Heterogeneity effects in micro-beam XRF scanning spectroscopy of binary powdered mixtures and lake sediments

Abstract

Micro-beam X-ray fluorescence (XRF) scanning spectroscopy is very useful for analyzing heterogeneous samples from earth and environmental sciences with high spatial resolution. However, XRF intensity depends on grain size and mineralogical composition, known as the heterogeneous effect, although little quantitative evaluation has been conducted. In this study, the binary powdered mixtures of Fe2O3 in the CaCO3 or SiO2 matrix with nine grain size fractions (0.3–605 μm), four Fe2O3 concentrations (5–20 wt%), and fine-grained sedimentary cores from Lake Baikal were used for analysis, using a scanning X-ray analytical microscope (SXAM), and then evaluated by making a simple modification of the Berry–Furuta–Rhodes model. The results of binary powdered samples show that the Fe intensity decreases as the grain size increases, but if the grain size is constant, its intensity has a linear relationship with the composition of Fe2O3. The experimental data are in good agreement with theoretical curves. The theory predicts that if a phase that contains fluorescent elements has narrow ranges of concentration and grain size, the micro-beam XRF spectroscopy enables highly precise calibration from the XRF intensity to element concentration. The theoretical curves of the Lake Baikal sediment core suggest that the Fe intensity has about a 30% maximum difference in the median grain size range of 3.9–28.2 μm. This variation appears in a scatter of regression between Fe intensity and concentration, but it scarcely affected the XRF intensity variability of sediment composition.