Technologies for single chip integrated optical gyroscopes

Abstract

Optical gyroscopes are being employed for navigational purposes for decades now and have achieved comparable or better reliability and performance than rotor-based gyroscopes. Mechanical gyros are however generally bulky, heavy and consume more power which make them unsuitable for miniaturized applications such as cube satellites and drones etc. Therefore, much effort is being expended worldwide to fabricate optical gyros having tactical grade robustness and reliability, small size, weight, cost and power consumption with minimal sacrifice of sensitivity. Integrated optics is an obvious approach to achieving this. This work comprises detailed comparative analysis of different types and structures of integrated optical gyroscopes to find out the suitable option for applications which require a resolution of <10 o/h. Based on the numerical analysis, Add-drop ring resonator-based gyro is found to be a suitable structure for integration which has a predicted shot noise limited resolution of 27 o/h and 2.71 o/h for propagation losses of 0.1 dB/cm and 0.01 dB/cm respectively. An integrated gyro is composed of several optical components which include a laser, 3dB couplers, phase/frequency modulators, sensing cavity and photodetectors. This requires hybrid integration of multiple materials technologies and so choices about which component should be implemented in which technology. This project also undertakes theoretical optimization of few of the above-mentioned optical components in materials systems that might offer the most convenient/tolerant option, this including 3dB coupler, thermo-optic phase modulator and sensing cavity (resonator and waveguide loop). In particular, the sensing element requires very low propagation loss waveguides which can best be realised from Si3N4 or Ta2O5. The optimised Si3N4 or Ta2O5 waveguides however are not optimal for other functions and this is shown and alternatives explored where the Si3N4 or Ta2O5 can easily be co-integrated. The fabrication process of low loss Si3N4 and Ta2O5 waveguides are also reported in this thesis. Si3N4 films were grown by using low pressure chemical vapor deposition (LPCVD) technique. Dry etching of Si3N4 films have been optimized to produce smooth and vertical sidewalls. Experimental results predicted that the propagation loss of 0.009 dB/cm is achievable by using optimum waveguide dimensions and silica cladding with the relatively standard processes available within the Laser Physics Centre at the Australian National University. A CMOS back end of line compatible method was developed to deposit good quality Ta2O5 films and silica claddings through ion beam sputtering (IBS) method. Plasma etching of Ta2O5 waveguides has been demonstrated by using a gas combination of CHF3/SF6/Ar/O2. Oxygen was introduced into the chamber to produce non-vertical sidewalls, so the waveguides could be cladded without voids with IBS silica. Average propagation losses of 0.17 dB/cm were achieved from Ta2O5 waveguides which appeared after extensive investigation to be limited by the spatial inhomogeneity of the processing. Lastly, a detailed theoretical and experimental analysis was performed to find out the possible causes of the higher average propagation loss in Ta2O5 waveguides, some sections being observed with 0.02 dB/cm or lower losses.