Enhancing Polyphenol Recovery and Stability in Lemon Myrtle (Backhousia citriodora) and River Mint (Mentha australis) in Herbal Tea Infusions Using Ultrasound-Assisted Extraction and Spray-Dried Microencapsulation

This study evaluated water-based ultrasound-assisted extraction (UAE) and spray-dried microencapsulation for producing phenolic-rich powders from Australian lemon myrtle (LM; Backhousia citriodora ) and river mint (RM; Mentha australis ). UAE was screened across time, amplitude and material-to-solvent ratio, with total phenolic content (TPC) used to identify the strongest tested condition. At 20 min, 30% amplitude and 1:20 m/v, method-development TPC was 25.10 ± 0.61 mg GAE/g for LM and 51.98 ± 4.28 mg GAE/g for RM. Applying these conditions to production materials, including a separately sourced commercial LM batch, yielded feed TPCs of 64.35 ± 2.72 and 37.67 ± 2.40 mg GAE/g, respectively. Spray drying with maltodextrin and maltodextrin–gum Arabic carriers gave production yields of 72.00–78.92% (LM) and 62.09–65.63% (RM), and process yields of 53.37–69.83% and 68.76–70.65%, respectively. RM showed higher TPC-based encapsulation efficiency (80.60–86.50%) than LM (59.71–68.02%), whereas LM powders had higher TPC (45.33–48.05 vs 24.25–25.63 mg GAE/g) and retention efficiency (68.8–74.9 vs 61.3–70.4%). Compound-specific encapsulation efficiency was approximately 78–85% for gallic acid in LM and 88–96% for rosmarinic acid in RM. Moisture ranged from 5.69 to 9.37%, while DPPH activity ranged from 3.31 to 10.52 mg TE/g for LM and 3.48–6.26 mg TE/g for RM. Powders had water activity of 0.42–0.47, solubility of 85.3–88.4%, Carr Index values of 39.69–46.73 and Hausner ratios of 1.66–1.88. Under the tested conditions, LM delivered greater phenolic density, while RM provided stronger proportional biomarker retention.