An improvement of the calibration results for grey analytical system in high performance liquid chromatography applying constrained background bilinearization method based on genetic algorithm optimization strategy

Abstract

Constrained background bilinearization (CBBL) method was applied for multivariate calibration analysis of the grey analytical system in high performance liquid chromatography (HPLC). By including the variables of the concentrations and the retention time of the analytes simultaneously, the standard CBBL was modified for the multivariate calibration of the HPLC system with poor retention precision. The CBBL was optimized globally by genetic algorithm (GA). That is to say, both the concentrations and the retention times of the analytes were optimized globally and simultaneously by GA. The modified CBBL was applied in the calibration analysis for both simulated and experimental HPLC system with poor retention precision. The experimental data were collected from HPLC separation system for phenolic compounds. The modified CBBL was verified to be useful to prevent the inherent limitation of the standard CBBL, which means that the standard CBBL may result in poor calibration results in the case of poor retention precision in chromatography system. Moreover, the modified CBBL can give not only the concentrations but also the retention time of the analytes. i. e., more useful information of the analytes can be generated by the modified CBBL. Subsequently, nearly ideal calibration results were obtained. On the other hand, comparing with the calibration results by the classical rank annihilation factor analysis (RAFA) and residual bilinearization (RBL) method, the results given by the modified CBBL were also improved significantly for the HPLC systems studied in this work.