Comparing ultrasonic- and microwave-assisted methods for extraction of phenolic compounds from Kabkab date seed (Phoenix dactylifera L.) and stepwise regression analysis of extracts antioxidant activity

This study aimed to extract bioactive compounds (phenolics and flavonoids and condensed tan-nins) from roasted date seed (Phoenix dactylifera L. cv Kabkab) using various solvent systems (W: water, AE: aqueous-ethanol, AA: aqueous-acetone) and extraction method (ultrasonic-assisted (UAE), microwave-assisted (MAE) and the combination of these two methods (UMAE) macera-tion (ME) and Decoction-Infusion (DIE) Extraction). Moreover, the feasibility of antioxidant ac-tivity prediction was investigated based on stepwise regression analysis and phytochemical prop-erties. Extraction yield, Total phenolic content (TPC), total flavonoid content (TFD) and antioxi-dant activity (DPPH*, ABTS*+, FRAP and averaged antioxidant activity: AAA) of the extracts were evaluated. The main effect of solvent systems and extraction methods on phytochemical compounds and antioxidant activity were significant (P < 0.01). Water and aqueous-ethanol sol-vents extracted higher phytochemical compounds than aqueous-acetone (P < 0.05). Although the highest extraction performance was observed for the ME method, the novel methods show an acceptable result in a much shorter time. Among novel methods, the highest and lowest perfor-mances were recorded for UAE and MAE. There was no significant difference between novel -green methods (e.g., UAE and UMAE). Although the lowest phytochemical yield was obtained for MAE, this performance was obtained in less than 5 min. The highest and lowest correlation be-tween the phytochemical compositions and antioxidant activity parameters were for DPPH* and AAA, respectively. Stepwise-regression analysis showed the weakest and strongest prediction models for AAA (10-fold S = 0.098 and 10-fold R2 = 87.07) and ABTS*+ (10-fold S = 0.129 and 10-fold R2 = 78.25), respectively.