On-line X-ray fluorescence analysis applied to industrial processes and environmental monitoring
MetadataVis full innførsel
Abstract X-ray stimulated fluorescence analysis is a powerful and widely used tool for determining the concentration of chemical species in materials. Presently, this is done by taking and preparing samples for off-line laboratory measurements. This project is a feasibility study of an on-line X-ray fluorescence system for continuous and direct analysis of industrial and environmental processes. The feasibility of this measurement principle depend on the accuracy with which the concentration can be measured within a given response time. Therefore this project has been focused on investigating the fundamental properties, such as fluorescence yield, X-ray source emission spectrum and stability of X-ray source intensity. Mathematical models to calculate X-ray fluorescence intensities and detection limits have been developed. Further, four different experimental arrangements were designed to measure absolute and relative fluorescence yield from samples with liquid and “gaseous” hosts containing a specie. In these experimental set-ups ionizing radiation from both X-ray tubes and gamma-ray sources (241Am) were used. Comparison of the experimental results to the results predicted by the mathematical model were carried out where possible. Gamma-ray stimulated fluorescence measurements were carried out in three different controlled experimental setups. Firstly, absolute fluorescence yield measurements were conducted on a liquid sample in a closed vessel. This proved not to be feasible due to intensity limitations of the 241Am source, which resulted in too low intensities of fluorescent radiation to be detected. Secondly, relative fluorescence yield measurements were conducted on the surface of a liquid sample contained in an open vessel with the radioactive source above the sample surface. This also proved not to be feasible again due to intensity limitations of the 241Am source, which resulted in very low intensities of fluorescent radiation. Thirdly, relative fluorescence yield measurements were conducted on the surface of a liquid sample in an open vessel with the radioactive source submerged into the sample. This resulted in detectable fluorescence peaks, but in a very limited concentration range. Characterization of the X-ray tubes and X-ray stimulated fluorescence measurements have been investigated in a fourth experimental arrangement. The characterization of the X-ray tubes consisted of measuring stability of radiation intensity, emission spectra and total or integrated intensity. Measurements of X-ray stimulated fluorescence were performed on different specie types and concentrations in liquid and “gaseous” hosts. The results from these measurements confirm that the highest potential lies in measurements on species in a gaseous host, due to its lover density. Effects of different penetration and escape depths for incident and fluorescent radiation were also investigated for a liquid host. The results from these measurements confirm that for an optimized system it is more important to reduce the escape depth for fluorescent radiation than the penetration depth for ionizing radiation. The mathematical model was applied to the results from X-ray stimulated fluorescence generated in a liquid host. Calculations obtained from the two models regarding fluorescence intensities and detection limits support the trends observed in the experimental results. Though, the theoretical results underestimate the fluorescence yield and overestimates the detection limits compared to the experimental results. These deviations can be explained by examining the experimental arrangements and the uncertainties in the measured characteristics of the X-ray tube.
UtgiverThe University of Bergen
Copyright the author. All rights reserved