The basic requirements for the detection and the recording of faint optical images are discussed and the performance of various types of faint image detectors are considered. die superiority of the vacuum device employing photoemissive surfaces as the primary photon detector is indicated. Some photoelectronic devices are reviewed and the advantages of using cascade image intensifier are pointed out. The basic design of the standard cascade image intensifier tube is described and its theoretical performance is estimated. A short description of the technique employed in the construction is given and then the methods of preparation of the phosphor screens and the photo-cathodes are described. A novel design of cascade tube for use in spectroscopy is described. A detailed description of the techniques used in the manufacture of such tubes, together with their practical development is given. Their experimental performance is studied and important operating characteristics are measured. It is concluded that the new tube design is very satisfactory from different points of view and should be a useful device in spectroscopic work.

We use the XMM-Newton Serendipitous Source Catalogue to create a catalogue of non-nuclear extragalactic X-ray sources that is both large, containing 1,464 sources, and clean, with steps taken to minimise contamination. This catalogue contains 372 ultraluminous X-ray sources (ULXs), objects with extreme luminosities Lx > > 10³⁹ erg s⁻¹ higher than expected for a stellar-mass black hole (BH) accreting at its Eddington limit, and thought to mainly be super-Eddington accreting compact objects. We find the colour properties of ULXs to be similar to lower-luminosity sources, distinct from active galactic nuclei, and similar between populations in spiral and elliptical galaxies. Our catalogue contains eight new candidate intermediate-mass BHs (IMBHs) amongst the most luminous sources, 33 transient ULXs, and five ULXs with high short-term variability. Of these highly variable ULXs, we closely examine M51 ULX-7, which exhibits a persistently hard spectrum and evidence of band-limited noise, arguments in favour of it being an IMBH. However, a possible high-energy spectral turnover and its high dynamic range of flux bear similarities to properties observed in neutron star ULXs discovered to date. A useful subpopulation of our catalogue are sources at the Eddington Threshold (10³⁸ < L_x < 3 x 10³⁹ erg s⁻¹), the luminosity regime in which stellar-mass BHs may be accreting at or just below the Eddington limit. Our catalogue contains 666 such objects, four of which we studied in detail and found to demonstrate the heterogeneous nature of this luminosity regime, including high-luminosity sub-Eddington accretion states, the ultraluminous supersoft accretion regime, and sources in intermediate accretion modes. Finally, we searched for evidence of the most highly obscured and down-scattered super-Eddington accreting sources by looking for ultraluminous ultraviolet (ULUV) sources in UV source catalogues. We discovered six ULUV candidates in the Swift UVOT catalogue that are ideal for follow-up observations using HST.

There have been many successful THz instruments used in ground and space based telescopes that operate over a range of frequencies from 10s of GHz up to 10s of THz. For some instruments, their operation depends on precise ltering of the received radiation, which can be achieved by using metal mesh lters. These lters can also be referred to as frequency selective surface, which are part of a broad category of metamaterials. By controlling the geometry of these surfaces and combining multiple layers, it is possible to steer the radiation or control the polarisation. This thesis expands on using existing knowledge and fabrication techniques of metal mesh structures to build new quasi-optical devices, which could be a lightweight replacement for existing components, such as lenses, or new devices with properties that are not possible using traditional materials. Initially an analysis of suitable modelling methods to predict the behaviour of the individual gird structures is performed. It is then shown that multiple girds can be combined to create an arti cial dielectric. By using the developed modelling tools a parameter space is found which sets a practical limit on devices that can be built. This theory is also applied to rectangular grids, which have the same properties, but are polarisation dependent. With this framework in place, a series of devices are designed and characterised. This includes a broadband gradient index at lens, which was characterised using a VNA from 90 to 220GHz. This design was modi ed to produce a lens that is capable of separating out two orthogonal linear polarisations and focusing them separated on the same focal plane. Finally an analysis of di erent of arti cial waveplates is performed, to determine which metamaterial structures are best suited for replacing traditional sapphire waveplates.

No description available.

This thesis studies the broadband behaviour of GRBs by fitting a detailed spectral/temporal model to both the prompt and afterglow hard and soft X-ray emission observed by the Swift satellite. The prompt emission is decomposed into pulses which are fitted individually while the afterglow is modelled using a smoothly varying broad pulse which evolves into a power-law decay at late times. Using this model a comprehensive study of GRB 080310 is presented and followed by similar analyses of GRB 061121, GRB 080810 and GRB 081008. The optical behaviour is found to be inconsistent with the high-energy model: a spectral break between the X-ray and optical band is necessary and for many prompt pulses the self-absorption mechanism is required. The latter three bursts have optical afterglows that are shown to be inconsistent with those fitted to the X-ray regime, peaking earlier in the lower energy bands and requiring a low-energy spectral break. The prompt optical emission seen from GRB 061121 has pulse-like features which match reasonably well with contemporaneous high-energy features, but have longer durations. The same model was used to study the expected evolution of GRB properties when moved to higher redshifts. Using a sample of bright Swift GRBs, the changes in measured duration with redshift were found to be driven by a combination of time dilation, gradual loss of pulse tails and sudden loss of pulses as the flux falls below instrumental sensitivity. A realistic sample of synthetic bursts is produced which, when simulated at high redshift, are found to be significantly longer in duration that the observed high redshift GRBs. Also demonstrated is that several bright bursts seen by Swift could be detected if they occurred at a redshift > 10 encouraging the use of GRBs as probes of the early Universe.

No description available.

This thesis primarily comprises simulations of SKA1-Low, the low-frequency aperture array of the Square Kilometre Array (SKA) radio telescope. Simulations assess how different design specifications would affect the telescope's performance and ability to undertake one of its key science projects: observing the Epoch of Reionisation (EoR). The simulations are based on the previous baseline design of ~ 260 000 dipole antennas in 1024 stations of 256 antennas each. It is found that errors in the gain and phase of signals at each antenna can significantly affect observations, with the greatest impact at the lowest frequencies. The EoR signal will be detectable at limited redshifts with gain errors of 10% and phase errors of 6 degrees, however to ensure the systematic noise contribution from these errors is always weaker than the EoR signal between z ~ 10 - 21 and spherical harmonic multipole coefficients l ~ 102 - 104 , the gain and phase errors may need to be less than or equal to approximately 5 x 10-6 and less than or equal to approximately 3 x 10-4 degrees, respectively. Alternative telescope layouts to the SKA1-Low baseline design of 1024 stations containing 256 antennas each are evaluated. Some telescope models which share antennas between stations, allowing stations to be formed in a flexible manner, are found to provide superior thermal noise sensitivity than the current design, but suffer greater detrimental effects from gain and phase errors when imaging. The implementation of the digital signal processing of a station beamformer is also discussed. Calculations show that five or nine Field Programmable Gate Arrays (FPGAs) with specifications comparable to the Virtex UltraScale+ XCVU13P could be used to beamform a single 256-antenna SKA1-Low station, depending on the utilisation rate of the FPGAs and the number of channels required.

This thesis describes the pre-flight preparation and analysis of results from two scientific satellites engaged in the study of cosmic X-ray Astronomy. The places occupied by Ariel V and Copernicus in current X-ray Astronomy are reviewed and areas of current interest are noted. Supernovae and their remnants as x-ray sources are reviewed and the ability of x-ray measurements to probe the energetics of such systems is discussed. The Copernicus satellite provides a stabilised platform from which high resolution soft x-ray sources can be studied. The study of two strong soft supernova remnants are described in detail. Puppis A, providing the greatest surface brightness at 1 KeV, was mapped in two dimensions with the grazing incidence telescopes aboard Copernicus. The map, with the highest resolution to date, indicates the existence of a bright extended source with unresolved softer components. The data is consistent with a thermal source resulting from the interaction of a shock wave with a massive interstellar cloud. The Vela X observations indicate excess 0.5 - 1.5 KeV emission near, but not coincident with, the pulsar. The pulsar contribution must be less than previously reported. An unexpected 2.5 - 7.5 KeV flux was recorded from Vela X and this can be explained by an extended non-thermal region. The Ariel V Observing Programme is discussed and the Leicester Sky Survey Experiment is described in detail. Finally, Ariel V data was used to follow up the Copernicus observations of Vela X and show that the region contains several variable sources and yet still allows the extended harder x-ray source suggested.

This work characterises PIRATE’s primary camera, the SBIG STX-16803, as well as assessing the usefulness and impact of a small-aperture semi-autonomous facility in Mallorca for exoplanet studies and studies of transient sources. Additionally, a method for exploring the Roche lobe of an exoplanet and the effects this has on the shape and density of the planet is also described. PIRATE is a small aperture photometric facility that can be operated remotely or autonomously, is constructed from commercially available hardware and utilised by The Open University for research, education and outreach. The camera gain measurements are within the manufacturer specifications while the read noise deviates quite significantly. The camera shutter is also verified to evenly illuminate the CCD which may be suffering from a form of residual image. Regarding exoplanets, PIRATE has helped to identify 108 false positives as well as 24 real and plausible planets. Combined, this is 67% of PIRATE’s total exoplanet candidate observations. Several of these targets are explored in further detail. Still on the subject of exoplanets, Roche calculations are applied to 207 known exoplanets and highlight that WASP-12b, WASP-19b and WASP-103b are likely to have strong distortions from a spherical shape. Look-up charts were also generated for mass ratios of 10^-6 through to 10^-1. These look-up charts are intended to provide a quick reference volume correction to exoplanets (or any other system of similar mass ratios). As part of the Gaia transient preparatory and to demonstrate the capabilities of PIRATE, SN 2014J is provided as an example. A peak B apparent magnitude, Bmax of 11.80 +/- 0.14 mag is observed at a tmax of 56690.79 +/- 0.01 MJD and provides Dm15 = 0.98 +/- 0.20 mag. A peak E(B-V) extinction of 1.24 +/- 0.16 mag (E(B-V)host = 1.19 +/- 0.16 mag) is determined.

The manufacture of large off-axis aspheric optics for the next generation of extremely large telescopes presents a number of unique challenges. For example, the European Southern Observatory (ESO) Extremely Large Telescope (E-ELT) requires the manufacture of one 1.4 m class ultra-precision mirror segment per week in order to satisfy the first-light deadline. One of the factors limiting the pace of manufacture is metrology. Many of the tasks associated with measurement, such as optic positioning, alignment and acquisition are carried out manually. It is also common for the optic to be removed to a laboratory for measurement, which can be time consuming and risk damage. This thesis presents research into the development of new on-machine metrology techniques, which allow measurement to be carried out in the manufacturing environment. This work is supported by a software application developed by the author to allow the design and control of on-machine metrology. This application uses the computer numerical control (CNC) polishing system as part of the positioning and alignment system. The inclusion of CNC has enabled the development of a close-loop control system which facilitates automatic alignment and acquisition of metrology data. The software presented uses a modular architecture, allowing many different types of metrology to be planned and control using a single application. This is demonstrated using two case studies, which allow automatic on-machine sub-aperture stitching metrology using a metrology tower placed over the machine, and automatic on-machine texture measurement. The use of a closed loop software application to control automatic on-machine texture measurement is a novel step. It is also demonstrated that on-machine metrology in the manufacturing environment can produce measurement data of comparable quality to that of the laboratory. Automatic measurements systems such as those presented are likely to play an increasing role in the large and high-volume optical fabrication sectors.