Applications of x-ray computed microtomography to material science: devices and prespectives

The three-dimensional visualization of the inner microstructural features of objects and materials is an aspect of relevant interest for a wide range of scientific and industrial applications. X-ray computed microtomography (?-CT) is a powerful non-destructive technique capable to satisfy these needs. Once the complete reconstruction of the sample is available, a quantitative characterisation of the microstructure is essential. Through digital image processing tools, image analysis software or custom developed algorithms, it is possible to obtain an exhaustive geometrical, morphological and topological description of the features inside the volume, or to extract other particular parameters of interest (e.g. porosity, voids distribution, cell size distribution, average struts length, connectivity between the cells, tortuosity). This thesis was carried out at the third-generation Elettra Synchrotron Radiation Facility (Trieste, Italy), where a hard X-ray imaging beamline is available. The experience developed at this beamline has leaded scientists to design a complementary state-of-the-art ?-CT facility based on a micro-focus X-ray source, working both in absorption and phase contrast mode. In this dissertation a detailed description of this facility is given together with a rigorous characterization of the imaging system capabilities, in terms of the actual achievable spatial resolution, in order to optimize the working parameters for the different experiments. The main artefacts that concur to the degradation of the quality of the reconstructed images have been considered (e.g. beam hardening effects, ring artefacts, uncertainness associated with the cone-beam geometry): procedures are presented in order to eliminate, or at least to reduce, the causes of these artefacts. The aspects related to the digital image processing of the reconstructed data are intensively developed in this study: appropriated methodologies have been elaborated capable to deal with the different three-dimensional data of complex porous media, providing a correlation between the microstructure and the macroscopic behaviour of the observed materials. Three representative examples obtained with the described methods are used to demonstrate the application of ?-CT, combined with the developed image processing tools, to material science: the geometrical and morphological characterisation of polyurethane foams employed in the automotive industry due their vibro-acoustic properties; a new approach to characterize the resonance spruce wood microstructure in order to study its acoustical behaviour; finally, the possibility of revealing defects in hybrid-friction stir welded aluminium joints, guiding the optimization of the process parameters.