A mathematical model based on the limit dilution method to obtain linear calibration curves which eliminate the matrix effect in quantitative analysis by X-ray fluorescence

Abstract

We propose a mathematical model from an analytical application viewpoint inspired in the limit dilution method. The theoretical development of the model and its results are given. The model shows that there is a linear relation between the inverse of fluorescence intensity and the inverse of the dilution factor; each analytic system (sample, diluent and analyte) is characterised by a general linear function which is easily obtained. The analytical applications arising from this linearity are of great importance in X-ray fluorescence analysis. The following immediate applications are proposed: direct procurement of the total correction factor Y/ H, rapid calculation of the fluorescence intensity of the analyte in a sample ( I i s) and direct calculation of the corrected fluorescence intensity ( I i sF). The suggested model makes it possible to deduce a linear function between the fluorescence intensity of the analyte and the analyte concentration in successive dilutions of a standard; this straight line behaves as a calibration curve with direct application in X-ray fluorescence analysis. The proposed model may be applied to complex samples of geological origin, with elimination of the matrix effect. The results obtained in the determination of Ca, K, Fe and Ti in a standard soil show complete agreement with the certified reference values with a relative error about 0.5%, even using a standard shale with very different chemical composition as reference sample.