Yuta Nito

A Modified Semivectorial BPM Retaining the Effects of the Longitudinal Field Component and Its Application to the Design of a Spot-Size Converter

For the analysis of a z -variant optical waveguide with high-index contrast by the beam-propagation method (BPM), the error often grows up during the propagation process. In this paper, the formulation of the BPM is revisited, taking into account the effects of the longitudinal field component. The improvement in accuracy is demonstrated through the analysis of a vertically tapered rib waveguide. The power expression based on the Poynting vector is also employed to improve the power conservation property. As an application, the present technique is applied to the design of a spot-size converter composed of a multicore structure. It is numerically demonstrated that the use of a curvilinearly tapered core leads to a conversion length of 570 mum, which is 42% less than the length of a linearly tapered core, with a coupling efficiency of 97% being maintained at a wavelength of 1.55 mum.

Reduction in Bend Losses of a Buried Waveguide on a Silicon Substrate by Adjusting the Core Location

A new technique for reducing pure bend and transition losses in a buried dielectric waveguide is discussed, using the imaginary-distance beam-propagation method. It is numerically shown that only the adjustment of a core location enables us to reduce the bend losses. The reason why a proper location of the core offers minimum bend losses regardless of a bending radius is explained in terms of the generation and suppression of a leaky wave. Significant loss reduction is achieved over a wide spectral range from 1.3 to 1.65

Full-Vectorial Beam-Propagation Methods Based on a Fundamental Scheme—Design of a Short Polarization Converter

\mu {\rm m}

Bent Embedded Optical Waveguide With a Loaded Metal Film for Reducing a Polarization Dependent Loss

. It is also revealed that a minimum pure bend loss is obtained regardless of the relative refractive index difference, when the air-cladding interface is placed at the position, where the field amplitude decays to approximately 12% of its peak.

Transverse-Magnetic BPM Analysis of a Step-Index Slab Waveguide Expressed by a Sigmoid Function

A fundamental scheme is utilized to reformulate the full-vectorial beam-propagation methods (BPMs). First, the fundamental scheme is introduced into the Fresnel equations which are discretized using the Crank-Nicolson method (FCN-BPM). The present FCN-BPM has the advantage that the total number of arithmetic operations is extremely reduced when compared with the conventional CN-BPM, while maintaining identical accuracy. Next, the fundamental scheme is applied to the alternating-direction implicit BPM (FADI-BPM). The choice of the refractive index at each split step is discussed paying attention to the commutativity of coefficient matrices. With the present method, computation time is reduced to approximately 74% for the analysis of a mode-evolution-based z-varying polarization converter. Power expression based on both electric and magnetic fields is also adopted to improve the power conservation. Finally, the present method is applied to the design of a short polarization converter. It is revealed that the use of a curvilinearly tapered core leads to an extinction ratio of more than 15 dB with a device length of 100 μm, over a wide wavelength range from 1.2 to 1.7 μm.

A Beam-Propagation Method Using Both Electric and Magnetic Fields

The beam-propagation method is applied to the analysis of a bent embedded waveguide with a loaded metal film. As a basic study, coupled bent slab waveguides are treated to discuss their mechanism for reducing the pure bend and transition losses. Based on this study, a metal film is loaded on the bent embedded waveguide. The metal film serves to generate the surface plasmon-polariton mode for the transverse-magnetic (TM) wave, so that the embedded waveguide with the metal film can be regarded as coupled bent waveguides for the TM wave. This leads to reduction in the bend losses for the TM wave. As a result, a low polarization dependent loss (PDL) can be obtained over a wide spectral range, since the pure bend loss of the embedded waveguide for the TM wave is larger than that for the transverse-electric wave. Furthermore, the application to a trench structure with a depressed refractive index contributes not only to a low PDL but also to a low pure bend loss.

A Modified Semivectorial Beam-Propagation Method Retaining the Longitudinal Field Component

Power conserving property of the propagating transverse-magnetic wave in a strongly guiding waveguide is demonstrated using the beam-propagation method with a transformed field. A step-index profile is expressed by a sigmoid function. A tilted, asymmetrical waveguide and a Y-branching waveguide with high index contrast are analyzed and discussed.

Bend Loss Reduction of a Buried Waveguide on a Silicon Substrate by Adjusting the Core Location

The power evaluation in the beam-propagation method is reconsidered to overcome the problem that the power is not conserved during the propagation process. The Fresnel equations based on the electric and magnetic fields are solved in parallel, and both fields are employed for the power expression based on the Poynting vector. A spot-size and a polarization converter with a high-index contrast are analyzed and discussed.

A waveguide configuration for reducing both pure bend and polarization dependent losses

The formulation of the beam-propagation method is revisited, taking into account the effects of the longitudinal field component. The improvement is demonstrated through the analysis of a vertically tapered rib waveguide.

Bend Loss Reduction of a Buried Waveguide Using Reflection Waves Generated from Cladding Interfaces

Leaky waves generated in a bent waveguide are suppressed by an appropriate choice of the distance between the core and the upper (lower) interface of the cladding. Loading a metal film reduces polarization dependent loss.