Optical and plasmonic coupling in linear and nonlinear regimes

Surface plasmon polaritons are electromagnetic surface waves that may be excited at the interface between metallic and dielectric materials. They confine light to subwavelength dimensions, and provide large field enhancement, making them useful for creating compact devices and sensing applications. Unfortunately, surface plasmons are difficult to utilise due to their characteristically large losses. They typically have propagation lengths from microns to hundreds of microns. Surface plasmon-to-optical coupling is also challenging, due to the inherent disparity in the size and wavevector of optical and plasmonic modes. In this thesis,we take two approaches to addressing these challenges. The first approach is to explore second-order nonlinear effects. Nonlinear effects benefit from the plasmon field-enhancement, and can potentially be used to provide gain via parametric amplification. In our study, we experimentally consider the nonlinearity of gold in plasmon-to-plasmon second harmonic generation, as well as the dielectric nonlinearity in lithium niobate. In gold-coated lithium niobate crystals, we explore optical-to-plasmonic processes, using quasi-phase-matching for second harmonic generation, and birefringent phase-matching for parametric down conversion. We find that the large size disparity in optical and plasmonic modes strongly inhibits the efficiency of the nonlinear response, and diminishes the contributions made by phase-matching. We were unable to overcome the plasmonic absorption losses using second-order nonlinear effects in the nanosecond domain. In our second approach, we consider directional coupling between optical and longrange surface plasmon waveguides using numerical methods. We then consider the design of a three-waveguide device, an adiabatic passage coupler with a plasmonic intermediate waveguide. By the use of a dark state, we show that light can be transported through the plasmonic waveguide without suffering plasmonic loss. This is achieved by supressing the surface plasmon amplitude. An analogous device, a digitised adiabatic passage coupler, is characterised experimentally, verifying the tolerance to loss in that design.