Study of poled lithium niobate waveguide devices by confocal defect spectroscopy

V, Dierolfnrid Chr. Snridnranri Leliigh Universifq., Physics Depnminit and Center for Opticnl Technologies, Bethlehem, PA 18015 USA Nonlinear light sources based on periodically poled LiNbO, (PPLN) have gained a great deal of attention due to low thresholds, ease of operation and wide tuning range. In particular, the integrated optical versions of them, achieve, due to the confinement of light over long interaction lengths, very high conversion efficiencies. In order to optimize these integrated optical devices, both the Ti-profile and the domain structure have to he engineered precisely and effects of interaction have to be under control. To achieve this goal, we developed a novel characterization tool based on the luminescence of well characterized "designer defect"(e.g.: E?' ) that we use as probes for the local environment. We applied this concept to the two components of a PPLN waveguide device. ( I ) Characrerizaation ofdomain strirctiires nnd domain walls. It has been shown that the domain inversion process, most notably the coercive field, is strongly influenced by the presence of intrinsic and extrinsic defects. We investigated this effect by addressing the reverse question: How does domain inversion influence the intrinsic defects and dopants? Performing optical spectroscopy with high resolution (both spectrally and spatially) on trivalent rare earth ions (E?' and Yb'7 in congruent and stoichiometric LiNbO, we find drastic changes in the defect luminescence. Based on a detailed analysis of these changes, using earlier extensive studies of the defect system [ I ] ; we conclude that a defect reconfiguration takes place. Obviously, some o f the required charge compensators do not reposition themselves according to the reversed orientation of the spontaneous polarization vector at room temperature, while others do. Furthermore, we were able to detect a change in the intrinsic electric field experienced by the ion. On the other hand, we use the emission changes for non-destructive 3D-domain imaging (Fig. I ) in a home-built confocal fluorescence microscope. For this technique, we scan the excitation laser over the sample and measure the corresponding emission spectra. Evaluation of the moments of the spectra enables us to image domains with high contrast and high spatial resolution (<500nm). This method can be employed in situ, allowing precise control of the domain structures and time resolved investigations of domain inversion. For instance, we are studying the spatial extent and dynamics of domain walls. Especially, we focus-on the question at which stage the defect rearrangement occurs. "E. '' uOnrnlnS rrnngea ' y cOn~Oc*' fluorescence microscope.