近红外光谱法定量分析水溶液中葡萄糖、麦芽糖和β-环糊精

Quantitative Analysis of Glucose,Maltose and β-Cyclodextrin in Aqueous Solution by Near Infrared Spectroscopy

本文应用正交设计法配制了葡萄糖、麦芽糖和伊环糊精混合水溶液的29个样品（其中25个用作校正集样品，4个用作预测集样品），扫描此混合水溶液样品的近红外（NIR）光谱，并用Unscrambler定量分析软件将近红外光谱与对应的化学成分值相关联，在不同波长范围下建立了水溶液中葡萄糖、麦芽糖和p环糊精的近红外原始光谱、一阶导数光谱的偏最小二乘法（PLS）定量分析数学模型，根据模型评价参数选择了最佳模型，并且用这些模型对未知样品进行了预测。结果显示，对浓度范围分别在22.73-104.17mg·mL^-1、22.73-104.17mg·mL^-1和1.85-8.41mg·mL^-1的葡萄糖、麦芽糖和仔环糊精水溶液，其NIR数学模型的相关系数分别为0.9946、0.9976和0.9036，模型的预测相对偏差（RMSEP）分别为2.52mg·mL^-1、1.95mg·mL^-1和0.87mg·mL^-1，预测标准偏差（SEP）分另4为2.58mg·mL^-1、2.01mg·mL^-1和0.90mg·mL^-1。

In this paper, 29 mixed samples of glucose, maltose and β-cyclodextrin in aqueous solution were designed by orthogonal design, in which 25 samples were used for the calibration set and 4 samples for the prediction set. The partial least square mathematical model for simultaneous determination of glucose, maltose and β-cyclodextrin in aqueous solution was built by the Unscrambler software associating with near-infrared spectroscopy and the corresponding value of the chemical constituents. The original near-infrared spectra and the first derivative spectra of the mathematical model were compared in different wavelength range. The best model parameters were chosen by the model evaluation, and unknown samples were predicted using these models. The results showed that the concentration range were 22.73-104.17mg·mL^-1 for glucose, 22.73-104.17mg·mL^-1 for maltose and 1.85-8.41mg·mL^-1 for β-cyclodextrin, respectively. The correlation coefficients （r） of the mathematical model were 0.9946, 0.9976 and 0.9036, respectively. The root mean square errors of prediction（RMSEP） were 2.52mg·mL^-1, 1.95mg·mL^-1 and 0.87mg·mL^-1, and the standard errors of prediction （SEP） were 2.58mg·mL^-1 , 2.01mg·mL^-1 and 0.90mg·mL^-1, respectively.