Use of multivariate statistical techniques to optimize the separation of 17 capsinoids by ultra performance liquid chromatography using different columns.

In this work a multivariate statistical tool (Derringer and Suich optimization) was proposed for the separation of seventeen capsinoids (natural and synthetic) using the UHPLC-DAD chromatography. Capsinoids were analyzed at 280 nm. The variables optimized were the mobile phase (water (0.1% acetic acid as solvent A) and acetonitrile (0.1% as solvent B)), gradient time and flow rate. Two columns with different length (50 and 100 mm) were used for the chromatographic separation. The two columns used properly separated the seventeen capsinoids, however the 100 mm column length showed a better chromatographic separation with a shorter run time and smaller peak widths. These results provided better values of limit of detection and quantification for the 100 mm column length. The better conditions of separation with the 100 mm column length were established with: initial mobile phase with 41.8% of solvent B; 3.96 min of linear gradient time to reach 100% of solvent B; flow rate of 0.679 mL min(-1). A validation of the method has been done with excellent values of repeatability (RSD<1.92) and intermediate precision (RSD<3.92). The developed method has been applied to real samples. Capsiate has been identified and quantified in some varieties of peppers.