Determination of fluorine by prompt γ-radiation from proton bombardment. Part II. Results

Publisher Website
Google Scholar
Abstract
The practical application of the nuclear reaction 19F(p,αγ)16O to the determination of fluorine is outlined. Analysis in liquid and gaseous phases suffers from the need to use a proton transmission foil. In principle, solids can be analysed inside the accelerator vacuum, but it is shown that powder targets can be handled more easily by using a foil. Lack of independent standards has prevented absolute calibrations, but it has been demonstrated that fluorine in gases can be determined at relatively high levels (5 to 100 per cent.) by a simple system that could easily be adapted, by the use of thicker foils and higher bombarding energies, to reach much lower levels. Determinations in liquids have been carried out in the range 20 p.p.m. to 10 per cent. of fluorine in aqueous solutions. There is no obvious reason to rule out acidic and alkaline solutions that do not attack the proton window or its holder. Pellet targets were tried for analysis in solids, and it is shown how such targets, made by mixing fluorine salts with iron powder, are usually inhomogeneous. Thus, without independent solid standards the method appears to present difficulties, especially in view of the “thin target” behaviour of the pelleted samples. Bombardment of powders, through a foil window, is shown to be a practical alternative, and it has been demonstrated in a selection of ground rock samples that fluorine is distributed homogeneously throughout a given sample. The importance of the stopping power, and hence a knowledge of the molecular form in which the fluorine occurs, has been emphasised for solids; the situation is far less critical for solutions, and for very weak solutions it has been demonstrated that chemical form is immaterial. Interferences are not important, except for aluminium and possibly lithium. The theoretical limits of detection are in good agreement with those observed in practice, being about 20 p.p.m. for the particular background and experimental arrangement used.

This publication has 0 references indexed in Scilit: