A Validated Ultrasound-Assisted UPLC-PDA-QDa Method for Safflower Flavonoid Glycosides: Box-Behnken Optimization and Drying-Induced Compositional Changes

Safflower (Carthamus tinctorius L.) is widely recognized for its therapeutic and nutritional value, primarily attributed to flavonoid glycosides with strong antioxidant activities. To establish a rapid, green, and reliable analytical approach, an ultrasound-assisted extraction (UAE) method coupled with UPLC-PDA-QDa-MS was developed, optimized, and validated. The extraction parameters, including solvent concentration, solid-to-liquid ratio, temperature, and time, were optimized using Box-Behnken design, yielding the best extraction efficiency at 50 methanol, 0.3:20 (g mL⁻¹), 30 °C, and 15 min. The method demonstrated good linearity (R2 = 0.9927), satisfactory precision (CV < 11), and accuracy (recoveries 90-100), with limits of detection and quantification of 2.967 and 8.992 ppm, respectively, meeting ICH Q2 criteria. Five major flavonoid glycosides and two HSYA-derived compounds were identified, and the flavonoid glycosides were determined on a semi-quantitative basis as HSYA equivalents. Application of the validated method to safflower samples subjected to different drying treatments revealed significant compositional variations, where freeze-drying best preserved hydroxysafflor yellow A (HSYA), while hot-air drying favored other glycosides. The method provides a robust analytical tool for comparative profiling and quality control of safflower-derived products and supports green extraction and validation strategies in functional food and phytopharmaceutical research. © 2026, Walailak University. All rights reserved.