Controlled light propagation in random, periodic, and superperiodic silicon nanophotonic materials
Due to the COVID-19 crisis the PhD defence of Manashee Adhikary will take place (partly) online.
The PhD defence can be followed by a live stream.
Manashee Adhikary is a PhD student in the research group Complex Photonic Systems (CPS). Her supervisor is prof.dr. W.L. Vos from the Faculty of Science and Technology (S&T).
This thesis explores the possibilities of controlled light propagation in different nanophotonic media, namely: randomly scattering, periodic and super-periodic media. Various instances of these types of materials have been experimentally investigated for the different possible ways of (controlled) interaction of near-infrared (NIR) light with such materials. The ultimate goal is to build up completely new knowledge and expertise about the special mode of transportation of light in superperiodic systems. These special structures consist of a three-dimensional (3D) grid of coupled optical cavities. Such a configuration provides access to a form of light transport that has been known only in electronic systems. These special conditions can give access to never-before-seen phase transitions like Anderson localization. The structures are also CMOS compatible and functional in the telecom bands that are used for wireless communication. This gives the possibility to combine these structures in the future with other existing electro-optical components.
First, we provide a detailed overview of the experimental setup that is used in all the measurements shown in this thesis. With this setup, it is possible to measure broadband reflection, transmission as well as lateral scattering of various optical media. Furthermore, a spatial light modulator (SLM) allows us to perform wavefront shaping to steer light. The specialty of the setup is to measure samples optically over a wide spectral range from 950 to 2000 nm, with a narrow spectral linewidth of 0.6 nm. All relevant technical details are covered, and the limitations of the setup are explored. In addition, an overview is given of the CMOS-compatible optical samples measured in this thesis, namely 2D photonic crystals, 3D photonic crystals, and 3D superperiodic crystals.
Optical speckles are produced when different light paths in a multiply scattering medium interfere coherently. In experiments, speckles are affected by the interaction between light and the scattering material on the one hand, and the properties of the light source and the detector on the other. Typically, speckle is measured with a narrow-band light source because such source has a long coherence length. However, the source used in our experimental setup has a comparatively broad linewidth than typical narrowband laser sources. The large linewidth reduces the temporal coherence of the light and thus influences the speckle statistics. We examine how this linewidth influences the speckle statistics using measurements on a sample consisting of zinc oxide (ZnO) nanoparticles. We show that the speckle statistics deviate from the expected exponential distribution. This deviation is due to both the light source and the high noise of the detector used. This conclusion has been validated with measurements with approximately monochromatic light sources and a detector with less noise. This quantifies the limitations of the setup and that the setup properties are sufficient for meaningful measurements described in this thesis.
By changing the spatial phase of the wavefront of the incident light, it is possible to realize a focus behind a scattering medium. This process is known as wavefront shaping (WFS). Typically, the optimized wavefront is unique to the frequency of the incoming light, but there exists some correlation with optimized wavefronts for the neighboring frequencies. This is given by the speckle correlation function. We investigate for the entire range of telecom wavelengths the spectral width of this correlation. It follows that for weak scattering media it is possible to reach the spectral bandwidth of WFS to tens of THz (about 100 nm) in the NIR. These results indicate the opportunity to implement WFS with scattering media for focusing light over a very wide frequency range.
3D photonic crystals with the inverse woodpile structure, fabricated by CMOS-compatible methods were studied and found to exhibit a complete photonic band gap, whose frequency range depends on the radius of nanopores and the lattice parameter. The crystals exhibit wide photonic band gaps in various ranges of the telecom bands. Up to 96% reflectivity within the band gap was measured, which is to our knowledge record the highest band gap reflectivity of 3D photonic crystals. Such high-quality band gap crystals are ideal for shielding active sources when placed inside and to confine defect modes within the band gap.
We performed spectral measurements of reflectivity and lateral scattering from 2D photonic crystals that reveal the directional photonic gaps of such structures. Within the gap, where otherwise no Bloch modes are allowed, the disorder-induced scattered light within the gap was steered deep into the crystal far beyond the depth called Bragg length that is allowed conventionally by crystal diffraction. Wavefront shaping (WFS) of telecom light in the L-band on 2D photonic crystals to send light as deep as 8× the Bragg length of the crystal.
A structure of periodically placed coupled cavities embedded in a 3D photonic crystal forms a superperiodic photonic medium. Such a medium gives the possibility to make light ‘hop’ between the various neighboring cavities. This hopping takes place within the band gap of the underlying 3D crystal, where normally no light propagation is possible. Both broadband reflection and scattering measurements on such superperiodic media reveal the existence of these cavity modes inside the band gap of the 3D photonic crystals. This is the very first observation of a completely new form of light transport in a 3D super-periodic photonic crystal.