Evaluation of minerals, toxic elements and bioactive compounds in rose petals (Rosa spp.) using chemometric tools and artificial neural networks

Rose (Rosa spp.) petals have been used for the preparation of new cooking recipes, due to their sensory properties. In this paper were evaluated the micro (Co, Cu, Fe, Mn, Mo, Ni and Zn), macro (Ca, Mg, Na and P) and toxic elements (As, Cd and Cr) of organic cultivation rose by ICP OES and ICP-MS. The bioactive compounds (total phenolics, total flavonoids, total anthocyanins and antioxidant capacity in DPPH) were determined by UV-VIS spectrophotometry. Chemometric tools were used in method optimization, such as full factorial design for acid digestion and Box-Behnken design for total phenolics, in addition to the application of Artificial Neural Networks with Kohonen Neural Network for bioactive compounds. The application of PCA and HCA showed that the mineral composition of rose petal samples had differences between summer and winter. The accuracy of the method was confirmed by the analysis of three certified reference materials: apple leaves (1515), spinach leaves (1570a) and peach leaves (1547), from the National Institute of Standards and Technology (NIST). Digestion efficiency was evaluated by residual carbon content. The developed methods were promising to evaluate the minerals, toxic elements and bioactive compounds in rose petals, and the results indicate that they can be added as a potential nutritional supplement in the human diet.