Multivariate curve resolution alternating least squares in the quantitative determination of sulfur using overlapped S(Kα)–Mo(Lα) emission peaks by wavelength dispersive X‐ray fluorescence spectrometry

Spectral line overlap is a serious problem in quantitative X‐ray fluorescence analysis. In this study multivariate curve resolution alternating least squares (MCR‐ALS) approach was used to resolve the effect of overlapping S(K_α)–Mo(L_α) emission lines generated by standard‐less software of a wavelength dispersive X‐ray fluorescence spectrometer (WDXRF) for the quantitative monitoring of sulfur in mineral samples. Scan channel set contained Ge crystal, 550‐µm collimator, flow detector (Ar + CH₃) and rhodium (Rh) tube. The 18 calibration and 10 validation samples contain 0.00%–10.98% sulfate (SO₃) and 0.00%–92.40% MoO₃. The digitized spectral data were extracted in the range between 109° and 113.9° (2θ) at every 0.1 degree. Lack of fit percentage (LOF%) for experimental data and the variance explained at the optimum condition () were 2.32 and 99.94, respectively. The values of the root mean square error of prediction (RMSEP) for analyzing of sulfur were 0.23. MCR‐ALS was also compared with partial least squares (PLS) method for determination of sulfur in the presence of molybdenum. To evaluate the resolution and quantification performance of MCR‐ALS procedure, the method was used to determine sulfur in presence of molybdenum in two synthetic soil samples. Copyright © 2015 John Wiley & Sons, Ltd.