Application Of Near Infrared Spectroscopy Technology In Quality Analysis Of Stevia Production Process

Stevia has become an ideal sweetener because of its high sweetness,low calorie,good taste,no toxic side effects,and has the advantages of regulating blood sugar and lowering blood pressure.It is increasingly valued and favored by the food and medicine industry.However,our country’s stevia industry still has defects such as high energy and low efficiency in production technology,large fluctuations in product quality,and lack of quality monitoring of the production process,resulting in low competitiveness of my country’s stevia industry in the international market.Near infrared spectroscopy(NIRS),as an effective process analysis technology,is fast,non-destructive,and suitable for online advantages,which can realize real-time monitoring of the production process and ensure uniform and stable product qualityThis research carried out the rapid quality analysis of stevia production process,mainly from the three aspects of Stevia rebaudiana Bertoni leaves extraction process,the macroporous resin elution process and the finished product quality control,and established the quality analysis method of stevia production process based on NIRS,which made a breakthrough in stevioside Industrial production and quality monitoring technology are insufficient,and a quality monitoring system based on process analysis technology is established to lead the development of the industry and lay the foundation for the modernization of my country’s natural product production.The specific content of this research includes the following three aspects:(1)Research on the application of NIRS in the quality analysis of Stevia rebaudiana Bertoni leaves extraction process.Aiming at the insufficiency of the fuzzy end point of the extraction process,this study simulated the extraction process in the laboratory,using the Antaris II Fourier transform near-infrared spectrometer combined with related chemometric methods.Firstly,the quantitative analysis of the index components in the stevia leaf water extract were established,which could realize the rapid determination of rebaudioside A(RA)and stevioside(STV).Then,the process characterization based on water spectromics was completed,and visualized the extraction process at the molecular level.Finally,different endpoint judgment methods were investigated and compared,and the moving block of standard deviation method was chosen at last.This study provides a reference for the quality analysis of the stevia leaf extraction process by quickly determining the fluctuation of the index components during the extraction process(2)Research on the application of NIRS in the quality analysis of macroporous resin elution process.Aiming at the lack of effective process analysis methods for the macroporous resin elution process,this study first simulated the elution process in the laboratory,verified the feasibility of online monitoring of RA content through Bruker MATRIX-F,and then transferred the site from the laboratory to the factory.The portable MicroNIR1700 was used to establish a linear model and a non-linear model of the RA content in the eluate to realize the accurate prediction of the RA content in the elution process.This study can be used to quickly determine the content of index components in the elution process,and to achieve accurate collection of the effective components during the elution process(3)Research on the application of NIRS in the quality analysis of stevia products.Aiming at the hygroscopic properties of stevia,this study firstly used NIRS combined with the second derivative,principal component analysis,two-dimensional correlation and gravimetric analysis to characterize the hygroscopicity process of stevia to reveal the adsorption mode,bond and effect of water.Then,the external parameter orthogonalization algorithm(EPO)was used to extract the external matrix of the moisture absorption process,thereby effectively eliminating the influence of moisture absorption by the sample,and finally the prediction performance of the model was improved.This study has an important role in understanding the moisture absorption process and making accurate and rapid determination of RA content.