Abstract
Iron oxide nanoparticles (IONPs) are an ideal sorbent for magnetic dispersion extraction due to their superparamagnetic properties and developed and active surface. This work aims to use IONPs, obtained by chemical co-precipitation, to purify 100% acetone and 50% acetone extracts from hop cones (Humulus lupulus L.) obtained by ultrasonic-assisted solvent extraction. The extracts were purified from bitter acids (i.e., humulones, lupulones) to isolate xanthohumol. The sorption conditions were optimized depending on the composition of the extraction mixture, specifically the mass of IONPs and the time needed to achieve effective sorption using response surface methodology (RSM). An analysis of variance (ANOVA) was performed to assess the adequacy of the developed model, and a good agreement was found between the experimental data and the proposed model. The polynomial equation describing the model is highly significant (p < 0.05), with a precision of Adeq (above 4). This indicates the usefulness of the polynomial regression model for prediction in experimental design. The final products of the purification for 100% acetone extracts and 50% acetone contain 40.58 ± 2.84 µg mL-1 and 57.64 ± 0.83 µg mL-1 of xanthohumol, respectively. The use of 50% acetone extract provides more favorable conditions due to the smaller amount of nanoparticles required for extract purification and a higher recovery of xanthohumol. The development of a reliable multivariate model allowed for the optimization of the extract purification process, resulting in high-purity xanthohumol from natural sources.