Modelling and Optimization of Ultrasound Assisted Extraction of Polyphenols Using Response Surface Methodology

Abstract

Mango seed kernels are by-products of the consumption and transformation of mango fruits (Mangifera indica L.). Many ways of valorisation have been proposed, and among them, their phenolic compounds extraction. To increase the extraction yield, ultrasound-assisted extraction was modelled and optimized. The 4 factors Central Composite design associated with the Response Surface Methodology (RSM) were used to achieve that goal. The effect of extraction time, temperature, stirring rate and the Ultrasound Amplitude, on the total phenolic compound extraction yield and the total reducing power of the extract, were studied and modelled. The modelling allows us to do a multi-response optimization to identify the best-operating conditions to achieve at the same time the highest extraction yield and antioxidant capacity. The optimal operating conditions achieved were 41.82 min of extraction time, 54.75?C as extraction time, under 266.67 rpm as stirring rate, and 100% ultrasound amplitude. With an expected extraction yield of 71.35 mg GA/g, and 123.058 mg AA/g of total reducing power. 2 extraction cycles, under these conditions, are enough to extract a maximum of the phenolic content, under the described conditions.