论文部分内容阅读
目的:采用星点设计-效应面法优化地不容超声提取工艺。方法:以乙醇浓度、提取时间以及溶媒比为自变量,以左旋延胡索乙素提取率和干膏收率的综合评分为因变量,用星点设计-效应面法优选地不容提取工艺,对所得结果建立多元线性回归和二项式方程拟合等数学模型,并进行预测分析。结果:最佳提取工艺为乙醇浓度55%;提取时间20min;溶媒比20mL·g~(-1);提取2次,二项式拟合的复相关系数r=0.971 0,工艺验证的实际值与预测值的偏差为-3.91%。结论:所建立的数学模型精密度高,优化后的超声提取工艺操作简单,重复性、预测性良好。
OBJECTIVE: To optimize the ultrasonic extraction process by using the star point design-response surface method. Methods: Taking ethanol concentration, extraction time and solvent ratio as independent variables, taking the comprehensive score of levodropropin extraction rate and dry extract yield as the dependent variable, the star point design-response surface method Results The mathematical models of multivariate linear regression and binomial equation fitting were established and predicted. Results: The optimal extraction process was ethanol concentration 55%, extraction time 20 min, solvent ratio 20 mL · g -1, extraction twice, the complex correlation coefficient of binomial fitting r = 0.971 0, the actual value of process validation The deviation from the predicted value is -3.91%. Conclusion: The mathematical model established has high precision and the optimized ultrasonic extraction process is simple, reproducible and predictive.