Runyao Gu

Soil water is a major interference in the on-site analysis of soil by energy dispersive X-ray fluorescence. Apparent consequences of this interference include lowered readings for elemental concentrations and significant changes in spectral characteristics in wet soils com-pared with dry soils. A rigorous interpretation on this issue remains unresolved. Thus, this study evaluated the impact of soil water on the detection of Ca, Ti, Mn, Fe, Cu, Zn, As, Rb, Sr, and Pb. Specimens were prepared from 11 certified reference soils and 3 field soils with water contents from 0 to ~40 wt.%. Results from three commercial models revealed that the readings were subjected to respective internal quantification algorithms; therefore, they could not provide a fundamental perspective of this issue. We analyzed the spectra to ex-amine the mechanism underlying this phenomenon. The spectra of wet soils feature elevated baseline, increased Compton and Rayleigh scatter peaks, and lowered characteristic peaks of elements. Previous studies attributed the lowered characteristic peaks to the absorption of fluorescent X-rays by water and considered soil water and dry soil as separate layers in the calculation. This work argued that wet soils should be treated as mixtures. Water becomes part of the soil matrix and leads to lower attenuation capability, which could be explained by the matrix effect. Meanwhile, the mass fraction of analytes is lowered because of dilution. Results confirmed that dilution lowers the characteristic peaks, whereas the matrix effect heightens them. When estimating the elemental concentrations on a wet weight basis, the matrix effect becomes the major interference. The Compton compensation method provided satisfying results on correcting the matrix effect caused by soil water on Zn, As, Rb, Sr, and Pb.