Optimization of Ultrasound-Assisted Extraction of Phenolic Compounds from Salvia tomentosa via Response Surface Methodology: Evaluation of Antioxidant Activity and Phenolic Content

Objective: This study aimed to optimize the ultrasound-assisted extraction (UAE) of total phenolic content (TPC) and DPPH radical scavenging activity from aerial parts of Salvia tomentosa (S. tomentosa) using Response Surface Methodology (RSM) based on a Box-Behnken design (BBD). Materials and Methods: Extraction time (X₁), solvent-to-solid ratio (X₂), and methanol concentration (X₃) were selected as independent variables, with TPC and DPPH as response variables. Each factor was evaluated at three levels (X₁: 20–120 min; X₂: 20–40 mL/g; X₃: 0–100%), and the phenolic profile of the optimum extract was further characterized by high-performance liquid chromatography (HPLC).Results: The models showed high explanatory power with R² = 0.98 and adjusted R² = 0.96 for both responses, confirming the suitability of the developed second-order polynomial models in the experimental domain under investigation. The Lack-of-Fit was not significant for either model (p > 0.05), further confirming the model fit. Analysis of variance (ANOVA) of the selected independent variables revealed statistically significant effects on both TPC and DPPH (p < 0.05). Among the quadratic terms, methanol concentration (X₃²) had the most significant effect on the response for both TPC and DPPH (F = 365.72 and 350.71, respectively; p < 0.0001). The presence of negative coefficients confirmed a curvilinear relationship and a well-defined maximum within the experimental range. The interaction between extraction time and methanol concentration (X₁X₃) was found to be significant only for TPC (p < 0.05). This suggests that the synergistic effect of these parameters influences the solubility of phenolic compounds. Optimal extraction conditions were as follows: extraction time (X₁) = 20 min, solvent-to-solid ratio (X₂) = 37.31 mL/g, methanol concentration (X₃) = 44.86%. The maximum TPC and DPPH values obtained under optimum extraction conditions were 105.75 ± 5.22 mg GAE/g and 74.17 ± 4.06 mg AAE/g, respectively. HPLC analysis of the optimized extract showed that rosmarinic acid and hesperidin were the most abundant phenolic compounds, at 13550.99 and 996.18 mg/g dry sample, respectively. Conclusion: These findings demonstrate that RSM optimization using BBD is an effective and systematic approach for the recovery of bioactive phenolic compounds from Salvia plants, and that the UAE method helps shorten extraction time without compromising bioactive compounds.