Multiresponse simultaneous optimization applied to epinastine determination in human serum by using capillary electrophoresis

GOICOECHEA, HÉCTOR C; AC OLIVIERI,; VERA CANDIOTI, LUCIANA

Campinas

Congreso; 10th Internacional Conference on Chemometrics in Analytical Chemistry (CAC 2006); 2006

CAC

The design of the experiments and the optimization of the experimental procedures for chemical measurements constitute an important area of chemometrics. The objectives of their application involve the study of the influence of different factors over a certain objective function, in order to find the experimental conditions for the simultaneous optimization of the response. This allows to develop quantitative methods with properties such as high resolution, sensitivity and specificity, as well as decreasing costs and operational time. On the other hand, it is known that separation by capillary electrophoresis depends on many different factors. As a consequence, the simultaneous study of all the potential factors becomes a complex task, mainly because it requires to spend a long time to carry out the analyses.1 In the present work, a Plackett-Burman experimental design was applied to the determination of the main factors affecting the peak resolution between the target analyte (epinastine) and lidocaine hydrochloride, used as an internal standard. The analyzed factors were: concentration and pH of the buffer, voltage and time of injection, and voltage of separation. An ANOVA test was applied to the experimental data with the following results that injection voltage, injection time and pH of the buffer are all significant (p < 0.05). Subsequently, three responses: resolution, wide and area of the epinastine peak were simultaneously optimized by resorting to the desirability function, which allows one to find experimental conditions where the system generates desirable results. 2,3 A central composite design consisting of twenty experiments was used. It was built with combinations of the independent variables in the following ranges: pH 3.00 – 4.00, voltage of injection 5 – 10 kV and time of injection 10 – 20 seconds. The latter ranges were selected based on the prior knowledge of the system under study. On the other hand, the buffer concentration and the separation voltage were set at 20 mmol. L–1 and 20 kV, respectively. The wavelength detector was set at 214 nm and the temperature was fixed at 25 ºC. The experimental conditions corresponding to a maximum of the desirability function (D = 0.713) are:  20 mmol. L–1 of buffer concentration, pH 4.00, voltage of separation of 20 kV, voltage of injection of 5 kV and 20 seconds of time of injection.  Finally, the selected conditions were applied to the analysis of epinastine in human serum samples.