A study of the uses of scattered x-rays for internal standardization in x-ray spectroscopic analysis

Abstract

The use of scattered x-rays as an internal standard is evaluated in detail. The physical bases of secondary emission and scattering are described using the isolated atom model of photon-matter interactions. Unique quantitative expressions relating analyte concentration, intensity of analyte fluorescence, and intensity or scattering from the specimen are derived. Predictions based upon the isolated atom model are compared with experimental results for several types of aqueous matrices. Deviations of the experimental results from ideality are rationalized in terms of molecular scattering from the specimen. Evaluation of the method includes a complete review of its use in dealing with various analytical problems (i.e., matrix absorption effects, variations in the physical condition of the specimen, etc.). Many examples of its use in the analysis of a wide variety of specimen types are described, while establishing also that comparatively little has been written regarding its theoretical justification. '!he publications relative to this topic have tended to set very narrow conceptual restrictions upon the arguments. As a result, the scattered radiation method, which involves the comparison of analyte emission with specimen scattering, has come to be regarded as an analog of the purely empirical "peak to background" ratio method widely used in optical emission spectroscopy. Quantitative expressions derived from the isolated atom model are used to predict the degree to which matrix absorption effects can be overcome. The most significant theoretical result is that the degree of internal standardization should be essentially independent of the scattered wavelength used, as well as the character (i.e., Rayleigh/ Compton ratio) of the radiation. It is also demonstrated, however, that the foregoing generalization is valid only in the absence of sample matrix absorption discontinuities and anomalous dispersion effects. Experimentally, the internal standardization is most accurate for wavelengths composed mainly of Compton radiation, e. g. Compton bands and short bremsstrahlung continuum. The experimental results suggested that the isolated atom model, especially as it deals with x-ray scattering, does not accurately describe the spectroscopic behavior of aqueous specimens. It is hypothesized that deviations in the observed scattered intensity are caused by molecular scattering between adjacent water molecules. This hypothesis is tested and confirmed by using a collimated primary beam to observe the scattering from water in the soft x-ray region. The efficacy of Compton scatter for internal standardization is therefore explained on the basis that it is not susceptible to diffraction effects.