豆腐木广泛分布于秦巴山区，其叶果胶含量高且酯化度低，具有良好的应用开发前景。为充分利用豆腐木叶果胶资源，考察了pH、提取温度和超声功率三个因素对豆腐木叶果胶得率的影响，在单因素试验基础上进行了3因素3水平的Box-Behnken中心组合试验，利用响应面分析法对超声波辅助提取技术进行了优化，并对优化后的提取产物进行了30%和50%乙醇分级沉淀分离，测定分析了这两个沉淀组分的理化性质和红外光谱特征。试验结果表明，以果胶得率为响应值的曲面回归方程拟合性良好，在pH 1.9、温度70 ℃、超声功率630 W的条件下，果胶实际得率可达23.16%，接近理论预测值23.20%。优化条件下提取的豆腐木叶果胶经30%和50%乙醇分级沉淀，两种沉淀组分的pH、灰分指标值均符合QB 2484-2000要求，红外光谱分析结果也表明其均具有果胶典型结构特征。

Premna microphylla leaves were wildly existed in Qinba Mountain with high pectin content and low degree of esterification. In order to make full use of Premna microphylla leaves, the effects of different solvents-pH, extraction temperature and ultrasonic power on the yield of pectin extraction were investigated. Response surface methodology was employed to optimize the pectin extraction conditions from Premna microphylla leaves. The mathematical model was established according to Box-Behnken central composite design. Besides, the fractionally precipitates with ethanol concentration of 30% and 50% respectively were produced from the optimized extraction products, and their physicochemical properties and infrared spectrum were studied. The results showed that the fitted quadratic regression model was valid. The optimum extraction parameters were pH 1.9, extraction temperature 70 ℃, and ultrasonic power 630 W. Under these conditions, the maximal rate of pectin extraction yield was 23.16%, which was close to the predicted value with 23.20%. The pH and ash value of 30% and 50% fractionally precipitates were in accordance with the requirement of QB 2484-2000 and their typical characteristics were also shown by infrared spectrum.

陕西省农业科技创新与攻关项目（2014K02-13-01）；安康市科技计划项目（2014AK01-11）；陕西省大学生创新创业计划项目（2015sxjy014）