Relationship between elemental content determined

Abstract

Context Studies comparing the elemental contents obtained via portable X-ray fluorescence (pXRF) spectrometry under different scanning conditions and traditional time-consuming concentrated acid-digestion-based methods are rare in tropical soils. Aims (1) To compare the contents of Al, Ca, Cu, Fe, K, Mn, Zn, Cr, Ni, Ti, V and Zr obtained by pXRF with the results of the USEPA 3051a method; and (2) to evaluate the impacts of different conditions on the pXRF results. Methods Soil samples were analysed with pXRF under five conditions: field (F), post-field (PF), air-dried fine earth (ADFE, <2 mm), macerated (M), and macerated and sieved (MS, <150 μm). Linear regressions were adjusted between the USEPA 3051a and pXRF results for the five conditions. Coefficient of determination (R2), root-mean-square error (RMSE), and residual prediction deviation (RPD) were used as validation parameters for the models. Key results The different scanning conditions provided contrasting results for the total elemental contents via pXRF. The validation generated good models, as indicated by the condition, R2, and RPD values, and the results were, respectively: Ca (M; 0.88; 3.00), Cu (F; 0.91; 3.29), Fe (ADFE; 0.94; 4.14), Mn (F; 0.85; 2.65), Cr (ADFE; 0.86; 2.77), and Ni (ADFE and M; 0.74 for both; 2.10 and 2.08). Conclusions PXRF can accurately determine the contents of Ca, Cu, Fe, Mn, Cr and Ni in tropical soils compared with the 3051a method. Implications Results of the USEPA 3051a method can be accurately predicted via pXRF and regression models, reducing time, cost and work required.