Process development for isolation of dietary eugenol from leaves of basil (Ocimum sanctum) in combination of optimization of process variables and modeling by artificial neural network

Abstract

Dietary eugenol helps prevent free radical-induced and lifestyle-related chronic illnesses such as cancer, autoimmune disorders, cardiovascular disease, and aging. A technique for extracting eugenol from green basil (Ocimum sanctum) leaves is created using a combination of extraction variable optimization and the organization of an artificial neural network (ANN) model. For thermally degradable bioactive eugenol, solvent extraction is the recommended separation method. With the following optimum variables: polarity of the solvent of 0.009, the solid-solvent ratio of 1.0 ​g/20 ​mL, stirring speed of 200 ​rpm, extraction temperature of 40 ​°C, and extraction duration of 40 ​min, a yield of 5.39 × ​10−3 ​kg eugenol per kilogram dried leaves of basil was found. At 10 ​min of batch extraction, the highest throughput of eugenol was found to be 5.4 ​× ​10−3 ​kg ​m−3 ​s−1. Additionally, experimental data are used to construct the yield prediction model. The statistical parameters that are obtained in model evaluation encourage the use of the predicted model for the commercialization of eugenol isolation.