ABSTRACT
Since the early 1970s, there has been an explosion of research activities in realizing planar integrated
waveguide devices fabricated by the technologies of ion-exchange in glass (Izawa and Nakagome 1972;
Giallorenzi et al. 1973), Ti diffusion into LiNbO3 crystals (Schmidt and Kaminow 1974) and proton
exchange in LiNbO3 substrates (Jackel et al. 1982). Early efforts involving diffusion processes were
directed toward conceptualization and experimental realization of such devices, which have great
signal processing capabilities in optical fiber communication and sensor applications. Some of these
capabilities include power division, wavelength division multiplexing/demultiplexing, switching,
modulation, polarization splitting, and so on. However, to obtain high performance in these
devices, it is important, first, to develop accurate designs. For this purpose, detailed and accurate
information on the characteristics of slab and channel guides in respective substrates must be known
in relation to their fabrication conditions because slabs and channels often form the basic units in the
more complicated structures of waveguide devices. Therefore, “design methodology” does not merely
involve a mathematical or computational process. It is more important, first, to establish an accurate
refractive-index model for the waveguide, fabricated from the results of experimental characterization,
before mathematical and computer-aided design procedures can become effective. It goes without
saying, however, that clever mathematics and efficient, accurate numerical algorithms would help
provide accurate designs.