Fengwen Wang

Topology Optimized Architectures with Programmable Poisson's Ratio over Large Deformations

Topology optimized architectures are designed and printed with programmable Poissons ratios ranging from -0.8 to 0.8 over large deformations of 20% or more.

Robust topology optimization of photonic crystal waveguides with tailored dispersion properties

A robust topology optimization method is formulated to tailor dispersion properties of photonic crystal waveguides, with consideration of manufacturing uncertainties. Slightly dilated and eroded realizations are considered as well as the real structure, and by worst-case optimization, we also ensure a satisfactory performance in the case of an under- or overetching scenario in the manufacturing process. Two photonic crystal waveguides facilitating slow light with group indexes of ng=25 and ng=100 and bandwidths of Δω/ω=2.3% and 0.3%, respectively, are obtained through the proposed robust design procedure. In addition, a novel waveguide design with two different constant group index waveguide regions is demonstrated. The numerical examples illustrate the efficiency of the robust optimization formulation and indicate that the topology optimization procedure can provide a useful tool for designing waveguides that are robust to manufacturing uncertainties such as under or overetching.

Fundamental limitations to gain enhancement in periodic media and waveguides

A common strategy to compensate for losses in optical nanostructures is to add gain material in the system. By exploiting slow-light effects it is expected that the gain may be enhanced beyond its bulk value. Here we show that this route cannot be followed uncritically: inclusion of gain inevitably modifies the underlying dispersion law, and thereby may degrade the slow-light properties underlying the device operation and the anticipated gain enhancement itself. This degradation is generic; we demonstrate it for three different systems of current interest (coupled-resonator optical waveguides, Bragg stacks, and photonic crystal waveguides). Nevertheless, a small amount of added gain may be beneficial.

Systematic design of loss-engineered slow-light waveguides

This paper employs topology optimization to systematically design free-topology loss-engineered slow-light waveguides with enlarged group index bandwidth product (GBP). The propagation losses of guided modes are evaluated by the imaginary part of eigenvalues in complex band structure calculations, where the scattering losses due to manufacturing imperfections are represented by an edge-related effective dissipation. The loss engineering of slow-light waveguides is realized by minimizing the propagation losses of design modes. Numerical examples illustrate that the propagation losses of free-topology dispersion-engineered waveguides can be significantly suppressed by loss engineering. Comparisons between fixed- and free-topology loss-engineered waveguides demonstrate that the GBP can be enhanced significantly by the free-topology loss-engineered waveguides with a small increase of the propagation losses.

Benchmarking five numerical simulation techniques for computing resonance wavelengths and quality factors in photonic crystal membrane line defect cavities

We present numerical studies of two photonic crystal membrane microcavities, a short line-defect cavity with a relatively low quality (Q) factor and a longer cavity with a high Q. We use five state-of-the-art numerical simulation techniques to compute the cavity Q factor and the resonance wavelength λ for the fundamental cavity mode in both structures. For each method, the relevant computational parameters are systematically varied to estimate the computational uncertainty. We show that some methods are more suitable than others for treating these challenging geometries.

Benchmarking state-of-the-art numerical simulation techniques for analyzing large photonic crystal membrane line defect cavities

In this work, we perform numerical studies of two photonic crystal membrane microcavities, a short line-defect L5 cavity with relatively low quality (Q) factor and a longer L9 cavity with high Q. We compute the cavity Q factor and the resonance wavelength λ of the fundamental M1 mode in the two structures using five state-of- the-art computational methods. We study the convergence and the associated numerical uncertainty of Q and λ with respect to the relevant computational parameters for each method. Convergence is not obtained for all the methods, indicating that some are more suitable than others for analyzing photonic crystal line defect cavities.

Comparison of five computational methods for computing Q factors in photonic crystal membrane cavities

Five state-of-the-art computational methods are benchmarked by computing quality factors and resonance wavelengths in photonic crystal membrane L5 and L9 line defect cavities. The convergence of the methods with respect to resolution, degrees of freedom and number of modes is investigated. Special attention is paid to the influence of the size of the computational domain. Convergence is not obtained for some of the methods, indicating that some are more suitable than others for analysing line defect cavities.

Optimization of photonic crystal cavities

We present optimization of photonic crystal cavities. The optimization problem is formulated to maximize the Purcell factor of a photonic crystal cavity. Both topology optimization and air-hole-based shape optimization are utilized for the design process. Numerical results demonstrate that the Purcell factor of the photonic crystal cavity can be significantly improved through optimization.

Benchmarking five computational methods for analyzing large photonic crystal membrane cavities

We benchmark five state-of-the-art computational methods by computing quality factors and resonance wavelengths in photonic crystal membrane L5 and L9 line defect cavities. The convergence of the methods with respect to resolution, degrees of freedom and number of modes is investigated. Convergence is not obtained for some of the methods, indicating that some are more suitable than others for analyzing line defect cavities.

Comparison of five numerical methods for computing quality factors and resonance wavelengths in photonic crystal membrane cavities

The photonic crystal (PhC) membrane represents a platform for planar integration of components, where cavities and waveguides may play a key role in realizing compact optical components with classical functionality such as switches, lasers, and amplifiers or quantum optical functionality such as integrated sources of quantum light. By leaving out a row of holes in an otherwise perfect PhC membrane lattice, a line defect is created in which light may be guided. If the waveguide is terminated at both ends, the finite-length waveguide forms an Ln cavity, where n denotes the length of the cavity. Such Ln cavities support spectrally discrete optical modes, and the fundamental cavity mode profile of an L9 cavity is shown in Fig. 1. Light may be confined to such an Ln cavity for extended periods, as quantified by the quality (Q) factor. For laser applications, the Q factor governs the onset of lasing, and for cavity quantum electrodynamics applications, it governs the onset of strong coupling. The Q factor thus represents a key parameter in the design of a PhC membrane cavity.