Sprays and other industrially relevant turbid media can be quantitatively and qualitatively characterized using modern optical diagnostics. However, current laser based techniques generate errors in the dense region of sprays due to the multiple scattering of laser radiation e ected by the surrounding cloud of droplets. In most industrial sprays, the scattering of light occurs within the so-called intermediate scattering regime where the average number of scattering events is too great for single scattering to be assumed, but too few for the di usion approximation to be applied. An understanding and adequate prediction of the radiative transfer in this scattering regime is a challenging and non-trivial task that can significantly improve the accuracy and e ciency of optical measurements. A novel technique has been developed for the modelling of optical radiation propagation in inhomogeneous polydisperse scattering media such as sprays. The computational model is aimed to provide both predictive and reliable information, and to improve the interpretation of experimental results in spray diagnostics. Results from simulations are verified against the analytical approach and validated against the experiment by the means of homogeneous solutions of suspended polystyrene spheres. The ability of the technique to simulate various detection conditions, to di erentiate scattering orders and to generate real images of light intensity distributions with high spatial resolution is demonstrated. The model is used for the real case of planar Mie imaging through a typical hollow cone water spray. Versatile usage of this model is exemplified with its applications to image transfer through turbid media, correction of experimental Beer-Lambert measurements, the study of light scattering by single particles in the farfield region, and to simulate the propagation of ultra-short laser pulses within complex scattering media. The last application is fundamental for the development and testing of future optical spray diagnostics; particularly for those based on time-gating detection such as ballistic imaging.

A novel, disaggregated approach to land cover survey is developed on the basis of land cover attributes; the parameters typically used to delineate land cover classes. The recording of land cover attributes, via objective measurement techniques, is advocated as it eliminates the requirement for surveyors to delineate and classify land cover; a process proven to be subjective and error prone. Within the North York Moors National Park, a field methodology is developed to characterise five attributes: species composition, cover, height, structure and density. The utility of land cover attributes to act as land cover ‘building blocks’ is demonstrated via classification of the field data to the Monitoring Landscape Change in the National Parks (MLCNP), National Land Use Database (NLUD) and Phase 1 Habitat Mapping (P1) schemes. Integration of the classified field data and a SPOT5 satellite image is demonstrated within per-pixel and object-orientated classification environments. Per-pixel classification produced overall accuracies of 81%, 80% and 76% at the field samples for the MLCNP, NLUD and P1 schemes, respectively. However, independent validation produced significantly lower accuracies. These decreases are demonstrated to be a function of sample fraction. Object-orientated classification, exemplified for the MLCNP schema at 3 segmentation scales, achieved accuracies approaching 75%. The aggregation of attributes to classes underutilises the potential of the remotely sensed data to describe landscape variability. Consequently, classification and geostatistical techniques capable of land cover attribute parameterisation, across the study area, are reviewed and exemplified for a sub-pixel classification. Land cover attributes provide a flexible source of field data which has been proven to support multiple land cover classification schemes and classification scales (sub-pixel, pixel and object). This multi-scaled/schemed approach enables the differential treatment of regions, within the remote sensing image, as a function of landscape characteristics and the users’ requirements providing a flexible mapping solution.

This thesis describes the processes used to manufacture optical fibre tapers and tapered long period gratings (TLPGs) using a CO2 laser. A semi-automated system for fabricating adiabatic and non-adiabatic tapers with repeatable physical dimensions has been developed. The tapers had waist diameters which were reproducible to within ± 0.5 μm. This system has also been used to fabricate TLPGs with periods ranging from 378 μm to 650 μm. Novel techniques to monitor the process of fabricating tapers were also explored. These techniques included; monitoring the transmission of the fibre using a spectrophotometer, using an in-line fibre Bragg grating (FBG) to measure the strain experienced by the optical fibre and the use of a near infra-red (NIR) camera to aid fibre alignment and laser power optimisation. The spectrophotometer allowed the optical properties of the tapers to be tailored for specific applications and the FBG provided strain data for process optimisation. The use of a NIR camera and an FBG as an in-line strain sensor are a novel use of these devices in a fibre tapering process. Tapers were also thin-film coated using sputtering techniques to form surface plasmon resonance sensors and their refractive index sensitivity was measured. A novel protein sensor based on gold nanoparticles deposited on a fibre taper is also reported, together with a lossy mode resonance taper sensor. The TLPGs which were fabricated, comprised of between 6 to 18 periods. The refractive index sensitivity of a 6 period TPLG was measured and was 372 nm/ RI. Their resonance bands had twice the bandwidth and exhibited a higher extinction, compared to UV-written long period gratings of a similar number of periods.

A cw titanium sapphire laser pumped by an argon ion laser was used for experiments both in second harmonic generation using a non-linear crystal and for sum-frequency generation in an atomic vapour. In addition the laser was stabilised to sub-MHz levels. Using a crystal of potassium niobate (KNbO₃), the Ti:sapphire laser was frequency doubled over the range 860nm-905nm using an intracavity scheme. The crystal was temperature tuned to achieve near non-critical phase-matching and powers of up to 50mW were obtained, with a constant conversion efficiency per Watt over the doubling range. The thermal properties of potassium niobate in frequency doubling were examined and the effect of the crystal on the ring cavity during temperature tuning was investigated. The laser cavity was frequency stabilised using a 'side of fringe' locking scheme with a confocal Fabry-Perot etalon as the reference discriminator. The cavity length control elements were two Brewster-angled tilt plates and a piezo mounted mirror. Using this scheme the frequency noise was reduced from tens of MHz to 550kHz. Methods of obtaining a 30GHz frequency scan were also evaluated and an experimental 10GHz scan achieved. It was concluded that only dither and lock schemes are good enough to achieve reliable 30GHz scans. Using both a Ti:sapphire laser and a dye laser, two contrasting schemes for sum-frequency generation in sodium vapour were investigated, in which a magnetic field was used to break the symmetry of the medium. Using one of these resonantly enhanced routes, powers of 17μW in the UV were obtained from a classically 'forbidden' quadrupole transition. The effects of phase-matching on each of the transitions was examined in detail, and it was found that in general there are at least six factors which affect the phase-matching behaviour of the frequency mixing scheme. It was also found that on the route which had two sources of dispersion there was a significant distortion of the output line profiles at high temperatures due to a variation in phase-matching across the line profile itself.

Our increased understanding of the molecular biology of disease has had a significant impact on healthcare. Genetics has allowed the identification of hereditary diseases and predisposition for others, such as cancer. However, for many diseases is it also necessary to monitor the expression of panels of proteins. The need to monitor protein expression presents a significant technological challenge requiring a highly multiplexed analytical technology with a sensitivity down to femtomolar-range. Low-cost photonic devices are highly sensitive to changes in their local environment and can be chemically modified to exhibit high specificity detection towards, for instance, proteins or DNA oligonucleotides. However, even if small-footprint photonic biosensors can be engineered in silicon microarrays, approaches to realise the very high-density, multiplexed sensing potential of photonic biosensors have yet to be demonstrated. This study aims to develop and demonstrate, for the first time, a dual-mode electro-photonic technology capable of highly multiplexed detection at the submicron scale and multiparameter profiling of biomolecules on the silicon photonics platform. Furthermore, the technology integrates electrochemical and photonic measurements in a single sensor platform. By combining the complementary information revealed by each of the domains it is possible to broaden the range of systems that are accessible for silicon photonics. Our dual-mode technology consists of microring resonators optimally n-doped (doping density of 7.5 x 10 16 cm −3 ) to support high-Q resonances (Q-factor ≈ 50, 000) alongside electrochemical processes in situ. This combination of sensing mechanisms enables the application of electrochemical methods for site-controlled immobilisation of receptor molecules. Furthermore, electrochemical characterisation of molecules bound to the sensor surface also provides direct quantification of binding density and unique insight into chemical reactivity, which is unavailable with photonic detection alone. This unique technology, based on the combination of electrochemical and photonic sensing on a silicon platform, not only enables detection of multiple biological molecules required for future clinical diagnostics, but also has the potential to impact on fundamental biochemical research.

This thesis describes methods for temporal contrast enhancement and phase characterisation of ultra-short laser pulses using nonlinear optical processes. Ultra-short laser pulses generated using the chirped-pulse amplification technique usually have unwanted emission that precedes and follows the main peak. The preceding part can cause significant changes in the conditions and the results of experiments implemented with these pulses. The first part of this thesis is devoted to the description and implementation of two techniques for enhancing the temporal contrast of ultrashort laser pulses. The first technique employs a second harmonic generation process linked to a low-gain optical parametric amplification stage. The second temporal contrast technique is one that uses cross-polarised wave generation. The second part of the thesis describes methods for phase characterisation of ultrashort laser pulses, based on the dispersion scan technique. In this method, the pulse is chirped with a set of chirp values and used to pump a nonlinear process. The phase of the pulse is extracted by linking the output of the process with simulated spectra via a phase function guessed by an iterative algorithm. This part will firstly describe and characterise a dispersion scan technique that uses second harmonic generation. Secondly, a new variant of the technique that employs self-phase modulations is introduced and characterised. The performance of both methods is optimised using a new method for phase representation, based on summing a set of Gaussian functions.

The hyperfine structure of the odd-even ⁵¹⁻⁶³Mn isotopes (N = 26 - 38) were measured using bunched beam collinear laser spectroscopy with the COLLAPS experimental setup at ISOLDE, CERN. The properties of these nuclei were investigated over the course of two experiments. During the first experiment, nuclear spins and magnetic dipole moments were extacted from spectroscopy on manganese atoms. These nuclear properties were then compared to the predictions of two large-scale shell model effective interactions (GXPF1A [1, 2] and LNPS [3]) which use different model spaces. In the case of 61,63Mn, these results show the increasing importance of neutron excitations across the proposed N = 40 subshell closure, and of proton excitations across the Z = 28 shell gap. These measurements provide the first direct proof that proton and neutron excitations across shell gaps are playing an important role in the ground state wave functions of the neutron-rich Mn isotopes. The electric quadrupole moment provides complementary information to the magnetic dipole moment, since it is able to directly probe the degree of collectivity and deformation in a nucleus, however sensitivity limitations during the first experiment prevented the quadrupole moments from being accurately determined. To overcome these limitations, it was necessary to find an electronic transition with sensitivity to the quadrupole interaction. A suitable transition was found in the ion from a metastable state and optical pumping in ISOLDE's cooler and buncher, ISCOOL, was used to populate this state for the second manganese experiment. This is the first use of the in-cooler optical pumping technique at ISOLDE, and required an overhaul of ISCOOL. During the second experiment, the spectroscopic quadrupole moments of the isotopes were extracted from the measured hyperfine spectra of manganese ions. The sensitivity and spectroscopic efficiency were increased as compared to the atomic experiment. The extracted quadrupole moments were again compared to predictions from the GXPF1A and LNPS effective interactions. The inclusion of the 1 vgg/2 and 2 vd5/2 orbitals in the model space were shown to be necessary to reproduce the experimentally observed onset of quadrupole deformation.

This thesis describes research work undertaken in the field of image mining (particularly medical image mining). More specifically, the research work is directed at 3-D image classification according to the nature of a particular Volume Of Interest (VOI) that appears across a given image set. In this thesis the term VOI Based Image Classification (VOIBIC) has been coined to describe this process. VOIBIC entails a number of challenges. The first is the identification and isolation of the VOIs. Two segmentation algorithms are thus proposed to extract a given VOI from an image set: (i) Volume Growing and (ii) Bounding Box. The second challenge that VOIBIC poses is, once the VOI have been identified, how best to represent the VOI so that classification can be effectively and efficiently conducted. Three approaches are considered. The first is founded on the idea of using statistical metrics, the Statistical Metrics based representation. This representation offers the advantage in that it is straightforward and, although not especially novel, provides a benchmark. The second proposed representation is founded on the concept of point series (curves) describing the perimeter of a VOI, the Point Series representation. Two variations of this representation are considered: (i) Spoke based and (ii) Disc based. The third proposed representation is founded on a Frequent Subgraph Mining (FSM) technique whereby the VOI is represented using an Oct-tree structure to which FSM can be applied. The identified frequent subtrees can then be used to define a feature vector representation compatible with many classifier model generation methods. The thesis also considers augmenting the VOI data with meta data, namely age and gender, and determining the effect this has on performance. The presented evaluation used two 3-D MRI brain scan data sets: (i) Epilepsy and (ii) Musicians. The VOI in this case were the lateral ventricles, a distinctive VOI in such MRI brain scan data. For evaluation purposes two scenarios are considered, distinguishing between: (i) epilepsy patients and healthy people and (ii) musicians and non-musicians. The results indicates that the Spoke based point series representation technique produced the best results with a recorded classification accuracy of up to 78.52% for the Epilepsy dataset and 84.91% for the Musician dataset.

Graphene, despite its many remarkable material properties, is fundamentally limited for many photonic and microelectronic applications due to its semi-metallic nature. Chemical functionalisation of graphene affords one route toward opening an energy gap, potentially extending its utility to these areas. Graphene oxide strongly absorbs in the ultraviolet range and reduction by various chemical treatments has been demonstrated to shift the absorption peak toward the visible spectrum. Photoluminescence emission has also been observed across the spectrum from ultraviolet to infrared, further suggesting the possibility of tuning optical properties. However, such methods produce highly defective graphene oxide, with hydroxyl, carboxyl, and carbonyl moieties being left behind in addition to the desirable epoxy functional groups. Considerable damage to the graphene sub-lattice is also caused. More recently, chemical deposition of atomic oxygen on graphene has been shown to form epoxy functional groups on graphene without causing this damage. Ab initio and hybrid density functional theory and time-dependent density theory studies of graphene oxide and reduced graphene oxide are carried out to investigate its structural, electronic, and optical properties. Patterned removal of oxygen to form graphene quantum dots embedded in the graphene oxide lattice is shown to permit tuning of the energy gap and optical absorption from ultraviolet through to infrared wavelengths, with long calculated radiative relaxation times. A simple relationship between the predicted gap and size of the most symmetric quantum dot structures, which are also the most thermodynamically stable, is demonstrated.

In this thesis, a Spatial Light Modulator (SLM) and a nano-structured S-waveplate were integrated with a picosecond laser system and employed to generate complex vector optical fields. Precise tailoring of optical vector beams is demonstrated, shaping their focal electric fields. Using novel optical setups, radial and azimuthal polarizations with and without vortex phase wavefront were created, producing complex laser micro patterning on a polished metal surface. Imprinting Laser Induced Periodic Surface Structures(LIPSS) elucidates the detailed vector fields around the focal region, which shows clearly how the Orbital Angular Momentum(OAM) associated with a helical wavefront induces rotation of vector fields along the optical axis of a focusing lens. In addition, unique, variable logarithmic spiral micro-structures were imprinted on the metal surface. These are the first experimental observations of such micro-structured spirals created by multi-pulse exposure with spiral vector fields which are shown to be due to superpositions of plane wave radial and azimuthal polarizations. A comparative analysis of micro-drilling with radially, azimuthally, circularly and linearly polarized beams with various wavefront was also carried out. The results are compared in terms of quality and efficiency, illustrating how the distinct machining properties associated with each beam make it best suited for specific processes or materials. A radially polarized beam was considered as the most efficient at drilling high-aspect-ratio features. By applying an axicon phase map on the SLM, annular shaped laser beams with adjustable dimension were created and employed in semiconductor thin film selective removal, which enhanced processing speed to a new level. Furthermore, annular beams with radial and azimuthal polarization were generated, imprinting complex concentric ring LIPSS pattern on a polished stainless steel surface. Side white light illumination tests demonstrated distinguishable bright strips on these rings due to diffractive properties of LIPSS, which offers a novel way in variable information encoding technologies. Finally, synchronization of vector field polarizations with micro-positioning at a material surface was demonstrated, leading to potential industrial applications of this work.