Rapid Large-Scale Assembly and Pattern Transfer of One-
Dimensional Gold Nanorod Superstructures
Rana Ashkar,*,†,‡,§,◆ Michael J. A. Hore,*,∥,◆ Xingchen Ye,⊥,¶ Bharath Natarajan,#
Nicholas J. Greybush,○ Thomas Lam,# Cherie R. Kagan,⊥,∇,○ and Christopher B. Murray⊥,○
†Center for Neutron Research, National Institute of Standards and Technology (NIST), Gaithersburg, Maryland 20899, United States
‡Materials Science and Engineering Department, University of Maryland, College Park, Maryland 20742, United States
§Biology and Soft Matter Division, Neutron Sciences Directorate, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831,
United States
∥Department of Macromolecular Science & Engineering, Case Western Reserve University, 10900 Euclid Avenue, Cleveland, Ohio
44106, United States
⊥Department of Chemistry, University of Pennsylvania, 231 South 34th Street, Philadelphia, Pennsylvania 19104, United States
#Center for Nanoscale Science and Technology, National Institute of Standards and Technology (NIST), Gaithersburg, Maryland
20899, United States
∇Department of Electrical and Systems Engineering, University of Pennsylvania, 200 South 33rd Street, Philadelphia, Pennsylvania
19104, United States
○Department of Materials Science & Engineering, University of Pennsylvania, 3231 Walnut Street, Philadelphia, Pennsylvania 19104,
United States
*S Supporting Information
ABSTRACT: The utility of gold nanorods for plasmonic
applications largely depends on the relative orientation and
proximity of the nanorods. Though side-by-side or chainlike
nanorod morphologies have been previously demonstrated, a
simple reliable method to obtain high-yield oriented gold nanorod
assemblies remains a significant challenge. We present a facile,
scalable approach which exploits meniscus drag, evaporative self-
assembly, and van der Waals interactions to precisely position and
orient gold nanorods over macroscopic areas of 1D nano-
structured substrates. By adjusting the ratio of the nanorod
diameter to the width of the nanochannels, we demonstrate the
formation of two highly desired translationally ordered nanorod patterns. We further demonstrate a method to transfer the
aligned nanorods into a polymer matrix which exhibits anisotropic optical properties, allowing for rapid fabrication and
deployment of flexible optical and electronic materials in future nanoscale devices.
KEYWORDS: gold nanorods, end-to-end assembly, controlled positional order, plasmonics, polymer nanocomposites
■ INTRODUCTION
Gold nanorods (Au NRs) possess highly tunable optical
absorption due to the dependence of their surface plasmon
resonances (SPRs) on aspect ratio. Because of two distinct
length scales along the transverse and longitudinal directions of
the nanorod, two surface plasmon resonances exist at
wavelengths that depend on the diameter and aspect ratio of
the nanorod, respectively,1 which allows tuning the SPR by
varying these two dimensions. In fact, unlike spherical gold
nanoparticles, gold nanorods can be tuned to absorb light over
a wide wavelength range from around 600 nm to well into the
near-infrared spectrum (>1100 nm). This tunability and broad
range of optical absorption renders Au NRs as attractive
candidates for use in optical devices, surface-enhanced Raman
spectroscopy (SERS) substrates, and biological or molecular
recognition techniques.2 In addition to the size-dependent
extinction, the plasmonic response of gold nanorods is highly
sensitive to the proximity and orientation of neighboring
nanorods, which has driven significant research efforts on the
design of nanorod assemblies for advanced technological
applications. In many such applications, there is a growing
need for precise positioning3 and directionality4 of gold
nanorods on substrates or in matrices if they are to enjoy
widespread use in nanoscale devices.
To date, obtaining highly ordered gold nanorods within a
material remains an outstanding challenge. Two types of
Received: May 5, 2017
Accepted: July 7, 2017
Published: July 7, 2017
Research Article
www.acsami.org
© 2017 American Chemical Society 25513 DOI: 10.1021/acsami.7b06273
ACS Appl. Mater. Interfaces 2017, 9, 25513−25521
assembly are primarily considered for plasmonic applications:
side-by-side and end-to-end arrangements, which, respectively,
exhibit increasing blue and red shifting of the SPR with
decreasing rod−rod separation.5 While several methods, mainly
based on depletion−attraction forces, have been demonstrated
for side-by-side alignment of nanorods in solution6,7 and in
polymer melts,8,9 a quick and reliable method for precise
positioning of close-proximity nanorods in parallel or end-to-
end assemblies remains a challenge.10 The largest hurdle in
obtaining these structures is devising methods to overcome the
strong rod−rod interactions that favor compact side-by-side
NR assembly. The case of end-to-end assembly is particularly
challenging as it requires site-specific linking of nanorods, but a
few techniques have been successfully developed for end-
linking Au NRs, including driving the assembly by end-
functionalizing Au NRs and subsequently tuning solvent
quality,11−13 using molecular bridging,14,15 or taking advantage
of the preferential bonding of capping molecules.16,17 Ferrier
and co-workers16 demonstrated that, once linked in solution,
nanorod chains could be transferred into a polymer matrix;
however, the chains did not have a significant persistence
length, which reduces both the ability of the nanocomposite to
harvest polarized light and the anisotropy of the optical
properties of the nanocomposite. Chain assemblies of nanorods
were also recently demonstrated on lithographically structured
polymers18 and patterned block copolymer films,19 but these
assemblies lacked well-defined orientational order of the
nanorods within the chains, which can be detrimental for
material designs requiring strong plasmonic field coupling.
Nepal and co-workers20 were able to obtain long-range
translational order of Au NRs cast from solution onto oriented
poly(styrene)-b-poly(2-vinylpyridine) (PS-b-P2VP) films by
exploiting preferential interactions between the nanorods and
the P2VP block. However, the resulting chainlike assemblies
were characterized by a noncompact nanorod assembly and had
larger-than-desired distances between adjacent nanorods.
Ahmed et al.21 adopted a similar approach to direct the self-
assembly of Au NRs to hydrophilic stripes, while Liz-Marzań’s
group22 used a poly(dimethylsiloxane) (PDMS) template to
confine Au NRs and induce vertical, side-by-side alignment as
drying occurred. Composto’s group used the lamellar phase of a
PS-b-poly(methyl methacrylate) (PMMA) diblock copolymer
to confine and orient Au NRs in the transverse direction of a
polymer film, but the in-plane orientation was still isotropic.23
To date, this method has not yet been successfully applied to
other block copolymer morphologies. Other methods, such as
mechanical stretching of polymer films24 and electrospinning of
fibers,25 have successfully oriented Au NRs, but these methods
were not able to obtain a high degree of nanorod order across
macroscopic length scales. One of the most promising methods
to date involves exploiting capillary forces to place and align Au
NRs on a templating substrate.18,22,26−29 A notable example of
such an approach is the recent work by Fery et al.27 in which
the plasma etching of PDMS was used to create wrinkled
PDMS substrates with a thin surface layer of SiO2, and a dip
coating method was used for nanorod patterning along the
direction of the wrinkles. They showed that the nanorods
formed multilayered structures, side-by-side-oriented Au NRs,
or single-file aligned Au NRs depending on the amplitude of
the wrinkle. The aligned Au NRs could then be transferred
onto indium tin oxide (ITO) substrates using wet-contact
printing. The authors noted, however, that as-synthesized Au
NRs could not be aligned, requiring Au NRs that were
functionalized with proteins such as bovine serum albumin
(BSA). While this approach successfully produced optical
metamaterials with highly aligned Au NRs, key drawbacks
include the requirement to functionalize Au NRs with proteins,
the presence of dislocations in the wrinkled PDMS, the sole
reliance on wrinkle geometry to control ordered structures, and
side-by-side clustering of the nanorods in wrinkles wider than
the nanorod diameter. Flauraud et al. recently took a similar
approach, and created lithographically defined templates with
which they could assemble Au NRs into predefined patterns.30
Motivated by previous work, and taking advantage of recent
advances in Au NR synthesis, we hypothesized that, by
manipulating geometry and taking advantage of strong Au−Si
interactions, it would be possible to align and position Au NRs
over macroscopic length scales, tune the characteristics of the
assembled nanorod structures, and transfer the particles into a
polymeric material that would possess anisotropic optical
properties.
In this work, we investigate the long-range alignment of Au
NRs on nanostructured silicon gratings, and present a method
which exploits meniscus drag, evaporative self-assembly, and
short-range van der Waals forces to obtain highly oriented Au
NRs over large length scales. We demonstrate that, by changing
the ratio of the nanorod diameter (DNR) to the groove width
(w) of the grating, two distinct nanorod assemblies can be
obtained. For Au NRs with diameters that are much smaller
than w, we observe segregation of the nanorods to the walls of
the grooves, and end-to-end alignment of Au NRs along the
groove direction, which can be tuned to form two distinct files
of nanorods in each unit cell. For Au NRs with diameters
commensurate with w, nanorods form discrete, single-file chains
along the groove direction, with persistence lengths that can be
micron-scale. Finally, we demonstrate that the aligned nanorods
can be readily transferred into a polymeric material within
minutes to form a polymer nanocomposite with polarization-
dependent light extinction, presenting many exciting oppor-
tunities for rapid device fabrication in the future.
■ EXPERIMENTAL METHODS
Materials. Gold nanorods were synthesized using a seed-mediated
growth method following the procedures by Ye et al.31,32 In this work,
we used two nanorod batches with sizes (20 × 73) nm and (58 × 110)
nm, which we refer to as thin and thick nanorods, respectively. The
nanostructured silicon substrates were purchased from LightSmyth
Technologies and were used as received without any surface treatment
or coating. However, silicon is known to naturally form an oxide layer,
the thickness of which was obtained by neutron reflectometry to be ∼2
nm. Reflectometry measurements (not shown here) were performed
on beamline 4B at the Spallation Neutron Source at Oak Ridge
National Laboratory. The substrates used in this work are periodic
linear nanostamps (gratings) with a rectangular profile, a period of 140
nm, a channel depth of 50 nm, and a channel width of 70 nm. A cross-
sectional SEM micrograph of the substrates is shown in Figure S1 in
the Supporting Information. It is important to point out that gratings
with different channel widths can be obtained, commercially or by e-
beam lithography, for use with rods of different dimensions. This offers
the possibility of tuning both channel width and rod dimensions for
the generation of different nanorod assemblies. In the current work,
the use of two nanorod sizes with a single channel width was sufficient
for the comparative studies intended here. The grating identified above
was chosen on the basis of the smallest channel width that is
commercially available and that would provide the maximum possible
confinement for the alignment of the nanorods described hereafter.
Small-Angle Neutron Scattering (SANS). SANS measurements
were performed on aqueous suspensions of the CTAB-coated
ACS Applied Materials & Interfaces Research Article
DOI: 10.1021/acsami.7b06273
ACS Appl. Mater. Interfaces 2017, 9, 25513−25521
25514
(cetyltrimethylammonium bromide-coated) nanorods used in the
solution casting method described in the main text. The experiments
were conducted on the NG3 30m SANS beamline at the NIST Center
for Neutron Research (NCNR). The suspensions were loaded in 2
mm demountable titanium cells available at the NCNR. The
suspensions were sonicated for 30 min just before loading and the
start of SANS measurements to break down aggregates and ensure
good dispersion of the nanorods in solution. The SANS measurements
were done in standard configuration covering a Q range 0.002−0.6
Å−1, where Q is the wave vector transfer. Data reduction was
performed using the Igor NCNR data reduction protocols which
account for background subtraction, detector sensitivity, and empty
beam corrections to obtain the absolute intensity pattern shown in
Figure S2. The SANS data indicate highly monodisperse nanorods, as
evident from the well-defined oscillations in the scattering pattern. The
scattering data were fit to a core−shell cylinder model to extract both
the nanorod and surfactant bilayer dimensions.
Scanning Electron Microscopy (SEM). SEM measurements of
the NR-decorated gratings were performed using a Zeiss Supra-55VP
field emission SEM in variable pressure mode and an FEI Helios
Nanolab 650. Conducting copper thongs were applied to the surface of
the grating to minimize charging effects. For the PMMS elastomer, the
front “nanostructured” face of the stamp was coated with a thin layer
of carbon using a Gatan high-resolution ion beam coater (Model 681)
prior to the SEM measurement to eliminate charging effects, which are
even more pronounced in the elastomer compared to those in silicon.
The coating was applied using an ∼8 keV ion beam on a carbon target
for 1 min at a rate of 0.15 Å/s, yielding a carbon coating ∼1 nm thick.
SEM measurements were then performed using an FEI Helios
Nanolab 650.
Dark Field Scattering Spectroscopy. Transmission dark field
scattering spectra were collected by a CRAIC 308 PV micro-
spectrophotometer integrated on an Olympus BX51 microscope.
White light from a 100 W tungsten halogen lamp was focused onto the
sample using a substage dark field condenser lens, and dark field
scattering from the sample was collected by a 50× IR objective
(Olympus LCPlan N IR, NA 0.65). The incident light was linearly
polarized by a broad-band visible−NIR polarizer (Melles Griot) on a
rotatable mount. The polarizer was rotated in increments of 30° over a
180° range. At 50× magnification, the collection aperture of the
microspectrophotometer corresponded to a sample region of 5.3 μm ×
5.3 μm. Spectra were collected over the wavelength range 350−1000
nm. The measured scattering intensity spectra were normalized by the
lamp spectrum as measured in transmission bright field mode.
Orientational Analysis. Two-dimensional orientational analysis
was performed on the SEM micrographs using different grid sizes, r. At
each r, multiple boxes of size r are used to obtain statistically
representative values of the 2D order parameter, S2D(r), as shown in
Figure S6a.33 The 2D order parameter is defined as
∑ θ=
=
S r
N
( )
1
cos 2
i
N
i2D
rod 1
rod
where Nrod is the number of nanorods within a grid size r in the
micrograph, and θi is the angle between nanorod i and the average
orientation of the ensemble of rods within the region of size r
(Supporting Information, Section 5). The uncertainty is obtained from
Figure 1. (a) Solution casting geometry showing the grating tilt (α) relative to a horizontal surface such that the grating lines are along the
anticipated flow direction of the cast droplets. Parts b−d show schematic depictions of the forces acting on the nanorods. (b) Comparison of the
short-range attraction forces on a nanorod on top of the mesa, Fnp, vs those of a nanorod in the groove, Fgroove, showing the energetic preference for
nanorods to locate within the groove. (c) Capillary (blue lines) and evaporation-mediated (red lines) forces contributing to the nanorod assembly in
the grooves. (d) Van der Waals interactions (W) on a thin nanorod interacting with a single surface near the top of the grooves vs those interacting
with two surfaces near the corners of the grooves.
Figure 2. Oriented Au NRs with average diameter DNR = 20.5 nm confined to grooves with width w = 70 nm. (a) Large-area scanning electron
micrographs showing that Au NRs preferentially segregate to the walls of the silicon grooves, and are aligned along the direction of the groove. The
schematic illustrates the alignment and location of Au NRs in the groove. (b) Higher-magnification image of the same sample, showing highly
oriented, end-to-end-aligned, Au NR structures that persist over long length scales. (c) Image analysis of multiple micrographs showing an average of
1.20 ± 0.6 rods/chain and 4−6 rods in the longest observed chains.
ACS Applied Materials & Interfaces Research Article
DOI: 10.1021/acsami.7b06273
ACS Appl. Mater. Interfaces 2017, 9, 25513−25521
25515
the variation in these values. The measured values and the
corresponding uncertainties are shown in Figure S6b for thin and
thick nanorod samples. The average order parameter was obtained
from the analysis of multiple micrographs (n > 3) with different grid
sizes.
■ RESULTS AND DISCUSSION
Au NRs were synthesized using a seed-mediated growth
method that has been extensively investigated in the literature,
resulting in cetyltrimethylammonium bromide-coated (CTAB-
coated) nanorods. High yields of monodisperse Au NRs were
produced by the incorporation of small amounts of aromatic
additives21 or a cosurfactant sodium oleate22 into the nanorod
growth solution, as previously described by Ye and co-
workers.31,32 Small-angle neutron scattering (SANS) measure-
ments confirm that the seed-mediated synthesis yields highly
monodisperse nanorods with a CTAB coating of 3.61 ± 0.05
nm thickness, in accord with previous measurements by
Goḿez-Graña et al.34 and Hore et al.35 It is noted that all
reported errors correspond to ±1 standard deviation. The
nanorod assemblies were generated by solution casting onto 1
× 1 cm2 silicon gratings with a 140 nm period, 70 nm groove
width (w), and 50 nm groove depth (d). The gratings were
inclined at approximately 5° with respect to a horizontal
surface, as shown in Figure 1a. Aqueous solutions of Au NRs
(∼20 μL) were pipetted onto the grating surface near its
elevated edge, allowing capillary forces to wick the liquid along
the grating lines. The gratings were then covered with a glass
Petri dish and allowed to dry slowly in a fume hood over the
course of 2 h. A combination of short-range (Figure 1b),
capillary, and evaporation-mediated (Figure 1c) forces resulted
in the self-assembly of Au NRs along the groove direction.
While SEM imaging on the gratings (Figure 2) cannot
definitively determine the vertical location of the nanorods
within the channels, we hypothesized that they are located in
the bottom corners of the channels because of increased van
der Waals interactions in this configuration. As shown
schematically in Figure 1d, there are two interacting surfaces
near the corners of the channel as opposed to the single surface
close to the top or in the middle of the channel. This provides a
thermodynamic driving force for positioning the nanorods deep
in the channels and along the channel walls. In fact, the
positioning of nanorods was later confirmed by SEM on the
PMMS imprints of the assembled structures, as described later
in the paper and shown in Figure S3 of the Supporting
Information.
The tilted substrate geometry is essential for the assembly
approach adopted in this work, and results in a concentration
gradient of Au NRs along the groove direction. Although
capillary forces naturally drag the cast fluid along the grooves,
the tilt geometry ensures unidirectional flow of the solution
along the lines of the grating and reduces the amount of excess
solution on the top surface of the substrate. The shallow tilt
angle used here ensures a slow flow of the solution along the
grooves and allows the nanorods to reassemble at the meniscus
front as it sweeps across the grating. In studying the coffee-ring
effect36 in suspensions of elongated colloids, Yodh et al.37
found that ellipsoidal and rodlike particles are drawn to the
edge of the meniscus along the flow direction and thus tend to
deposit perpendicular to the meniscus front as the solvent
evaporates. By tuning the speed of the meniscus front and the
translational and rotational diffusion of the nanorods in
solution, it is possible to deposit nanorods in chainlike patterns
as the meniscus recedes. The deposition and assembly of
particles depend on the wetting properties of the solution onto
the grating material, the evaporation rate, as well as the
concentration and viscosity of the nanorod solution.
Because of the concentration gradient in the sample,
measurements were performed over three different regions of
the substrate: low-nanorod-concentration regions near the top
of the substrate, intermediate-concentration regions in the
middle, and high-concentration regions close to the bottom.
Figure 2a shows scanning electron micrographs of the long-
range order of the assembly of the thin nanorods (DNR = 20.5
nm) at an intermediate concentration, clearly demonstrating
that the nanorods are confined within the grooves with
remarkable segregation to the sidewalls. The segregation of
particles to the sidewalls, to form two parallel channels, is a
phenomenon that has not been observed in other capillary-
force-based techniques, and is likely the result of van der Waals
interactions between the Au NRs and the silicon grooves. The
Au NRs remain almost perfectly oriented along the grooves
over many micrometers, although a small number of regions
exist where tilted orientations of Au NRs are observed, as
indicated by yellow circles in Figure 2b which shows a selected
higher-magnification micrograph of the sample in the same
region. Image analysis of the corresponding micrographs
(Figure 2c) shows that, on average, the Au NRs formed
aligned chains with a maximum of 4−6 rods per chain (average
of 1.2 ± 0.6 rods/chain). Regions with lower and higher
nanorod concentrations are shown in Figures S3 and S7,
respectively, in the Supporting Information. More detailed
image analysis (Supporting Information, Figure S8) shows that
as the filling fraction of Au NRs increases from approximately
9% to 29%, the percentage of Au NRs in two-file orientations
decreases from 66% to 15%, while the percentage of side-by-
side orientations increases from 29% to 85%. The observed
two-file Au NR assembly opens up new possibilities not only
for assembling nanorods in a chainlike structure but also for
controlling the spacing between nanorod chains. Such a control
over nanorod spacing was shown in a recent theoretical finite
difference time domain (FDTD) calculation38 to be of great
potential in amplifying plasmon coupling and enhancing near-
infrared tunability. While such patterns have been observed on
spherical nanoparticles with chemically modified surfaces,39 the
gratings used in this work were not modified in any form.
Going forward, chemical modification of the silicon surface
using silane chemistry, for example, may present opportunities
for further tuning of the nanorod assembly within the channels.
Several factors concomitantly contribute to the formation of
the assembled structures depicted in Figure 2. Capillary forces
and van der Waals interactions between the gold nanorods and
the silicon substrate are expected to draw the nanorods into the
channels. As shown schematically in Figure 1b,c, nanorods in
the channels experience stronger net attractive forces compared
to nanorods close to the grating mesas, which preferentially
drives more nanorods into the channels. These forces favor NR
alignment along the walls of the channels, thus lowering their
energy state by maximizing their contact area. Simultaneous
fluid-phase NR rearrangements near the grating surface
contribute to the rod alignment in accord with Onsager’s
theory,40 which describes spontaneous ordering of rigid
colloidal rods by a simultaneous minimization of excluded
volume and maximization of entropy. Pasquali and co-workers
reported similar effects in the preferential orientation of Au
NRs on carbon nanotube bundles.41 Finally, upon sample
ACS Applied Materials & Interfaces Research Article
DOI: 10.1021/acsami.7b06273
ACS Appl. Mater. Interfaces 2017, 9, 25513−25521
25516
drying, the evaporation of the water from the channels
generates flow fields that carry the entrained NRs to the air−
liquid−substrate interface and deposit them along the channel
walls (cf. Figure 1c). The evaporation-mediated hydrodynamic
forces in the channels are analogous to those observed in
evaporating cylindrical droplets with pinned contact lines.
Grogan et al. previously observed in situ alignment of Au NRs
along contact lines,42 and Strano et al. demonstrated the
efficacy of flow fields within evaporating cylindrical droplets in
aligning and positioning the suspended carbon nanotubes on a
substrate along the pinning direction.43 They showed that the
normal component of the fluid velocity perpendicular to the
pinned edge is directly dependent on the evaporation rate per
unit length. If the evaporation rate is faster than the rotational
diffusion of the nanorods within the channels, the nanorods
may not have sufficient time to reorient within the channels and
can be trapped in a tilted configuration relative to the channel
walls, as shown in the circled regions of Figure 2a. This is
particularly important in the current confinement configuration
since it is known that the rotational diffusion of rods near walls
can be significantly slower, especially in the presence of
electrostatic interactions,44 emphasizing the importance of the
evaporation rate when using gratings to create aligned nanorod
assemblies. In fact, effects of slower evaporation were observed
near the bottom of the grating (cf. Figure 1a) where the
meniscus took longer to dry. In these regions, denser
assemblies of nanorods were observed, as shown later, but a
two-file configuration was preserved up to a filling factor of
∼25%, beyond which side-by-side NR assemblies start to
dominate (Supporting Information, Figure S8).
The case of thin nanorods within channels that are much
wider and deeper than the rod thickness is particularly
interesting in that the rods not only can occupy different
positions along the width of the channels, but also have the
Figure 3. Confinement and orientation of Au NRs with DNR = 58 nm into channels (w = 70 nm). (a) Low-magnification SEM image in an
intermediate concentration region demonstrating alignment of Au NRs over several μm. (b) SEM image analysis reveals significantly higher
population of longer chains for the thick nanorods with an average of 3.0 ± 1.7 rods/chain and 9−11 rods in the longest chains. (c) When washed
from the channels, some nanorods maintain their end-to-end alignment, as shown in the regions highlighted in yellow.
Figure 4. (a) Schematic representation of the PMMS elastomer reaction and pattern-transfer procedure. (b) SEM image of PMMS elastomer films at
regions of medium and high nanorod concentration showing high fidelity of the nanorod pattern after transfer from the grating. (c) Image analysis of
SEM images showing the average (top) and maximum (bottom) number of Au NRs per chain as a function of the Au NR filling fraction.
ACS Applied Materials & Interfaces Research Article
DOI: 10.1021/acsami.7b06273
ACS Appl. Mater. Interfaces 2017, 9, 25513−25521
25517
ability to adhere to the sides of the channels at different vertical
locations. Inspired by this, we hypothesized that if the Au NR
diameter was commensurate with the channel width, we could
achieve significantly better order as rotational diffusion and
horizontal positioning of Au NRs within the channel would be
restricted. For Au NRs with DNR = 58 nm, the nanorods order
in a single file within the groove since the channel can no
longer accommodate multiple nanorods. Here too, measure-
ments were performed over regions of low, intermediate, and
high nanorod concentrations. Representative SEM images of
these regions are shown in the Supporting Information, Figure
S9. Figure 3a represents an intermediate nanorod concentration
region and shows that the nanorods primarily arrange into end-
to-end assemblies over micron-scale areas with approximately 3
nanorods/chain (Figure 3b). In regions that have higher
concentrations of Au NRs, superstructures of end-to-end
nanorod chains with tens of rods each can be obtained, as
shown later. Interestingly, these end-to-end assemblies appear
to be stable even after being washed out of the channels. Figure
3c shows several intact chains of Au NRs that remain after the
sample was rinsed with water, an hour after it was dried. This
process suggests new avenues for harvesting end-to-end
nanorod assemblies to use in solutions or in polymeric
materials. Moreover, such a close proximity of the nanorods
within the chains opens up new possibilities for the design of
ultrasmall optical resonators for tunable antennae, filters, and
biosensors.45 While it is not completely clear why the chains
retain their integrity after being washed from the channels, we
hypothesize that this could result from lower surfactant
coverage at the nanorod tips, and strong van der Waals
interactions between adjacent particles.
The silicon gratings that template the self-assembly present
several obstacles for using the Au NR structures in many
applications, primarily because of their lack of optical
transparency and high elastic modulus. For a transfer of the
Au NRs into a flexible polymeric material that is suitable for
optical applications, a solution of triallyl cyanurate (TAC),
bisphenol A dimethacrylate (BPADMA), and poly-
(methylmercaptosiloxane) (PMMS) was drop-cast on the
gold-patterned gratings and UV-photocured for 2 min using
2,2-dimethoxy-2-phenylacetophenone (DMPA) as a radical
initiator, as shown in Figure 4a. The low viscosity of the PMMS
solution allows the effective wetting of both the grating and rod
structures, and the thiol groups ensure strong binding to the
nanorods, resulting in an excellent pattern transfer. The cured
PMMS elastomer was immediately separated from the silicon
grating, resulting in a Au NR/polymer nanocomposite. The
SEM images in Figure 4b show the precision of the imprinted
grating structure and the effectiveness of the thiol-terminated
PMMS in lifting the gold nanorod assemblies without
disturbing the long-range order at medium (top) and high
(bottom) concentrations. These concentrations correspond to
fill factors of approximately 18% and 24%, respectively. Image
analysis (Figure 4c) indicates that both the average and
maximum number of Au NRs per chain increase as the fill
factor increases from approximately 10% to 24%. The average
number of Au NRs per chain increases from 2 to 6, while the
maximum number per chain increases from 8 to 53. In
principle, the same grating can be reused to generate additional
polymer imprints, and the mechanical properties of the
nanocomposite can be tuned by varying the relative amounts
of PMMS, TAC, and BPADMA, as shown previously by
Campos and co-workers.46 When poly(dimethylsiloxane)
(SYLGUARD 184, PDMS) was used in place of PMMS, no
Au NRs were transferred, highlighting the importance of the
PMMS thiol group in the pattern-transfer process. However,
the choice of the elastomer is determined by its affinity to the
employed nanoparticles, and other suitable elastomers could be
used for other nanoparticle compositions.
The optical properties of the Au NR composites were probed
using dark field scattering spectroscopy. Measurements of the
polymers containing the transferred Au NRs show polarization
dependence in the observed dark field scattering intensity.
Integrated dark field scattering intensities, from 350 to 1000
nm, for nanocomposites containing the thin (dashed lines) and
thick (solid lines) Au NRs are shown in Figure 5, along with
the scattering intensity of a patterned PMMS elastomer
containing no nanorods (bottom). The signal from the blank
PMMS sample shows no angular dependence in the scattering
intensity, demonstrating that the anisotropic scattering
observed in the nanocomposite samples is not due to the
textured substrate. For the nanorod composites, the scattering
exhibits a minimum at 90° for light that is polarized
perpendicular to the direction of the Au NR alignment,
implying that the localized surface plasmon resonance (LSPR)
of the nanorod assemblies is not being excited.
The degree of Au NR alignment can be quantified by
defining an orientation order parameter S = (A∥ − A⊥)/(A∥ +
2A⊥), where A∥ and A⊥ refer to the scattering intensity
measured for polarizer angles which are parallel and
perpendicular to the direction of the Au NR alignment,
respectively, measured at the LSPR wavelength. For perfectly
oriented Au NRs, S ≈ 1, whereas S = 0 for isotropic assemblies.
To account for uncertainties in the polarizer angles and
quantitatively obtain values for A∥ and A⊥, we fit the dark field
scattering intensities to a squared cosine function, and the
maximum and minimum values of the function were assigned
to A∥ and A⊥, respectively. Under this approach, the calculated
order parameter S ≈ 0.04 for both Au NR composites, which is
considerably lower than expected given the large degree of
alignment observed in the SEM images shown in Figures 2−4.
However, we note that the integration of the scattering
intensity over all wavelengths and the position of the LSPR in
the assemblies all contribute to an artificially low difference
between the scattering intensity at 0° and 90°(for more
information see the Supporting Information, Section 4).
Though this low difference results in a value of S that is not
representative of the true orientational order within the sample,
the dark field scattering measurements demonstrate the
anisotropic optical properties that the Au NR composites
possess. Indeed, 2D orientational analysis of the SEM images
reveals remarkable alignment of the nanorods along the
channels (see Figure 5b) with average 2D order parameter of
S2D = 0.83 ± 0.02 for the thin nanorod sample and S2D = 0.92 ±
0.01 for the thick nanorod sample. This proximity of the order
parameter to unity confirms the effectiveness of our procedure
in obtaining 1D alignment of nanorods. The S2D values
calculated for the two samples are in agreement with the fwhm
of the angular dependence of the normalized orientational
probabilities (Figure S7), and quantitatively demonstrate the
higher degree of alignment obtained in samples with
commensurate rod and channel widths.
■ CONCLUSION
In summary, we have demonstrated a simple, yet scalable,
approach for positioning and orienting Au NRs over large
ACS Applied Materials & Interfaces Research Article
DOI: 10.1021/acsami.7b06273
ACS Appl. Mater. Interfaces 2017, 9, 25513−25521
25518
length scales using capillary action in drop-cast suspensions on
nanostructured linear stamps. Au NRs that are much thinner
than the channel width form a unique two-file assembly along
the channel walls over a wide concentration range before
transitioning into a side-by-side assembly that fills the entire
width of the channel, as observed in previous studies. Au NRs
with widths that are commensurate with the channel width
form single-file assemblies. We hypothesize these configura-
tions are due to strong van der Waals interactions between Au
and Si, which may explain why two-file Au NR orientations
have not been observed in previous studies. Future experiments
which vary both the nanorod material (e.g., Ag or Si), as well as
the stamp material, will be important in definitively establishing
the mechanism which drives the formation of these assemblies.
From an application standpoint, such assemblies have been
theoretically shown to amplify plasmon coupling and enhance
near-infrared tunability, which are of the utmost importance in
the design and fabrication of nanoantennae. We validated the
efficacy of transferring the aligned assemblies to a polymer
matrix for later use in devices and applications. In the future, we
envision applying a selective coating that preferentially adheres
to the nanorods (and not to the grating material), where soft
coatings can be used for applications requiring flexible Au NR
chains and hard coatings for generating rigid chain structures.
By the additional employment of dissolvable substrates, the
nanorod chains can be retrieved and resuspended, thus offering
a myriad of possibilities for easy harvesting and utilization of
long Au NR chains. For example, in applications requiring well-
defined chain lengths, the resuspended nanorod chains can be
length-sorted using ultracentrifugation techniques that have
been previously utilized in the fluid-phase sorting of carbon
nanotubes.47,48 While beyond the scope of the present study,
these techniques are anticipated to play a vital role in
synthesizing high-quality polymer-based optical materials.
■ ASSOCIATED CONTENT
*S Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acsami.7b06273.
Nanostamp cross section, small-angle neutron scattering
(SANS) characterization, SEM of PMMS imprint,
discussion of dark field scattering spectroscopy results,
and orientational analysis of SEM micrographs (PDF)
■ AUTHOR INFORMATION
Corresponding Authors
*E-mail: ashkarra@ornl.gov.
*E-mail: hore@case.edu.
ORCID
Rana Ashkar: 0000-0003-4075-2330
Michael J. A. Hore: 0000-0003-2571-2111
Nicholas J. Greybush: 0000-0001-5784-654X
Present Address
¶Xingchen Ye: Department of Chemistry, Indiana University,
Bloomington, IN 47405, United States.
Author Contributions
◆R.A. and M.J.A.H. contributed equally to this work.
Notes
The authors declare no competing financial interest.
■ ACKNOWLEDGMENTS
The identification of commercial products or experimental
methods does not imply endorsement by the National Institute
of Standards and Technology nor does it imply that these are
the best for the purpose. This work utilized neutron scattering
facilities supported in part by the National Science Foundation
(NSF) under Agreement DMR-0944772. Research at ORNL’s
Spallation Neutron Source was sponsored by the Scientific User
Facilities Division, Office of Basic Energy Sciences, U.S.
Department of Energy. C.B.M., C.R.K., and N.J.G. thank
NSF/MRSEC (DMR-1120901). R.A. acknowledges the sup-
port of ORNL’s Neutron Sciences Directorate Clifford G. Shull
Fellowship program. R.A. thanks Dr. Christian Long for his
help with preliminary AFM measurements. The authors thank
Figure 5. (a) Dark field scattering intensity as a function of
polarization angle for elastomers with thick (, top) and thin
(−−−, middle) nanorod samples, and for the pure elastomer (,
bottom). The intensity for both nanorod samples shows a minimum at
polarization angles of 90°, demonstrating a large degree of alignment
of the nanorods within the elastomers along the channel direction.
Colored lines represent measurements from different regions of the
sample. (b) Color-coded directionality of thin (top) and thick
(bottom) nanorods, showing remarkable alignment of the rods along
the channel direction.
ACS Applied Materials & Interfaces Research Article
DOI: 10.1021/acsami.7b06273
ACS Appl. Mater. Interfaces 2017, 9, 25513−25521
25519
Prof. Lüis M. Campos (Columbia University) for fruitful
discussions regarding PMMS elastomer compositions.
■ REFERENCES
(1) Jain, P. K.; Eustis, S.; El-Sayed, M. A. Plasmon coupling in
nanorod assemblies: Optical Absorption, Discrete Dipole Approx-
imation Simulation, and Exciton-Coupling Model. J. Phys. Chem. B
2006, 110 (37), 18243−18253.
(2) Hore, M. J. A.; Composto, R. J. Functional Polymer
Nanocomposites Enhanced by Nanorods. Macromolecules 2014, 47
(3), 875−887.
(3) Kosako, T.; Kadoya, Y.; Hofmann, H. F. Directional Control of
Light by a Nano-Optical Yagi-Uda Antenna. Nat. Photonics 2010, 4
(5), 312−315.
(4) Tabor, C.; Van Haute, D.; El-Sayed, M. A. Effect of Orientation
on Plasmonic Coupling between Gold Nanorods. ACS Nano 2009, 3
(11), 3670−3678.
(5) Funston, A. M.; Novo, C.; Davis, T. J.; Mulvaney, P. Plasmon
Coupling of Gold Nanorods at Short Distances and in Different
Geometries. Nano Lett. 2009, 9 (4), 1651−1658.
(6) Park, K.; Koerner, H.; Vaia, R. A. Depletion-Induced Shape and
Size Selection of Gold Nanoparticles. Nano Lett. 2010, 10 (4), 1433−
1439.
(7) Nepal, D.; Park, K.; Vaia, R. A. High-Yield Assembly of Soluble
and Stable Gold Nanorod Pairs for High-Temperature Plasmonics.
Small 2012, 8 (7), 1013−1020.
(8) Hore, M. J. A.; Frischknecht, A. L.; Composto, R. J. Nanorod
Assemblies in Polymer Films and Their Dispersion-Dependent Optical
Properties. ACS Macro Lett. 2012, 1 (1), 115−121.
(9) Frischknecht, A. L.; Hore, M. J. A.; Ford, J.; Composto, R. J.
Dispersion of Polymer-Grafted Nanorods in Homopolymer Films:
Theory and Experiment. Macromolecules 2013, 46 (7), 2856−2869.
(10) Zhang, S.-Y.; Regulacio, M. D.; Han, M.-Y. Self-Assembly of
Colloidal One-Dimensional Nanocrystals. Chem. Soc. Rev. 2014, 43
(7), 2301−2323.
(11) Nie, Z.; Fava, D.; Kumacheva, E.; Zou, S.; Walker, G. C.;
Rubinstein, M. Self-Assembly of Metal-Polymer Analogues of
Amphiphilic Triblock Copolymers. Nat. Mater. 2007, 6 (8), 609−614.
(12) Nie, Z. H.; Fava, D.; Rubinstein, M.; Kumacheva, E.
″Supramolecular″ Assembly of Gold Nanorods End-Terminated with
Polymer ″Pom-Poms″: Effect of Pom-Pom Structure on the
Association Modes. J. Am. Chem. Soc. 2008, 130 (11), 3683−3689.
(13) Wang, Y.; DePrince, A. E.; Gray, S. K.; Lin, X.-M.; Pelton, M.
Solvent-Mediated End-to-End Assembly of Gold Nanorods. J. Phys.
Chem. Lett. 2010, 1 (18), 2692−2698.
(14) Caswell, K. K.; Wilson, J. N.; Bunz, U. H. F.; Murphy, C. J.
Preferential End-to-End Assembly of Gold Nanorods by Biotin−
Streptavidin Connectors. J. Am. Chem. Soc. 2003, 125 (46), 13914−
13915.
(15) Zhou, X.; Wang, Y.; Zhong, L.; Bao, S.; Han, Y.; Ren, L.; Zhang,
Q. Rational Design of Oriented Assembly of Gold Nanospheres with
Nanorods by Biotin-Streptavidin Connectors. Nanoscale 2012, 4 (20),
6256−6259.
(16) Ferrier, R. C.; Lee, H. S.; Hore, M. J. A.; Caporizzo, M.;
Eckmann, D. M.; Composto, R. J. Gold Nanorod Linking to Control
Plasmonic Properties in Solution and Polymer Nanocomposites.
Langmuir 2014, 30 (7), 1906−1914.
(17) Leung, F. C.-M.; Leung, S. Y.-L.; Chung, C. Y.-S.; Yam, V. W.-
W. Metal−Metal and π−π Interactions Directed End-to-End Assembly
of Gold Nanorods. J. Am. Chem. Soc. 2016, 138 (9), 2989−2992.
(18) Holzner, F.; Kuemin, C.; Paul, P.; Hedrick, J. L.; Wolf, H.;
Spencer, N. D.; Duerig, U.; Knoll, A. W. Directed Placement of Gold
Nanorods Using a Removable Template for Guided Assembly. Nano
Lett. 2011, 11 (9), 3957−3962.
(19) Lai, F.; Borca-Tasciuc, T.; Plawsky, J. Controlling Directed Self-
Assembly of Gold Nanorods in Patterned PS-b-PMMA Thin Films.
Nanotechnology 2015, 26 (5), 055301.
(20) Nepal, D.; Onses, M. S.; Park, K.; Jespersen, M.; Thode, C. J.;
Nealey, P. F.; Vaia, R. A. Control over Position, Orientation, and
Spacing of Arrays of Gold Nanorods Using Chemically Nanopatterned
Surfaces and Tailored Particle-Particle-Surface Interactions. ACS Nano
2012, 6 (6), 5693−5701.
(21) Ahmed, W.; Glass, C.; Kooij, E. S.; van Ruitenbeek, J. M.
Tuning the Oriented Deposition of Gold Nanorods on Patterned
Substrates. Nanotechnology 2014, 25 (3), 035301.
(22) Hamon, C.; Novikov, S.; Scarabelli, L.; Basabe-Desmonts, L.;
Liz-Marzan, L. M. Hierarchical Self-Assembly of Gold Nanoparticles
into Patterned Plasmonic Nanostructures. ACS Nano 2014, 8 (10),
10694−10703.
(23) Deshmukh, R. D.; Liu, Y.; Composto, R. J. Two-Dimensional
Confinement of Nanorods in Block Copolymer Domains. Nano Lett.
2007, 7 (12), 3662−3668.
(24) Murphy, C. J.; Orendorff, C. J. Alignment of Gold Nanorods in
Polymer Composites and on Polymer Surfaces. Adv. Mater. 2005, 17
(18), 2173−2177.
(25) Roskov, K. E.; Kozek, K. A.; Wu, W.-C.; Chhetri, R. K.;
Oldenburg, A. L.; Spontak, R. J.; Tracy, J. B. Long-Range Alignment of
Gold Nanorods in Electrospun Polymer Nano/Microfibers. Langmuir
2011, 27 (23), 13965−13969.
(26) Ahmad, I.; Jansen, H. P.; Zandvliet, H. J. W.; Kooij, E. S.
Hydrodynamic Confinement and Capillary Alignment of Gold
Nanorods. Nanotechnology 2016, 27 (2), 025301.
(27) Tebbe, M.; Mayer, M.; Glatz, B. A.; Hanske, C.; Probst, P. T.;
Muller, M. B.; Karg, M.; Chanana, M.; Konig, T. A. F.; Kuttner, C.;
Fery, A. Optically Anisotropic Substrates via Wrinkle-Assisted
Convective Assembly of Gold Nanorods on Macroscopic Areas.
Faraday Discuss. 2015, 181, 243−260.
(28) Kuemin, C.; Nowack, L.; Bozano, L.; Spencer, N. D.; Wolf, H.
Oriented Assembly of Gold Nanorods on the Single-Particle Level.
Adv. Funct. Mater. 2012, 22 (4), 702−708.
(29) Kuemin, C.; Stutz, R.; Spencer, N. D.; Wolf, H. Precise
Placement of Gold Nanorods by Capillary Assembly. Langmuir 2011,
27 (10), 6305−6310.
(30) Flauraud, V.; Mastrangeli, M.; Bernasconi, G. D.; Butet, J.;
Alexander, D. T. L.; Shahrabi, E.; Martin, O. J. F.; Brugger, J.
Nanoscale Topographical Control of Capillary Assembly of Nano-
particles. Nat. Nanotechnol. 2016, 12 (1), 73−80.
(31) Ye, X.; Jin, L.; Caglayan, H.; Chen, J.; Xing, G.; Zheng, C.;
Doan-Nguyen, V.; Kang, Y.; Engheta, N.; Kagan, C. R.; Murray, C. B.
Improved Size-Tunable Synthesis of Monodisperse Gold Nanorods
through the Use of Aromatic Additives. ACS Nano 2012, 6 (3), 2804−
2817.
(32) Ye, X.; Zheng, C.; Chen, J.; Gao, Y.; Murray, C. B. Using Binary
Surfactant Mixtures To Simultaneously Improve the Dimensional
Tunability and Monodispersity in the Seeded Growth of Gold
Nanorods. Nano Lett. 2013, 13 (2), 765−771.
(33) Hore, M. J. A.; Composto, R. J. Nanorod Self-Assembly for
Tuning Optical Absorption. ACS Nano 2010, 4 (11), 6941−6949.
(34) Gomez-Grana, S.; Hubert, F.; Testard, F.; Guerrero-Martinez,
A.; Grillo, I.; Liz-Marzan, L. M.; Spalla, O. Surfactant (Bi)Layers on
Gold Nanorods. Langmuir 2012, 28 (2), 1453−1459.
(35) Hore, M. J. A.; Ye, X.; Ford, J.; Gao, Y.; Fei, J.; Wu, Q.; Rowan,
S. J.; Composto, R. J.; Murray, C. B.; Hammouda, B. Probing the
Structure, Composition, and Spatial Distribution of Ligands on Gold
Nanorods. Nano Lett. 2015, 15 (9), 5730−5738.
(36) Deegan, R. D.; Bakajin, O.; Dupont, T. F.; Huber, G.; Nagel, S.
R.; Witten, T. A. Capillary Flow as the Cause of Ring Stains from
Dried Liquid Drops. Nature 1997, 389 (6653), 827−829.
(37) Yunker, P. J.; Still, T.; Lohr, M. A.; Yodh, A. G. Suppression of
the Coffee-Ring Effect by Shape-Dependent Capillary Interactions.
Nature 2011, 476 (7360), 308−311.
(38) Sharma, Y.; Dhawan, A. Hybrid Nanoparticle−Nanoline
Plasmonic Cavities as SERS Substrates with Gap-Controlled Enhance-
ments and Resonances. Nanotechnology 2014, 25 (8), 085202.
(39) Liddle, J. A.; Cui, Y.; Alivisatos, P. Lithographically Directed
Self-Assembly of Nanostructures. J. Vac. Sci. Technol., B: Microelectron.
Process. Phenom. 2004, 22 (6), 3409−3414.
ACS Applied Materials & Interfaces Research Article
DOI: 10.1021/acsami.7b06273
ACS Appl. Mater. Interfaces 2017, 9, 25513−25521
25520
(40) Onsager, L. The Effects of Shape on the Interaction of Colloidal
Particles. Ann. N. Y. Acad. Sci. 1949, 51 (4), 627−659.
(41) Dan, B.; Wingfield, T. B.; Evans, J. S.; Mirri, F.; Pint, C. L.;
Pasquali, M.; Smalyukh, I. I. Templating of Self-Alignment Patterns of
Anisotropic Gold Nanoparticles on Ordered SWNT Macrostructures.
ACS Appl. Mater. Interfaces 2011, 3 (9), 3718−3724.
(42) Grogan, J. M.; Schneider, N. M.; Ross, F. M.; Bau, H. H. The
Nanoaquarium: A New Paradigm in Electron Microscopy. J. Indian
Inst. Sci. 2012, 92 (2), 295−308.
(43) Sharma, R.; Lee, C. Y.; Choi, J. H.; Chen, K.; Strano, M. S.
Nanometer Positioning, Parallel Alignment, and Placement of Single
Anisotropic Nanoparticles Using Hydrodynamic Forces in Cylindrical
Droplets. Nano Lett. 2007, 7 (9), 2693−2700.
(44) Haghighi, M.; Tahir, M. N.; Tremel, W.; Butt, H.-J.; Steffen, W.
Translational and Rotational Diffusion of Gold Nanorods Near a Wall.
J. Chem. Phys. 2013, 139 (6), 064710.
(45) Li, Z.; Butun, S.; Aydin, K. Touching Gold Nanoparticle Chain
Based Plasmonic Antenna Arrays and Optical Metamaterials. ACS
Photonics 2014, 1 (3), 228−234.
(46) Campos, L. M.; Truong, T. T.; Shim, D. E.; Dimitriou, M. D.;
Shir, D.; Meinel, I.; Gerbec, J. A.; Hahn, H. T.; Rogers, J. A.; Hawker,
C. J. Applications of Photocurable PMMS Thiol−Ene Stamps in Soft
Lithography. Chem. Mater. 2009, 21 (21), 5319−5326.
(47) Arnold, M. S.; Green, A. A.; Hulvat, J. F.; Stupp, S. I.; Hersam,
M. C. Sorting Carbon Nanotubes by Electronic Structure Using
Density Differentiation. Nat. Nanotechnol. 2006, 1 (1), 60−65.
(48) Fagan, J. A.; Becker, M. L.; Chun, J.; Hobbie, E. K. Length
Fractionation of Carbon Nanotubes Using Centrifugation. Adv. Mater.
2008, 20 (9), 1609−1613.
ACS Applied Materials & Interfaces Research Article
DOI: 10.1021/acsami.7b06273
ACS Appl. Mater. Interfaces 2017, 9, 25513−25521
25521