月桂酸单甘油酯是用途广泛的食品添加剂,在其制备过程中经分子蒸馏得到的制备品中会有月桂酸、甘油等杂质。用化学滴定或气相色谱等传统方法检测制备品中的月桂酸单甘油酯纯度及其杂质含量过程相当繁琐。为了对月桂酸单甘油酯制备品的品质进行快速定量,先利用气相色谱法确定不同工艺下的月桂酸单甘油酯产品中各成分的含量,再利用傅立叶红外光谱仪对月桂酸单甘油酯制备品进行分析,得到它们的光谱数据曲线,并结合主成分分析和反向传播神经网络建立回归模型。通过对实验结果的均方根误差预测值PRMSE以及相关系数r辨析,预测月桂酸单甘油酯含量的结果为PRMSE＝3.6376,r＝0.9950,预测甘油含量的结果为PRMSE＝1.4764,r＝0.9795,预测月桂酸含量的结果为PRMSE＝1.2859,r＝0.9247。结果表明,应用光谱分析方法能够较好检测月桂酸单甘油酯、月桂酸和甘油含量。

Glycerol monolaurate (GML) is a food additive which is widely used. The products prepared by esterification of glycerol and lauric acid followed by molecular distillation have lauric acid, glycerol and other impurities. To detect the glycerol monolaurate, lauric acid and glycerol in the products is a very tedious process using traditional methods such as chemical titrimetry and gas chromatography. For fast quality detection of the products that has been analyzed on gas chromatography, Fourier transform infrared spectroscopy (FTIS) was used in this study. Principal component analysis (PCA) and back-propagation artificial neural networks (BP-PANN) were used to build the regression model. Then this model was used to detect the glycerol monolaurate, lauric acid and glycerol of samples in the validation set according to their spectra. Root mean square error prediction (RMSEP) and correlation coefficient (r) of prediction were used as the evaluation standards, the predicted results of GML was PRMSE＝3.6376, r＝0.9950, the predicted results of glycerol was PRMSE＝1.4764, r＝0.9795 and the predicted results of lauric acid was PRMSE＝1.2859, r＝0.9247. The results indicated that it was feasible to detect glycerol monolaurate, lauric acid and glycerol content using spectroscopic technique.