A method of XRF spectrochemical analysis based on some geometrical properties of the X-ray fluorescent intensity

In previous (J.E. Fernandez and M. Rubio, subm. to X-Ray Spectrom (1989); J.E. Fernandez, subm. to Comput. Phys. Commun. (1989)) it was found out that the primary-XRF (X-ray fluorescence) intensity remains invariant under variation of the tilt angle α of the propagation plane whilst the secondary one vanishes at the limit /α/-> Π/2. As a consequence the detected fluorescence is only composed of primary photons, simplifying the increasing complexity, for multicomponent samples, of the matematical dependence of the XRF intensity on their composition. This feature is exploited to elaborate analytical methods based on the resulting simplified XRF-intensity expression. The sample composition is calculated as solution of a linear set of equations when the excitation spectrum is monochromatic, and as iterative solution of a nonlinear set of equations for the more realistic polychromatic excitation spectrum. For polychromatic excitation an additional method is devised, which uses some nonlinear least-squares coefficients and tabulated data to build the coefficients matrix of a linear system of equations whose solution is the required concentration. Measured intensities under this scheme were used to determine the major-elements composition of some NBS standard steels. comparison with NBS reported values evidences that this method is reliable and precise