Sphere-on-pillar optical nano-antennas
ABSTRACT We propose an optical nano-antenna consisting of a pair of sphere-on-pillar structures. Experiments show that the controlled fabrication of metallic nanospheres on the tip of carbon nanotubes (CNTs) is effective, and numerical investigation revealed that a pair of such structures are capable to convert free space radiation into an intense near-field; hence can function as an optical antenna. The fabrication process, electron-beam-induced bubbling (EBIB) and electromigration-based bubbling (EMBB), are based on nanofluidic mass delivery at the attogram scale using metal-filled CNTs. Under the irradiation of a high energy electron beam of a transmission electron microscope (TEM), the encapsulated metal is melted and extruded out from the tip of the nanotube; generating a metallic sphere. In the case that the encapsulated materials inside the CNT have a higher melting point than that of the beam energy can reach, electromigration-based mass delivery is an optional process to apply. Under a low bias (2-2.5V), spherical nanoparticles are formed on the tips of nanotubes. The optical properties of the nano-antenna are analyzed numerically using the finite element method. Our investigations have revealed that the field enhancement, the resonances, and the radiation patterns can be easily tuned since all these quantities strongly depend on the size of the nanotubes and the metallic spheres, as well as on their material properties. Sphere-on-pillar optical antennas carry a great potential for bio-sensing, tip-enhanced spectroscopy applications, and interfacing integrated nanophotonic circuits.
- SourceAvailable from: Dieter W. Pohl
Article: Resonant optical antennas.[show abstract] [hide abstract]
ABSTRACT: We have fabricated nanometer-scale gold dipole antennas designed to be resonant at optical frequencies. On resonance, strong field enhancement in the antenna feed gap leads to white-light supercontinuum generation. The antenna length at resonance is considerably shorter than one-half the wavelength of the incident light. This is in contradiction to classical antenna theory but in qualitative accordance with computer simulations that take into account the finite metallic conductivity at optical frequencies. Because optical antennas link propagating radiation and confined/enhanced optical fields, they should find applications in optical characterization, manipulation of nanostructures, and optical information processing.Science 07/2005; 308(5728):1607-9. · 31.20 Impact Factor
Article: Nanoantennas for light emissionScience. 01/2005; 308:1561-1563.
- [show abstract] [hide abstract]
ABSTRACT: The optical properties of coupled metallic nanorods are studied to investigate the use of coupled plasmonic structures in field-enhanced spectroscopies. Light scattering by coupled nanorods is calculated with the bound-ary element method, including retardation. The modes of coupled nanorod systems are calculated by the boundary charge method and discussed in terms of their symmetry. Similar scattering behavior for isolated nanorods and pairs of nanorods can mask the very different local responses that produce near-field enhance-ment. The response of isolated rods redshifts with increasing rod length because intrarod restoring forces are reduced. The near-and far-field responses increase monotonically with increasing rod length increasing polarization along the rod. For coupled nanorods, coupling localizes charge at the gap between the rod ends and splits degenerate modes. The localized charge depolarizes the intrarod response and provides an additional redshift. Moreover, the near-field enhancement in the gap between the nanorods is dramatically increased by coupling-induced charge localization at the gap. For short nanorods, the near-field response in coupled systems is determined by the geometry of the rod ends that define the gap. For longer nanorods, the response in coupled systems is determined by the rod length. Changing the dimensions and geometry of the nanorods to modify the interrod coupling has a major effect on the local-field enhancement. The effects of the environment and the actual metallic material do not have as big an influence on the field enhancement.01/2005; 20.
Xudong Cui1,*, Zheng Fan2, Xinyong Tao3, Weihua Zhang4, Daniel Erni5, Xudong Fan2, Xiaobin Zhang6, and
1China Academy of Engineering Physics, Mianyang, Sichuan 621900, China
2Michigan State University, East Lansing, Michigan 48864, USA
3Zhejiang University of Technology, Hangzhou 310014, China
4EPFL Lausanne, CH-1015, Lausanne, Switzerland
5University of Duisburg-Essen, D-47048, Duisburg, Germany
6Zhejiang University, Hangzhou 310014, China
*Email: email@example.com, firstname.lastname@example.org
Abstract—We propose an optical nano-antenna consisting
of a pair of sphere-on-pillar structures. Experiments show that
the controlled fabrication of metallic nanospheres on the tip of
carbon nanotubes (CNTs) is effective, and numerical
investigation revealed that a pair of such structures are capable
to convert free space radiation into an intense near-field; hence
can function as an optical antenna. The fabrication process,
electron-beam-induced bubbling (EBIB) and electromigration-
based bubbling (EMBB), are based on nanofluidic mass
delivery at the attogram scale using metal-filled CNTs. Under
the irradiation of a high energy electron beam of a transmission
electron microscope (TEM), the encapsulated metal is melted
and extruded out from the tip of the nanotube; generating a
metallic sphere. In the case that the encapsulated materials
inside the CNT have a higher melting point than that of the
beam energy can reach, electromigration-based mass delivery
is an optional process to apply. Under a low bias (2-2.5V),
spherical nanoparticles are formed on the tips of nanotubes.
The optical properties of the nano-antenna are analyzed
numerically using the finite element method. Our investigations
have revealed that the field enhancement, the resonances, and
the radiation patterns can be easily tuned since all these
quantities strongly depend on the size of the nanotubes and the
metallic spheres, as well as on their material properties.
Sphere-on-pillar optical antennas carry a great potential for
bio-sensing, tip-enhanced spectroscopy applications, and
interfacing integrated nanophotonic circuits.
Index Terms – Carbon
nanostructure, optical antenna, nanofluidics, nanophotonics
radiation to a sub-wavelength sized near-field, provided the
possibilities to tailor the optical properties of the underlying
nanostructure. The feasibility of optical nano-antennas has
been demonstrated in the context of enhanced single
molecule light emission and absorption , field-enhanced
spectroscopy , and nanolithography . All these antenna
structures are made of noble metals and thus the localized
surface plasmon resonances are playing a central role in the
optical responses of functional nanostructures, where the
incident light is efficiently absorbed by the metallic
nanostructure and/or redirected as scattered light at
resonance. Due to the highly dispersive nature of the
underlying material system, it becomes a major challenge to
Optical antennas [1-3] can efficiently couple far-field
engineer such optical nano-antennas according to the desired
specifications, not to mention the challenges that are faced in
fabrication technology, such as material compatibility in
composite nanodevices, as well as the difficulty to access
experimental data from such nano-sized functional
topologies. Moreover, these plasmonic nanostructures are
quite sensitive with respect to the structure morphology and
the surrounding environment
controllable properties (materials, geometries etc.) either
during fabrication or at the post processing stages are
therefore anticipated to be capable of overcoming some of
those difficulties. Besides noble metallic particles, carbon
nanotubes (CNTs) have been demonstrated as nanooptical
antennas [6, 7]. Nevertheless, the antenna effect in CNTs is
not as pronounced as in its classical microwave counterpart
because of the large material losses. Even though, CNTs
provide an ideal system to study optical antennas because of
their well-adapted shape, but also due to their tunable
material properties . Recently, copper-filled CNTs have
been demonstrated for attogram level mass deposition and
spot welding using nanorobotic manipulation . We expect
that such a combination of metallic parts with nanotubes
would provide significant degrees of freedom to realize a
new type of optical nano-antenna. Hence, we propose a dual-
segmented optical nano-antenna (Fig. 1), where a pair of
metallic spheres is bubbled onto the ends of CNTs.
[1-5]. Designs with
Fig. 1 Optical nano-antennas. The spheres are joint to two identical
cylindrical parts (L1 = L2) to mimick the support provided by the
corresponding nanotubes. The cylinder axes coincide while forming a direct
connection between the centers of the two spheres. The computation
window is truncated using perfectly matched layers (PMLs) to avoid
Sphere-on-Pillar Optical Nano-Antennas
2010 IEEE Nanotechnology Materials and Devices Conference
Oct 12 - 15, 2010, Monterey, California, USA
978-1-4244-8897-1/10/$26.00 ©2010 IEEE 171
A verity of techniques, such as inversed self-assembly
grafting, wet-chemistry surface assembly, water-flow
suction, photocatalytic deposition, and optical trapping, have
been developed to attach metallic nanoparticles on a
cantilever tip or an optical fiber , however, the
controlled attachment of individual nanoparticles on
nanopillars has been shown infeasible .
Here we propose two processes, electron-beam-induced
bubbling (EBIB) and electromigration-based bubbling
(EMBB), for the fabrication of the structures of nanospheres
on pillars. The technique resembles blowing a balloon using
a pipe and is technically based on nanofluidic mass delivery
at the attogram scale using metal-filled carbon nanotubes
(m@CNTs). In previous works, electron-beam-induced
expansion, melting, and flowing of the encapsulated
materials inside a nanotube have been observed, and
potential applications such as thermometers and extruders
have been demonstrated [11-13]. Investigations on the intra-
and inter-nanotube mass melting, flowing, evaporation, and
delivery based on electromigration [9, 14] have enabled new
techniques such as nanorobotic spot welding  and devices
such as archival memories . Bubbling involves several
novel aspects such as pressure accumulation inside
nanotubes, nanotube tip breaking or opening, and shaping
and sizing of the particles.
A. EBIB of Sn-filled CNTs
Sn-filled CNTs (Sn@CNTs)  were used to show the
process of EBIB. The experiments were performed in a
transmission electron microscope (TEM, JEOL 2200FS)
with a field emission gun. In the experiments, individual
Sn@CNTs were exposed under the high energy electron-
beam. The current density of the electron beam transmitting
through the CNTs was adjusted by changing the beam
convergence as well as the magnification and the incident
area. Figure 2 (a) to (j) show the formation of a Sn sphere on
the tip of a CNT (The setup is schematically shown in Fig. 2
(k)) at a current density of 20 A/cm? with a magnification of
×300K and an irradiation area of 1.3×10-14 cm2. In our
previous investigations, we have observed melting and
expansion of tin inside CNTs when the current density
reached 0.4 A/cm? , here we observed that a bubbling
process will follow (Fig. 2 (e) to (j)). At the beginning of the
process, polyhedral nanoparticles (Fig. 2(g) and (h)) were
formed. By increasing temperatures, they were developed
into spheres (Fig. 2(j)) . We attribute the bubbling and
the shape conversion to the electron irradiation and the
secondary effects including the carbon shell reconstruction
and the surface tension of the molten metal. The
encapsulated materials were melted and then squeezed out
by the carbon shells as spheres onto the top of nanotubes.
Applying an image processing method, the mass of the
sphere is analyzed. The CNT has an external diameter of
~40nm. The diameter of the final sphere is 54 nm, and the
mass of the resulted sphere is 0.6 fg (femtograms) according
to the density of the tin (7.31 g/cm?). The influence of the
current densities on the formation speed and the diameter of
the spheres are depicted in Fig. 2 (l).
Fig. 2. Fabrication of sphere-on-pillar nanostructures using EBIB. (a) A
Sn-filled CNT was brought to under the exposure of electron beam at t
= 0s. (b, c) Under the current density of 20A/cm? from 240s to 480s, a
molten section of tin appeared and moved to the tip of the tube, which
deformed the tip of the tube into a quasi-spherical shape. (d) At t=720s,
the inner metal first broke out from the tube. (e, f) Metal started to flow
out. The carbon shell near the tip of tube was deformed more, which
attributed to the squeezing out of the Sn sphere from the nanotube shell.
(g) At t=1440s, a sphere is visible on the tip of the CNT. (h-j) The
shape changed with time and a sphere was formed at t=2880s. The
shrinkage of the carbon shell prevented more metal to flow from the
bottom to the tip of the tube. (k) Schematic drawing of EBIB. (l) shows
the influence of the current densities on the formation speed and the
diameter of the spheres.
B. EMBB of Cu-filled CNTs
A scanning tunnelling microscope built in a TEM holder
(Nanofactory Instruments AB) is adopted for EMBB. By
delivering the encapsulated materials from the carbon shells,
nanospheres are bubbled over the CNT tips (Fig. 3(a)).
Fig. 3 Fabrication of sphere-on-pillar nanostructures using EBIB. (a)
Schematic of the fabrication of spheres on CNTs by flowing out
encapsulated metal from a nanotube using electromigration. (b) EMBB
from a Cu-filled CNT. Inset shows that the nanotube has an opened cap
originally. (c) Detailed series of (b). (d) A sphere formed on a CNT without
an initial opening. The diameter of the sphere is 2R = 140 nm, and the
diameter of the CNT D is about 50 nm.
0600 1200 1800 2400 3000 3600 4200
Sphere Diameter [nm]
During the experiments, the intensity of the electron
beam has been kept in the range for regular imaging, which
is several orders of magnitude lower than the above-
mentioned values. Results show that under a low bias (2-2.5
V), spherical particles can be formed on the tip of nanotubes.
At a low temperature, polyhedral nanoparticles (Fig. 3(b)
and (c)) rather than spheres (Fig. 3(d)) will be formed. By
increasing temperatures, it is possible to convert polyhedral
nanoparticles into spheres.
III. MODELING AND SIMULATIONS
The nano-antenna under consideration is displayed in
Fig. 1, where a metallic sphere that is attached at the near
end of the nanotube has been mirrored accordingly to form a
paired nano structure (L1 = L2) with a feed gap. The feed gap
g is initially selected to be 20 nm, which indicates the
smallest distance between the two surfaces of spheres we
could experimentally achieve. The entire structure is
embedded in vacuum. Other structural parameters, such as
the diameter of the two spheres and the dimensions of the
joint nanotubes are adapted from our fabricated structure
shown in Fig. 3(d). A plane wave incident from below [i.e.
from the –y-direction, cf. Fig. 1] is used as excitation within
a wavelength range from 400 nm up to 1200 nm. The
polarization is chosen to be parallel to the axes of the
nanotubes and the electric field strength amounts to
Ex = 1 V/m. The simulations are carried out using the Finite
Element Method (FEM) included in the COMSOL
Multiphysics modeling platform. We adopt the FEM mainly
for its compatibility with curved geometry due to its
adaptive meshing capabilities. Meshing becomes particularly
critical when dealing with structures with feature sizes
differing in several orders of magnitude. Mesh sizes as small
as 0.5 nm are used throughout the calculations when
compared to the overall size of the antenna structure,
resulting in 5-9105 degrees of freedom (indicating the
number of parameters required for the derived field in the
simulation space). The field was calculated in the framework
of the scattered field formulation by using perfectly matched
layers (PMLs) to avoid spurious reflections from the
surrounding medium boundaries [Fig. 1]. Since FEM is
memory-hungry for 3D simulations, the direct solver
PARDISO is used and the computation time for the paired
geometry amounts to about 58-90 hours on an Intel quad-
core processor (3 GHz) with 32 GB RAM.
From numerical considerations, a MWNT is usually set
to be equivalent to a single-walled CNT (SWNT) with a
much larger effective radius and an effective conductance
. For example, the complex dielectric function for an
infinitely long SWNT is expressed as a function of effective
conductance cn and the frequency as follows:
Where is the angular frequency, 0 is the permittivity of
free space, cn is the complex dielectric function for the
SWNT, S is the surface area of the nanotube and T is the
volume density of the proper tube . The effective
conductance for an N-layer MWNT is then given by the
simple relation cn
MWNT is defined as Reff = R = (R1 + R2) / 2, where R1 is the
radius of inner wall and R2 is the radius of the outer wall. It
should be noted that the optical conductance could be varied,
due to different factors that are affecting the electronic
conductivity of SWNT, including the doping level of the
semiconducting tubes, the sample purity, and the metal-to-
semiconductor volume fraction [9, 19]. In addition, since the
MWNTs basically consist of multi-layered concentrically
aligned SWNTs, the intertube interactions should be taken
into account with respect to the applied field polarizations
. Referring to the antenna application we expect an
electric polarization, which is parallel to the tube axis. In this
case, the intertube interactions play a less important role and
can therefore be neglected in the simulations. This
polarization direction also supports a significant field
enhancement, while virtually no enhancement is evident
when the electric field is not parallel to the axis of nanotube
[cf. Fig. 1].
The nanotubes can either be metallic or semiconducting
depending on the fabrication technology, which has a
considerable influence on the resonant features of our
antenna structures. For this reason, we will only discuss one
specific case, namely a nanotube with a fixed dimensional
size (tube radius) that is modeled as being metallic. The data
for the nanotube used in the calculations is shown in
Fig. 3(d). For the metallic spheres, we used measured data of
silver according to the well-known reference of Johnson and
M = Ncn. The effective radius of the
A. Optical properties of the paired structure
In this section, we investigate the optical properties of the
nano-antenna as depicted in Fig. 1. The resulting computed
spectral responses of the electric field in the center of the
gap (antenna feed) is shown in Fig. 4(b) for different
nanotube lengths and for a plane wave excitation with
Ex = 1 V/m. For comparison, the spectral response of the
antenna structure without nanotubes (L1 = L2 = 0) is
displayed in the same figure. One resonance at 686 nm is
identified for the sole paired sphere structure, with the
corresponding field enhancement factor of 10.75 [cf.
Fig. 4(b)]. The large field enhancement between the two
spheres is attributed to near-field interactions that increase
the surface charge density, and hence, the field strengths
associated to the present surface plasmon (SP) modes at
resonance . The resonant field pattern of the structure
(L1 = L2 = 0) at 686 nm is characterized as a fundamental
dipole mode, based on the fact that the antenna response is
maximized when the antenna length is an odd multiple p of
half the effective wavelength eff (i.e. L = p·eff /2, with p = 1
for the dipole mode) . Note that due to material dispersion
at optical frequencies, the effective wavelength of an optical
antenna is usually very short and strongly material-
dependent. An alternative way to identify the mode order of
an antenna is to analyze the current distribution in the proper
structure . The resonant electric and magnetic field
distributions for the sole paired structure without CNTs are
shown in Fig. 4(c-d). As observed here, the fields primarily
concentrated in the gap region with maximal values in the
gap center and die off toward both antenna ends. When the
nanotubes (L1 = L2 < 100 nm) are added into the structure,
the electric field in the gap increases at the resonant
wavelengths compared with the sole paired structure. The
field enhancement is also promising at shorter wavelength,
i.e., in the wavelength range 300-550 nm. In general, the
field increases [cf. Fig. 4(a)] for a nanotube length of
L1 = L2 < 500 nm at shorter wavelengths. This further shows
that the field enhancement is a function of the length of
CNT, where the presence of CNT affects the charge density
on the spheres, resulting in varied coupling strength within
the gap. Another fact to be noted is that the field
enhancement is associated with the material properties of
both, silver and CNT, where the material losses are small at
long wavelengths for both materials.
It is well known for plasmonic structures that the SP
resonances strongly depend on the shape and size of the
structure, as well as on the material losses . A more
efficient way to improve the field enhancement in our
antenna topology is to increase the lightning rod effect by
reducing the size of the metallic spheres while keeping a
fixed gap width. This reduction can be realized by precisely
controlling the parameters of the nanoscopic welding
process , where the sphere can be made even as small as
the diameter of the nanotube (~50 nm). An alternative
scheme to increase the field enhancement in the gap also
consists in altering the metallic doping level in the CNT
fabrication to reduce the losses of CNT. Notably, the electric
field in the gap drops exponentially with increasing gap size,
indicating that small gap widths are necessary in order to
keep a large field enhancement and, hence a high optical
intensities in the antenna feed. A huge field enhancement in
the gap is therefore observed for two nearly touching spheres
. However, it is nearly impractical to accurately
assemble two spheres with such a small gap size (i.e.
g = 1 nm or even less). Therefore, additional methods have
to be applied to create such small gaps in order to generate
so-called “hot spots” . Hot spots, namely, large optical
intensities are beneficial when exploiting nonlinear
properties such as e.g. optical frequency mixing on metallic
surfaces . More exotic applications such as single-
photon sources may be realized by placing individual
quantum emitters inside the gap to enhance light-matter
The resonance shift of nano-antenna structures is also of
interest, since the related effective antenna length is an
important indicator of various antenna effects . When
the nanotubes are added to the spheres, the spectral
resonances of the structure undergo a red-shift as shown in
Fig. 4(b). For example, the introductions of a 50 nm-long
nanotube gives rise to two resonances in the wavelength
range of interest (400-900nm), which occurs at the
wavelengths of 526 nm and 842 nm (the related field
enhancement factors at these two resonances are 9.5 and
12.6, respectively). Compared to the fundamental resonance
of the sole paired silver sphere structure (at 686 nm), the
fundamental resonance wavelength of the resulting nano-
antenna (with L1 = L2 = 50 nm) is now red-shifted by 156 nm
reaching a value of 842 nm. Increasing the length of the two
nanotubes further (i.e., L1 = L2 = 100 nm, 200 nm), provides
even larger red-shifts (~ 500 nm). The resonant mode
appearing in the structure (with L1 = L2 = 50 nm) at 526 nm
is identified as high-order dipole-like modes based on their
current distributions within the structure and the associated
radiation patterns. This additional resonance is significantly
narrower than the fundamental one at 842 nm. It is even
more interesting that the introduction of a 10 nm-long CNT
would cause 8 nm resonance shift as shown in Fig. 4(b),
implying that this type of multi-segmented structure might
provide an efficient way to tune the resonance of plasmonic
structures without increasing the structural size too much. In
addition, only small resonance shifts are observed if the
length of the nanotube becomes longer than several
wavelengths (i.e., 5), while the size of both, the gap and
silver sphere are kept unchanged (g = 20 nm, R = 70 nm).
Some resonances – which remain nearly unchanged (i.e.,
shift less than 3 nm) when the CNTs are introduced to the
structure – are observed at short wavelengths (at 378-
381 nm). Our antenna structure is thus best suited for
applications where small resonance shifts are beneficial,
such as e.g. tip-enhanced applications. For instance,
negligible small resonance shifts can inhibit the spectral
mismatch between the laser source and the tip resonance if
the structure is properly designed. Further numerical studies
have revealed that the size of spheres and the gap width are
the most relevant geometric parameters for tuning the
resonance based on the configuration shown in Fig. 1: some
Fig. 4. (a) Permittivity of the MWNT as a function of wavelength used
in our calculations; (b) Electric field strength in the center of the gap as
a function of the wavelength for various MWNT lengths. Some of the
resonant wavelengths for the paired structure are indicated with
corresponding arrows. (c) Electric field distribution for the sole paired
structure at the resonant wavelength of 686 nm; (d) Magnetic field
distributions for the sole paired structure at the resonant wavelength of
686 nm. The red and bright colors in the figure indicate high field
resonances will red-shift while increasing the gap size and
finally leak out from the wavelength range of interest,
rendering the spheres more and more uncoupled. These
findings comply well with similar studies on pure metallic
bow-tie antennas  and dimer plasmonic antennas .
B. Radiation Properties
From the above investigations, we can see that the added
CNTs do have a strong influence on the field enhancement,
as well as on the resonance of the structure. The antenna
effects associated to the nanowires are arising from photon
confinement in the structures, where internal standing wave
photon modes are generated in the wire under proper light
illumination . It is well-known from antenna theory that
antennas radiate from currents and while tailoring the
current distribution in the structure enables to design a
desired radiation pattern . Therefore, currents in
antennas play an important role on the antenna effects.
Figure 5(a) shows the magnetic field distributions at three
resonance wavelengths of the structure with two 200 nm
nanotubes. Note that the corresponding current distribution
can be obtained when integrating the magnetic field over the
antenna surface. At the resonance wavelength of 382 nm the
dominant field strength is found in the gap (i.e. the antenna
feed) and the field distributions at the two joints and the
CNT are barely visible; while at the resonance wavelengths
of 568 nm and 708 nm, the fields are predominantly
distributed on the CNT arms and the gap regions. This
implies that the interacting fields between the two antenna
arms can vary as a function of the excitation wavelength. We
also plot the normalized magnetic field strength along the
antenna axis. At each resonance it displays a sort of
alternating modulation across the structure [cf. Fig. 5(b)],
which gets more pronounced for shorter wavelength and
shorter CNTs. Nevertheless, this field modulation will
become weaker if both CNTs yield a length larger than the
wavelength (e.g. 5). An increasing length is obviously
associated with a red-shift of the aforementioned resonance,
where the wavelength shift becomes small if the CNTs are
both exceeding the aforementioned limit of 5.
Above facts imply that the presence of the CNTs may
impact the radiation properties, despite the antenna effect of
CNT itself is relatively weak due to high material losses .
In order to validate the influence of CNT on the emission
characteristics, we numerically investigated the radiation
patterns at the corresponding resonances of the antenna
structure. The radiation patterns are obtained for the upper
half x-0-y plane [i.e. the E-plane as indicated in Fig. 1]. The
radiation characteristics in Fig. 6(a) are displayed for the
sole paired sphere structure without the two CNTs. The
resonance at 686 nm is maintained by the dipole-like
fundamental mode; while some other resonance such as e.g.
the one at 386 nm has a multipole high-order mode pattern,
with a main lobe having a beamwidth of around 30°. This
multipolar nature of the emission pattern owes to the
frequency dependence of the near-field coupling as well as
to the fact that the minor lobes are arising from the spectrally
varying charge distributions at the far-end of the antenna
topology. Therefore, it is not surprising that the main and
minor lobes of the far-field characteristics have different
beamwidths as shown in Fig. 6 due to the different
bandwidths of the corresponding resonances. More
interestingly, different modes show correspondingly
different radiation efficiencies. For example, the far-field
radiation is stronger at 386 nm than at 686 nm [cf. Fig. 6(a)],
where a stronger field enhancement was found at 386 nm.
When two 50 nm CNTs are introduced into the structure,
minor lobes still appear in the radiation patterns [cf.
Fig. 6(b)] but with different beam widths and field strengths
compared to the radiation patterns of the bare sphere
structure [cf. Fig. 6(a)]. However, the bandwidth differences
for the multipolar modes are graphically indistinguishable
from the comparison between Fig. 6(a) and Fig. 6(b), since
the resonance bandwidths are only slightly different at those
resonances. For the dipolar modes, the bandwidth difference
is graphically discerned, due to the improved suppression of
the minor sidelobes, which are originating from multiple
mode interactions with the far-end of both, the metallic parts
and the CNT parts. As the nano-antenna is operated in a
distinct sub-wavelength regime the minor lobes can be
further suppressed by increasing the length of CNT; or then
by optimizing the overall antenna shape. Both measures are
Fig. 5. (a) Distribution of the magnetic field strength in the antenna
structure with two 200 nm nanotubes at two different resonances. The
unit used in the color bar is A/m; (b) Magnetic field strength along the
axis of the structure at the two resonances; the axis is indicated as the
dashed line shown in Fig. 5(a).
Fig. 6. Far-field radiation characteristics of the structure with respect to
the x-y-plane (E-plane): (a) Two spheres without CNTs at resonance
wavelengths of 386 nm and 686 nm, the mode at 386nm is characterized
as multipolar mode, the mode at 686 nm is dipole-like mode; (b) Two
spheres with two 50 nm long semiconductor CNTs at the resonant
wavelengths of 381 nm, 525 nm (multipolar mode) and 842 nm (dipole-
like mode), respectively. The relevant dimensions of the two spheres are
R = 70 nm and g = 20 nm.
equivalent to the tailoring of the current flow in the proper
structure, which actually corresponds to a well-established
strategy in RF antenna design . In addition, the
narrowing in beam width that comes along with the sidelobe
suppression in the structure owes definitely to the CNT
antenna effect, which refers to the current modulations in
both antenna arms. The antenna performance can thus be
further improved by reducing the radius of both, the CNT
and the sphere. Due to the increased field enhancement,
higher radiation field strength
correspondingly tailored radiation pattern can be obtained.
together with a
antenna that encompasses a pair of sphere-on-pillar
nanostructure. EBIB and EMBB have been shown effective
for the fabrications of nanospheres on nanopillars. EBIB
does not need to make a contact of the nanotube to an
electrode; hence featured with simplicity. However, the
process is time-consuming and involves high energy beam.
To the materials with a high melting point, EMBB is a
method to be applied. The copper-filled CNTs in the
experiment has a 40-nm diameter, and the flowing has
continued for 17 s at the flow rate of 82.3 nm/s and the bias
threshold for the flowing is 2.4 V. The sphere-on-pillar
optical antenna has proven successful to interrelate free
space radiation with intense near fields in the feed point
addressing both, the aspect as receiving antenna as well as
the characteristics of the nano-emitter. The antenna
resonances, the associated field enhancement, as well as the
radiation pattern can be efficiently tuned since they strongly
vary with respect to structural parameters like the length and
the material properties of the nanotubes, the size of the
spheres, and the gap size. It is worth noting that the proposed
antenna design is not restricted to the materials we used
here. Other noble metals, such as gold, copper are applicable
to yield different antenna performance. Since our
nanostructure can be precisely tailored during fabrication
and post-processing stages, the remarkable tuning properties
allow us to promote it as a highly efficient optical nano-
antenna with tailored radiation characteristics applicable to
the various fields in the realm of functional nano-optics.
In conclusion, we proposed a new type of optical
 P. Muhlschlegel, H. J. Eisler, O. J. F. Martin, B. Hecht, and D. W.
Pohl, "Resonant optical antennas," Science, vol. 308, pp. 1607-1609,
 J. J. Greffet, "Nanoantennas for light emission," Science, vol. 308, p.
1561, Jun 2005.
 L. Novotny, "Effective wavelength scaling for optical antennas,"
Physical Review Letters, vol. 98, art. no. 266802, Jun 2007.
 J. Aizpurua, G. W. Bryant, L. J. Richter, F. J. G. de Abajo, B. K.
Kelley, and T. Mallouk, "Optical properties of coupled metallic
nanorods for field-enhanced spectroscopy," Physical Review B, vol.
71, art. no. 235420, Jun 2005.
 A. Sundaramurthy, P. J. Schuck, N. R. Conley, D. P. Fromm, G. S.
Kino, and W. E. Moerner, "Toward nanometer-scale optical
photolithography: Utilizing the near-field of bowtie optical
nanoantennas," Nano Letters, vol. 6, pp. 355-360, Mar 2006.
 K. Kempa, J. Rybczynski, Z. P. Huang, K. Gregorczyk, A. Vidan, B.
Kimball, J. Carlson, G. Benham, Y. Wang, A. Herczynski, and Z. F.
Ren, "Carbon nanotubes as optical antennae," Advanced Materials,
vol. 19, pp. 421-426, Feb 2007.
 J. Rybczynski, K. Kempa, Y. Wang, Z. F. Ren, J. B. Carlson, B. R.
Kimball, and G. Benham, "Visible light diffraction studies on
periodically aligned arrays of carbon nanotubes: Experimental and
theoretical comparison," Applied Physics Letters, vol. 88, art. no.
203122, May 2006.
 P. Avouris, Z. H. Chen, and V. Perebeinos, "Carbon-based
electronics," Nature Nanotechnology, vol. 2, pp. 605-615, Oct 2007.
 L. X. Dong, X. Y. Tao, L. Zhang, X. B. Zhang, and B. J. Nelson,
"Nanorobotic spot welding: Controlled metal deposition with attogram
precision from copper-filled carbon nanotubes," Nano Letters, vol. 7,
pp. 58-63, Jan. 2007.
 Y. Gan, "A review of techniques for attaching micro- and
nanoparticles to a probe's tip for surface force and near-field optical
measurements," Review of Scientific Instruments, vol. 78, art. no.
081101, Aug 2007.
 Y. H. Gao and Y. Bando, "Carbon nanothermometer containing
gallium - Gallium's macroscopic properties are retained on a miniature
scale in this nanodevice," Nature, vol. 415, p. 599, Feb 2002.
 L. Sun, F. Banhart, A. V. Krasheninnikov, J. A. Rodriguez-Manzo, M.
Terrones, and P. M. Ajayan, "Carbon nanotubes as high-pressure
cylinders and nanoextruders," Science, vol. 312, pp. 1199-1202, May
 X. Y. Tao, L. X. Dong, W. K. Zhang, X. B. Zhang, J. P. Cheng, H.
Huang, and Y. P. Gan, "Controllable Melting and Flow of β-Sn in
Flexible Amorphous Carbon Nanotubes," Carbon, vol. 47, pp. 3122-
3127, Nov. 2009.
 L. X. Dong, X. Y. Tao, M. Hamdi, L. Zhang, X. B. Zhang, A.
Ferreira, and B. J. Nelson, "Nanotube fluidic junctions: Inter-nanotube
attogram mass transport through walls," Nano Letters, vol. 9, pp. 210-
214, Jan 2009.
 G. E. Begtrup, W. Gannett, T. D. Yuzvinsky, V. H. Crespi, and A.
Zettl, "Nanoscale reversible mass transport for archival memory,"
Nano Letters, vol. 9, pp. 1835-1838, May 2009.
 X. D. Feng, D. C. Sayle, Z. L. Wang, M. S. Paras, B. Santora, A. C.
Sutorik, T. X. T. Sayle, Y. Yang, Y. Ding, X. D. Wang, and Y. S. Her,
"Converting ceria polyhedral nanoparticles into single-crystal
nanospheres," Science, vol. 312, pp. 1504-1508, Jun 2006.
 J. Hao and G. W. Hanson, "Optical scattering from a planar array of
finite-length metallic carbon nanotubes," Physical Review B, vol. 75,
art. no. 165416, Apr 2007.
 J. Zhou, T. Koschny, M. Kafesaki, E. N. Economou, J. B. Pendry, and
C. M. Soukoulis, "Saturation of the magnetic response of split-ring
resonators at optical frequencies," Physical Review Letters, vol. 95,
art. no. 223902, Nov 2005.
 C. A. Balanis, Antenna Theory: Analysis and Design. New York:
 I. Romero, J. Aizpurua, G. W. Bryant, and F. J. G. de Abajo,
"Plasmons in nearly touching metallic nanoparticles: singular response
in the limit of touching dimers," Optics Express, vol. 14, pp. 9988-
9999, Oct 2006.
 P. B. Johnson and R. W. Christy, "OPTICAL-CONSTANTS OF
NOBLE-METALS," Physical Review B, vol. 6, pp. 4370-4379, 1972.
 M. Danckwerts and L. Novotny, "Optical frequency mixing at coupled
gold nanoparticles," Physical Review Letters, vol. 98, art. no. 026104,
 E. Peter, P. Senellart, D. Martrou, A. Lemaitre, J. Hours, J. M. Gerard,
and J. Bloch, "Exciton-photon strong-coupling regime for a single
quantum dot embedded in a microcavity," Physical Review Letters,
vol. 95, art. no. 067401, Aug 2005.
 D. P. Fromm, A. Sundaramurthy, P. J. Schuck, G. Kino, and W. E.
Moerner, "Gap-dependent optical coupling of single "Bowtie"
nanoantennas resonant in the visible," Nano Letters, vol. 4, pp. 957-
961, May 2004.
 G. Chen, J. Wu, Q. J. Lu, H. R. H. Gutierrez, Q. Xiong, M. E. Pellen,
J. S. Petko, D. H. Werner, and P. C. Eklund, "Optical antenna effect in
semiconducting nanowires," Nano Letters, vol. 8, pp. 1341-1346, May