基于近红外光谱的甜菊糖吸湿过程表征及莱鲍迪苷A含量快速测定的研究

Characterization of Moisture Absorption Process of Stevia and Rapid Determination of Rebaudioside a Content by Using Near-Infrared Spectroscopy

作为一种绿色安全的食品药品配料,甜菊糖具有广阔的应用前景,然而吸湿性是其面临的一大难题,同时也是大多数制剂原辅料普遍存在的问题,研究分析吸湿的过程状态并提出针对性解决办法具有重要的理论意义和应用价值。利用近红外光谱分析技术结合化学计量学方法对甜菊糖吸湿过程进行表征、解析,从而揭示吸湿过程中水的吸附方式和键合作用;通过吸湿过程光谱并采用外部参数正交(EPO)算法消除样品中水分的影响,建立甜菊糖中莱鲍迪苷A(RA)含量的快速分析方法。研究表明,甜菊糖吸湿过程初始,水分子迅速吸附在甜菊糖粉末表面形成单分子层;之后,表面吸附位点变少,吸湿速率明显变慢,水分子将同时吸附在单分子层之上;最后,甜菊糖整体吸湿达到饱和状态,含水量保持稳定。揭示水分吸收规律后,利用EPO算法建立RA定量模型,模型外部测试集的均方根误差、决定系数和预测相对标准偏差分别为0.6695%,0.9570和4.3368,与未使用水分校正所建模型相比有较大提升,表明EPO算法可以有效去除吸湿的影响。该研究利用近红外光谱技术对甜菊糖吸湿过程中水的变化进行了表征,同时利用EPO算法有效消除吸湿的影响,实现对甜菊糖成品中RA的快速测定,为其进一步的研究和使用提供参考。

As a green and safe food and drug ingredient,stevia has broad application prospects.However,moisture absorption is a major problem it faces,which is also a common problem in most preparations raw materials.Research and analyze the process state of moisture absorption,then proposed targeted solutions have important theoretical significance and application value.Near-infrared spectroscopy analysis technology combined with chemometric methods were used to analyze the moisture absorption process of stevia in this study.On this basis,the External Parameter Orthogonalisation(EPO)algorithm was used to eliminate the influence of sample moisture,and to establish a rapid analysis method for Rebaudioside A(RA)content in stevia.The results showed that at the beginning of the moisture absorption process of stevia,water molecules were rapidly adsorbed on the surface of the stevia powder to form a monomolecular layer;after that,the surface adsorption sites became fewer,the moisture absorption rate became significantly slower,and water molecules would be adsorbed on top of the monomolecular layer at the same time;finally,the overall moisture absorption of stevia reached its saturated state,and the water content remained stable.After revealing the law of moisture absorption,the RA quantitative model was established using the spectrum preprocessed by the EPO algorithm,the root mean square error,coefficient of determination,and predicted relative standard deviation of the external test set of the model were 0.6695%,0.9570 and 4.3368,respectively.Compared with the model built before EPO treatment,there was a big improvement,indicating that the EPO algorithm could effectively remove the influence of moisture absorption.In this study,near infrared spectroscopy was used for the first time to characterize the water changes during the moisture absorption of stevia,at the same time,the EPO algorithm was used to effectively eliminate the influence of moisture absorption and realize the rapid determination of RA in stevia products,which provides a reference for its further research and use.