Abstract: Trace amounts of heavy metals were preconcentrated on discs of polyurethane (PU) foam loaded with ammonium diethyldithiocarbamate (DDTC), to increase the sensitivity of XRF. Since the diameter of the cavities in the open-cell PU foams reaches 30300 μm, considerable heterogeneity effects might occur, which could reduce the accuracy of the method. The possible systematic error introduced by considering the sample to be homogeneous is calculated using the Berry et al. model. The calculations show that the underestimation of the absorption correction factor when considering the sample as being homogeneous is less than 2% for high-Z elements when the sample thickness is at least 2 mm. The detection limits are also around the minima for this thickness.

Abstract: Expressions are presented for calculating the matrix effect and the pure particle size effect in the XRF analysis of particulate samples with a discrete particle size. The equations are based on the absorption-weighted radiometric diameter concept. Two excitationdetection geometries are considered, with the angles between the sample plane and both the incident and emerging radiation being either 90° (π geometry) or 45° (π/2 geometry). Calculations were made for different sample loadings and exciting radiation energies. The influence of these parameters on the matrix and pure particle size effects is shown. From the results, it is possible to predict the performances of alternative experimental correction procedures for the particle size effect, involving dual measurements at different excitation energies or in different excitationdetection geometries.

Abstract: An automated electron microprobe, equipped with an energy-dispersive X-ray spectrometer and an additional backscattered electron signal digitalization system, can allow rapid sizing and major element analysis on numerous particles. A software package has been developed to exploit the particle size and shape information to achieve quantitative analysis of single particles, and to compare the performance of the different matrix correction procedures.

Abstract: Most computer XRF-spectrum deconvolution routines make use of fixed intensity ratios for the lines from one element. The magnitude of the error that fixed ratios imply has been quantitatively evaluated for samples with a varible thickness or matrix. A procedure for routinely adapting the line ratios according to the matrix effect in every sample (by making use of the matrix information present in the scatter peaks) enhances the accuracy of the spectrum evaluation.

Abstract: A method has been investigated which allows calculations from the X-ray fluorescence spectra of the absorption coefficients at any energy for any sample, without any additional measurement. Use is made of the ratio of the characteristic X-ray signals from a Zr wire positioned in front of the sample and from a Pd foil placed behind the sample, both in a fixed geometry. From the experimentally measured absorption coefficient at the Pd L energy (2.9 keV), the coefficients for higher energies are calculated. By the use of an iterative computer routine in which corrections for the enhancement of the Pd foil by the sample are also included, an accuracy of 2% or better on the absorption coefficient determination can be reached for homogenous samples in one measurement. Grain-size and heterogeneity effects induce inaccuracies on the absorption coefficient determinations which might well reach 20% for particulate samples like intermediate thickness deposits of geological materials. This approach thus has the same limitations as the classical transmission method for such heterogeneous samples.