1. |
- Benisty, H., et al.
(author)
-
Low-loss photonic-crystal and monolithic InP integration : Bands, bends, lasers, filters
- 2004
-
In: Photonic Crystal Materials and Devices II. - : SPIE - International Society for Optical Engineering. - 0819452688 ; , s. 119-128
-
Conference paper (peer-reviewed)abstract
- Practical realizations of 2D (planar) photonics crystal (PhC) are either on a membrane or etched through a conventional heterostructure. While fascinating objects can emerge from the first approach, only the latter approach lends itself to a progressive integration of more compact PhC's towards monolithic PICs based on InP. We describe in this talk the various aspects from technology to functions and devices, as emerged from the European collaboration "PCIC". The main technology tour de force is deep-etching with aspect ratio of about 10 and vertical sidewall, achieved by three techniques (CAIBE, ICP-RIE, ECR-RIE). The basic functions explored are bends, splitters/combiners, mirrors, tapers, and the devices are filters and lasers. At the end of the talk, I will emphasize some positive aspects of "broad" multimode PhC waveguides, in view of compact add-drop filtering action, notably.
|
|
2. |
- Benisty, H., et al.
(author)
-
Models and measurements for the transmission of submicron-width waveguide bends defined in two-dimensional photonic crystals
- 2002
-
In: IEEE Journal of Quantum Electronics. - 0018-9197 .- 1558-1713. ; 38:7, s. 770-785
-
Journal article (peer-reviewed)abstract
- One of the essential building-blocks of miniature photonic crystal (PC)-based photonic integrated circuits (PICs) is the sharp bend. Our group has focused on the 2-D photonic crystal based on a triangular lattice of holes perforating a standard heterostructure. The latter, GaAlAs-based or InP-based, is vertically a monomode waveguide. We consider essentially one or two 60 bends defined by one to five missing rows, spanning both cases of monomode and multimode channel waveguides. From intensive modeling and various experimental measurements (both on GaAs and InP), we point out the origin of the present level of bend insertion losses and discuss the merits of the many roads open for improved design.
|
|