Forthcoming, imaging X-ray telescopes cannot detect photons with energies above ~10KeV due to the limitations of current optics and silicon detectors. However, there are known to be X-ray emissions at higher energies from neutron stars, pulsars, black hole candidates, supernova remnants, active galactic nuclei and galactic clusters. In order that future telescopes may spectroscopically image these emissions, research is being conducted into new optics, and the use of new semiconductors for detector arrays. This thesis covers work to assess gallium arsenide (GaAs) as an X-ray detection medium. Detector fabrication is discussed and results are presented demonstrating the effect of carrier trapping on spectral resolution. By measuring charge collection efficiency, the activation energy of the dominant electron trap is calculated to be ~0.73eV. This trap may be 'frozen in' at temperatures below -85°C giving a resolution of 2.4keV FWHM at 59.5keV. Trapping is modelled, and a Monte-Carlo simulation is developed. By spectral fitting, the activation energy of the dominant hole trap is calculated to be ~0.2eV. These energy levels are associated with the native EL2 defect. Proton spectra demonstrate that the depletion region width varies linearly with bias at room temperature which is consistent with the existence of a quasi-neutral region. A series of large scale Liquid Phase Epitaxial (LPE) growth trials was conducted to assess this potential source of high purity material. Layers in excess of 100microm thick were grown, but the resultant devices could not be operated as detectors. It is concluded that the density of native EL2 defects in bulk GaAs is still too high to allow effective room temperature high resolution spectroscopy, and epitaxial material is difficult and expensive to produce. New sources of material will be sought and other semiconductors such as CdTe will be investigated, as will methods of operating simple high Z detector arrays.

No description available.

No description available.

Efficient identification and follow-up of astronomical transients is hindered by the need for humans to manually select promising candidates from data streams that contain many false positives and prioritise those candidates for follow-up. These difficulties are faced by most ground-based time domain surveys. This dependence on humans is unsustainable for next generation all-sky surveys and significant effort is now being invested to solve the problem computationally. In this work we investigate machine learning approaches to difference image artefact rejection and contextual classification of transient discoveries. We construct training sets from data gathered during the course of Pan-STARRS1 survey operations. In all cases the feature representation is carefully selected such that it is independent of the specific Pan-STARRS1 image processing pipeline and survey strategy, with the aim of designing solutions that can be easily applied to other current and future transient searches. We explore a number of machine learning algorithms and carefully evaluate their performance on these tasks. This thesis has developed working code that is now applied on the live PS1 data stream on a nightly basis. This acts as a useful testbed for application to future surveys such as LSST.

This thesis presents long-tenn X-ray and TeV y-ray observations from 2000 to 2006 of three TeV blazar type Active Galactic Nuclei (AGN): Mrk 421, Mrk 501, and H1426i428. Standard emission models for TeV blazars predict correlated and highly variable X-ray and TeV y-ray radiation from accelerated electrons in a jet orientated along our line of sight By using a large sample . of near simultaneous observations, this thesis examines potential flux and spectral correlations between the X-ray and TeV y-ray energy bands. Joint nightly observations ofMrk 421 with the X-ray instrument RXTEPCA, and at 0.5-10 TeV y-ray energies with the Whipple 10 m telescope during periods ranging from a week to 6 months revealed complicated, high amplitude flaring. Spectral variability was investigated for the rising and decay phases of large isolated day timescale flares. Generally, the X-ray and TeV y-ray spectra hardened with increasing integral flux, but a consistent spectral evolution was not shown between individual flares. Absorption of the TeV y-ray energy spectrum by the extragalactic background light (EBL) was corrected for. During a large flare in \. March 2001; tentative evidence was found for a highly curved intrinsic TeV y-ray energy spectrum peaking at 0.85 ± 0.22 TeV simultaneous to a curved X-ray spectruI)1 peaking at 2.14 ± 0.19 keY. RXTEPCA and Whipple io m campaigns on Mrk 501 and H1426i428 showed significant X-ray spectral variability, however the TeV y-ray integral flux remained near the detection limit of the Whipple 10 m. The broadband 0.6-20 keY X-ray spectrum of the 3 TeV blazars was investigated with XMM-Newton and RXTE peA observations. In addition to simultaneous day timescale variability, the large sample of Whipple 10 m observations from 2000 to 2006 allowed for a detailed study of long-tenn y-ray variability. For Mrk 421, a weak correlation, with large spread is shown for the Whipple 10 m integral flux above 0.6 TeV and RXTEASM X-ray rate on month timescales. From 2001 to 2006, Mrk 501 was recorded by the Whipple 10 m in a low flux state of 27% of the integral flux from the Crab nebula supernova remnant. In July 2005, the MAGIC telescope recorded large TeV y-ray flaring in Mrk 501 by a factor> 3 of the Crab nebula flux. This high flux state occured after the Whipple 10 m observing period on Mrk 501 in June 2005, and so could not be verified in this work. The detection of H1426-t428 at TeV y-ray energies by the Whipple 10 m in 2001 is confirmed in this work, however the source was not detected again by the Whipple 10 m over a 5 year period. The detailed study ofTeV blazar X-ray and y-ray variability in this work highlights the need for high sensitivity observations to better resolve complicated and unpredictable flaring states.

No description available.

No description available.

The work in this thesis is focused on the study of CdTe/CdZnTe pixellated detectors. During this research, three main aspects have been covered in the development and application of these detectors. These broadly describe the fabrication process, characterisation of energy spectrum and imaging performance, and application of these detectors in industrial X-ray tomography. The first topic to be discussed in detail is the fabrication process of CZT pixellated detectors. This is a process that involves many steps, such as cutting the sample, cleaning the substrate, writing a masking pattern with e-beam lithography, photolithography, cleaning the surface using oxygen ash, metallisation and removing the resin in the development stage. However, these stages were grouped into three major steps to give a clearer image on what was going on throughout the fabrication process. They are grouped as sample preparation, pixellated structure deposition and sample passivation. The dry ash processing had been carried out in the James Watt Nano Fabrication Centre at Glasgow University. The motivation for the development of the dry ash process was to reduce the level of leakage current on the surface of the device so that a good energy spectrum resolution could be obtained. The research showed that a passivation technique using a layer of SiN as well as an oxygen plasma oxidation layer provided low leakage current in the pixel and lower interpixel leakage current. A good performance in energy spectrum was achieved after performing passivation. The FWHM of Co-57 is 10.78 keV where energy resolution is 8.8%. The second area covered in this thesis is the energy spectrum and imaging performance of pixellated CdTe Timepix detector. Two pitches of pixel have been studied namely, 110 µm pitch and 55 µm pitch. The work on characterisations has been performed with polychromatic and monoenergetic beams of X-rays. In both the Counting mode and ToT (Time over Threshold) mode it was discovered that there were large effects on performance due to charge sharing and fluorescence. In terms of image quality the result of MTF was found to be as expected, around half the contrast transfer (30%) recorded for 110 µm compared to 55 µm pixel detector at 4 lp/mm. Another field of interest explored was the defect factor. For this to be achieved studies had been performed in positive and negative bias of CdTe Timepix detector. The results showed that the presence of defects in the material affects the performance of the energy resolution of CdTe Timepix detector. The third area covered in this thesis was the study of CdTe Timepix detector for use in Industrial X-ray tomography. To give the sensor the best performance it could offer in capturing images for industrial use, work had been carried to optimise the technique for producing good quality images. The hardware setup involved the careful alignment all of the components of the system and the introduction of a collimator. In addition the sofware software it dealt with the flat field correction, noise filter and data projection algorithm. The results show that the CdTe Timepix detector can be used as a good detector for X-ray tomography. Here, the 3D of volume rendering studies had been performed to a pencil as a sample and voltage regulator. The result of the pencil image in volume rendering showed that CdTe Timepix detector had successfully provided good image for different density profile of material (paint coating, wooden and carbon). Not only that, another important finding besides the excellent image produced was that the effect of beam hardening, ring artifact and metal shrank artifact in tomogram of voltage regulator had been eliminated successfully.

SuperSpec is an innovative, fully planar, compact spectrograph for mm/sub-mm astronomy. Its very small size, wide spectral bandwidth, and highly multiplexed detector readout will enable construction of powerful multi-object spectrometers for observations of galaxies at high redshift. SuperSpec is based on a superconducting filterbank consisting of a series of planar half-wavelength filters to divide up the incoming, broadband radiation. The power in each filter is coupled into a titanium nitride (TiN) lumped element Kinetic Inductance Detector (KID), facilitating the read out of a large number of filter elements with minimal cryogenic electronics. We present electromagnetic simulations of the various components that make up the first generation prototype device. We then present a characterisation of the performance of the TiN detectors and compare these to the standard Mattis-Bardeen prediction. We then demonstrate the operation of the filterbank at 250 GHz through a spectral characterisation made using a Martin-Puplett interferometer measuring a minimum filter bandwidth that corresponds to a spectral resolution of R = 700. From blackbody measurements of the most responsive filter channels, we estimate a system noise equivalent power of 2X10 to power -15 WHz to power -1/2.

No description available.