Search

Partial least squares (PLS) regression-based methods have been proven to be a good alternative for quantification in X-ray fluorescence spectroscopy. These methods are fast and easy to use though giving satisfactory results under certain conditions. One of these conditions is the necessity of having a great number of spectra to build the model (training set). The choice of the constituent concentration range in the training set has a big influence on the accuracy of the model. Better accuracy is obtained if the model is built in relatively narrow regions containing (or close to) the real concentration value. In the present work, Monte Carlo (MC) simulated spectra are used to form the training set. The advantage to use MC generated training spectra is the unlimited availability of perfect standards. This paper aims to improve the accuracy of the method by introducing a multiple step procedure in order to build the PLS model using narrow concentration range close to (or containing) the real concentration values in the samples to be measured. This approach consists of an initial guess of the constituents' concentrations and a preliminary PLS model before building the final model. The prediction of ten MC simulated alloy standard samples containing Ti, Mn, Fe, Co, Cu, Zn, Sr, Zr, and Mo using this method allowed to have average relative prediction errors less than 5% for elements with narrow concentration ranges. Copyright © 2013 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

Abstract: X-ray scattering may contribute significantly to the spectral background of X-ray fluorescence (XRF) spectra. Based on metrological measurements carried out with a scanning electron microscope (SEM) having attached a well characterised X-ray source (polychromatic X-ray tube) and a calibrated energy dispersive X-ray spectrometer (EDS) the accuracy of a physical model for X-ray scattering is systematically evaluated for representative samples. The knowledge of the X-ray spectrometer efficiency, but also of the spectrometer response functions makes it possible to define a physical spectral background of XRF spectra. Background subtraction relying on purely mathematical procedures is state-of-the-art. The results produced by the analytical model are at least as reliable as those obtained by Monte-Carlo simulations, even without considering the very challenging contribution of multiple scattering. Special attention has been paid to Compton broadening. Relevant applications of the implementation of the analytical model presented in this paper are the prediction of the limits of detection for particular cases or the determination of the transmission of X-ray polycapillary lenses. [ABSTRACT FROM AUTHOR]

Abstract: A systematic series of measurements has been carried out with monochromatic X-ray excitation with synchrotron radiation in order to check a physical model on X-ray scattering. The model has recently been successfully tested for the case of polychromatic, unpolarised excitation emitted by an X-ray tube. Our main purpose is the modelling of a physical background in X-ray fluorescence spectra, so that improved quantitative results can be achieved especially for strongly scattering specimens. The model includes single Rayleigh and Compton scattering in the specimen, the effect of bound electrons, the challenging Compton broadening and the polarisation degree. Representative specimens, measurement geometries and excitation energies have been selected with synchrotron monochromatic light at BAMline/BESSY II. Monte-Carlo simulations have been also carried out in order to evaluate the quality of the results achieved with the model. [Copyright &y& Elsevier]

A Monte Carlo code for the simulation of X-ray imaging and spectroscopy experiments in heterogeneous samples is presented. The energy spectrum, polarization and profile of the incident beam can be defined so that X-ray tube systems as well as synchrotron sources can be simulated. The sample is modeled as a 3D regular grid. The chemical composition and density is given at each point of the grid. Photoelectric absorption, fluorescent emission, elastic and inelastic scattering are included in the simulation. The core of the simulation is a fast routine for the calculation of the path lengths of the photon trajectory intersections with the grid voxels. The voxel representation is particularly useful for samples that cannot be well described by a small set of polyhedra. This is the case of most naturally occurring samples. In such cases, voxel-based simulations are much less expensive in terms of computational cost than simulations on a polygonal representation. The efficient scheme used for calculating the path lengths in the voxels and the use of variance reduction techniques make the code suitable for the detailed simulation of complex experiments on generic samples in a relatively short time. Examples of applications to X-ray imaging and spectroscopy experiments are discussed. [Copyright &y& Elsevier]