A new Monte Carlo code for simulation of the effect of irregular surfaces on X-ray spectra

Generally, quantitative X-ray fluorescence (XRF) analysis estimates the content of chemical elements in a sample based on the areas of the fluorescence peaks in the energy spectrum. Besides the concentration of the elements, the peak areas depend also on the geometrical conditions. In fact, the estimate of the peak areas is simple if the sample surface is smooth and if the spectrum shows a good statistic (large-area peaks). For this reason often the sample is prepared as a pellet. However, this approach is not always feasible, for instance when cultural heritage or valuable samples must be analyzed. In this case, the sample surface cannot be smoothed. In order to address this problem, several works have been reported in the literature, based on experimental measurements on a few sets of specific samples or on Monte Carlo simulations. The results obtained with the first approach are limited by the specific class of samples analyzed, while the second approach cannot be applied to arbitrarily irregular surfaces. The present work describes a more general analysis tool based on a new fast Monte Carlo algorithm, which is virtually able to simulate any kind of surface. At the best of our knowledge, it is the first Monte Carlo code with this option. A study of the influence of surface irregularities on the measured spectrum is performed and some results reported. © 2014 Elsevier B.V. All rights reserved.