Eco-friendly extraction of Pistacia lentiscus bioactives: Supercritical CO2 technology and artificial neural networks predictions

Abstract

In the present study, supercritical CO 2 extraction was used as green extraction techniques to recover extracts from P. lentiscus both fresh and dried leaves. The fresh and dried leaf at different temperature ranging between 30 °C and 70 °C, was analyzed in a various of experiments at extraction temperatures between 35 °C and 55 °C and extraction pressures between 100 bar and 200 bar. The process efficiency was determined basing on the extraction yield, total phenol content, antioxidant and antidiabetic activity using artificial neural network. As a result, the fresh leaves extracted at 100 bar and 35 °C presented the most interest extraction yield (48.11 ± 0.56 %), followed by the dried leaves extract at 200 bar and 55 °C (39.39 ± 1.13 %). Regarding the total phenol content, the important amount was observed at 100 bar, 35 °C and 200 bar, 55 °C for the fresh and dried leaves at 30 °C respectively. Concerning antioxidant activity via DPPH test, the highest observed in fresh leaves was with an IC 50 of 37.13 ± 2.7 mg/mL, whereas for dried leaves, the best IC 50 was 37.18 ± 0.99 mg/mL for leaves dried at 30 °C. Similarly, for antidiabetic activity using α-amylase, the significant IC 50 are observed at low pressure and low temperature in fresh samples. The ANN model has a higher predictive potential with higher correlation coefficients of 0.9810, compared to other models. • Promoting eco-extraction for cleaner, sustainable processing. • Studying operational factors influencing extraction and bioactive yield. • Enhancing the effectiveness of the bioactive properties of extracted compounds. • Employing ANN modeling to predict and optimize the extraction process.