Clays are sedimentary rocks with a very diverse elemental composition which is mainly dependent on the mineral content. In Latvia, the typical clay mainly consists of SiO2, Al2O3, Fe2O3, K2O, MgO, TiO2, CaO un Na2O (if the composition of the elements is expressed in oxide formulas). For determination of chemical composition more often are used X-ray fluorescence spectrometry (XRFS). The method is non-destructive and can analyze a large number of samples relatively quickly. However, previous studies [1] have shown that analysis of samples with such a complex elemental matrix by XRFS is difficult. It is estimated that the elements in the sample matrix can mutually affect the X-ray fluorescence analytical signal either strengthening or weakening it.
The aim of this study was to evaluate direct action and interaction coefficients using the standard additive method. Mathematical processing is performed assuming that the expected linear relationship between the amount of the additive and the analytical signal. The direct action coefficient shows the effect of the added element oxide on the analytical signal of the same element. In turn, interaction coefficients show the added element oxide effect on the other elements analytical signals. The additives of Al2O3, Fe2O3, K2CO3, MgO, TiO2, CaO and NaF were used to determine the initial composition of the element oxides of the clay. It should be noted that the SiO2 content in the clay sample is higher than 50%, so its determination by the standard additive method will not give accurate results. Therefore, the silicon dioxide is calculated by taking into account previously determined element oxides in combination with thermogravimetric analysis of the sample.
Based on the interaction results, it was found that the addition of Al2O3 reduces the analytical signal of K2O. Reduction of the analytical signal was also observed for Al2O3 and MgO using Fe2O3 and TiO2 as additives. Therefore, the resulting direct action and interaction coefficients were used to develop a multi-parameter optimization method that would allow recalculating element oxide content from XRF data without implementing the additive method in the future. It should be noted that such calculations will be possible for clay samples with a similar composition. Comparison of XRFS, standard additive, and multi-parameter optimization methods are summarized in Table 1. Standard additive and multi-parameter optimization methods give very similar results.
| Element oxide | XRFS data | Method | |
|---|---|---|---|
| Standard additive | Multi-parameter optimization | ||
| W, % | |||
| SiO2 | 57 ± 2 | 51.6 ± 1.9 | 51 ± 2 |
| Al2O3 | 19.3 ± 1.2 | 22.0 ± 1.5 | 22.6 ± 1.8 |
| Fe2O3 | 7.16 ± 0.19 | 7.23 ± 0.19 | 7.09 ± 0.17 |
| K2O | 4.5 ± 0.3 | 4.1 ± 0.4 | 4.1 ± 0.4 |
| CaO | 4.34 ± 0.15 | 4.48 ± 0.13 | 4.45 ± 0.16 |
| Na2O | 2.29 ± 0.17 | 3.72 ± 0.15 | 3.65 ± 0.15 |
| MgO | 1.6 ± 0.4 | 1.15 ± 0.03 | 1.11 ± 0.03 |
| TiO2 | 0.90 ± 0.07 | 0.77 ± 0.05 | 0.78 ± 0.04 |