Title: Optical Pattern Formation and the Detection of Phase Singularities
Submitted: November 1990
Supervisors: Profs Norman Heckenberg, Halina Rubinsztein-Dunlop & Dr Carl Weiss
Examiners: Profs John Mainstone & Gerard Milburn

This thesis was published in the days when "cut and paste" was more than a metaphorical phrase and so there is currently no electronic version available. However if anyone is really interested, email me and I'll scan a version for you.

A high gain, high loss, sodium dimer laser was built to confirm the existence of the recently predicted and observed cooperatively frequency locked stationary patterns. These patterns were observed and a preliminary investigation undertaken of their stability. A striking feature in most of the steady state patterns observed was the presence of dark points, which have previously been proposed as corresponding to phase singularities. A possible interferometric technique to allow direct experimental detection of these singularities was investigated. Experimental results from this technique directly reveal the presence of phase singularities for the first time. The advantages and possible uses for this technique are discussed. Calibration of the interferometric technique via astigmatic imaging was attempted with limited success. Future directions for research in this area are proposed.

Title: Classical and quantum dynamics of optical frequency conversion
Submitted: March 1997
Supervisors: Profs Hans-A. Bachor & David McClelland
Examiners: Prof. Keren Bergman, Assoc-Prof. Craig Savage and Dr Andreas Sizmann

Apologies for the appalling English in this abstract—it's literally the worst part of the thesis! I wrote this very early in the morning on the day of hand- in and boy does it show...

A second harmonic generator is constructed to investigate the power and noise behaviour of optical frequency conversion.

Strong squeezing of the second harmonic is demonstrated. It is found that pump noise critically affects the squeezing, with attenuation of the pump noise significantly improving the squeezing. A modular modelling approach is used to describe and quantify this effect, and excellent agreement is found between theory and experiment.

Two methods of SHG are possible, passive (occurs external to a laser) and active (occurs within a laser). Theoretically exploring the possible squeezing regimes, the effect of laser noise on both methods is considered. It is concluded that active SHG is not feasible, as the high dephasing of practical lasers totally destroys the squeezing.

It is shown that the second harmonic generator can simultaneously support multiple, interacting, second order nonlinear processes. Two categories of interaction are identified: competing, where the interacting processes do not share all of the modes; and cooperating, where they do.

Competing nonlinearities are evident in the system as triply resonant optical parametric oscillation (TROPO): where second harmonic generation (SHG) and non- degenerate optical parametric oscillation (NDOPO) occur simultaneously. Power clamping of the second harmonic and nondegenerate frequency production in both the visible and infrared are observed and explained, again obtaining good agreement between theory and experiment. Design criteria are given that allow TROPO to be avoided in future efficient SHG systems.

The second harmonic squeezing is observed to be degraded by TROPO, with a maximum value occurring just before the onset of TROPO -- in contrast to predictions for closely related systems. A model is developed that shows this is due to two effects: a noise eating effect related to the second harmonic clamping; and low frequency noise added by the additional TROPO modes.

A model of cooperating nonlinearities is developed that shows a wide variety of third order effects, including cross- and self phase modulation (Kerr effects) and two photon and Raman absorption, are in principle possible in the second harmonic generator. A strong third order effect is demonstrated experimentally: the system is phase mismatched and optical bistability is observed that is shown to be due to the Kerr effect.

Arguments are presented to prove that, in principle, the system acts as a Kerr medium even at the quantum level. A model of Kerr squeezing is developed that allows consideration of the effect of pump noise: it is shown that the predicted squeezing is sensitive to both the amplitude and phase quadratures of the pump. Strong classical noise reduction (but no squeezing) is observed on light reflected from the cavity. It is speculated that the squeezing is masked by excess phase noise from the laser.

Due to the quantitative and qualitative agreement between experiment and theory, and the experimental reliability of the system, it is concluded that SHG is now a well understood and practical source of squeezed light. The potential for future systems, given the availability of new nonlinear materials, is discussed.