Resonant Inelastic X-ray Scattering (RIXS) is a modern soft X-ray spectroscopy technique used to investigate the structure of and excitations in materials. It requires high resolution spectrometers and a brilliant, tunable, X-ray source and therefore is carried out at spectrometers such as SAXES at the Swiss Light Source Light, a synchrotron at the Paul Scherrer Institut. SAXES uses a grating to disperse X-rays scattered from a sample across a position sensitive detector, a Charge-Coupled Device (CCD). It has been recognised that the spatial resolution of the CCD currently limits the spectrometer resolution and therefore the investigations described in this thesis focus on developing and testing methods of improving the detector resolution. Whilst this thesis addresses improving the resolution of the detector at SAXES specifically, the methods and results are applicable to other applications requiring high spatial resolution soft X-ray detection. After an introduction, Chapters 2 and 3 describe the importance of RIXS, operation of SAXES and background of soft X-ray detection in CCDs. A comparison of models that describe charge spreading in a CCD is in Chapter 4, and the best model is implemented in a simulation package that generates populations of soft X-ray events. Chapter 5 explores the resolution improvements possible through applying centroid algorithms to simulated X-ray events, and Chapter 6 begins by describing experimental work undertaken to verify simulation results. Due to the limitations of applying centroiding algorithms to the current SAXES camera, a small-area Electron Multiplying CCD (EM-CCD) is experimentally tested (Chapter 7). Results with the EM-CCD proved positive, therefore in Chapter 8 the spatial resolution achievable with a large area EM-CCD is verified for a future SAXES camera upgrade. Due to the successful results presented in this thesis, negotiations to develop a new camera system for SAXES are underway, and interest from other RIXS beamlines in the community may lead to the work also being applied elsewhere. The detection of soft X-rays with high spatial resolution is applicable to other future CCD and EM-CCD instruments, such as astronomical X-ray observatories.

This thesis details studies to develop methods for the generation of ultra-short pulse of light to enable the study of ultra-fast phenomena. The main contents are: an overview of methods of light sources and applications; a review of the physics of the free-electron laser, and its resultant properties; a review of existing and proposed methods of generating ultra-short pulses from free-electron lasers; a study into the physics of the mode-locked amplifier FEL technique, including its use to amplify an HHG source while retaining the pulse train structure, an explanation of the role of electron beam modulation in the technique, specification of alternative modulation methods, and a simplified model to describe the behaviour; and lastly a proposal for a method to generate x-ray laser pulses with duration approaching the zeptosecond range, two orders of magnitude shorter than the current record, including detailed modelling and analytic description.

No description available.

No description available.

No description available.

Pulsed laser ablation has been investigated as a method for the creation of thick segmented scintillator arrays for high-energy X-ray radiography. Thick scintillators are needed to improve the X-ray absorption at high energies, while segmentation is required for spatial resolution. Monte-Carlo simulations predicted that reflections at the inter-segment walls were the greatest source of loss of scintillation photons. As a result of this, fine pitched arrays would be inefficient as the number of reflections would be significantly higher than in large pitch arrays. Nanosecond and femtosecond pulsed laser ablation was investigated as a method to segment cadmium tungstate (CdWO_4). The effect of laser parameters on the ablation mechanisms, laser induced material changes and debris produced were investigated using optical and electron microscopy, energy dispersive X-ray spectroscopy and X-ray photoelectron spectroscopy for both types of lasers. It was determined that nanosecond ablation was unsuitable due to the large amount of cracking and a heat affected zone created during the ablation process. Femtosecond pulsed laser ablation was found to induce less damage. The optimised laser parameters for a 1028 nm laser was found to be a pulse energy of 54 μJ corresponding to a fluence of 5.3 J cm^-2 a pulse duration of 190 fs, a repetition rate of 78.3 kHz and a laser scan speed of 707 mm s^-1 achieving a normalised pulse overlap of 0.8. A serpentine scan pattern was found to minimise damage caused by anisotropic thermal expansion. Femtosecond pulsed ablation was also found to create a layer of tungsten and cadmium sub-oxides on the surface of the crystals. The CdWO_4 could be cleaned by immersing the CdWO_4 in ammonium hydroxide at 45°C for 15 minutes. However, XPS indicated that the ammonium hydroxide formed a thin layer of CdCO_3 and Cd(OH)_2 on the surface. Prototype arrays were shown to be able to resolve features as small as 0.5 mm using keV energy X-rays. The most efficient prototype showed low detective quantum efficiency of 0.08±0.01 at 0 lp/mm using a tube voltage of 160 kVp.

No description available.

No description available.

No description available.

No description available.