Breast cancer is a prevalent disease within today’s modern society, affecting 1 in 8 women and 1 in 870 men within a lifetime. With the introduction of mammographic breast screening in 1987 and marked improvements to targeted therapies, mortality rates declined, highlighting the need for early diagnosis and tailored treatment to halt disease progression in its foremost stages. Histology assessed biopsies, alongside initial two-view mammographic imaging, are paired as the current diagnostic “gold standard”. The need to incorporate several techniques, applying an “all-angles” approach to diagnostics, provides an effective, streamlined diagnostic pathway, reducing patient wait times between testing and results – crucial in preventing disease progression. Optical spectroscopic techniques for the characterisation of biomolecular compounds and structures present within tissue are fast becoming the biomedical analysis tools of choice, coming to the forefront of clinical applications. Raman spectroscopy is one such technique providing highly chemically specific results, in a non-ionising and non-invasive way. When used in conjunction with metal nanoparticle probes, the inherently weak Raman signals of the biomolecules surrounding the nanoparticle surface undergo extensive levels of enhancement – an eponymous technique, Surface Enhanced Raman Scattering (SERS). This thesis is split into two principal areas of study. The first explores extrinsic SERS nanoparticles at depth within optical phantoms, “imaged” in a Transmission orientation, mimicking the composition of the breast within a cranio-caudal mammographic imaging position. The second concerns the micro-Raman quantification of gold nanoparticles, functionalised to a biocompatible level for the active targeting of hydroxyapatite – a calcium apatite form which, when dominant within breast microcalcifications, act as a biomarker for malignancy. Key aspects drawn from the results include a greater understanding of Raman reporter gold nanoparticles at depth, and how the absorption profile of the sample material affects the garnered intensity profile. The synthesis of a novel nanoparticle probe was also founded, with promising future applications in terms of targeting and theranostic capabilities. Furthermore, a protocol into the implementation of an automated mapping system within an open optical set up is given, detailing the software, hardware and electrical installation requirements. The advantages of Raman spectroscopy integration within current diagnostic practices are highlighted, with limitations such as nanoparticle biocompatibility issues, the inherent optical properties of biological tissues, and system conditions touched upon.

In this thesis the cross-matching problem is considered. One of the most fundamental processes in astrophysics, the cross-matching of two photometric catalogues is the assignment of an object in one catalogue and one object from the second catalogue as pairs, i.e., different detections of the same physical source in the sky. I present here new methods for considering such a problem, including the additional magnitude information available in photometric catalogues to break degeneracies between astrometric matches. I also generalise the Astrometric Uncertainty Function (AUF), usually assumed to be a Gaussian, to allow for the inclusion of systematic astrometric perturbations, such as those from blended sources of contamination. The separations of sources in several widely used photometric catalogues with respect to the much more precise Gaia positions are considered. I find that the separations are described by a combination of a Gaussian distribution and a large non-Gaussian wing, and show that this is caused by flux contamination from blended stars not treated separately. At least one in three of the stars in the faint half of a given catalogue will suffer from flux contamination above the 1% level when the density of catalogue objects per point-spread function area is above approximately 0.005. I then introduce a new method to use the additional photometric information from both catalogues in the process of accepting or rejecting counterparts, providing approximately a factor 10 improvement in Bayes' factor with its inclusion. The method uniquely combines photometric information from both catalogues while avoiding the use of prior astrophysical knowledge. Additionally, I formally describe the probability of two sources being the same astrometric object, allowing systematic effects of astrometric perturbation (by, e.g., contaminant objects) to be accounted for. I apply this method to two key match cases, of two catalogues of similar wavelength coverage but differing dynamic ranges, and of two catalogues with approximately equal astrometric precision, discussing the importance of the inclusion of the magnitude information in each case. Finally, the extension to the inclusion of perturbations due to faint contaminant stars in the AUFs of catalogues is combined with the improved cross-matching method for the specific case of the Wide-field Infrared Survey Explorer (WISE) catalogue. I describe the rigorous construction of the description of astrometric offsets due to faint stars, and then apply the method to Gaia-WISE matches in the Galactic plane. I analyse several test cases and discuss the photometric effects of the blended star contamination, showing that stars with significant astrometric perturbation are detectably photometrically compromised. I discuss the implications this has on derived parameters in several areas of astrophysics.

This thesis focuses on two key aspects of light-matter interactions: absorption and spatial dispersion, both of which are described on a macroscopic scale by the electromag- netic susceptibility. The first part of this thesis contains an investigation into the microscopic origin of absorption in dielectric models, providing a detailed calculation for a long-held assumption of the Hopfield model that has formed the basis of many key works on the subject. While previous work has either focused on the quantum regime or used phenomenological methods which lack a clear relationship to the underlying physics, the microscopic model and calculations presented here are purely classical in nature, matching Hopfields initial proposal. A discrete model of a dielectric is developed, containing nonlinear interaction terms between polarizable dipoles and lattice vibrations. The lattice vibrations are found to act as a pseudo-reservoir, leading to broadband absorption of electromagnetic radia- tion that naturally emerges from the model, without the need to add damping terms to the dynamics. The effective linear susceptibility is calculated using a perturbative iteration method and is found to match the form of a model that is widely used for real dielectrics. The second half of the thesis presents a series of modifications to the Halevi-Fuchs susceptibility model, which is used to calculate the electromagnetic reflection and trans- mission coefficients of a spatially-dispersive half-infinite medium. The initial model, valid only for an idealized single-resonance scalar susceptibility with a specific wave vector dependence, is extended to include many more of the susceptibility features found in real materials, including unequal transverse and longitudinal components, multiple res- onances, anisotropy and alternate wave vector dependences. In each case, the effect of the boundary is characterized by a set of phenomenological reflection coefficients for the polarization waves in the medium, with specific values corresponding to various addi- tional boundary conditions for Maxwell’s equations. The exact expressions derived for the electromagnetic reflection and transmission coefficients can be used in the calculation of a range of physical phenomena near the boundary of the medium. This thesis consid- ers the spectral energy density of thermal and zero-point radiation outside the medium, with the key result that the inclusion of spatial dispersion naturally removes an unphysical divergence associated with the use of a spatially local susceptibility model.

This thesis illustrates the contribution of the author to experiments using liquid argon Time Projection Chambers (LAr TPCs), a technology already widely used, that is becoming the dominating detection technique in dark matter (DM) and neutrino searches.

Ce travail de thèse est consacré à l’étude de la croissance de fines couches organisées d’azaacènes par microscopie à effet tunnel (STM) spectroscopie de différence de réflectance (RDS) sous ultravide. Ces molécules, en plus de présenter des propriétés de semi-conducteur en couches minces, ont été fonctionnalisées afin de permettre la croissance de structures hautement organisées via la formation de liaison hydrogène. La croissance d’une monocouche de 6,17-dihydro-6,8,15,17-tetraazaheptacène (DHTA7) a été étudiée sur Au(111) et comparée avec la croissance du 5,14-dihydro-5,7,12,14-tetraazapentacène (DHTAP), réalisée précédemment au laboratoire. Ainsi le rajout d’un groupement phényle supplémentaire à chaque extrémité du DHTAP provoque, notamment, une diminution des interactions intermoléculaires. La croissance du DHTAP a été étudiée sur différentes surfaces : Cu(110), Cu(110)-(2x1)O et Al2O3/Ni3Al(111). Il a ainsi été montré que la molécule s’adsorbe à plat sur Cu(110) le long de la direction [1 ̅10] et forme une liaison covalente entre les atomes N et les atomes de Cu. Sur Cu(110)-(2x1)O, les molécules de DHTAP sont absorbées le long de la direction [001] du substrat. Les molécules de DHTAP forment ici une phase HOC (High Order Commensurate en anglais) unidirectionnelle composée d’un bloc de 7molécule dans la direction [1 ̅10]. Sur Al2O3/Ni3Al(111) l’adsorption en première couche est dominée par les défauts de l’oxyde. Bien que la nature du substrat joue un rôle important sur l’autoassemblage moléculaire en monocouche, des structures ordonnées similaires ont été observées sur Cu(110), Cu(110)-(2x1)O et Al2O3/Ni3Al(111) en multicouche par STM et RDS. / This thesis is devoted to the study of the growth of organized thin layers of azaacenes by scanning tunneling microscopy (STM) and reflectance difference spectroscopy (RDS) spectroscopy in ultra-high vacuum. These molecules, in addition to having semiconductor properties in thin films, have been functionalized to allow the growth of highly organized structures via hydrogen bond formation. The growth of a monolayer of 6,17-dihydro-6,8,15,17-tetraazaheptacene (DHTA7) was studied on Au (111) and compared with the growth of 5,14-dihydro-5,7,12,14-tetraazapentacene (DHTAP), previously performed in the laboratory. Thus the addition of an additional phenyl group at each end of the DHTAP causes, in particular, a decrease in intermolecular interactions. The growth of DHTAP was further studied on different surfaces: Cu(110), Cu(110)-(2x1)O and Al2O3/Ni3Al(111). It has been shown that the molecules adsorb flat on Cu (110) along the [1 ̅10] direction and form a covalent bond between N atoms and Cu atoms. On Cu(110)-(2x1)O, the DHTAP molecules are absorbed along the [001] direction of the substrate, which in this case is the direction of the dense CuO rows. Here, the DHATP molecules form a unidirectional High Order Commensurate (HOC) phase composed of a 7-molecule block along the [1 ̅10] direction. On Al2O3/Ni3Al(111) the adsorption in the first layer is dominated by the defects of the oxide layer. Although the nature of the substrate plays an important role in molecular self-assembly in the monolayer, similar ordered structures have been observed on Cu(110), Cu(110)-(2x1)O and Al2O3/Ni3Al(111) for DHTAP multilayers by STM and RDS.

Recent interest in conical diffraction (CD) has led to a large increase in experimental and theoretical investigations over the last two decades, a marked change from the previous 160 quiet years in the field. Once dismissed as an optical curiosity, the phenomenon has emerged as a fascinating area with potential for a large number of practical applications many of which have been realised while others are still being discovered. In this thesis a number of aspects of the theory as recently described are experimentally investigated with a view to strengthening the current theoretical understanding of the phenomenon. Developing from single crystal CD (the simplest case), through cascade CD, nonlinear CD, and with a particular emphasis on the polarisation effects of the phenomenon, a number of areas are investigated. Single crystal CD with circularly (CPL), linearly (LPL), azimuthally and radially polarised light (APL and RPL) is examined. The effect of LPL in removing a section of the ring orthogonally polarised to the incident beam is shown, along with the first investigation into the effects of RPL and APL polarisation effects in CD. The effect of the incident beam spot size on the pattern developed is also investigated and shown to conform to the theory. All findings show good agreement with the current theory. Cascade CD with various numbers of crystals and incident beam polarisations is investigated. Included in these experiments are a variable two crystal cascade and the first demonstration of the different patterns produced for a three crystal cascade when left and right circularly polarised light (LCPL and RCPL, respectively) are used, as recently predicted. As with the single crystal case, results are in agreement with theory. In both the single and cascade cases a cross section of the beam is captured to demonstrated the free space evolution of a CD beam. Simultaneous second harmonic generation (SHG) and CD from a single crystal is described in an update of Bloembergen et al.'s pioneering 1970's investigations with added emphasis on polarisation. SHG in non-phase matched conditions, as well as the influence of incident polarisation on the pattern and type of SHG, are observed. And finally a sensor based on CD demonstrating a simple, practical application of the phenomenon is outlined and a prototype device made and trialled. Using the effect of LPL described earlier to determine the polarisation angle of an incident beam, the device's use as a polarimeter is also tested to determine the specific rotation introduced by optically active liquids with the initial prototype showing results comparable to current methods. This work contains 7 chapters. The first is an introduction to the history of the phenomenon along with a thesis statement. Chapter 2 deals with single crystal CD, chapter 3 contains the expansion from single to cascade CD. The complexities introduced by various types of polarisation are described in chapter 4 for both single and cascade setups. Chapter 5 deals with SHG using a CD crystal and chapter 6 outlines the design and operation of the novel sensor based on the phenomenon. The final chapter is a summary of the work and outlook on the future of the field.

In this thesis, the smartphone is installed on vehicle as a tool to detect potholes on the road. The data collected and pre-processed by smartphone is then analysed to obtain the result. The history and status of pothole detection study is firstly discussed. The traditional pothole detection method has two disadvantages: 1, low efficiency; 2, limited detection area. To work out a new method, the suspension models and tyre models are studied. Also the axis-correction and de-noise in signal process discussed for data analysis. Based on these, the requirement of detection system is analysed, and the features of embedded systems are discussed, which leads to the conclusion that smartphone is the best hardware for pothole detection. Then the smartphone (3 mobiles are chosen: Samsung Note 1, Nexus 7 and iPhone 5) is fixed on vehicle (Nissan Micra K11 and Saab 93) as experiment platform. The Tay Road Bridge is chosen as experiment road. The software Sensor Insider Pro and Sensor Data are chosen to collect data. Simulations about vehicle and tyre are done by MATLAB Simulink. From the simulation the relationship between the bouncing height and the speed of vehicle is obtained, and is used as threshold in pothole detection. The experimental data from acceleration sensor and gyroscope is processed by MATLAB using axis-correction and wavelet transform. Together with velocity data gathered by GPS and threshold (calculated from speed), the position of pothole is obtained. Comparing with road, the detection has good accuracy, which proves the feasibility of this pothole detection method. At the end of thesis, other potential field of application of smartphone is discussed.

Despite the tremendous advances witnessed in light microscopy over the past two decades, non-invasive optical imaging is still limited to penetration depths smaller than 1 mm into tissue. Multiple scattering caused by the refractive index inhomogeneities of biological matter rapidly distort any optical wavefront prop-agating through, rendering tissues opaque. Such turbidity restricts imaging, as well as other biophotonics techniques, to the most superficial layers of tissue. A perspective strategy to overcome the turbidity of living matter exploits holographic light control in multimode optical fibres. This allows devising min-imally invasive imaging probes with footprints far bellow those of conventional endoscopes, as well as enhanced spatial resolution up to the diffraction limit de-termined by the numerical aperture (NA) of the fibre. In this Thesis, high-resolution focussing is demonstrated with unprecedented ability across novel specialty fibres offering very-high NAs, by devising a system and methodologies which allow counteracting the severe mode-dependent loss affecting such fibres. The high quality and NA of the generated foci is capable of 3D optical confinement of dielectric microparticles, thus enabling the deliv-ery of holographic optical tweezers introduced through a bare optical fibre with cross-section comparable to a single cell. The holographic methods developed allow the manipulation of complex 3D arrangements of particles, as well as their independent positioning with nanometre-scale precision in all three dimensions. Separately, a multimode fibre based deep-brain fluorescence imaging is demonstrated in animal models in vivo, allowing the identification of neuronal structures at depths exceeding 2 mm and resolving fine details down to ≈1 µm resolution.

Under the ever-increasing requirements of higher power and higher radio-frequency in future power electronics and telecommunication applications, the thermal management of gallium nitride (GaN) based devices becomes crucial, this can be significantly improved by integrating high thermal conductivity diamond into the devices to enhance the extraction of waste heat. An important consideration is the thermal boundary resistance (TBR) of the GaN/diamond interface, which forms a bottleneck for heat transport. For incorporation as a substrate, the thermal properties of this interface and of the polycrystalline diamond (PCD) grown onto GaN using various controlled barrier layers under different growth conditions are investigated and systematically compared; SiN barrier layers were found normally producing lower TBR with a smoother interface formed. For integration of PCD as a top-side heat spreader onto AlGaN/GaN-on-Si HEMT, its thermal performance was systematically evaluated by time-domain thermoreflectance and ANSYS simulation; at best a 15% reduction in peak temperature was obtained when only the source-drain opening of a passivated AlGaN/GaN-on-Si HEMT is overgrown with PCD. Meanwhile, next generation higher compacted electronics for future communications or computing require sub-10-nm or even atomic dimension scaling, incorporation of a new two-dimensional (2D) materials channel then has emerged as a highly attractive solution to address this challenge. Gallium telluride (GaTe) is a 2D layered material that recently raised considerable interests due to its unique optoelectronic properties but is still under extensive exploration of its fundamental properties for potential applications. The pressure-dependent solid-state properties of GaTe multilayers up to 46 GPa were firstly investigated. A strong Raman mode anisotropic splitting started at ~6.5 GPa originating from phase transition was first-time revealed and understood through first-principles calculations. Then the thermal properties of free-standing GaTe multilayers were studied mainly by micro-Raman opto-thermography, displaying an anisotropic and very low thermal conductivities along the in-plane armchair and zigzag orientations. Moreover, the mechanical properties of both SiO2/Si substrates supported and free-standing GaTe multilayers were investigated mainly through nanoindentation. Concurrence of multiple pop-ins and load-drops in the loading curve were found, likely originating from interlayer sliding within the GaTe multilayer. These pressure-tuned behaviors, thermal and mechanical properties of GaTe multilayers enable new insights for investigating and manipulating the anisotropic solid-state properties of potential device applications and other low symmetry layered materials.

In this thesis both the static and dynamic magnetic behaviour of complex three-dimensional nanoscale commercial hard disk drive write heads and thin film structures of interest to emerging spintronic devices have been investigated using a plurality of experimental techniques. The magneto-optical Kerr effect (MOKE) provides the basis for an optical microscopy technique sensitive to the magnetisation of a sample, detectable as a change in polarisation of light reflected from the sample surface. With a modelocked laser light source, synchronised electrical pulse generator and lock-in amplifier (LIA), a stroboscopic technique has been used to observe the magnetisation dynamics of hard disk drive write heads at 600 nm spatial resolution and 10 ps time resolution in response to a driving electrical pulse. The equilibrium magnetic state of these devices has been directly imaged by x-ray photo-emission electron microscopy (XPEEM), as well the stability of the equilibrium state in response to the application of an external bias field. Direct images of the equilibrium state obtained by XPEEM were found to agree with inferences made from MOKE images. Time-resolved scanning Kerr microscopy (TRSKM) images of magnetisation dynamics showed that flux does not form in ‘beams’ as commonly believed, but instead nucleates in separate sites across the writer. Static and time-resolved x-ray techniques have also been used to investigate a number of thin films of interest to spintronics. Spin pumping and spin transfer torque in Co2MnGe / Ag / Ni81Fe19 spin valves were explored using time-resolved x-ray ferromagnetic resonance (XFMR) carried out at Diamond Light Source (DLS), a as well as static x-ray magnetic circular dichroism (XMCD) for sample characterisation. This has provided element-specific measurements of the spin state in the source and sink layers of the spin valve, revealing a clear sign of spin transfer torque, while also investigating the role of sink layer thickness in spin pumping and damping. Ferrimagnetic yttrium iron garnet (Y3Fe2(FeO4)3) (YIG), a material of great interest in spintronics, has been studied by static and dynamic XMCD in comparison with ferromagnetic Co. While static and dynamic spectra for Co were identical, those for YIG differed markedly. While this may hint at a phase difference between the precession of Fe moments on different lattice sites, the true source of this difference has not been identified. Comparisons between vector network analyser ferromagnetic resonance (VNA-FMR) and XFMR measurements further suggest the presence of long-range inhomogeneities in the YIG. The spin dynamics of an antiferromagnet being driven by a ferromagnet have also been investigated using XMCD and x-ray magnetic linear dichroism (XMLD). A CoO / Fe / Ni81Fe19 trilayer wherein the thickness of the CoO layer varies across the sample has been thoroughly characterised by static XMCD and XMLD, providing information necessary to fully interpret time-resolved MOKE measurements on these samples. Measurements have shown that even small amounts of ordered CoO significantly modify the resonant field and linewidth of the adjacent ferromagnetic layers. Phase-resolved measurements of CoO spins have shown these spins to precess in phase with those of the adjacent Fe. The viability of dynamic XMLD measurements has also been confirmed. Finally, potential directions for future work in each project are discussed.