Quantitative Analysis of Overlapping X-Ray Fluorescence Spectra for Ni, Cu, Zn in Soil by Orthogonal Signal Correction and Partial Least Squares Algorithm

In this paper, the combination method viz. orthogonal signal correction (OSC) and partial least squares (PLS) algorithm is applied to analyze overlapping X-ray fluorescence spectra. The experiment is carried out by using 16 prepared soil samples which are divided into 4 groups according to their different combinations. Prediction results of these 4 groups are respectively calculated using OSC and PLS algorithm. Influence factors of the model are discussed according to specific samples. The experiment result shows that correction factor of OSC can be increased more than one and OSC-PLS algorithm can effectively eliminate the influence of overlapping XRF spectra and improve the predictive accuracy.