Optical Properties of Lithium Niobate Waveguides, and Prism Coupling for Optimizing Waveguide Fabrication.

I investigated the accuracy and precision of effective index of refraction measurements using prism coupling. I found that the coupling gap's geometry is critical to predicting the accuracy and resolution of measurements. I provide experimental evidence that there are advantages to using high coupling pressures that yield long coupling lengths. I present and evaluate all experimental technique aimed at reducing the effects of perturbation and find that it may not be applicable to all situations. I extend a previous model of prism coupling to include a finite coupling length. Also, I introduce an alternative model of prism coupling based upon a local normal-mode expansion and present its preliminary numerical results. I have observed more extraordinary polarized guided modes near cut-off in titanium in-diffused lithium niobate planar-waveguides than are predicted by titanium in-diffusion atmospheres. I also observe these nearly cut-off modes in lithium niobate specimens fabricated without titanium under a variety of diffusion atmospheres. Values of the effective index of these nearly cut-off modes found in plain lithium niobate plarlar waveguides do not agree with those calculated using the lithium out-diffusion model for either vacuum-diffused or oxygen-diffused lithium niobate waveguides. I found good agreement between measured values of the effective mode index in specimens fabricated in either dry oxygen or air and calculated values found using an empirical diffusion model of a step index profile whose depth increases as the square root of diffusion time. I report on polarization coupling from x-propagating TE-polarized waveguide modes to TM-polarized radiation modes in y-cut titanium in-diffused lithium niobate slab waveguides. The polarization coupling is linear with respect to the TE-polarized input power and depends on the titanium diffusion time. Furthermore, the polarization coupling was not affected by white light flooding the waveguide. I observe no turn-on delay greater than 1 s between the excitation of the TE waveguide mode and the observation of the TM radiation. Intentional amplitude modulations of the TE -polarized waveguide mode were registered as equivalent amplitude modulations by the TM-polarized radiation modes.