Multivariate optimization of an ultrasound-assisted extraction method of bioactive phenolic compounds in malagueta peppers (Capsicum frutescens)

The use of statistical tools in the optimization of bioactive extraction processes increases efficiency and reduces process costs, indicating the optimal conditions for obtaining the best results. The objective of this work was to optimize an ultrasound-assisted extraction procedure and to quantify phenolic compounds in Capsicum frutescens samples. In the optimization of the extraction solution, simplex-centroid mixture design for the proportions optimization of the extractor components (water, methanol, and ethanol) was used. A two-level factorial design (23) was used for the preliminary evaluation of the factors that influenced the extraction process; before this step, a Box-Behnken design was carried out to optimize the factors of solvent volume, sonication time, and temperature with the desirability function as a response. The best mixture composition was 95% methanol and 5% water and the best experimental conditions for the process variables were 16 mL, 15 min, and 55 °C. Applying the method to pepper samples, twelve compounds were determined by HPLC–DAD. The relative standard deviations (RSD) found for all analytes were lower than 3.0%. The accuracy was evaluated using addition/recovery test with recoveries between 81 and 95%. The Box-Behnken and mixture designs are efficient tools for optimizing the analytical method using ultrasound to extract phenolic compounds from C. frutescens. The procedure was applied for extraction phenolics compounds in three papers samples and the concentrations found were trans-cinnamic acid (