Optimizing Anthocyanin Extraction and Sensory Quality of Roasted Black Bean Tea Using Quadratic Modeling and Response Surface Methodology

Black beans ( Phaseolus vulgaris L. ) are rich in anthocyanins and have potential for functional beverage development. This study aimed to optimize anthocyanin extraction and evaluate the sensory quality of roasted black bean tea using response surface methodology (RSM). The effects of roasting time (8-14 min at 200 °C), infusion time (5-20 min), and infusion temperature (70-100 °C), with a brewing ratio of 5 g of raw material per 100 mL of water, on total anthocyanin content (TAC), total soluble solids (TSS), and sensory quality were first evaluated using single-factor experiments and fitted models. The results showed that all three factors significantly affected anthocyanin extraction, with optimal conditions of 12 min roasting time, 15 min infusion time, and 90 °C infusion temperature. Quadratic regression models were used to describe these trends, with infusion time showing a greater effect compared to the other factors on TAC. The interactions among these variables were further analyzed using RSM based on a Box–Behnken design. The quadratic model was statistically significant (p < 0.05) with a coefficient of determination (R 2 = 0.9403). The optimal conditions predicted by the model were roasting time of 11.73 min, infusion time of 16.11 min, and infusion temperature of 92.22 °C, resulting in a predicted TAC of 4.27 mg/mL. Experimental validation produced a TAC value of 4.23±0.21 mg/mL, showing reasonable agreement with the predicted value. Overall, the optimized conditions enhanced anthocyanin extraction while maintaining desirable sensory attributes, demonstrating the potential for developing high-quality functional beverages from black beans.