Evolutionary Optimization of Optical Antennas

Abstract

The design of nano-antennas is so far mainly inspired by radio-frequency technology. However, material properties and experimental settings need to be reconsidered at optical frequencies, which entails the need for alternative optimal antenna designs. Here a checkerboard-type, initially random array of gold cubes is subjected to evolutionary optimization. To illustrate the power of the approach we demonstrate that by optimizing the near-field intensity enhancement the evolutionary algorithm finds a new antenna geometry, essentially a split-ring/two-wire antenna hybrid which surpasses by far the performance of a conventional gap antenna by shifting the n=1 split-ring resonance into the optical regime.PACS numbers: 84.40. Ba, 73.20.Mf, 78.67.Bf Light-matter interaction, i.e. absorption and emission of light as well as the control of its spectral and directional properties, can be optimized by means of antennalike plasmonic nano structures [1, 2]. This is of immediate importance in diverse fields of research ranging from solar energy conversion [3], photocatalytic [4] and sensing applications [5] to single-particle manipulation [6,7] and spectroscopy [8] as well as quantum optics and communication [9][10][11][12].RF-antenna designs are usually optimized for thin, infinitely good conducting wires that only support surface currents and are typically fed by transmission lines connected by infinitely narrow gaps [13]. For antennas at optical frequencies the general operation conditions deviate substantially from such ideal behaviour: (i) Antenna wire diameters are comparable to the electromagnetic penetration depth into the wire material leading to volume currents [14]. In the case of noble metals, such wires therefore exhibit plasmon resonances in the visible spectral range resulting in a reduced effective wavelength of wire waves [15]. (ii) Feeding (excitation) of optical antennas is often achieved by focused laser beams or quantum emitters. (iii) high-frequency-related effects such as the 'kinetic inductance' become significant [16]. It can therefore not be taken for granted that RF-inspired antenna designs, like dipole [17], bow tie [18,19] and Yagi-Uda antennas [20,21], represent 'optimal' geometries also at optical frequencies, although they provide a reasonable performance.Evolutionary algorithms (EAs) find optimized solutions to highly complex non-analytic problems by creating subsequent generations of individuals coded by their respective genomes that compete for the right to pass on their properties, according to a fitness parameter [22]. These optimized solutions can then be analyzed to foster the understanding of underlying physical principles. Evolutionary optimization has successfully been applied in various fields of research, including pulse shape optimization in coherent control of chemical reactions [23] and field localization in plasmonic structures [24,25]. Furthermore, evolutionary optimization has been used to aid the development of radio-wave antennas [26,27]. First attempts to employ such met...

Plasmonic nanoantenna design and fabrication based on evolutionary optimization

“…The first generation consisted of random structures with a filling factor of 0.7. The antennas were ranked according to their fitness and the best eight structures were taken as parents of the subsequent generation (more details about the mechanism of the EA can be found in [10] as well as in the supplementary material). Three methods were employed to create the next 30 individuals: mutation (creation of random structures), as well as linear and spiral genome crossing (see Supplementary Fig.…”

“…Yagi-Uda antennas [9]. However, this approach does not always lead to the best possible antenna performance and, as we have shown before, unexpected designs found by evolutionary optimization can perform much better [10].…”

“…Here we adapted our EA based on finite difference time domain (FDTD) simulations combined with a Matlab code [10] to find planar optical antennas for which the near-field localization and intensity enhancement is optimized at r 0 , the very center of the antenna. Accordingly, the fitness parameter was chosen to be the near field intensity enhancement…”

“…The FEA current pattern can be described by a constructive superposition of two fundamental modes [10] (see Fig. 2(e)): (i) dipolar antenna currents comparable to the lowest order bonding-mode current pattern of linear two-wire antennas [3] with the well-known benefits of good far-field coupling as well as accumulation of opposite charges at either side of the gap as well as; (ii) the current pattern of fundamental split-ring like modes [17] above and below the gap, leading to additional charge accumulations at the center and resulting in a larger NFIE as for a plain dipolar antenna.…”

Artificial intelligence in nanotechnology

“…The analysis of the fittest antenna geometry reveals that it merges the features of the fundamental magnetic resonance of split-ring with the electric one of a linear dipole antenna. [64] To obtain the best configuration by a GA, authors decided to use as fitness parameter the normalized near-field intensity enhancement in the focus of an illuminating Gaussian beam. The GA used generations of 20 to 30 individual matrix antennas, where the five best structures were selected as parents for the next generation.…”