Feasibility for chemometric energy dispersive X‐ray fluorescence and scattering (EDXRFS) spectroscopy method for rapid soil quality assessment

Abstract

Soil quality assessment (SQA) is important for modulating agricultural productivity and thus requires simple and rapid analysis of soil (macro & micro) nutrients (here called soil quality indicators – SQIs). We report proof of concept of a chemometrics‐assisted energy dispersive X‐ray fluorescence and scattering (EDXRFS) spectroscopy technique for direct rapid analysis of SQIs. The EDXRFS method exploits, in addition to fluorescence, the X‐ray scatter peaks obtained non‐invasively from soils to develop a calibration strategy for quantitative analysis of SQIs in model clay soils doped with micronutrients (Fe, Cu, and Zn) and macronutrients (NO3−, SO42−, and H2PO4−). The soil samples and certified reference materials IAEA Soil‐7 and IAEA Soil‐1 (used to build spectral library for soil classification) were irradiated at various live times (to simulate different signal‐to‐noise ratios of analyte signals and analysis speed) in a teflon holder and were analyzed using a 109Cd‐excited XRF spectrometer. Principal components analysis was used for spectral data compression and pattern recognition (including for those SQI spectral signatures without any visibly discernible characteristic X‐ray lines), whereas partial least squares regression and artificial neural networks were used to build a calibration and quantitative analysis strategy. The method furnishes soil micronutrient and macronutrient information simultaneously and rapidly (t = 200 s). To the best of the authors' knowledge, this is the first time that an XRF method has demonstrated spectroanalytical potential for quantitative macronutrients analysis in soils applicable to routine SQA. Coupling EDXRFS with multivariate chemometrics enables rapid and reliable assessment of chemical SQIs. Copyright © 2011 John Wiley & Sons, Ltd.

Multivariate chemometrics as a key tool for prediction of K and Fe in a diverse German agricultural soil-set using EDXRF

“…Similar correlation between PC-1 and Fe with a high variance was proposed by Kaniu et al ., who used energy-dispersive X-ray fluorescence and scattering spectroscopy for assessment of soil quality. Moreover, soils were clustered according to their soil type and not to their texture 25,35 . The second PC is dominated by calcium with a peak in the positive area of the loading plot and a residual variance of 4%.…”

“…The model was created by the Y-variables (mass fraction of the analyte) as a function of the X-variables (intensity of the fluorescence) 25,41 . In this work the used dataset for calibration with 15 CRM was a 75 × 1234 matrix and for prediction of German agricultural soils a 2990 × 1234 matrix (598 samples, 5 repetitive measurements).…”

“…For calibration with chemometric tools, Kaniu et al . has tested the performance of PLSR and artificial neural network (ANN) in combination with energy-dispersive X-ray fluorescence and scattering spectroscopy for soil quality assessment in Kenja 11,25,26 . It was also possible to predict the sand or clay content with chemometric tools such as PLSR 27,28 .…”

Method for Sediment Texture Characterization Using Spectroscopy Techniques and Multivariate Analysis

“…Therefore, the magnitude of complex problems that are solved by spectroscopy could be increased by the use of multivariate analysis methods. 20,21 Multivariate statistical methods are useful for spectra separation based on non characteristics information (implicit information). In this case, the application of unsupervised methods, like principal component analysis (PCA) and Hierarchical Cluster Analysis (HCA) are appropriate.…”

Using chemometric methods for overlap correction of sodium-zinc spectral lines generated by wavelength dispersive X-ray fluorescence in mineral samples

“…The goal of the support vector regression method is to approximate a function that accepts a data point and returns a continuous value . Some papers report the use of chemometric methods such as PLS to develop the evaluation of XRF spectrometry results in order to correct the inter‐element effects …”