Microstructuring of lithium niobate

Barry, I.E. (2000) Microstructuring of lithium niobate. University of Southampton, Department of Electronics and Computer Science, Doctoral Thesis, 186pp.

Record type: Thesis (Doctoral)

Abstract

This thesis presents the results from an investigation into methods for micron-scale relief structuring of lithium niobate. A wet etch consisting of HF and HNO₃ was applied, and directed by 1) patterning the ferroelectric domain structure of the samples and 2) illuminating the crystals with patterned 488nm light. Post-etch treatment of the structures resulted in ridge waveguides and alignment grooves, while pre-etch manipulation achieved an etch-stop. Ablation was investigated as a method of directly structuring the crystal and for patterning photoresist.

The etch was found to leave the +z face untouched. The -z face was etched at a rate, k, in µm/hour given by

k = e 20.37 - 6300/T

where T is the absolute temperature. This differential etch rate reveals a pattern induced in the ferroelectric domain structure by the technique of electric field patterning. The structures had walls with roughness < 5nm. Straight walls were easily achieved aligned along the y-direction at 120o to this. Other directions can result in facetted walls. Ridge waveguide losses <1dBcm^-1, fibre alignment grooves and an etch stop were demonstrated using appropriate pre- and post-etch treatments.

The etch was found to be affected by illumination with 488nm radiation. In Fe:LiNbO₃ complete and partial frustration of the etch was induced on the -z face. Characteristic features of the partial frustration were sub-micron ridges and triangular pillars, separated by gaps as small as 500nm. In LiNbO₃ the etch rate was found to increase on the -z face. The etch rate on the +z face was unaffected in both.

Direct ablation with an excimer laser produced relief structures. Aspect ratios > 1:1 resulted in a dendritic structure in the ablated area. Direct ablation was suitable for patterning the photoresist. Surface damage was intentionally induced when producing large (>100µm) openings, however, the effect of surface damage on electric field poling could not be conclusively tested. Submicron openings were also created and subsequent poling produced sub-micron domains, revealed by etching.