Abstract
Tuneable photonic devices have drawn attention from both research and industry for wide ranging applications that include displays, sensors, and communication. Optical devices made from traditional materials, such as silicon and glass, usually have a fixed optical response. Incorporating materials with tuneable optical properties provides a sophisticated way of tuning photonic devices. In recent years, chalcogenide phase change materials (PCMs) have been widely studied for tuneable photonics applications because they can have a large optical contrast between the amorphous and crystalline states. Switching between the two states is fast and reversible. Moreover, the switch is nonvolatile. Therefore, PCMs are promising materials for reconfigurable photonic devices. Indeed, one of the most common PCMs, Ge2Sb2Te5, has been widely studied in tuneable photonic devices. However, Ge2Sb2Te5 has a large absorption in the visible and Near Infrared (NIR) spectrum, which limits the performance in many photonics applications. To overcome this drawback, a wide bandgap PCM is needed. The research question to address in this thesis is which photonics applications would a wide bandgap PCM be most useful? In this work, Sb2S3, which has a bandgap of 2.0 eV and 1.7 eV for the amorphous and crystalline states respectively, was studied for tuneable photonics applications. It has a low absorption in the visible and NIR spectral range. The objective of this thesis is to establish whether Sb2S3 can be used to tune photonic devices, and to understand which applications it is best suited. Tuning plasmonic devices using PCMs has been widely studied for sensing and light manipulation. Typical plasmonic designs usually consist of metal and dielectric layers. However, the inter-diffusion between PCMs and the plasmonic metals is rarely considered. I found that inter-diffusion can affect the iv optical performance of the tuneable photonic devices. However, I also report that this interfacial damage can be prevented using a diffusion barrier, such as Si3N4, or a stable plasmonic metal, such as TiN. This understanding allowed me to develop a hyperbolic metamaterial (HMM) that exploits Sb2S3 and Ag, with a Si3N4 diffusion barrier. By changing the structural phase of Sb2S3, the hyperbolic dispersion spectral range in the visible could be programmed. I demonstrated that spontaneous emission of quantum dots (QDs) can be enhanced on HMMs, and the decay rate can be tuned using the structural phase transition of the Sb2S3 HMM layers. Due to the intrinsic losses in plasmonic metals at visible and NIR frequencies, tuneable plasmonic devices tend to absorb light strongly. This renders plasmonics impractical for many low-loss photonics applications. One promising solution to this loss-problem is to make all-dielectric photonic devices. I report that Sb2S3 nanoantennas with a large refractive index can support electric and magnetic dipole resonances. Thus, Sb2S3 is a promising material for tuneable all-dielectric metasurfaces in the visible and NIR spectrum. Nanoantenna array metasurfaces using Sb2S3 were designed for two different applications: beam steering and tuneable colour displays. By changing the structural phase of Sb2S3, these nanoantenna metasurfaces exhibit beam bending and tuneable colour. In addition to all-dielectric tuneable metasurfaces, on-chip light routing and modulation using Sb2S3 as an active dielectric were demonstrated. Due to the low loss in the NIR spectrum, Sb2S3 is a promising material for reconfigurable waveguide components. Sb2S3 was incorporated into InP directional couplers, ring resonators and Mach-Zehnder nterferometers. Changing the structural phase of Sb2S3 allowed control of the light propagation in the waveguide and optical switching. By studying the low loss and tuneable optical properties of Sb2S3 in a variety of tuneable visible-NIR photonic devices, I found that Sb2S3 is most well-suited to tuneable nanoantenna metasurfaces and reconfigurable waveguide components.