A Selectivity Study on the Use of Caffeine and Theobromine Imprinted Polypyrrole Surface Electrodes

Molecularly imprinted polymers (MIPs) are proving to be very effective in development of synthetic recognition systems and are of great interest to those interested in the field of sensor technology. The use of MIPs is receiving considerable interest due to the ability to prepare recognition matrices that possess high substrate selectivity and specificity. Conducting polymers (CP) have proved to be an excellent tool for the preparation of nano-structured biologically selective systems. Polypyrrole (Ppy) is one such CP that is extensively used for the construction of bioanalytical sensors. Ppy has shown great promise primarily due to its biocompatibility and thermal stability under a variety of environmental conditions. In this study, caffeine imprinted electrodes (CIE) and theobromine imprinted electrodes (TIE) were prepared. This research project subsequently focused on three main aspects: 1) to determine the selectivity of a caffeine and theobromine imprinted MIP using Ppy as the conducting polymer matrix, 2) comparing pulsed amperometric detection (PAD) and electrical impedance spectroscopy (EIS), for their value as potential detection schemes and 3) to determined the applicability of the molecularly imprinted polypyrrole by analyzing commercial samples of instant coffee and tea and comparing results to that obtained from established HPLC procedures. In summary, the following conclusions are stated: - Both PAD and EIS measurements taken from CIE and TIE MIPs showed no statistical difference in response at the 95% confidence level using a standard paired t-test. - Reproducibility for both MIPs was estimated by calculating an average percent relative standard deviation (%RSD) for the corresponding MIPs and was determined to be less than 3%. - The degree of selectivity was estimated by calculating a % relative error for the CIE and TIE electrodes using both PAD and EIS analysis. These results revealed percent relative errors typically less than 5% for equimolar amounts of “interfering” analyte. - A ruggedness revealed that over the concentration range and time interval tested (1-20 mM and 5 days), the average percent relative standard deviation was determined to be less that 7%. - The caffeine content in the coffee sample analyzed, as determined by PAD and EIS, was consistent with results obtained by HPLC analysis however, the theobromine content determined in tea using PAD and EIS was significantly different from that determined by HPLC at the 95% confidence level