Search

The nanoscale systems are of great importance in modern science and technology, offering a wide range of possible daily life applications. In order to characterize the electromagnetic (EM) properties of metallic nanotopologies, the commercialized computational tools COMSOL, CST, and Lumerical have been widely used. In this work, the performance of the above tools is compared, through the investigation of the plane wave response of canonical spherical and cubical nanoparticles. It is demonstrated that not all EM field solvers provide an accurate description of the nanoparticle's behavior.

The interaction between light and plasmonic structures at the deep-nanometer scale, which is essentially governed by nonclassical effects, has been increasingly investigated by the hydrodynamic model. Several hydrodynamic models have been introduced, in order to describe the motion of the free electron gas in metals. In this work, by employing the hard wall hydrodynamic model (HW-HDM) and the quantum hydrodynamic model (Q-HDM), the nonlocal response of an isolated metallic nanoparticle, excited by a dipole emitter, is investigated. The analysis is performed for the local density of states of an emitter coupled with a spherical nanoparticle, which may be treated analytically. Both radial and tangential emitters are studied, while varying the emitter–nanoparticle separation, and the nanoparticle’s size and material. It is shown that HW-HDM and Q-HDM may generate striking spectral features, which do not arise within the classical approach.

In order to describe light-matter interaction at deep-nanometer scale, which is governed by nonclassical phenomena, a semiclassical hydrodynamic model with additional boundary conditions has been introduced. This model investigates the motion of the charged gas in metals, characterized by nonlocal material parameters. Based on different treatments of the nonlocality, several hydrodynamic models have been proposed. This work employs the hard wall hydrodynamic model (HW-HDM) and the quantum hydrodynamic model (Q-HDM). The study is performed for the mode structure (natural modes) of a spherical metallic nanoshell, supplemented by the plane wave response of the system. It is shown that the main transverse resonances of HW-HDM and Q-HDM experience nearly identical blueshifts in comparison to the classical resonances. The corresponding longitudinal resonances show noticeable spectrum shifts.

The electromagnetic properties of plasmonic nano-antennas and scatterers with the characteristic dimensions at deep-nanometer scale, governed by quantum mechanical effects, have been extensively studied by a hydrodynamic approach. Several hydrodynamic models, together with additional boundary conditions, have been proposed to deal with the collective motion of the free electron gas in metals. In this work, four hydrodynamic models, namely the hard-wall hydrodynamic model, the curl-free hydrodynamic model, the shear forces hydrodynamic model, and the quantum hydrodynamic model are employed. The study is performed for a deep-nanometer metal core–dielectric shell sphere, excited by a plane wave. It is demonstrated that the far field characteristics of the core–shell nanosphere are largely affected by the choice of a specific hydrodynamic model.

Much attention has been attracted by the nonlocal hydrodynamic model, which provides the first-order approximation of non-classical effects, essentially governing the light-matter interaction at deep-nanometer scale. Several hydrodynamic models, together with additional boundary conditions, have been introduced to describe the behavior of the free electron gas in metals. This work investigates the performance of the following models: the hard wall hydrodynamic model (HW-HDM), the quantum hydrodynamic model (Q-HDM), the curl-free hydrodynamic model, and the shear forces hydrodynamic model. The analysis is conducted for the mode structure (natural or quasi-normal modes) of a canonical spherical nanoparticle. In contrast to HW-HDM and Q-HDM, the other two models break down, due to artifactual mode interference.

In this work, we present a unified approach to model the optical response from a classical local and a nonclassical nonlocal hydrodynamic interface. The modeling is based on a boundary integral equation (BIE) representation. This representation is general for both 2D and 3D problems and can be numerically solved by the Method of Moments (MoM) algorithm. An example is then demonstrated for the 2D and 3D case. A very good agreement with the generalized Mie theory has been demonstrated.

Within the framework of the Boundary Element Method (BEM), we develop a computational scheme to account for nonlocal hydrodynamic effects in three -dimensional metallic nanoparticles of arbitrary shape. In contrast to conventional field-based methods, the reported scheme is formulated based on vector and scalar potentials. Especially, an extra scalar potential is introduced to describe longitudinal fields due to the hydrodynamic motion of electrons. The computational scheme is demonstrated for a canonical structure, i.e. sphere, for far field properties, e.g. scattering cross sections, and an excellent agreement is found with generalized Mie theory.

In this paper, a new way is introduced to define the concept of recoverable energy for general radiating structures. We also show that this recoverable energy can be maximized, that is, the maximum energy can be extracted from a certain electromagnetic field distribution by minimizing the radiative energy. This maximum recoverable energy obviously never is negative and it is totally coordinate independent. A loop antenna is employed to demonstrate the notions introduced.

Nowadays nanotechnology allows us to size down every single object in our daily life at the nanoscale. Especially, conventional microwave antennas/scatterers are reproduced at nanoscale and have found many applications, including bio-sensing [1], energy harvesting [2], nanoscale communications [3, 4], etc. Most recently, a Nanoparticle-on-Mirror (NPoM) topology with a sub-nm gap size (See Fig. 1, a nanosphere-on-mirror, as an example) is proposed [5], soon draws many interests due to its capability of confining extremely intensive field to the sub-nm gap between the nanoparticle and the gold mirror, and promises a new generation of Surface-enhanced Raman Scattering (SERS) substrates. Computationally, the NPoM structure is more often than not handled by an Finite difference Time Domain (FDTD) implementation, for example, Lumerical [5–7]. However, applying a FDTD solver to the analysis of the NPoM system may not be an optimal choice and can have several disadvantages. First of all, to model the subnanometric nature of the gap and the detailed geometric feature of a nanoparticle whose diameter is around tens of nanometers, an extremely fine mesh has to be employed, which inevitably increases the computational time. Secondly, for the NPoM system, the optical response of an oblique incident wave is of much greater interests than the usual normal incidence. Since in an FDTD algorithm, an oblique incident wave suffers from frequency/wavelength dispersion, in order to understand the optical response of the NPoM system over a wide wavelength range, multiple narrow band simulations have to be run, which worsens the computational time consideration mentioned previously and leads to inaccurate simulation results as the FDTD is ideal for a wide band simulation. Last but not least, a FDTD run renders very little physical understanding to the considered NPoM system where the following two questions are of great interests: 1) which modes are excited by the incident field and 2) how the presence of the mirror and the nanometric/sub-nanometric gap size affects the optical response of the system. All the above issues can be resolved by employing an integral equation formalism. In this work, we focus on the Electric Field Volume Integral Equation (EF-VIE) and its solution given by a Volumetric Method of Moments (V-MoM) algorithm. When we construct the solver, we have the following two design goals. Firstly, this solver should be able to deliver physical understandings as much as possible. Secondly, under the condition that the first design goal is fulfilled, this solver must be as much computationally efficient as possible. The first goal is achieved by clearly separating the contributions from primary wave and reflected wave from the mirror. Since s-(TE) and p-(TE) waves are the eigenmodes of a multilayer structure [8], their effects are further distinguished in the reflected wave contribution. Considering the computational efficiency, we extract the near field behavior for the s- and p-waves in closed form and finish the following volume integral as analytically as possible.

A wearable UWB antenna is proposed and analyzed in this paper, which is totally fabricated with textile materials, except the feeding connector for testing. By implementing a multilayered structure, the antenna obtains a good matching over the whole UWB band while keeping its low profile. The performance of the antenna is unaffected when worn on the human body due to the shielding by the full ground plane placed under the patch. Besides that, high on-body fidelity and a robust performance are produced when the antenna is studied under several types of deformations.

A low profile dual-band antenna based on the magneto-electric dipole concept is investigated on-body. An efficient space-utilization configuration of the conductive textiles and substrate layers ensured the compactness of the structure for on-body application. The two U slots in the dipole combine the electric and magnetic resonances to produce a dual band behavior and wideband characteristic in each band, as validated via simulations and measurements. Low backward radiations in both the lower and the upper bands are observed, rendering it insensitive to the coupling to the human body. To validate this, the antenna is first simulated in a bent configuration using a simplified, cylindrically-shaped human phantom and was found to be robust. It was then simulated on a detailed voxel model to assess its bent performance. Besides the slightly degraded reflection performance, the studied specific absorption rate (SAR) is proven to be well below the European limit of 2 W/kg.

The far field of a plasmonic nano dipole antenna topology is studied systematically. The important metals used in plasmonics are considered (silver, aluminum, chromium, gold, and copper), the length of the dipole is varied, and a very wide frequency band is considered. The results concern mainly radiation patterns, gains, and radiation efficiencies.

Due to the relationship between charge dimensionality and singular field behavior, the stand-alone half rooftop is allowed to exist in the volumetric description of a topology carrying a three dimensional current. This leads to the possibility of using multi-level meshing to describe nanotopologies. The meshing scheme is validated by numerical calculations and experiments.

The formulation of Integral Equations for a problem, to be solved with the Method of Moments, has been a cornerstone in the field of computational electromagnetics for components and antennas in the microwave and millimeter wave frequency range for decades. In this invited paper, the applicability, advantages and disadvantages of this method are discussed in view of applying it at much higher frequencies, namely in the IR and optical frequency range. This invited presentation is based on an overview paper on this topic recently published.

Computational electromagnetics has become a real cornerstone in the design of microwave and millimeter wave components and systems. One of the major computational techniques used in these designs is to solve integral equations with the Method of Moments (MoM). Traditionally, MoM has been used extensively for antennas, transmission lines, filters, couplers, etc., mainly in the MHz and GHz range. However, MoM can also be applied at much higher frequencies. It will be shown that at these frequencies, this computational technique is just emerging. One of the reasons is that there are serious differences with the classical implementations, which prohibits the direct use of the well-known commercially available MoM solvers. For example, one of the differences is the fact that the concept of a surface current, confined to a very thin layer at the surface of a conductor, is not valid any more. In this presentation, the different computational techniques will be reviewed and compared, not only in the traditional microwave and millimeter wave range, but especially at THz, IR, and optical frequencies. In concreto, the MAGMAS MoM solver will be compared to a numerical FDTD solver well-known in the plasmonics research community: Lumerical. It will be shown that also at these very high frequencies, MoM has its proper place in the pallet of computational algorithms.

Recently, a great amount of research has been triggered by the prediction, formulated by Pendry and co-authors, that novel and enhanced nonlinear optical phenomena could be observed in metamaterials.[1] This prediction is based on the fact that, in metamaterials, local field enhancements can have a dramatic influence over the optical properties of the material. One of the largest contributions to such local field enhancements is attributable to surface plasmon resonances. Plasmons are collective oscillations of the electrons under the influence of light's electromagnetic field. Plasmons occur naturally on the surfaces of homogeneous metal films, where they usually dissipate quickly and cancel each other's influence. However, upon patterning the metal surface at the nanoscale, plasmons can be manipulated in a manner similar to classical waveguiding, whereby propagation or standing wave patterns can be achieved. In other words, artificial structuring allows for nanoengineering the position and intensity of the local fields. Incidentally, nonlinear optical effects, such as second-, third-, or forth-harmonic generation, scale as the second, third or fourth power of the electromagnetic intensity, respectively. Therefore, it is perfectly reasonable to assume that the large electromagnetic local field enhancements in metamaterials should yield large or previously unobserved nonlinear optical effects.

Clustering is one of the most important and well studied fields of data mining. Many clustering methods have been proposed in the last few decades with different backgrounds. An open problem in clustering field is the lack of a benchmark for contrasting all these algorithms as well as other available ones. In this paper, we use the well-known log marginal likelihood (LML) score and classification accuracy as two criteria of comparison. Experiment results on k-means, expectation maximization (EM) and farthest-first in the t-test show that EM outperforms other two algorithms on most of the benchmark data sets with respect to both criteria. Considering the performance and comprehensibility of EM, it is an ideal model that could be used in many real world applications

Recently, combining Naive-Bayes with the expectation maximization (EM) algorithm for unsupervised learning have received significant attention. AutoClass is a classical Bayesian clustering algorithm that uses Naive-Bayes in combination with EM algorithm to find the probability distribution parameters to best fit the data. In this study, we introduce a robust approach, which is similar to AutoClass, it can arbitrarily impose any Bayesian classifiers in combination with EM algorithm to enhance cluster's performance. This paper focuses on how clustering techniques can benefit from classification. We provide experimental evidence that more accurate than original results of clustering in the t-test on most of the benchmark data sets