Synthesis and application of molecularly impressed polymers for the extraction of caffeine from food and beverage samples

For the last decades Molecularly Impressed Polymers (MIP) have been a tendency for analyte separation from complex matrices, allowing to achieve adequate selectivity, as a function of their molecular recognition characteristics, and sensibility, due to the possibilities of analyte pre concentration they offer. This work intends to apply MIP to selectively extract caffeine, in order to make such a pre treatment able to eliminate interferences present in food and beverage samples. The molecularly impressed and control (non impressed, NIP) polymers synthesized in this work were prepared based on the method proposed by Theodoridis and Manesiotis1 and Farrington et al2. The synthesized polymers were washed in Soxhlet apparatus for 6 hours with a methanol/water 9:1 v v-1 mixture, dried, inserted in cartridges and evaluated as solid phases for solid phase extraction (MIP-SPE). The MIP-SPE method was conducted by conditioning the cartridges with 4,00 mL acetonitrile, adding 1,00 mL of 25,0 mg L-1 aqueous caffeine solution, washing with sodium phosphate buffer 50 mmol L-1 at pH 10,6 and eluting the analyte with 1,00 mL acetonitrile. The collected fractions were spectrophotometrically and chromatographically analyzed. Under these conditions, the analytical parameters of merit shown in Table 1 were obtained. By the study of possible interferents, it was found that other xanthines, such as theophylline and theobromine, did affect the quantification of the analyte. For that reason, samples that did not contain other xanthines, such as soft drinks or coffee powder, could be simply spectrophotometrically analyzed. Samples containing other xanthines, as energy drinks, chocolate powder and black tea leaves needed also chromatographic separation previously to quantification. The results of some of these quantifications and their respective recovery tests are shown in Table 2, revealing the accuracy and applicability of the method. It is valid to emphasize that even for the samples that were quantified by HPLC the use of MIP became indispensable for the elimination of interferences.