The development of a proton induced x ray emission system for biomedical analyses

Abstract

This dissertation is concerned with development work carried out on a Proton Induced X-ray Emission (PIXE) analysis system and its application to trace elements analysis of biomedical samples. Samples were bombarded with proton beams of 2.5 MeV, and the characteristic X-rays which were emitted were measured using a high resolution Si(Li)detector coupled to a multi-channel analyzer. The system as developed in this work has a low detection limit of typically (10-7 - 107°)g for elements with Z numbers between 15 and 42 for irradiations of 20 minutes. Calibration factors for the system were determined by measuring the proton induced X-ray yields from the K shells of elements contained in thin standard foils of KCl, Ti, Mn, Co, Cu, Ga, Ge, Y, Rb and Mo. The long term stability and reproducibility of the system were checked by repeated measurements of the X-ray yield from these standard foils over the whole period of the experimental work. The experimental precision and accuracy for biological sample analysis were estimated from the results obtained from at least 60 thin targets made from different solutions of digested NBS standard bovine liver material. A precision of 2-8% was achieved. Comparison of these results with those of NBS recommended values and results obtained from Atomic Absorption Spectroscopy showed that for medium weight element an accuracy of better than 6% had been obtained. Techniques available for sample preparation are critically reviewed. The use of thin targets is generally recommended for best quantitative results. Several backing materials onto which samples can be mounted, are evaluated. This work also demonstrates the applicability of PIXE for multi-elemental analysis of a wide variety of biomedical samples. The trace element content of hair for a group of mentally defective females was determined and studied. Quantitative analysis of heart and brain tissue sections were also performed and the homogeneity level of trace element distributions were investigated. Results of these studies indicate the need for a large set of samples due to the natural non-homogeneity of these tissues. Bladder and kidney stones were also analyzed, to investigate the existance of various elements at different zones of the stones. Applications of the findings of this study are discussed. Further improvements are suggested for optimizing and extending the analytical capabilities of the present PIXE system.