DETERMINATION OF INITIAL CONCENTRATION OF AN ANALYTE BY KINETIC DETECTION OF THE INTERMEDIATE PRODUCT IN CONSECUTIVE FIRST-ORDER REACTIONS USING AN EXTENDED KALMAN FILTER

Many analytical reagents react with analytes according to a rate law d escribed by a consecutive first-order reaction mechanism, A-->B-k1 --> C-k2, where a product formation step is followed by a product degradat ion step. In many cases, consecutive first-order reactions are used to determine the initial concentration of analyte A by kinetic detection of the intermediate product B. In this work, the factors that affect the determination of the initial concentration of analyte-and the kine tic parameters for this class of reactions have been investigated by c omputer simulations using an extended Kalman filter. The flip-flop phe nomenon exhibited by this kinetic model is discussed, and a unique app roach for the determination of the initial concentration of analyte is proposed. Empirical and residual methods for obtaining initial estima tes for the rate constants and the initial concentration from the kine tic data, based on the model for consecutive first-order reactions, ha ve been developed. The effects of changing the rate constants, the ini tial concentration of analyte, the mismatch between the measured time and the real time, the data density, the fitting range, and the initia l estimate for the background signal have been evaluated by using synt hetic data with Gaussian-distributed mise. The percent errors in the e stimated values for the parameters that are proportional to the initia l concentration of analyte have been evaluated for each of the differe nt variables under consideration. Plots of these errors as a function of the various effects mentioned above permit the methods to be comple tely characterized.