Pixe analysis of adjacent elements

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title

University of Cape Town

Proton induced X-ray emission (PIXE) analysis depends on the accurate stripping of the peaks of individual X-rays from the X-ray energy spectrum produced by the bombardment of a target material with charged particles. The energy separation between the Kᵦ X-ray of element Z and the Kα X-ray of element (Z + 1) increases with increasing Z. Accordingly, for adjacent lower Z elements, there is an overlap in energies between these two X-rays, and hence interference may be caused by one element in the determination of the other. The purpose of this investigation was to determine the extent of such possible interferences and to evaluate the accuracy and precision of the determination of adjacent elements, especially when one of the pair is present in overwhelming concentrations. Two elemental pairs, potassium-calcium, of biological significance, and vanadium-chromium, of metallurgical significance, were studied in detail. Mixtures of stock solutions of the two elements of each pair were made to provide samples with elemental ratios of the minor component decreasing from 10-¹ to 10-⁴ relative to the major one. A minimum of five of these samples were prepared on thin foils as well as on thick target pellets for each concentration level, and the PIXE spectra were recorded under bombardment with protons of 3 MeV. The spectra were analysed off-line by the program AXIL, which, in cases where the minor component could not be visually identified in the spectrum, was forced to evaluate the concentrations of both components. Under these conditions, low levels of the minor component were reported even though the component could not be resolved. All samples, the spectra of which showed the presence of both elements of the elemental pairs, could be analysed in mg/g concentration range. The precision and accuracy of such analysis were acceptable, except in the case of pellets onto which solutions containing both potassium and calcium were deposited. Chromatographic separation of potassium from calcium during the diffusion of the solution in the pellet matrix gave unacceptably erroneous results. When the concentration of the major component exceeded that of the minor by a factor of 200 or more, the presence of the minor component could no longer be recognised in the spectra. When such spectra were analysed by AXIL, forcing the program to determine both elements, results were a reflection of background levels in the energy region of the expected peak. It could therefore be concluded that the determination of the minor component in the presence of the major one for adjacent elements by PIXE is inaccurate for relative concentrations less than 1:200.

Bibliography: pages 58-60.