紫外-可见-短波近红外漫反射光谱结合化学计量学对丙氨酸的手性进行快速的定性定量分析研究

中文摘要: 以D-、L-和DL-丙氨酸为对象，首次研究了紫外-可见-短波近红外漫反射光谱结合化学计量学快速、损地定性、定量分析固体手性化合物的可行性。在粒度过100目、样品与光纤探头距离5.6mm的最优条件下测量漫反射光谱，经预处理后，采用主成份分析法（PCA）和判别偏最小二乘法（PLS-DA）建模进行定性分析，结果预测判别准确率为100%。定量分析上，16个不同比率的L-/D-丙氨酸二元混合粉末样的光谱经波长选择、光谱数据预处理后，以偏最小二乘回归法（PLR）和支持向量回归法（SVR）建模，校正集和预测集的最优决定系数（R2）均在0.99和0.98以上，绩效偏差率（RPD）值远高于2。研究表明，采用紫外-可见-短波近红外漫反射光谱法结合化学计量学可应用于手性固态丙氨酸的快速、损分析，并有望于运用到手性药物的质量检测上。
Abstract:UV-visible-shortwave near infrared diffuse reflectance spectroscopy (UV-vis-SWNIR DRS) combined with chemometrics was firstly used as a new method to analyze enantiomer and its racemic compound, using D-, L- and DL-alanine as model compounds. Under the optimal measuring conditions(powder particle sizes of 100-mesh and distance between fiber probe and sample of 5.6mm), after spectral pretreatment and/or wavelength selection, principal component analysis (PCA) and discriminant partial least squares (PLS-DA) resulted in a correct discrimination of chirality with 100 percent accuracy; partial least squares regression (PLR) and support vector regression (SVR) models allowed a satisfied quantitation of D/L binary mixture with determination coefficients (R2) above 0.99 and 0.98 for calibration set and prediction set respectively by both PLR and SVR, and ratios of performance deviation (RPD) far above 2. The results show that it is feasible to apply UV-vis-SWNIR DRS on chiral analysis of compounds.