Jump to ContentJump to Main Navigation
Show Summary Details
The one-stop-shop for nano science.

nano Online

Physics, Chemistry and Materials Science at the Nanoscale

More options …

Search

Search publication

Open Access

1 Introduction

In recent days, optical antennas have attracted the attention of many researchers because of their various physical properties and application possibilities in nanophotonics [1], [2], [3], [4], [5]. Similar to radiofrequency (RF) antennas, optical antennas, which operate in the visible and near-infrared (NIR) frequency regions, can pick up energies efficiently from propagating electromagnetic (EM) waves at the frequencies of interest and transfer EM energies from one point to another [6], [7], [8]. In real applications, RF antennas are usually used as wave transmitters or receivers. However, we cannot produce optical EM fields, which are fundamentally light, from optical antennas, as we generate radiowaves from antennas in the RF region by oscillating electrons in metal rods. Therefore, optical antennas should interact with light preexisting in the surrounding environment [9]. Since mid-2000, various researchers have proven that the interactions of optical antennas with free light shows the same properties as RF antennas. The representative characteristics of optical antennas are resonances, polarizations, and well-defined far-field radiations [10], [11], [12]. Recently, the first electrically operated optical antennas fabricated using gold nanowires and nanospheres using nanotechnologies such as nanopatterning and nanomanipulation were reported by Kern et al. [13], Kullock et al. [14], and Dorfmuller et al. [15].

The earliest studies on optical antennas have dealt with rod-type antennas because the geometry of metallic rods is directly analogous to RF antennas, and we can analyze the properties of metallic nanorods more conveniently by adopting and comparing the properties of RF antennas [15], [16], [17], [18], [19]. However, in real applications of RF antennas, slot-type antennas are widely used due to their practicality [20], [21]. At this time, we have two considerations concerning optical slot antennas. One is that metallic nanoslots have the properties of antennas similar to metallic nanorods, and they can be regarded as optical slot antennas (Figure 1). The other is that optical slot antennas have the advantage of higher applicability than optical rod antennas, similar to RF slot antennas in the RF region. In 2014, we have proven experimentally that metallic nanoslots show the properties of optical slot antennas [22]. Here, we present precise experimental data on the resonance, polarization, and bidirectional far-field radiation as proofs that metallic nanoslots are optical slot antennas. By extending our results to Babinet’s principle in the optical region, we show that an optical slot antenna can be regarded as magnetic dipoles just as optical rod antenna can be regarded as electric dipoles in the optical region [23]. Additionally, we manufactured optical slot Yagi-Uda antennas that change the omnidirectional far-field radiation of optical slot antennas to unidirectional radiations [22]. By incorporating optical slot Yagi-Uda antennas in the metal plane, we present the applications of optical interconnection using optical antennas, namely, plasmonic via in multilayered plasmonic circuits and light-emitting diodes (LEDs) with antenna-integrated electrode for controlling the direction and polarization of light emissions [24], [25].

Figure 1:

Metallic structures and optical antennas.

(A) Schematics of metallic nanorods (left) and optical rod antennas (right). Arrows show random polarization vectors. (B) Schematics of metallic nanoslots (left) and optical slot antennas (right). Arrows show aligned polarization vectors in antenna radiations.

2 Optical slot antennas

2.1 Metallic nanoslots as optical slot antennas

According to the antenna theory, RF antennas show many characteristic properties depending on their geometry. Dipole antennas are the most simple and generic type among various kinds of antennas. Dipole antennas have the shape of simple metallic rods and show many characteristic properties based on their geometry. We can determine the three properties of resonance, polarization, and omnidirectional far-field radiation pattern as the representative properties of dipole antennas [20], [21]. As with RF antennas, previous reports on optical rod antennas showed that metallic nanorods have linear resonance between the resonant wavelength and the length of the rods [17], [18], [19], only one polarization perpendicular to their length [26], [27], [28], and omnidirectional far-field radiation patterns [29] (Figure 2). Especially, the optical Yagi-Uda antenna as a representative directional antenna was also demonstrated successfully by adding the auxiliary elements that reflect or direct radiation near the optical rod antenna [8], [16], [29], [30], [31]. Besides the above three properties, the field localization and enhancement are worth being commented as the representative property of optical antennas [32], [33]. It is well known that, because of the inhomogeneous field distribution of dipole-like antenna, electric fields (E-fields) are localized at subwavelength volumes near the antenna and changes the radiation decay rate of interacting EM fields. This property has been used in subwavelength high-resolution microscopy and sensing.

Figure 2:

Properties of optical antennas.

(A) Resonance property of the optical rod antenna, (B) polarization property of the optical rod antenna, and (C) omnidirectional far-field radiation pattern of the optical rod antenna. (A) Reprinted with permission from Ref. [10]. Copyright 2007 American Chemical Society. (B) Reprinted with permission from Ref. [27]. Copyright 2011 American Chemical Society. (C) Reprinted with permission from Ref. [29]. Copyright 2010 AAAS.

As we asked in Section 1, can metallic nanoslots operate as optical slot antennas similar to metallic nanorods in the optical region? To answer this question, we should investigate if metallic nanoslots have three representative properties of antennas, which are resonance, polarization, and omnidirectional far-field radiation pattern. In this section, we will present the experimental results of this investigation. First, we fabricated nanoslots with different lengths on an Ag metal plane by focused ion beam [FIB (FEI Helios NanoLab, Hillsboro, OR, USA)] milling method. Because of the oxidation problem of Ag metal, we deposited the MgF2 layer with 10 nm thickness as the passivation layer on Ag metal. We measured the transmission spectra of the fabricated nanoslots to identify the relation between the length of the nanoslots and the resonant wavelengths, which is called resonance. While measuring the transmission spectra, we also examined the polarization dependence by changing the direction of the incident polarization from parallel to perpendicular to the length of the slots. Finally, we measured the Fourier space images of light transmitted from the fabricated nanoslots to determine the far-field radiation pattern of the metallic nanoslots [34]. To verify our results, we also executed 3D finite-difference time-domain [3D-FDTD Solutions (Lumerical, Vancouver, British Columbia, Canada)] simulation to obtain the transmission spectra and far-field radiation images of metallic nanoslots used in the experiments [35]. By comparing the experimental and simulated data, we concluded that metallic nanoslots have the properties of resonance, polarization, and omnidirectional far-field radiation pattern; therefore, they are optical slot antennas.

2.1.1 Resonance

In metallic nanorods, the peak wavelengths of the transmission spectra increase with their lengths [15], [36]. This phenomenon implies that metallic nanorods have a linear relation between their lengths and the resonant wavelengths of the transmission. Therefore, to increase the efficiency of scattering of incident light by optical rod antennas, we should use nanorods that have lengths that are resonant with the wavelength of interest. To determine if resonance property exists in the case of metallic nanoslots, we fabricated various nanoslots with different lengths from 120 to 200 nm by the FIB milling method, as shown in Figure 3A. By measuring the transmission in front of the fabricated samples, we obtained the spectra shown in Figure 3B (left). By redrawing the peak wavelengths of the measured spectra as functions of the slot length, as shown in Figure 3B (right), we could clearly observe a linear relation between the transmission peak and the slot length, as seen in the case of metallic nanorods. By fitting the data, we arrive at the equation λ=1.37l+353.87, where λ and l correspond to the wavelength of the main peak in the spectrum and the slot length in nanometers, respectively. With the help of 3D-FDTD simulation, similar spectra of metallic slots were obtained, as shown in Figure 3C (left). As in the case of the experimental data, the main peaks of the calculated transmission spectra were redrawn as functions of the slot length, as shown in Figure 3C (right). We can see that the two results are very similar to each other, which proves that metallic nanoslots have a linear resonance between their lengths and the resonant wavelengths as metallic nanorods do in the optical region.

Figure 3:

Resonance property of metallic nanoslots.

(A) Scanning electron microscope (SEM) images of metallic nanoslots of different lengths and (B) transmission spectra measured in the fabricated nanoslots. (Left) Wavelengths of the main peaks in the measured transmissions were redrawn as functions of the slot length and (Right) red line shows the linear fitting. (C) Transmission spectra of the nanoslots calculated by 3D-FDTD. (Left) Wavelengths of the main peaks in the calculated transmissions were redrawn as functions of slot length and (right) ted line shows the linear fitting.

2.1.2 Polarization

It is known that optical rod antennas have only one polarization, which depends on the orientation of metallic rods [37], [38]. The E-fields of radiation from optical rod antennas are aligned in the direction of their lengths. To investigate if the scattered radiations from the metallic slots also have only one polarization, we measured the transmission spectra again while changing the polarization direction of the incident light, as shown in Figure 4A. We fixed the slot length of the Ag sample as 145 nm and defined the direction perpendicular to the slot length as 0° of incident polarization. Figure 4B (left) shows the transmission spectra measured while changing the polarization direction of the incident light. The graph shows that the maximum intensities of the transmission decreased as the direction of polarization is changed from 0° to 90°. By redrawing the maximum intensities of the spectra as functions of the polarization angle, as shown in Figure 4B (right), we can see that the graph shows variation in its dependence on the cosine square of the polarization angle. This cosine square dependence can be guessed simply from the vector sum of the E-field components. To verify this, we also executed 3D-FDTD simulation while changing the polarization angle of the incident light, as was performed in the experiment. The simulated results were very similar to the experimental results, as shown in Figure 4C. This implies that metallic slots also have only one polarization perpendicular to their lengths, similar to metallic rods. Using this polarization property, we can distinguish the scattered light from the incident light and excite one slot from among several slots, as presented in Section 3.1.

Figure 4:

Polarization property of a metallic nanoslot.

(A) SEM image of single slot with 145 nm length and schematics of the incident polarization angle, (B) spectra measured while changing the polarization angle of the incident light (left) and the graph of the maximum intensity as a function of the polarization angle (right), and (C) spectra simulated by the 3D-FDTD simulation (left) and the redrawn graph (right).

2.1.3 Far-field radiation pattern

When the geometry of the antenna is given, we can accurately calculate the radiation pattern from an antenna by solving Maxwell’s equations. The rod-type RF and optical antennas show omnidirectional radiation patterns that are exactly the same as those of an electric dipole. Based on this result, we can regard rod-type antennas as electric dipoles operating in each wavelength regime. As noted in Section 1, Babinet’s principle implies that metallic nanoslots have the possibility of being regarded as magnetic dipoles similar to the case of metallic nanorods. The investigation of far-field radiation patterns from metallic nanoslots is closely related with this possibility, but we will discuss and present the proof for this in the next section.

For measuring far-field radiation patterns from metallic nanoslots, we created a set-up for measuring images in real space and Fourier (momentum) space simultaneously, as shown in Figure 5A. When the laser is incident on the sample stage normally, the objective lens collects the light transmitted through metallic nanoslot. By changing the flipping mirror, we can choose the direction of the transmitted light into an Newton electron multiplying charge-coupled device (EMCCD) (Andor, Belfast, UK) or INFINITY3 (Numenera Corp., Ottawa, Ontario, Canada). The Fourier space image is obtained by an EMCCD because a lens1 has the role of Fourier transformation that changes the light intensity distribution in a real space into a Fourier (momentum) space. On the contrary, a real space image is obtained by a CCD because a lens2 in front of the CCD changes transformed Fourier space image into a real space again. To extinguish far-field radiation pattern transmitted from a metallic nanoslot from the laser incidence, we used a polarizer perpendicular to the slot length in front of the EMCCD. We can interpret far-field radiation patterns measured in Fourier space as follows. When a sample that radiates light is located at the origin of the hemisphere, as shown in Figure 5B, the light from the sample propagates radially and is distributed on the surface of a 3D hemisphere. If the EM field distribution on the 3D hemisphere is projected on a 2D plane of a CCD, the obtained image in the 2D plane corresponds to a Fourier space image of the far-field radiation. Using a lab-made measurement set-up, we obtained the Fourier space image of the far-field radiation from a metallic slot of 145 nm length. Figure 5C shows the measured far-field radiation image of a single slot. The intensity is normalized by its maximum intensity. The measured Fourier space image of Figure 5C was obtained in the top side of the sample, not in the substrate side, as shown in Figure 2C. When the far-field patterns are measured in the substrate side, there is a critical angle where the radiation cannot propagate inside a certain solid angle [39]. In our experiments, far-field patterns were obtained in the upper side (air) of the sample and there is no constraint of critical angle collection. Therefore, field intensities were collected until the solid angle (71.8°) of objective lens with numerical aperture of 0.95. We can clearly see that the far-field radiation pattern of single slot is omnidirectional in the direction perpendicular to the slot length.

Figure 5:

Fourier space measurement of far-field radiation from a metallic nanoslot.

(A) Schematics of real and Fourier space image measurement set-up and (B) geometry of far-field radiation pattern and Fourier space image. The transmitted intensity of the red circle indicated by the red arrow and positioned at the coordinate qf(θ, ϕ) on 3D hemisphere is projected at the position of black circle in the 2D CCD plane. (C) Measured Fourier space image of the far-field radiation from a metallic nanoslot. The intensity was normalized by the intensity of incident light.

Based on the above experimental results of resonance, polarization, and omnidirectional far-field radiation pattern, we can conclude that metallic nanoslots operate as optical slot antennas similar to metallic nanorods in the optical region.

2.2 Optical slot antennas as magnetic dipoles

In the RF region, Babinet’s principle describes that EM fields radiated from a rod antenna are identical to those from the hole of the rod antenna, which is a slot antenna, except that the E-fields and magnetic fields (H-fields) replace each other [40]. As discussed in the previous section, the measured far-field radiation patterns of metallic slots are exactly the same as those of metallic nanorods. This means that Babinet’s principle can be extended to the optical region with some modifications because the Ohmic losses of metal materials are not negligible in the optical region. From the complementarity of Babinet’s principle, we can deduce the possibility that metallic slots are regarded as magnetic dipoles in the optical region because it is well known by many researchers that metallic nanorods can be regarded as electric dipoles in the optical region [15], [23]. In this section, we will present experimental proofs of the above possibility, as shown in Figure 6.

Figure 6:

Optical rod and slot antenna.

(A) Schematics of optical rod antenna as an electric dipole. (+) and (−) signs represent electron and hole charges, respectively. The blue arrow beside a metallic nanorod shows a virtual electric dipole formed in a metallic rod and the blue curved arrows show virtual E-field lines forming a loop generated by charge distributions in the nanorod. (B) Schematics of optical slot antenna as a magnetic dipole. Green arrow lines represent virtual current loops formed in a nanoslot and the red arrow shows the resultant virtual magnetic dipole due to the current loops. Red curved arrows show virtual H-field lines forming a loop generated by virtual magnetic dipole inside the nanoslot.

To verify the equivalence of metallic nanoslots and magnetic dipoles, we simulated and compared the far-field radiation patterns and field components of a virtual magnetic dipole and Ag slot with 145 nm length at the resonant wavelength of 660 nm using the 3D-FDTD method. We used near-to-far transformation method in FDTD to obtain far-field radiation patterns from the near fields of the FDTD results [41]. The near-to-far transformation was executed at a distance of r=1 m from the centers of the samples. The calculated far-field radiation patterns are shown in Figure 7. We plotted (from top left to bottom right) (1) |E|2, (2) |E|2 with an x-polarizer, (3) |E|2 with a y-polarizer, (4) |Ex|, (5) |Ey|, and (6) |Ez| in the far-field regime. Based on Maxwell’s equations, the H-field components can be calculated by considering the curl values of the E-field components. Therefore, we can determine that the H-field components must be parallel to the length of the dipole because the E-field components are perpendicular to its length, as expected from Babinet’s principle. The amplitudes of the far-field electric components are |Ex|≥|Ez||Ey|, and the contribution of the y-component of the E-field to the total intensity is negligible. Therefore, we can see in Figure 7A(1) that the total intensities of the simulated E-fields are exactly similar. Additionally, metallic nanoslots localize H-fields near the edge of slot as metallic nanorods do. As we have commented in Section 2.1, E-fields are localized at the edge of the dipole-like rod antenna. Similarly, we can see that H-fields are localized at the edge of the optical slot antennas, as shown in Figure 7C [42], [43]. These imply that the metallic nanoslots can be regarded as magnetic dipoles in the optical region.

Figure 7:

Equivalence of a metallic nanoslot and a magnetic dipole.

(A) Simulated electric near-field of a virtual magnetic dipole at 660 nm wavelength (left). Far-fields transformed from near field (right). (B) Simulated electric near-field of the Ag metal slot at 660 nm wavelength (left). Far-fields transformed from near field (right). The white arrows indicate the polarization of the detectors, and each plot is normalized. (C) Simulated E-field intensity of an optical rod antenna (left) and H-field intensity of an optical slot antenna (right). (C) Reprinted with permission from Ref. [42]. Copyright 2008 Wiley. and Ref. [43]. Copyright 2009 Optical Society of America.

Generally, optical rod antennas are driven by the E-fields parallel to the antenna’s length, as shown in Figure 8A. Similarly, we have driven optical slot antennas by the E-fields, which is perpendicular to the antenna’s length in the former experiments. However, if the Babinet’s principle is available in the optical regime, it is more natural to drive optical slot antenna by the H-fields parallel to the slot antenna’s length, as shown in Figure 8B. In 2016, Lee et al. reported and proved experimentally that the dominant incident components in optical slot antennas are H-fields parallel to the antenna’s length, not E-fields perpendicular to the length in accordance with Babinet’s principle [44], [45]. They measured the scattered intensities while changing the incident polarization and angles and calculated the ratio of the incident H-field and E-field contributions to the scattered intensities. When the p-polarization is incident to the metallic nanoslot obliquely, the tangential E-fields are very weak and the incident field configuration is very similar to Figure 8B (top). On the contrary, when the s-polarization is incident obliquely, the incident field configuration gets similar to Figure 8B (bottom). From the measured scattering intensities shown in Figure 8C and D, they proved that the scattering intensities for the p-polarization are much higher than those for the s-polarization and the ratio of |Is/Ip| corresponds to 1/|ε(λ)|, where ε(λ) is the relative permittivity of the metal. That means that the contributions of the incident E-field and H-fields to the scattering intensity are in the ratio of 1:|ε(λ)|. Based on these experimental results, we can deduce that optical slot antennas can be regarded as real magnetic dipoles in the optical region.

Figure 8:

Dominant incident fields in optical slot antennas.

(A) Schematics of rod antenna excitation by incident E-fields parallel in antenna length. (B) Schematics of slot antenna excitation by incident H-fields parallel in antenna length. Because a slot antenna is the Babinet complementary structure of a rod antenna, the roles of the H-fields and E-fields are reversed. (C) Polar plot of scattered intensity when the p-polarization (top) and s-polarization (bottom) are incident. The blue and red arrows represent the incident tangential E-fields and H-fields. (D) Ratio of scattering intensities between the s-polarization and the p-polarization as functions of the incident angle. The circles, rectangles, and triangles are experimental data. The measured ratios of |Is/Ip| can be fitted by 1/|ε(λ)|. Reprinted with permission from Ref. [38]. Copyright 2016 Springer Nature.

3 Optical slot Yagi-Uda antennas

We have known from the above experiments that optical slot antennas radiate omnidirectionally, not unidirectionally. However, unidirectional radiations are important and inevitable in real applications of the antenna for efficient beaming or receiving of light [2], [11], [46], [47], [48], [49], [50]. In the RF region, unidirectional radiations could be obtained by combining Yagi-Uda antennas with a reflector, a feed, and directors [21]. In 2010, Kosako et al. [8] and Hofmann et al. [51] adapted the geometry of RF Yagi-Uda antennas to optical antennas and proved successfully that unidirectional radiations could also be obtained in the optical region as with RF antennas as shown in the Figure 9. Based on these results, we investigated if unidirectional radiations achieved with rod-type antennas can also be achieved with optical slot antennas and if, consequently, the principle of Yagi-Uda antenna design applies to optical slot antennas.

Figure 9:

Optical rod Yagi-Uda antennas.

(A) Schematics of optical rod Yagi-Uda antenna (left) and the directional radiation (right). (B) Optical image of the coupling between quantum dot radiation and optical Yagi-Uda antenna (left) and the measured Fourier space image of the directional radiations from this coupling (right). (A) Reprinted with permission from Ref. [8]. Copyright 2010 Springer Nature. (B) Reprinted with permission from Ref. [28]. Copyright 2010 AAAS.

3.1 Slot-slot geometry

Yagi-Uda antennas are made of three components, namely, the reflector, the feed, and the directors. For adapting the geometry of Yagi-Uda antennas to the optical slot directional antennas, we should create multislots by combining the slots. Therefore, we have to study the interaction between the slots or at least between two slots to achieve unidirectional radiations using the Yagi-Uda geometry. For ease of analysis, we focused our study on the interaction between two slots.

To make directionality from multislot antenna, we have to excite only one slot among multislots and reflect or direct dipole radiation generated from the excited slot. This is called feeding in the RF antenna. However, light transmits through both slots in slot-slot geometry when it is incident from the backside of the sample composed of two slots and we cannot excite only one slot. To solve this problem, we rotated one slot by 45°, as shown in Figure 10A[22]. Based on the polarization property of the optical slot antenna wherein only light with polarization perpendicular to the slot length can transmit through the slot and light with polarization parallel to the slot length cannot, we could measure only the light transmitted through the unrotated slot using a polarizer parallel to the length of the rotated slot and placed in front of the slot-slot sample. By positioning the polarizer in the direction parallel to the rotated slot, light cannot transmit through the rotated slot and only the unrotated slot can radiate light omnidirectionally. For a given situation, the unrotated slot located on the right side fulfills the role of a feed and the rotated slot located on the left side can work as a reflector or a director depending on its length (L) and distance from the unrotated slot (D). When the phase difference between two slots is π/2, the destructive interference happens because of the wave’s round trip. The transmitted waves from the unrotated feed slot radiate omnidirectionally and they are reflected by the rotated slot. Resultantly, the waves radiated from the feed slot meet at feed position again having the round-trip phase difference of π. In this situation, the rotated slot works as a reflector. When the phase difference between two slots is π, the total round-trip phase would be 2π and the constructive interference happens. In this condition, the rotated slot has the role of a director. Simulation results obtained in simple model with various phase differences shows this phenomenon well, as shown in Figure 10B. To fabricate optical slot Yagi-Uda antennas with good unidirectionality, it is essential to determine the optimal L and D that make the rotated slot a good reflector or director. We can optimize the parameters by calculating the phase difference between the two slots.

Figure 10:

Optical multi-slot Yagi-Uda antennas.

(A) Schematics of the slot-slot geometry. L and D correspond to the length of the rotated slot and the distance from the unrotated slot, respectively. (B) Simple model calculation of various phase differences. (C) Ex-field intensities calculated by 3D-FDTD while changing L and D. The intensities are normalized by the maximum intensity. (D) Contour map of phase difference calculated by 3D-FDTD while changing L and D. The scale bar shows the phase difference between two slots in degrees. The red line corresponds to 90° phase difference. (E) Far-field radiation patterns calculated by 3D-FDTD while changing L and D. The intensities are normalized by the maximum intensity. (F) Schematics (top left) and SEM image of fabricated five-slot directional antenna (bottom left) and measured far-field radiation pattern of the fabricated sample (right). The white bar in SEM corresponds to 100 nm and the intensity of far-field radiation pattern is normalized by its maximum intensity.

In single slot antenna, only the x-components of the E-fields are dominant due to the polarization property described in Section 2.1.2. Using the 3D-FDTD simulation, we could calculate the phase of the Ex-field in a slot and observe that the phases of the Ex-fields are almost uniform in the slot structure because the intensity of the Ex-fields are almost uniform in the slot structure, as shown in Figure 10C. Therefore, we could define the phase of the Ex-field at the center of the slot as the phase of the entire slot antenna. After calculating the phases of two slots from the Ex-fields, the phase difference could be obtained by abstracting two phases. Figure 10D shows the contour map of the phase difference between the two slots calculated while changing the L and D parameters. When the phase difference corresponds to π/2, which is drawn as the red line in Figure 10D, the rotated slot reflects the light transmitted through the unrotated slot because of destructive interference. On the contrary, as the phase difference approaches π, the rotated slot directs the light transmitted through the unrotated slot toward the direction where the rotated slot is located because of constructive interference. To verify the analysis, we also calculated the far-field radiation patterns using near-to-far-field transformation method of the 3D-FDTD simulation while changing the L and D parameters. Figure 10E shows the calculated far-field radiation patterns with each L and D. In Figure 10E, we can see that the far-field radiation patterns are directed toward the right side when the parameters are coincident with the red line in the contour map shown in Figure 10D. This shows that phase analysis is a powerful tool for designing an antenna with directional radiation.

Based on the phase analysis, we chose the rotated slot with 155 nm length and 130 nm distance from the slot as the reflector and the rotated slot with 95 nm length and 130 nm distance from the slot as the director in the fabrication of the optical slot Yagi-Uda antenna. To increase the directionality, we added three rotated directors beside the unrotated feed slot and realized the optical slot Yagi-Uda antenna composed of one feed, one reflector, and three directors on a metal plane with 180 nm thickness using FIB milling. We measured the far-field radiation pattern of the fabricated sample using the set-up shown in Figure 5A. Figure 10F shows the measured result and we can see that the radiation pattern is directed to the right side and well coincident with the calculated patterns shown in Figure 10E. The intensity of the far-field radiation pattern is normalized by its maximum intensity. As a figure of merit for the directionality, we used a front-to-back (FB) ratio that is defined as the ratio of the power of the maximal radiation lobe (forward lobe) to the power of the lobe formed on the opposite side (back lobe) [29]. The FB ratio of measured Fourier space image in Figure 10F corresponds to 8.5 [22]. Thus, we fabricated an optical slot Yagi-Uda antenna and showed directionality successfully by adopting the geometry of the Yagi-Uda antenna.

3.2 Slot-groove geometry

The optical slot Yagi-Uda antenna with slot-slot geometry has two disadvantages compared to the conventional RF Yagi-Uda antenna. One is that we have to use the polarizer parallel to the tilted-slot direction when measuring the radiation for distinguishing the scattered light from the incident light. The other is the subtle tilting of the far-field radiation pattern direction. As shown in the calculation and measurement of Figure 10, the direction of the far-field radiation from the slot-slot geometry is slightly rotated from the x-direction, which is the direction of radiation from the single slot antenna, because of the rotated auxiliary slots. To avoid these two problems, we used the slot-groove geometry shown in Figure 11A[24]. This is the antenna version of the slit-groove structure demonstrated in plasmonics [50], [52], [53], [54], [55], [56], [57], [58]. We could avoid the radiation from multielements using the groove structure as the director or reflector because light cannot transmit through the groove structure. Therefore, it is not necessary to rotate the auxiliary elements for antenna feeding; thus, we can avoid the tilting of the far-field radiation patterns.

Figure 11:

Optical slot-groove Yagi-Uda antennas.

(A) Schematics of the slot-groove geometry. L and D correspond to the length of the groove and the distance from the slot, respectively. (B) Ex-field intensities calculated by 3D-FDTD while changing L and D. The intensities are normalized by the maximum intensity. (C) Contour map of phase difference calculated by 3D-FDTD while changing L and D. The scale bar shows the phase difference between the slot-groove in degrees. The red line corresponds to 90° phase difference. (D) Far-field radiation patterns calculated by 3D-FDTD while changing L and D. The intensities are normalized by the maximum intensity. (E) Schematics (top left) and SEM image (bottom left) of fabricated groove-slot-groove directional antenna and measured far-field radiation pattern of the fabricated sample (right). The white bar in SEM corresponds to 100 nm and the intensity of far-field radiation pattern is normalized by its maximum intensity.

As in the case of the slot-slot geometry, we calculated the phase difference between the slot and groove structure while changing the length of the groove (L) and the distance from the slot (D) and determined the optimal parameters for good directionality. Figure 11C shows the calculated phase difference between the slot and the groove. As in the slot-slot case, when the phase difference corresponds to π/2, the groove reflects light transmitted through the slot because of destructive interference. Additionally, when the phase difference approaches π, the groove directs light transmitted through the slot toward the direction where the slot is located because of constructive interference. The phase difference of π/2 is drawn as a red line in the contour map of Figure 11C. To verify the reflection of the groove, we also calculated the far-field radiation patterns while changing L and D. We can see in Figure 11D that the far-field radiation patterns are directed toward the right side when the parameters are coincident with the red line shown in Figure 11C. It is worth commenting that the direction of far-field radiation from the slot-groove structure is exactly aligned with the x-direction unlike the case of the slot-slot structure. We chose the groove with 260 nm length and 120 nm distance from the slot as the reflector and the groove structure with 110 nm length and 120 nm distance from the slot as the director based on the calculated maps of the phase difference and far-field radiation. The samples were fabricated by FIB milling on the Ag metal plane with thickness of 300 nm, and the entire structure was composed of a groove reflector, a slot feed, and a groove director. We measured the far-field radiation patterns from the fabricated samples using the set-up for measuring the Fourier space image shown in Figure 5A. Figure 11E shows the measured far-field radiation pattern directed toward the right side, which is exactly similar to the calculated one. The FB ratio of measured Fourier space image in Figure 11E corresponds to 5.57. However, it is worth noting that an FB ratio of 5.57 was obtained from the sample composed of three elements (reflector-feed-director), not five elements (reflector-feed-three directors) as the five-slot Yagi-Uda antenna in the above section. For a direct comparison between slot-slot and slot-groove geometries, we have measured the FB ratio of three-slot Yagi-Uda antenna and it corresponded to 3.08. Thus, we could conclude that the slot-groove geometry showed directionality superior to slot-slot geometry. Additionally, the measured pattern was obtained without the detection polarizer in front of the sample, and we could determine that the directional radiation was fully polarized in the x-direction by measuring the polarization. Because of the lens effect in Fourier space measurement set-up, the intensity of the far-field radiation was concentrated at the edge of the circle.

4 Applications of optical slot antennas

In terms of applicability, optical slot antennas have some advantages compared to optical rod antennas. Optical rod antennas should be isolated from other conductor components and passivated by insulating regions because when it is in touch with other plasmonic components made of metal the resonance condition collapses and the antenna fails to work. On the contrary, optical slot antennas are placed on a metal plane; therefore, any plasmonic component can be fabricated easily at any place on the metal plane, except for the area where the optical antenna is located. Therefore, optical slot antennas have high potential for integration and combination with other plasmonic components. Here, we present two integrations of optical slot antenna. One is the integration to metal-insulator-metal (MIM) plasmonic waveguides as passive components and the other is the integration to the electrodes of LEDs as active components.

4.1 Integration to plasmonic waveguides

Figure 12A shows the schematics of optical slot antennas integrated to MIM plasmonic waveguides. The sample was fabricated as follows. First, an Ag metal layer was deposited by an evaporator and a long slit for surface plasmon (SP) generation was fabricated by FIB milling. After slit fabrication, a thin dielectric layer with 50 nm thickness was spin coated and then the top Ag metal layer was deposited for slot antenna formation. Optical slot antennas were also fabricated by FIB milling on the top metal plane. When light is incident on the slit made in the bottom metal layer, transverse magnetic SP waveguide modes are formed in the insulator plane of the MIM structure, which propagate toward the direction perpendicular to the slit length [59]. When the propagating SP waveguide modes meet the slot antennas formed in the top metal layer, they radiate from the optical slot antennas to the free space, thereby generating omnidirectional dipole-like patterns. If the optical slot antennas fabricated on the top metal layer are replaced with the optical slot Yagi-Uda antennas, unidirectional radiation with the direction where the slot-groove structure forces light to go can be generated. Figure 12B shows the measured far-field radiation patterns from the fabricated optical slot Yagi-Uda antennas integrated to MIM plasmonic waveguides. We could see that the measured direction of the far-field radiation patterns was changed depending on the orientation of the slot-groove structure.

Figure 12:

Integration of optical slot antennas with waveguides.

(A) Schematics of optical slot antennas integrated to MIM plasmonic waveguides. The red arrows represent incident light launched from the bottom metal plane and radiation from the slot antennas to free space. (B) Schematics of optical slot Yagi-Uda antennas integrated to MIM waveguides (top left) and SEM images (top right) of the slot-groove structure that is parallel, rotated by 45°, and by −45° to the slit. The bottom images are the respective measured far-field radiation patterns. For ease of observation, thick white arrows are drawn on the measured images. White lines in the SEM images correspond to 200 nm.

These waveguide-integrated slot antennas have the possibility of an optical interconnection that transports signals in free space. Many researchers demonstrated that directional optical antennas can work as good receivers collecting signals in the desired direction [6], [60], [61]. Therefore, we can determine the optical interconnection using the optical slot Yagi-Uda antennas integrated to waveguides, as shown in Figure 13A. Huang et al. [62], [63] demonstrated that the slot antennas can radiate polarized and directional emissions in integrated plasmonic nanocircuits. Additionally, we suggest a “plasmonic via,” which is the equivalent of an electronic via in nanophotonics [64]. Via is an electrical connection between the layers in a physical electronic circuit that goes through the plane of one or more adjacent layers. In a multilayered structure comprising several stacked MIM plasmonic waveguides, the slot antennas can have the role of a connector between the vertical layers. Especially, with the help of directional radiation from optical slot Yagi-Uda antennas, we can efficiently change the direction of connection between the vertical layers. Figure 13B shows the schematics of the “plasmonic via” using the optical slot Yagi-Uda antennas integrated to MIM plasmonic waveguides and directional connections calculated by the 2D-FDTD method. From the simulation results, we could see that the SP waveguide modes were connected to another layer through the slot antennas and the coupled modes were well directed by the slot-groove structure. This indicates that the optical slot Yagi-Uda antenna integrated to MIM waveguides works as a via, specifically a plasmonic via, in a multilayered structure.

Figure 13:

Optical interconnections using optical slot antennas.

(A) Schematics of optical interconnection using optical slot antennas integrated to MIM plasmonic waveguide and (B) schematics of plasmonic via in multilayered structure with stacked MIM plasmonic waveguides: (left) 2D-FDTD simulation result showing directional coupling in a multilayered structure (right).

4.2 Integration to the electrodes of LEDs

Besides the integration to passive optical components such as waveguides, optical slot antennas can be combined with active devices such as LEDs, as shown in Figure 14A[25]. To integrate optical slot antennas with LEDs, we prepared colloidal quantum dot (CQD) LEDs with top metal n-contact layer and bottom indium tin oxide (ITO) p-contact layer. The fabricated LEDs have electroluminescence (EL) with emission peak of 614 nm wavelength and full-width at half-maximum (FWHM) of approximately 33 nm [65], [66], [67]. In this structure, the top metal layer simultaneously functions as the electrode of the LEDs and the base region, where the optical slot antennas are fabricated. Generally, CQD LEDs mostly use Al metal as the top electrode because it has low work function of approximately 4.24 eV compared to that of the ITO layer used as the bottom electrode, which is approximately 4.5 eV [68], [69], [70], [71]. However, Al metal is very absorptive in the visible range, including 614 nm of red CQD emission peak wavelength, and it is not suitable as the base metal for antenna fabrication [72]. With regard to optical property, Ag metal is a good candidate for base metal of the antenna because of its low absorption coefficient in the visible and NIR regions. Unfortunately, the work function of Ag metal corresponds to approximately 4.74 eV and is larger than that of the ITO bottom electrode. Therefore, carrier injection between the Ag electrode at the top of the LED and the ITO electrode at the bottom of the LED will not be efficient. To use the superior property of Al and Ag metals in the electrical and optical regimes, respectively, we fabricated a multilayer composed of Al with 50 nm thickness and Ag with 240 nm thickness and used it as the top contact electrode and the base metal for antenna fabrication. The entire structure is same as that of CQD LEDs used in the previous reports, except that the top electrode of Al metal is replaced with the multilayer comprising Al metal and Ag metal, as shown in Figure 14B. We operated the fabricated LEDs by supplying the voltage between the two electrodes and obtained the representative current-voltage curve, intensity-voltage curve, and EL spectra in the same manner as measured for conventional LEDs, as shown in Figure 14C and D [73], [74], [75], [76].

Figure 14:

Integration of optical slot antennas with light emitting diodes.

(A) Schematics of LEDs with antenna-integrated electrodes, (B) schematics of work functions in multilayered metal electrodes of LEDs, (C) current-voltage (black) and intensity-voltage (red) curves measured while increasing the voltage, and (D) EL spectra measured while increasing the voltage.

To confirm the coupling of LED emissions with optical antennas, we measured the transmission spectra and Fourier space images of LED emissions. The measured spectra of the LED emissions without antennas showed Gaussian forms with peak wavelength of 614 nm and FWHM of 33 nm, as shown in Figure 14D. On the contrary, we could see that the peak wavelength of the LED emissions with antenna-integrated electrodes changed from 614 to 600 nm due to the strong resonance of the optical slot antenna. From the linear relation between resonance wavelength and antenna length, we could guess that the effective length of the slot antenna engraved on the top metal electrode corresponds to approximately 172 nm because of trapezoidal milling in the FIB method. The peak change in the measured spectrum proves that the LED emissions were well coupled with the resonance of the slot antenna. Additionally, we could observe in Figure 15A and B that the emissions from the LEDs with antenna-integrated electrodes were polarized in the direction perpendicular to the antenna length by measuring the EL spectra with a polarizer placed in front of the samples. As proven in Section 2.1.2, optical slot antennas transmit only light with polarization perpendicular to their lengths. The polarized emissions from the LEDs with antenna are another proof that EL of LEDs with antenna is well coupled with slot antenna fabricated on the top metal electrode. To determine the far-field radiation patterns of LED emissions, we executed Fourier space imaging experiments already presented in Section 2.1.3. For a direct comparison of modification by antenna, we have measured a Fourier space image of emission from LED without antenna before measuring the EL from antenna-integrated LED. Figure 15C shows a typical Lambertian emission of LED as we expected. On the contrary, the measured Fourier space images of emissions from LEDs with single slot antenna showed omnidirectional dipole-like radiation patterns, whereas the emissions from LEDs with slot-groove antenna were unidirectional and directed to the opposite side where the grooves were located, as shown in Figure 15D and E. Apart from the polarized spectra, these measurements clearly showed that LED emissions were coupled to optical slot antennas and they re-radiated forming omnidirectional or unidirectional patterns depending on the structure of the antenna. From the above-mentioned results, we could know that emissions from LEDs with antenna-integrated electrodes were changed by three properties compared to emissions from LEDs without antenna. These properties are resonance, polarization, and far-field radiation patterns that originate from the properties of optical slot antennas. We expect that antenna-integrated LEDs can be promising optical sources in 3D display panels or holography that requires polarized light and well-controlled beam direction for the future.

Figure 15:

Properties of emission from antenna-integrated LEDs.

(A) EL spectra from antenna-integrated LEDs measured while changing the direction of detection of polarization. (B) SEM image of antennas with the x- and y-directions (top left) and optical image of EL from LEDs with antenna-integrated electrodes (top right). Optical images of EL from LEDs measured by the polarizer along the x-direction (bottom left) and y-direction (bottom right). (C) Measured Fourier space image of emissions from LEDs without antenna. (D) Measured Fourier space image of emissions from LEDs with single slot antenna. (E) Measured Fourier space image of emissions from LEDs with slot-groove antenna. The white lines in the SEM images correspond to 200 nm.

5 Conclusions and perspectives

In conclusion, we presented experimental results that showed that metallic nanoslots have the properties of resonance, polarization, and dipole-like far-field radiation patterns by measuring transmission spectra and Fourier space imaging. From these results, we conclude that metallic nanoslots can be regarded as slot antennas in the optical frequency. Based on Babinet’s principle that E-fields and H-fields in complementary structures replace each other, we concluded that optical slot antennas could be physically regarded as magnetic dipoles, and the intensity of tangential H-fields are higher by 1/|ε(λ)| than those of E-fields inside the slot structure. Additionally, we fabricated optical slot Yagi-Uda antennas with directional radiation by adopting the geometry of RF Yagi-Uda antenna. By analyzing the phase difference between two components, we could identify that interference plays an important role in ensuring directionality and realizing optimal structures with good directionality. Compared to optical rod antennas, slot antennas have the advantage of easy integration to other plasmonic or photonic components. To prove these possibilities, we presented experimental results of integrating optical slot antennas with MIM plasmonic waveguides and with the electrodes of LEDs. In slot antenna integrated to plasmonic waveguides, we showed that the propagating SP waveguide modes are well coupled with the slot antennas and the direction of radiations from the slot antennas is well controlled by the orientation of the Yagi-Uda slot antennas. We also suggested that slot antennas integrated to waveguides have the possibilities of optical interconnection in free space and as plasmonic via in multilayered photonic circuits. As another application, we showed experimentally that slot antennas integrated to the electrodes of LEDs could make the LED emissions linearly polarized and directed in nanometer scales. Based on these results, we expect that optical slot antennas will be integrated further with various photonic devices and play an important role in integrated nanocircuits in the future.

Despite the experimental results and applications presented here, optical slot antennas have some problems that need to be solved. The low coupling efficiency of optical antennas is the most important of these problems. Due to size in nanometer scale and metallic ohmic loss, optical plasmonic antennas have low efficiency in interaction with preexisting light. To avoid this problem, many researchers suggested and demonstrated dielectric optical antennas made of dielectric materials instead of metals and demonstrated their applications to photonic devices [77], [78], [79], [80]. Recently, dielectric metasurfaces, which change the amplitudes, polarizations, and wave vectors of light at the interfaces in nanometer scales and are composed of dielectric antenna arrays with different parameters or orientations, showed efficiencies of approximately 80% in hologram imaging and visible metalens [81], [82], [83], [84], [85]. Another important problem to be solved in optical antennas is the active tuning of resonant frequency during device operations. To change the frequency of optical antennas, several approaches, including carrier density modulation in high doping dielectrics [86], [87], usage of phase change materials [88], [89], [90] and liquids crystals [91], [92], and work function modulation of 2D materials [93], [94], have been suggested and investigated by many research teams. As in the case of rod antennas, if various methods for solving the problems in optical antennas, including dielectric antennas and tuning ability, are addressed in optical slot antennas, we expect that optical slot antennas will have more possibilities for practical applications and will become an essential and important element in several fields of nanophotonics.

References

  • [1]

    Mühlschlegel P, Eisler HJ, Martin OJ, Hecht B, Pohl DW. Resonant optical antennas. Science 2005;308:1607–9.Google Scholar

  • [2]

    Alda J, Rico-García JM, López-Alonso JM, Boreman G. Optical antennas for nano-photonic applications. Nanotechnology 2005;16:S230–4.Google Scholar

  • [3]

    Park QH. Optical antennas and plasmonics. Contemp Phys 2009;50:407–23.Google Scholar

  • [4]

    Novotny L, Van Hulst N. Antennas for light. Nat Photon 2011;5:83–90.Google Scholar

  • [5]

    Biagioni P, Huang JS, Hecht B. Nanoantennas for visible and infrared radiation. Rep Prog Phys 2012;75:024402.Google Scholar

  • [6]

    van Zanten TS, Lopez-Bosque MJ, Garcia-Parajo MF. Imaging individual proteins and nanodomains on intact cell membranes with a probe-based optical antenna. Small 2010;6:270–5.Google Scholar

  • [7]

    Taminiau TH, Stefani FD, van Hulst NF. Enhanced directional excitation and emission of single emitters by a nano-optical Yagi-Uda antenna. Opt Express 2008;16:10858–66.Google Scholar

  • [8]

    Kosako T, Kadoya Y, Hofmann HF. Directional control of light by a nano-optical Yagi-Uda antenna. Nat Photonics 2010;4:312–5.Google Scholar

  • [9]

    Bharadwaj P, Deutsch B, Novotny L. Optical antennas. Adv Opt Photonics 2009;1:438–83.Google Scholar

  • [10]

    Taminiau TH, Moerland RJ, Segerink FB, Kuipers L, van Hulst NF. λ/4 resonance of an optical monopole antenna probed by single molecule fluorescence. Nano Lett 2007;7:28–33.Google Scholar

  • [11]

    Taminiau TH, Stefani FD, Segerink FB, Van Hulst NF. Optical antennas direct single-molecule emission. Nat Photonics 2008;2:234–7.Google Scholar

  • [12]

    Knight MW, Sobhani H, Nordlander P, Halas NJ. Photodetection with active optical antennas. Science 2011;332:702–4.Google Scholar

  • [13]

    Kern J, Kullock R, Prangsma J, Emmerling M, Kamp M, Hecht B. Electrically driven optical antennas. Nat Photonics 2015;9:582–6.Google Scholar

  • [14]

    Kullock R, Grimm P, Ochs M, Hecht B. Directed emission by electrically-driven optical antennas. SPIE Opto 2018;10540:9.Google Scholar

  • [15]

    Dorfmuller J, Vogelgesang R, Khunsin W, Rockstuhl C, Etrich C, Kern K. Plasmonic nanowire antennas: experiment, simulation, and theory. Nano Lett 2010;10:3596–603.Google Scholar

  • [16]

    Li J, Salandrino A, Engheta N. Shaping light beams in the nanometer scale: a Yagi-Uda nanoantenna in the optical domain. Phys Rev B 2007;76:245403.Google Scholar

  • [17]

    Novotny L. Effective wavelength scaling for optical antennas. Phy Rev Lett 2007;98:266802.Google Scholar

  • [18]

    Holloway CL, Love DC, Kuester EF, Salandrino A, Engheta N. Sub-wavelength resonators: on the use of metafilms to overcome the λ/2 size limit. IET Microw Antennas Propag 2008;2:120–9.Google Scholar

  • [19]

    Bryant GW, Garcia de Abajo FJ, Aizpurua J. Mapping the plasmon resonances of metallic nanoantennas. Nano Lett 2008;8:631–6.Google Scholar

  • [20]

    Orfanidis SJ. Electromagnetic waves and antennas. New Brunswick, NJ, Rutgers University, 2002.Google Scholar

  • [21]

    Balanis CA. Antenna theory: analysis and design. Hoboken, NJ, Wiley-Interscience, 2005.Google Scholar

  • [22]

    Kim J, Roh Y-G, Cheon S, et al. Babinet-inverted optical Yagi-Uda antenna for unidirectional radiation to free space. Nano Lett 2014;14:3072–8.Google Scholar

  • [23]

    Jackson JD. Classical electrodynamics. New York, Wiley, 1962.Google Scholar

  • [24]

    Kim J, Roh Y-G, Cheon S, Kim UJ, Hwang SW, Park Y, Lee C-W. Directional radiation of Babinet-inverted optical nanoantenna integrated with plasmonic waveguide. Sci Rep 2015;5:11832.Google Scholar

  • [25]

    Park Y, Kim J, Cho K-S, et al. Metasurface electrode light emitting diodes with planar light control. Sci Rep 2017;7:14753.Google Scholar

  • [26]

    Gordon R, Brolo AG, McKinnon A, Rajora A, Leathem B, Kavanagh KL. Strong polarization in the optical transmission through elliptical nanohole arrays. Phys Rev Lett 2004;92:037401.Google Scholar

  • [27]

    Castro-Lopez M, Brinks D, Sapienza R, van Hulst NF. Aluminum for nonlinear plasmonics: resonance-driven polarized luminescence of Al, Ag, and Au nanoantennas. Nano Lett 2011;11:4674–8.Google Scholar

  • [28]

    Ren M. Linearly polarized light emission from quantum dots with plasmonic nanoantenna arrays. Nano Lett 2015;15:2951–7.Google Scholar

  • [29]

    Curto AG, Volpe G, Taminiau TH, Kreuzer MP, Quidant R, van Hulst NF. Unidirectional emission of a quantum dot coupled to a nanoantenna. Science 2010;329:930–3.Google Scholar

  • [30]

    Dregely D, Taubert R, Dorfmüller J, Vogelgesang R, Kern K, Giessen H. 3D optical Yagi-Uda nanoantenna array. Nat Commun 2011;2:267.Google Scholar

  • [31]

    Maksymov IS, Staude I, Miroshnichenko AE, Kivshar YS. Optical Yagi-Uda nanoantennas. Nanophotonics 2012;1: 65–81.Google Scholar

  • [32]

    Kühn S, Hakanson U, Rogobete L, Sandoghdar V. Enhancement of single-molecule fluorescence using a gold nanoparticle as an optical nanoantenna. Phys Rev Lett 2006;97:017402.Google Scholar

  • [33]

    Anger P, Bharadwaj P, Novotny L. Enhancement and quenching of single-molecule fluorescence. Phys Rev Lett 2006;96:113002.Google Scholar

  • [34]

    Saleh BEA, Teich MC. Fundamentals of Photonics. New York, Wiley, 1991.Google Scholar

  • [35]

    Taflove A, Hagness SC. Computational electrodynamics: the finite-difference time-domain method. Boston, MA, Artech House, 2005.Google Scholar

  • [36]

    Fang Z, Fan L, Lin C, Zhang D, Meixner AJ, Zhu X. Plasmonic coupling of bow tie antennas with Ag nanowire. Nano Lett 2011;11:1676–80.Google Scholar

  • [37]

    Zhao Y, Alù A. Manipulating light polarization with ultrathin plasmonic metasurfaces. Phys Rev B 2011;84:205428.Google Scholar

  • [38]

    Mandal P, Anantha Ramakrishna S, Patil R, Venu Gopal A. Polarization dependent color switching by extra-ordinary transmission in H-slit plasmonic metasurface. J Appl Phys 2013;114:224303.Google Scholar

  • [39]

    Lukosz W, Kunz RE. Light emission by magnetic and electric dipoles close to a plane dielectric interface. II. Radiation patterns of perpendicular oriented dipoles. JOSA 1977;67:1615–9.Google Scholar

  • [40]

    Born M, Wolf E. Principles of optics: electromagnetic theory of propagation, interference and diffraction of light. Cambridge, UK, Elsevier, 2013.Google Scholar

  • [41]

    Demarest K, Huang Z, Plumb R. An FDTD near-to far-zone transformation for scatterers buried in stratified grounds. IEEE Trans Antennas Propag 1996;44:1150–7.Google Scholar

  • [42]

    Pelton M, Aizpurua J, Bryant G. Metal-nanoparticle plasmonics. Laser Photonics Rev 2008;2:136–59.Google Scholar

  • [43]

    Kang JH, Choe J-H, Kim DS, Park Q-H. Substrate effect on aperture resonances in a thin metal film. Opt Express 2009;17:15652–8.Google Scholar

  • [44]

    Lee D, Kim DS. Light scattering of rectangular slot antennas: parallel magnetic vector vs perpendicular electric vector. Sci Rep 2016;6:18935.Google Scholar

  • [45]

    Yang H. Magnetic nature of light transmission through a 5-nm gap. Sci Rep 2018;8:2751.Google Scholar

  • [46]

    Pellegrini G, Mattei G, Mazzoldi P. Tunable, directional and wavelength selective plasmonic nanoantenna arrays. Nanotechnology 2009;20:065201.Google Scholar

  • [47]

    Pakizeh T, Kall M. Unidirectional ultracompact optical nanoantennas. Nano Lett 2009;9:2343–9.Google Scholar

  • [48]

    Shegai T, Chen S, Miljković VD, Zengin G, Johansson P, Käll M. A bimetallic nanoantenna for directional colour routing. Nat Commun 2011;2:481.Google Scholar

  • [49]

    Liu W, Miroshnichenko AE, Neshev DN, Kivshar YS. Broadband unidirectional scattering by magneto-electric core-shell nanoparticles. ACS Nano 2012;6:5489–97.Google Scholar

  • [50]

    Lin J. Polarization-controlled tunable directional coupling of surface plasmon polaritons. Science 2013;340:331–4.Google Scholar

  • [51]

    Hofmann HF, Kosako T, Kadoya Y. Design parameters for a nano-optical Yagi-Uda antenna. N J Phys 2007;9:217.Google Scholar

  • [52]

    Lezec HJ. Beaming light from a subwavelength aperture. Science 2001;297:820–2.Google Scholar

  • [53]

    Kim S, Kim H, Lim Y, Lee B. Off-axis directional beaming of optical field diffracted by a single subwavelength metal slit with asymmetric dielectric surface gratings. Appl Phys Lett 2007;90:051113.Google Scholar

  • [54]

    Lee B, Kim S, Kim H, Lim Y. The use of plasmonics in light beaming and focusing. Prog Quant Electron 2010;34:47–87.Google Scholar

  • [55]

    Jun YC, Huang KC, Brongersma ML. Plasmonic beaming and active control over fluorescent emission. Nat Commun 2011;2:283.Google Scholar

  • [56]

    Livneh N. Highly directional emission and photon beaming from nanocrystal quantum dots embedded in metallic nanoslit arrays. Nano Lett 2011;11:1630–5.Google Scholar

  • [57]

    Osorio CI, Mohtashami A, Koenderink AF. K-space polarimetry of bullseye plasmon antennas. Sci Rep 2015;5:9966.Google Scholar

  • [58]

    Feng J, Li D, Pacifici D. Circular slit-groove plasmonic interferometers: a generalized approach to high-throughput biochemical sensing. Opt Mater Express 2015;5:2742–53.Google Scholar

  • [59]

    Kihm HW, Lee KG, Kim DS, Kang JH, Park QH. Control of surface plasmon generation efficiency by slit-width tuning. Appl Phys Lett 2008;92:051115.Google Scholar

  • [60]

    Mironov EG, Khaleque A, Liu L, Maksymov IS, Hattori HT. Enhancing weak optical signals using a plasmonic Yagi-Uda nanoantenna array. IEEE Photonics Technol Lett 2014;26: 2236–9.Google Scholar

  • [61]

    Fan P, Huang KC, Cao L, Brongersma ML. Redesigning photodetector electrodes as an optical antenna. Nano Lett 2013;13:392–6.Google Scholar

  • [62]

    Huang KC, Seo MK, Huo Y, Sarmiento T, Harris JS, Brongersma ML. Antenna electrodes for controlling electroluminescence. Nat Commun 2012;3:1005.Google Scholar

  • [63]

    Huang KC, Seo MK, Sarmiento T, Huo Y, Harris JS, Brongersma ML. Electrically driven subwavelength optical nanocircuits. Nat Photonics 2014;8:244–9.Google Scholar

  • [64]

    Park Y, Kim J, Roh YG, Lee CW. Radiation direction control by optical slot antenna integrated with plasmonic waveguide. Proc SPIE 2016;9884:988410–7.Google Scholar

  • [65]

    Cho KS, Lee EK, Joo W-J, et al. High-performance crosslinked colloidal quantum-dot light-emitting diodes. Nat Photonics 2009;3:341–5.Google Scholar

  • [66]

    Jang E, Jun S, Jang H, Lim J, Kim B, Kim Y. White-light-emitting diodes with quantum dot color converters for display backlights. Adv Mater 2010;22:3076–80.Google Scholar

  • [67]

    Kim TH, Cho K-S, Lee EK, et al. Full-colour quantum dot displays fabricated by transfer printing. Nat Photonics 2011;5:176–82.Google Scholar

  • [68]

    Coe-Sullivan S. Optoelectronics: quantum dot developments. Nat Photonics 2009;3:315–6.Google Scholar

  • [69]

    Shirasaki Y, Supran GJ, Bawendi MG, Bulović V. Emergence of colloidal quantum-dot light-emitting technologies. Nat Photonics 2013;7:13–23.Google Scholar

  • [70]

    Michaelson HB. The work function of the elements and its periodicity. J Appl Phys 1977;48:4729–33.Google Scholar

  • [71]

    Park Y, Choong V, Gao Y, Hsieh BR, Tang CW. Work function of indium tin oxide transparent conductor measured by photoelectron spectroscopy. Appl Phys Lett 1996;68:2699–701.Google Scholar

  • [72]

    Johnson PB, Christy RW. Optical constants of the noble metals. Phys Rev B 1972;6:4370.Google Scholar

  • [73]

    Colvin VL, Schlamp MC, Alivisatos AP. Light-emitting diodes made from cadmium selenide nanocrystals and a semiconducting polymer. Nature 1994;370:354–7.Google Scholar

  • [74]

    Schlamp MC, Peng X, Alivisatos AP. Improved efficiencies in light emitting diodes made with CdSe (CdS) core/shell type nanocrystals and a semiconducting polymer. J Appl Phys 1997;82:5837–42.Google Scholar

  • [75]

    Schubert EF, Gessman T, Kim JK. Light emitting diodes. Hoboken, NJ, John Wiley & Sons, 2005.Google Scholar

  • [76]

    Xu J, Cui D, Lewis BA, Wang AY, Xu S, Gerhold M. Microcavity light-emitting devices based on colloidal semiconductor nanocrystal quantum dots. IEEE Photonics Technol Lett 2005;17:2008–10.Google Scholar

  • [77]

    Lin D, Fan P, Hasman E, Brongersma ML. Dielectric gradient metasurface optical elements. Science 2014;345:298–302.Google Scholar

  • [78]

    Yu YF, Zhu AY, Paniagua-Domínguez R, Fu YH, Luk’yanchuk B, Kuznetsov AI. High-transmission dielectric metasurface with 2π phase control at visible wavelengths. Laser Photonics Rev 2015;9:412–8.Google Scholar

  • [79]

    Arbabi A, Horie Y, Bagheri M, Faraon A. Dielectric metasurfaces for complete control of phase and polarization with subwavelength spatial resolution and high transmission. Nat Nanotechnol 2015;10:937–43.Google Scholar

  • [80]

    Genevet P, Capasso F, Aieta F, Khorasaninejad M, Devlin R. Recent advances in planar optics: from plasmonic to dielectric metasurfaces. Optica 2017;4:139–52.Google Scholar

  • [81]

    Ni X, Kildishev AV, Shalaev VM. Metasurface holograms for visible light. Nat Commun 2013;4.Google Scholar

  • [82]

    Zheng G, Mühlenbernd H, Kenney M, Li G, Zentgraf T, Zhang S. Metasurface holograms reaching 80% efficiency. Nat Nanotechnol 2015;10:308–12.Google Scholar

  • [83]

    Khorasaninejad M, Chen WT, Devlin RC, Oh J, Zhu AY, Capasso F. Metalenses at visible wavelengths: diffraction-limited focusing and subwavelength resolution imaging. Science 2016;352:1190–4.Google Scholar

  • [84]

    Lalanne P, Chavel P. Metalenses at visible wavelengths: past, present, perspectives. Laser Photonics Rev 2017;11: 1600295.Google Scholar

  • [85]

    Wang S, Wu PC, Su V-C, et al. A broadband achromatic metalens in the visible. Nat Nanotechnol 2018;13:227–32.Google Scholar

  • [86]

    Huang YW, Lee HWH, Sokhoyan R, et al. Gate-tunable conducting oxide metasurfaces. Nano Lett 2016;16:5319–25.Google Scholar

  • [87]

    Sherrott MC, Hon PWC, Fountaine KT, et al. Experimental demonstration of >230 phase modulation in gate-tunable graphene-gold reconfigurable mid-infrared metasurfaces. Nano Lett 2017;17:3027–34.Google Scholar

  • [88]

    Sámson ZL, Yen S-C, MacDonald KF, et al. Chalcogenide glasses in active plasmonics. Phys Status Solidi Rapid Res Lett 2010;4:274–6.Google Scholar

  • [89]

    Gholipour B, Zhang J, MacDonald KF, Hewak DW, Zheludev NI. An all-optical, non-volatile, bidirectional, phase-change meta-switch. Adv Mater 2013;25:3050–4.Google Scholar

  • [90]

    Wang Q, Rogers ETF, Gholipour B, et al. Optically reconfigurable metasurfaces and photonic devices based on phase change materials. Nat Photonics 2016;10:60–5.Google Scholar

  • [91]

    Sautter J, Staude I, Decker M, et al. Active tuning of all-dielectric metasurfaces. ACS Nano 2015;9:4308–15.Google Scholar

  • [92]

    Komar A, Paniagua-Domínguez R, Miroshnichenko A, et al. Dynamic beam switching by liquid crystal tunable dielectric metasurfaces. ACS Photonics 2018;5:1742–8.Google Scholar

  • [93]

    Kim UJ, Kim TG, Shim Y, et al. Modulation of the Dirac point voltage of graphene by ion-gel dielectrics and its application to soft electronic devices. ACS Nano 2015;9:602–11.Google Scholar

  • [94]

    Lee EJ, Choi SY, Jeong H, et al. Active control of all-fibre graphene devices with electrical gating. Nat Commun 2015;6:6851.Google Scholar

Back to top