The Development and Application of Novel X-ray Tomographic Characterisation Methods

Abstract

The aims of this dissertation are twofold. Firstly, a significant portion of the dissertation is devoted to exploring and improving the understanding of metrology and its application to defect identification in conventional X-ray computed tomography (CT). The development of quality indicators and their use for measuring internal defect detectability in additive manufacturing (AM) powders and components scanned in X-ray CT are the subject of intense investigation. The other key aim of this dissertation is to develop and apply tomographic imaging modalities beyond conventional X-ray CT. To this end, the use of X-ray diffraction computed tomography (XRD-CT) and hyperspectral digital tomosynthesis (DT) for performing 3D imaging that would not be possible with conventional X-ray CT are explored extensively. The developmental work on defect metrology in X-ray CT presented in this dissertation is driven by the need to understand tomographic internal defect detectability in scans of AM powders and components. The conventional method of determining tomographic spatial resolution relies on evaluating the blurring around the edges of reconstructions of a solid cylindrical phantom in the scan central plane. This approach does not capture the positional variation in tomographic image quality that is predicted by theory to occur across a CT reconstruction. Here, it is instead proposed that tomographic spatial resolution be evaluated from the contrast with which periodically spaced features of varying sizes contained within a novel phantom design are reproduced in a reconstruction. Spatial resolution measurements are performed at various radial distances from the scan axis of rotation, as well as at various positions along the axis of rotation itself. Applying this approach to CT scans of a batch of AM titanium powder feedstock reveals a positional bias in the detectability of internal pores identified in the powder. Pore detectability is observed to improve with increasing lateral displacement from the sample centre and to worsen with increasing displacement from the reconstruction central plane. DT is a form of limited angle tomography that, due to the speed with which it is able to generate 3D reconstructions, has promising applications in areas such as medical imaging and non-destructive testing. Despite its speed, the reliance of DT upon variations in attenuation coefficient as a mechanism of contrast generation can leave the technique ill-suited to distinguishing between materials of similar atomic numbers or densities. To overcome these limitations, a hyperspectral detector, the HEXITEC, is integrated into a DT instrument. Hyperspectral DT is shown to be capable of generating high quality 3D reconstructions from data gathered in acquisition times far shorter than would be possible in hyperspectral CT. It is demonstrated that analysis of hyperspectral DT reconstructions can allow the detection of elemental K-edges, enabling tissue identification in biological specimens treated with staining agents. Differentiation of closely spaced elemental K-edges is also shown to be possible using the technique, with determination of the elemental composition of a specimen made possible as a result. (Nb0.85, Zr0.15)4AlC3 is a MAX phase alloy that is seen as a promising candidate for accident tolerant fuel cladding (ATFC) use. Samples of (Nb0.85, Zr0.15)4AlC3 are irradiated with protons, with the aim of simulating the radiation damage likely to be encountered in a real reactor environment. Subsequent XRD-CT characterisation is used to evaluate the effects of the proton irradiation upon the MAX phase microstructure. Assigning relative strain measurements to every voxel in the XRD-CT reconstructions show the peak radiation dose is accompanied by a 2.43% increase in the c-parameter and a 0.55% decrease in the a-parameter of the MAX phase unit cell. A reduction in grain size is observed in the radiation damaged region of the specimen, a finding that is s

Date of Award	31 Dec 2023
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Philip Withers (Supervisor) & Timothy Burnett (Supervisor)