OPTIMIZATION OF ULTRASOUND-ASSISTED EXTRACTION OF PHENOLIC COMPOUNDS FROM JABUTICABA PEEL AND SEED

Jabuticaba (Plinia cauliflora) is rich in anthocyanins and ellagitannins, which are concentrated in the peel and seeds and are usually discarded during pulp processing. These by-products are therefore a promising and underused source of phenolic compounds, and recovering them adds value to the fruit production chain. Ultrasound-assisted extraction is a green technology that enhances mass transfer through acoustic cavitation, but its efficiency depends on operational conditions and their interactions. This study evaluated how temperature, solid:solvent ratio, and water:ethanol ratio affect ultrasound-assisted extraction of phenolic compounds from jabuticaba peel and seeds (JPSP) to maximize their recovery. JPSP obtained as a by-product of steam-extracted juice production were dried at 75 ºC for 22 h and ground. Extraction was performed by ultrasound (40 kHz, 100 W, 10 min) following a three-factor, three-level Box-Behnken design with 15 runs, including three center points. The factors were temperature (37 ºC, 48.5 ºC, and 60 °C), solid:solvent ratio (50 mg/mL, 225 mg/mL, and 400 mg/mL), and water:ethanol ratio (100:0, 65:35, and 30:70, v/v). Runs were performed in duplicate and center points in triplicate. Extracts were centrifuged, filtered, and analyzed by high-performance liquid chromatography with diode array detection. The data were evaluated by analysis of variance (ANOVA), Pareto charts, and the desirability function using Statistica^®. Six compounds were identified under all conditions: gallic acid, ellagic acid, cyanidin-3-O-glucoside (C3G), delphinidin-3-O-glucoside, castalagin, and vescalagin. Ellagic acid (20%), vescalagin (22%), and castalagin (22%) predominated, together accounting on average for 64% of the total phenolic compounds. The solid:solvent ratio was the most consistently significant factor, and 400 mg/mL yielded the highest contents. The water:ethanol ratio had the strongest effect on castalagin and vescalagin and interacted significantly with the solid:solvent ratio. Higher water proportions favored gallic acid, castalagin, and vescalagin but reduced ellagic acid and C3G, which reflects the different polarities of these compounds. The desirability function identified 60 °C, 82.5% water, and 400 mg/mL as the optimal condition (overall desirability of 0.86). This prediction was close to the experimental run with the highest total phenolic content (48.5 °C, 100% water, and 400 mg/mL; 131.7 mg/100 mL). The solid:solvent ratio was the main determinant of phenolic recovery, whereas solvent composition acted selectively according to the compound. Temperature had no significant effect within the tested range, so milder conditions could be used without loss of recovery and with lower energy demand. Overall, ultrasound-assisted extraction is an efficient and sustainable strategy for recovering ellagitannins and anthocyanins from jabuticaba by-products, with potential applications as natural ingredients in the food industry.

Acknowledgments: CAPES (Finance Code 001; 88887.285600/2026-00), CNPq (310912/2022-9), and FAPERJ (E-26/211.286/2021; E-26/210.082/2023; E-26/210.136/2023; E-26/200.027/2026; E-26/200/028/2026).