Studies On The Methodology Of Optimizing Analytical Condition Of Capillary Electrophoresis

Capillary electrophoresis (CE), a powerful separation technique, has been widely applied in pharmacology and life sciences due to its high resolution, high efficiency and short analysis time. For the analysis of complex matrices, it is difficult to determine a suitable CE condition, due to the fact that many factors (such as separation mode, capillary type, buffer system, temperature, applied voltage, injection condition, etc) and their interactions will affect the CE separation. The optimization of experimental condition is very important for the development of an appropriate CE method. Therefore, study on the methodology for CE optimization is of great importance to facilitate the application of CE technique, and this is helpful for the development of pharmaceutical analysis.In this thesis, experimental design, genetic algorithm and particle swarm optimization algorithm were developed to optimize the CE condition for the analysis of complex matrices. The outline of this study is listed as follows:(1) Using experimental design to optimize the separation condition of CE. Taking Scutellaria baicalensis as a typical example, central composite design (CCD) for five factors (borate concentration, NaH₂PO₄ concentration, SDS concentration, acetonitrile percentage and isopropanol percentage) was implemented to achieve an optimal condition of the CE separation. Total resolution was used to evaluate the CE separation. Under the optimal condition, eighteen samples of Scutellaria baicalensis from different sources were analyzed within fourteen minutes, moreover, eight co-possessing peaks were detected. An analytical method based on this CE condition for the quantitative analysis of Baicalin, Baicalein and Wogonin in Scutellaria baicalensis was developed.(2) Nucleosides in urine were difficult to be well separated by CE, an optimization method coupled with experimental design and genetic algorithm was developed to achieve the optimal CE condition. CCD was implemented to investigate the influence of the factors (borate concentration, SDS concentration, pH value andapplied voltage) on the CE separation. Chromatographic exponential function (CEF), in terms of the resolution and analysis time, was utilized to evaluate the quality of the CE separation and used as the fitness function. Genetic algorithm (GA) was utilized for the solution of the established optimization, and the predicted optimal separation condition was validated. Eight nucleosides were well separated using this optimized CE condition.(3) For the CE analysis of Salvia miltiorrhiza, four major effective factors (borate concentration, NaH2PC?4 concentration, acetonitrile percentage and applied voltage) of CE were screened by orthogonal design, and they were further investigated by CCD. The modified chromatographic exponential function was developed as the fitness function. Particle swarm optimization (PSO) algorithm was utilized to obtain the optimal CE separation condition. The phenolic acids contained in Salvia miltiorrhiza were well separated by the optimized CE condition. The result showed that the optimized CE condition was suitable for the analysis of three active compounds (Protocatechuic aldehyde, Danshensu and Salvianolic acid B) in Salvia miltiorrhiza.