Marzan, Angew. Chem. Int. Ed. 2009, 48, 9484.
[10] S. Umadevi, X. Feng, T. Hegmann, Adv. Funct. Mater. 2013, 23,
1393.
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
Acknowledgements
The authors would like to thank the Province of Manitoba for a grant
from the Science and Technology International Collaboration Fund, the
Centre for Chemical Innovation (CCI) (Caltech), the Natural Science
and Engineering Research Council (NSERC) of Canada, the Canada
Foundation for Innovation (CFI), and the Manitoba Research and
Innovation Fund (MRIF) for financial support. TH also acknowledges
financial support from Kent State University and the Government
of Ohio's Third Frontier Program for Ohio Research Scholars. T.M. is
grateful for financial support through a JSPS fellowship. L.S. and Y.S.
acknowledge JST for the finical support under the ALCA program. Special
thanks go to Dr. Kevin McEleney for his help with the SAXS experiments
and to Andre Dufresne for all the help and assistance with TEM imaging.
[11] M. Yamada, Z. R. Shen, M. Miyake, Chem. Commun. 2006, 2569.
[12] a) S. Kumar, S. K. Pal, P. S. Kumar, V. Lakshminarayanan, Soft
Matter 2007, 3, 896; b) S. Kumar, H. K. Bisoyi, Angew. Chem. Int.
Ed. 2007, 46, 1501.
[13] J. Mirzaei, M. Urbanski, H. S. Kitzerow, T. Hegmann, Philos. Trans.
R. Soc. A 2013, 371, 20120256.
[14] F. Nekelson, H. Monobe, M. Shiro, Y. Shimizu, J. Mater. Chem.
2007, 17, 2607.
[15] S. Kumar, M. Manickam, Synthesis-Stuttgart 1998, 1119.
[16] a) J. L. Schulte, S. Laschat, V. Vill, E. Nishikawa, H. Finkelmann,
M. Nimtz, Eur. J. Org. Chem. 1998, 11, 2499; b) H. Schonherr,
F. J. B. Kremer, S. Kumar, J. A. Rego, H. Wolf, H. Ringsdorf,
M. Jaschke, H. J. Butt, E. Bamberg, J. Am. Chem. Soc. 1996, 118,
13051.
[17] S. Umadevi, X. Feng, T. Hegmann, Ferroelectrics 2012, 431, 164.
[18] K. Mitamura, T. Imae, N. Saito, O. Takai, J. Phys. Chem. B 2007, 111,
8891.
Received: June 5, 2014
Revised: November 17, 2014
Published online:
[19] B. Nikoobakht, M. A. El-Sayed, Chem. Mater. 2003, 15, 1957.
[20] S. L. Pan, Z. J. Wang, L. J. Rothberg, J. Phys. Chem. B 2006, 110,
17383.
[21] M. R. Thomas, S. Klein, R. J. Greasty, S. Mann, A. W. Perriman,
R. M. Richardson, Adv. Mater. 2012, 24, 4424.
[22] D. V. Leff, P. C. Ohara, J. R. Heath, W. M. Gelbart, J. Phy.s Chem.
1995, 99, 7036.
[23] B. Bahadur, in Handbook of Liquid Crystals1st Ed., (Eds:
J. W. Goodby, D. Demus, G. Gray, H.-W. Spiess, V. Vill) , Vol. no. 2A,
Wiley-VCH, Weinheim, Germany 1998.
[1] a) T. Hegmann, H. Qi, V. M. Marx, J. Inorg. Organomet. Polym.
Mater. 2007, 17, 483; b) U. Shivakumar, J. Mirzaei, X. Feng,
A. Sharma, P. Moreira, T. Hegmann, Liq. Cryst. 2011, 38, 1495;
c) O. Stamatoiu, J. Mirzaei, X. Feng, T. Hegmann, Top. Curr.
Chem. 2012, 318, 331; d) G. L. Nealon, R. Greget, C. Dominguez,
Z. T. Nagy, D. Guillon, J. L. Gallani, B. Donnio, Beilstein J. Org.
Chem. 2012, 8, 349–370; e) S. Umadevi, V. Ganesh, T. Hegmann,
Handbook of Liquid Crystals2nd Ed., (Eds: J. W. Goodby,
P. J. Collings, T. Kato, C. Tschoerske, H. F. Gleeson, P. Raynes) Vol.
no. 6, Wiley-VCH, Weinheim, Germany 2014.
[24] Q. K. Liu, Y. X. Cui, D. Gardner, X. Li, S. L. He, I. I. Smalyukh, Nano
Lett. 2010, 10, 1347.
[25] H. Monobe, Y. Shimizu, S. Okamoto, H. Enomoto, Mol. Cryst. Liq.
Cryst. 2007, 476, 277.
[26] S. Kumar, NPG Asia Mater. 2014, 6, e82.
[27] H. Iino, J. Hanna, D. Haarer, R. J. Bushby, Jpn. J. Appl. Phys. 2006,
45, 430.
[28] J. Simmerer, B. Glusen, W. Paulus, A. Kettner, P. Schuhmacher,
D. Adam, K. H. Etzbach, K. Siemensmeyer, J. H. Wendorff,
H. Ringsdorf, D. Haarer, Adv. Mater. 1996, 8, 815.
[29] B. S. Avinash, V. Lakshminarayanan, S. Kumar, J. K. Vij, Chem.
Commun. 2013, 49, 978.
[2] a) M. Grzelczak, L. M. Liz-Marzan, Langmuir 2013, 29, 4652;
b) M. Zorn, R. Zentel, Macromol. Rapid. Commun. 2008, 29, 922;
c) C. M. Soukoulis, S. Linden, M. Wegener, Science 2007, 315, 47;
d) V. M. Shalaev, Nat. Photonics 2007, 1, 41; e) S. Meuer, P. Oberle,
P. Theato, W. Tremel, R. Zentel, Adv. Mater. 2007, 19, 2073;
f) S. Meuer, L. Braun, R. Zentel, Chem. Commun. 2008, 3166;
g) A. B. Golovin, O. D. Lavrentovich, Appl. Phys. Lett. 2009,
95, 254104; h) A. B. Golovin, J. Xiang, H. S. Park, L. Tortora,
Y. A. Nastishin, S. V. Shiyanovskii, O. D. Lavrentovich, Materials
2011, 4, 390.
[30] L. A. Holt, R. J. Bushby, S. D. Evans, A. Burgess, G. Seeley, J. Appl.
Phys. 2008, 103, 063712.
[31] a) K. Kishimoto, T. Suzawa, T. Yokota, T. Mukai, H. Ohno, T. Kato, J.
Am. Chem. Soc. 2005, 127, 15618; b) M. Yoshio, T. Mukai, H. Ohno,
T. Kato, J. Am. Chem. Soc. 2004, 126, 994.
[3] a) J. J. Zhou, G. X. Chen, E. Wu, G. Bi, B. T. Wu, Y. Teng, S. F. Zhou,
J. R. Qiu, Nano Lett. 2013, 13, 2241; b) T. Wang, X. R. Wang,
D. LaMontagne, Z. W. Wang, Y. C. Cao, J. Am. Chem. Soc. 2013,
135, 6022; c) Y. Zhao, A. Alu, Nano Lett. 2013, 13, 1086; d) X. P. Li,
T. H. Lan, C. H. Tien, M. Gu, Nat. Commun. 2012, 3, 1.
[32] K. H. Su, Q. H. Wei, X. Zhang, J. J. Mock, D. R. Smith, S. Schultz,
Nano Lett. 2003, 3, 1087.
[4] X. Y. Zheng, J. Fontana, M. Pevnyi, M. Ignatenko, S. Wang, R. Vaia,
P. Palffy-Muhoray, J. Mater. Sci. 2012, 47, 4914.
[33] a) J. Aizpurua, G. W. Bryant, L. J. Richter, F. J. G. de Abajo,
[5] Y. Q. Wang, B. Yan, L. X. Chen, Chem. Rev. 2013, 113, 1391.
[6] a) E. V. Shevchenko, D. V. Talapin, N. A. Kotov, S. O’Brien,
C. B. Murray, Nature 2006, 439, 55; b) S. C. Glotzer, M. J. Solomon,
Nat. Mater. 2007, 6, 557.
B. K. Kelley, T. Mallouk, Phys. Rev.
B 2005, 71, 235420;
b) V. M. Shalaev, W. S. Cai, U. K. Chettiar, H. K. Yuan, A. K. Sarychev,
V. P. Drachev, A. V. Kildishev, Opt. Lett. 2005, 30, 3356.
©
wileyonlinelibrary.com
Adv. Funct. Mater. 2014,
DOI: 10.1002/adfm.201401844
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
13