A Symphony of Noise in Nonlinear Optical Resonators

Understanding the effect of noise on light’s behavior is crucial, as light plays a central role in the study of many systems. This thesis explores the interplay between light, matter, and noise, demonstrating how this interaction can lead to fascinating phenomena and practical applications in sensing, signal transmission, and computing.

Chapter 1 provides an overview of the history of noise in physical systems and fundamental discoveries that shaped our understanding of noisy systems. The focus then shifts to optical systems, particularly nanophotonic sensors and devices, and introduces a basic model describing nonlinear optical resonators, pivotal in modern photonics.

Chapter 2 delves into the stochastic dynamics of light in a coherently-driven Kerr-nonlinear optical cavity, drawing analogies between light dynamics in the cavity and Brownian particles in a scalar potential. While this analogy holds in linear, on-resonance conditions, it breaks down in nonlinear or off-resonance scenarios due to non-conservative forces. An effective equilibrium description is proposed, providing accurate results in limited regimes but failing in deep bistability. This breakdown is marked by significant changes in the intra-cavity field fluctuation spectrum. Our findings lay the groundwork for an effective thermodynamic description of coherently driven cavities, bridging stochastic thermodynamics and resonant optics.

Chapter 3 introduces an innovative optical sensing approach using a single-mode Kerr-nonlinear resonator, achieving exceptional precision at a square-root singularity. This sensor leverages the dynamic hysteresis of the resonator, with intensity splitting at hysteresis endpoints scaling with the square root of perturbation strength. This novel method enhances sensing precision even in noisy environments and overcomes the traditional trade-off between precision and averaging time.

Chapter 4 presents the first experimental demonstration of non-Markovian stochastic resonance (SR), showing how memory effects from a non-instantaneous nonlinear response can broaden the SR bandwidth significantly. We imprint a signal on a laser by modulating the cavity length of a thermo-optical cavity. SR is evidenced by a peak in the signal-to-noise ratio as a function of noise variance. Our insights suggest broader implications for noise-assisted processes in various natural and engineered systems with a non-instantaneous response.

Chapter 5 and 6 explore complex dynamics emerging from memory effects in nonlinear resonators. Memory effects lead to stable limit cycles and self-sustained oscillations, both in nonlinear optical and mechanical systems. We demonstrate that memory effects can greatly enhance the ability of a nonlinear mechanical resonator to harvest energy from ambient vibrations.

Chapter 7 challenges the belief that enhancing clock precision requires more energy, demonstrating that precision can be altered at constant energy by varying laser-cavity detuning in a laser-driven thermo-optical dimer. Stochastic simulations show that detuning changes the limit cycle's confining potential and the phase diffusion strength, explaining the changes in precision.

Finally, Chapter 8 demonstrates enhanced performance of an all-optical flip-flop by leveraging noise, reducing switching energy by approximately 90% while maintaining 95% reliability. This approach uses noise as a resource, improving energy efficiency and reliability in optical systems and potentially extending to other platforms. Our findings pave the way for computational devices utilizing ambient fluctuations as a resource.