2.2.3. Sample Preparation With Normal-Phase SPE Clean-Up

Oil samples can be easily dissolved in hexane that allow testing the normal-phase (NP) SPE clean-up (Section 4.5) with silica cartridges (Strata-Si-1). The washing solvent was a mixture of *n*-hexane and ethyl acetate, while the elution solvent was a mixture of methanol and acetonitrile. An optimization was based on an experimental design using CCD. The factors and levels were the following: (I) ethyl acetate content of the washing solvent: 10%, 20%, and 30% (*v*/*v*); (II) acetonitrile content of the elution solvent: 0%, 25%, and 50% (*v*/*v*); and (III) sample pre-concentration: 3-fold, 4-fold, and 6-fold. The same naturally contaminated sunflower oil was used for the experimental design mentioned above (12.8 μg/kg TEA and 7.1 μg/kg TEN). The results showed that there is no significant difference in concentrations obtained for TEN under different conditions. However, the NP SPE considerably lowered the accuracy of TEA. The recovery of TEA was around only 10% in all

settings. The TEA could not be eluted from the silica cartridge with solvent containing only organic phase. Therefore, we added water into the elution solvent and tested the methanol/water mixture for sample elution with 10%, 20%, 30%, and 40% (*v*/*v*) water in methanol. Ten percent water in the elution solvent already resulted in ~ 70% recovery for TEA, which did not improve with a higher percentage of water. The drawback of having water in the elution solvent was that AME could not be eluted from the NP cartridge due to its lipophilic character. Hence, the NP SPE could not be used for all toxins involved.
