The capability of light goes well beyond illumination, yet it is so underused in our lives because the control of light still largely relies on clumsy bulk lenses. Less than 10 years ago, a type of revolutionary devices made of nanometer scale optical elements – metasurfaces – was invented to control the light propagation and its energy dissipation with arbitrary degree of freedom, at unprecedentedly small volumes (although some would argue that the advent of metasurfaces came in the 1990s). Vast diversity of new discoveries has since been made possible, and many more existing applications have seen significant performance enhancement with the aid of metasurfaces.<div><br><div> <div>In the scope of this work, I explore the use of a variety of metasurfaces to address several existing real-world challenges: sensing, optical heating, and data storage. Among these, three metasurfaces involve the world’s first two-dimensional material, graphene. I first investigate the graphene plasmonic resonator, which have been shown to be extremely sensitive single-molecule sensors. Graphene also has many intriguing properties in photodetection applications, such as lightweight, ultra-wide detection band, and ultrafast response speed. I have used two different metasurfaces to enhance the intrinsically low responsivity (sensitivity) of graphene photodetectors. Amidst the discussion of graphene photodetectors, I show the characterization result of plasmonic heating of metasurfaces, an essential process of the graphene photo-responsivity enhancement. Lastly, I present a multi-functional metasurface which can be used in optical steganography, encryption, and data storage. The proposed metasurface is compatible with large scale parallel readout, which outperforms current Blu-ray technology in both storage capacity and readout speed</div></div></div>

<div>In the 1970's, Arthur Ashkin developed a remarkable system called the ``optical tweezer'' which utilizes the radiation pressure of light to manipulate particles. Because of its non-invasive nature and controllability, optical tweezers have been widely adopted in biology, chemistry and physics. In this dissertation, two applications related to optical tweezers will be discussed. The first application is about the demonstration of multiple feedback controlled optical tweezers which let us conduct novel experiments which have not been performed before. For the second application, levitation of a silica nanodumbbell and cooling its motion in five degrees of freedom is executed.</div><div><br></div><div>To be more specific, the first chapter of the thesis focuses on an experiment using the feedback controlled optical tweezers in water. A well-known thought experiment called ``Feynman's ratchet and pawl'' is experimentally demonstrated. Feynman’s ratchet is a microscopic heat engine which can rectify the random thermal fluctuation of molecules to harness useful work. After Feynman proposed this system in the 1960’s, it has drawn a lot of interest. In this dissertation, we demonstrate a solvable model of Feynman’s ratchet using a silica nanoparticle inside a feedback controlled one dimensional optical trap. The idea and techniques to realize two separate thermal reservoirs and to keep them in contact with the ratchet is discussed in detail. Also, both experiment and simulation about the characteristics of our system as a heat engine are fully explored.</div><div><br></div><div>In the latter part of the dissertation, trapping silica nanodumbbell in vacuum and cooling its motion in five degrees of freedom is discussed. A levitated nanoparticle in vacuum is an extraordinary optomechanical system with an exceptionally high mechanical quality factor. Therefore, levitated particles are often utilized as a sensor in various research. Different from a levitated single nanosphere, which is only sensitive to force, a levitated nanodumbbell is sensitive to both force and torque. This is due to the asymmetry of the particle resulting it to have three rotational degrees of freedoms as well as three translational degrees of freedoms. In this dissertation, creating and levitating a silica nanodumbbell will be demonstrated. Active feedback cooling also known as cold damping will be employed to stabilize and cool the two torsional degrees of freedom of the particle along with the three center of mass DOF in vacuum. Additionally, both computational and experimental analysis is conducted on a levitated nanodumbbell which we call rotation-coupled torsional motion. The complex torsional motion can be fully explained with the effects of both thermal nonlinearity and rotational coupling. The new findings and knowledge of a levitated non-spherical particles leads us one step further towards levitated optomechanics with more complex particles.</div>

<p>Climate change and the energy crisis are substantial challenges facing the human species, and they are projected to threaten life on our planet. For millions of years, the sun has been the main source of energy for life on Earth; this inspires ongoing research efforts focusing on a “sunlight to fuel” energy solution. Photosynthesis is nature’s tool to derive energy from the sun. Hence, scientists focus on the biochemistry of this phenomenon to employ photosynthesis in a man-made device. Such a device is able to convert solar energy to chemical energy through a light-driven cycle of the chemical reactions which produce hydrogen gas, later used as fuel. This process, often called “artificial photosynthesis,” needs efficient catalysts which can be incorporated into a molecular assembly and other microscopic structures or immobilized on an electrode surface. </p><p>Additionally, evolution, in the course of billions of years, chose manganese as an abundant and effective metal to facilitate the process of photosynthesis. These manganese atoms formed a cluster and an optimized ligand field to maximize efficiency. The photochemistry and photo-physics process behind photosynthesis is yet to be fully understood and implemented in a man-made apparatus with comparable efficiency and durability. </p><p>Photosynthesis requires a source of electrons. Water is an abundant molecule on earth that can provide the electrons needed for the photosynthesis. Although water is ubiquitous, it is one of the most stable molecules; hence, splitting it demands a well-designed system with strong oxidizing capability. Because a single atom of oxygen is highly reactive, there should be at least four oxidation states in the system to remove four electrons and release molecular oxygen: O2. The O-O bond formation is one of the most important steps in photosynthesis to fully understand. Lacking a thorough knowledge of this step prevents design and fabrication of robust and active water oxidizing catalysts. To fully understand O-O formation, one should perform a comprehensive study of each of the intermediates of the system. In other words, we need an understanding of the structure and electronic configuration of the system (natural or artificial) from the moment that a water molecule attaches to the catalyst (usually a metal core, central in the complex), until the moment that oxygen released as an O2 molecule. </p><p>There are multiple possible mechanisms to explain O-O formation. Two mechanisms that were extensively studied in this thesis are water nucleophilic attack and radical coupling. The prevailing view about oxygen formation in the catalysts that we study here explains the O-O bond formation by nucleophilic attack of a water molecule to a highly oxidized ruthenium (RuV=O) species. In this hypothesis, all polypyridine ligands that are coordinated to ruthenium remain neutral during the water oxidation process, while the formation of RuV=O (the key intermediate) would require a relatively high free energy (about 1.8 to 2 eV); use of computational (numerical) calculations determine this to be thermodynamically inaccessible. Furthermore, the failure of spectroscopic techniques to confirm the presence of RuV=O calls the validity of this model into question.)</p><p>Alternatively, radical coupling hypothesis considers another pathway to oxygen bond formation. Here, one of the nitrogen atoms coordinated to ruthenium in polypyridine plays a crucial role. We hypothesize that after formation of RuIV=O (which is spectroscopically observed), one nitrogen decoordinates from the metallic core (ruthenium) and oxidizes to form Ru-ON species. This N-oxide ligand can be further oxidized to form a ligand cation radical. It has been shown that [ligand-NO]+• can have almost no energy barrier for O-O bond formation via spin alignment. The study of the role of N-oxide is one of the main focuses of this work. Since this hypothesis does not require RuV=O nor water nucleophilic attack, it explains the process of water oxidation and opens further avenues for the design of future catalysts.</p><p>To confirm our hypothesis, I employed several spectroscopic methods and computational calculations. This new pathway predicts new intermediates exclusive to this model. Our objective is to prove their presence by in situ spectroscopy and test the possibility of formation of each intermediate computationally, to see if their formation is thermodynamically feasible. </p><div><br></div>

A complex environment causes strong distortion of the field, inhibiting tasks such as imaging and communications in both the optical and radio-frequency (RF) region. In the optical regime, strong modal dispersion in highly multimode fiber (MMF) results in a scrambled output field in both space (intensity speckles) and time (spectral and temporal speckles). Taking advantage of the pulse shaping technique, spatial and temporal focusing has been achieved in this thesis, offering potential opportunities for nonlinear microscopy and imaging or space-division multiplexed optical communication through MMF. In the RF regime, multipath effect in wireless RF channel gives multiple echoes with random delay and amplitude attenuation at the receiver end. Static channel sounding and compensation with ultra-broadband spread spectrum technique resolves the issue by generating a peaking signal at the receiver, significantly improving the signal-to-noise/interference performance. However, the limited communication speed in the static approach makes it challenging for sounding and compensation in a dynamic channel. Here, we achieve real-time channel sounding and compensation for dynamic wireless multipath channel with 40 micro-seconds refresh rate by using a fast processing field programmable gate array (FPGA) unit, providing potential opportunities for mobile communications in indoor, urban, and other complex environments. Furthermore, by combining broadband photonics and RF radar technologies, a high depth and transverse resolution wide bandwidth (15 GHz) W-band (75 - 110 GHz) photonic monopulse-like radar system for remote target sensing is demonstrated, offering prospects for millimeter wave 3-D sensing and imaging.

<p>Optimization and Machine learning (ML) have emerged as two positively disruptive methodologies and have thus resulted in unprecedented applications in several domains of technology. In recent years, ML has forayed into physical sciences and provided promising outcomes thanks to its ability in representing and generalizing complex functions to reveal underlying relations among variables describing a system. By casting ML as an optimization task, we first focus on its application in solving quantum many-body problems. Leveraging the power of quantum computation, we develop hybrid quantum machine learning protocols and implement benchmark tests to calculate the band structures of two-dimensional materials. We also show how this method can be used to estimate the critical point for a quantum phase transition. One  hurdle in such techniques is related to parameter optimization, wherein to obtain the desired result, the parameters have to be optimized, which can be computationally intensive. For a particular class of problem and a choice of algorithm, we deduce a simple parameter setting rule. This rule is projected as a heuristic and is validated numerically for several problem instances. Finally, by venturing into thermal photonics, a framework that takes advantage of the spectral and spatial information of hyperspectral thermal images to establish a completely passive machine perception, titled HADAR is presented. A conventional deep neural network is developed that utilizes the governing equation of HADAR and its performance in semantic segmentation is demonstrated. Altogether, this report establishes the need for creative algorithms that exploit modern hardware to solve complex problems that were previously deemed unsolvable.</p>

<p>There is growing interest in the use of nanosecond surface dielectric barrier discharge (ns-SDBD) actuators for high-speed (supersonic/hypersonic) flow control. A plasma discharge is created using a nanosecond-duration pulse of several kilovolts, and leads to a rapid heat release and a complex three-dimensional flow field. Past work has been limited to qualitative visualizations such as schlieren imaging, and detailed measurements of the induced flow are required to develop a mechanistic model of the actuator performance. </p><p><br></p><p></p><p>Background-Oriented Schlieren (BOS) is a quantitative variant of schlieren imaging and measures density gradients in a flow field by tracking the apparent distortion of a target dot pattern. The distortion is estimated by cross-correlation, and the density gradients can be integrated spatially to obtain the density field. Owing to the simple setup and ease of use, BOS has been applied widely, and is becoming the preferred density measurement technique. However, there are several unaddressed limitations with potential for improvement, especially for application to complex flow fields such as those induced by plasma actuators. </p><p></p><p>This thesis presents a series of developments aimed at improving the various aspects of the BOS measurement chain to provide an overall improvement in the accuracy, precision, spatial resolution and dynamic range. A brief summary of the contributions are: </p><p>1) a synthetic image generation methodology to perform error and uncertainty analysis for PIV/BOS experiments, </p><p>2) an uncertainty quantification methodology to report local, instantaneous, a-posteriori uncertainty bounds on the density field, by propagating displacement uncertainties through the measurement chain,</p><p>3) an improved displacement uncertainty estimation method using a meta-uncertainty framework whereby uncertainties estimated by different methods are combined based on the sensitivities to image perturbations, </p><p>4) the development of a Weighted Least Squares-based density integration methodology to reduce the sensitivity of the density estimation procedure to measurement noise.</p><p>5) a tracking-based processing algorithm to improve the accuracy, precision and spatial resolution of the measurements, </p><p>6) a theoretical model of the measurement process to demonstrate the effect of density gradients on the position uncertainty, and an uncertainty quantification methodology for tracking-based BOS,</p><p>Then the improvements to BOS are applied to perform a detailed characterization of the flow induced by a filamentary surface plasma discharge to develop a reduced-order model for the length and time scales of the induced flow. The measurements show that the induced flow consists of a hot gas kernel filled with vorticity in a vortex ring that expands and cools over time. A reduced-order model is developed to describe the induced flow and applying the model to the experimental data reveals that the vortex ring's properties govern the time scale associated with the kernel dynamics. The model predictions for the actuator-induced flow length and time scales can guide the choice of filament spacing and pulse frequencies for practical multi-pulse ns-SDBD configurations.</p>

<p>Dipole-dipole interactions are ubiquitous fundamental physical phenomena that govern physical effects such as Casimir Forces, van der Waals forces, collective Lamb shifts, cooperative decay, and resonance energy transfer. These interactions are associated with real and virtual photon exchange between the interacting emitters. Such interactions are crucial in realizing quantum memories, novel super-radiant light sources, and light-harvesting devices. Owing to this, the control and modification of dipole-dipole interactions have been a longstanding theme. The electromagnetic environment plays a crucial role in enhancing the range and strength of the interactions. This work focuses on modifying the nanophotonic environment near interacting emitters to enhance dipole-dipole interactions instead of spontaneous emission. To this end, we focus on engineering the nanophotonic environment to enhance the strength and range of dipole-dipole interactions between an ensemble of emitters. We explore ordered and disordered nanophotonic structures. We experimentally demonstrate long-range dipole-dipole interactions mediated by surface lattice resonances in a periodic plasmonic nanoparticle lattice. Further, the modified electromagnetic environment reduces the apparent dimensionality of the interacting system compared to non-resonant in-homogeneous and homogeneous environments. We also develop a spectral domain inverse design technique for the accelerated discovery of disordered metamaterials with unique spectral features. </p> <p>Further, we explore the novel regimes of light localization at near-zero-index in such disordered media. The disordered near-zero-index medium reveals enhanced localization and near-field chirality. This work paves the way to engineer the electromagnetic nanophotonic environment to realize enhanced long-range dipole-dipole interactions.</p>