Study On Food Classification By Combining Chemometrics With Fluorescent Probes

As an important material basis for human survival and development,food provides the necessary nutrition and energy for the human body.With the increasing concern about food quality and safety,it is significant to explore fast and accurate methods for food origin traceability and adulteration identification analysis.In this paper,the combination of chemometrics and fluorescent probe technology was used to classify food.The details are as follows:In Chapter 2,a weakly selective fluorescent coumarin-like probe L1 was used to act on various metal ions in rice based on the different metal elements,and the three-dimensional fluorescence data were analyzed by combining multi-dimensional principal component analysis(M-PCA)and unfolded partial least squares discriminant analysis(U-PLS-DA)to achieve satisfactory classification results.Based on the comparative purpose,rice two-dimensional infrared data were simultaneously collected using infrared technology and further processed for authenticity identification study by combining principal component analysis(PCA)and partial least squares discriminant analysis(PLS-DA).The experimental results comparing the two methods showed that the weakly selective fluorescence-based probe for rice classification and identification was better than those based on infrared,indicating that the weakly selective fluorescence probe could be applied to rice authenticity identification studies.In Chapter 3,based on the differences in the types and contents of metal elements in different kinds of tea leaves,the responses of the new weakly selective quinoline-based probe L2 to the interaction with metal ions in tea leaves were different.Three-dimensional fluorescence fingerprints of green tea and adulterated tea of different origins were collected,and the three-dimensional fluorescence data of tea were analyzed by combining M-PCA,multi-dimensional partial least squares discrimination analysis(N-PLS-DA),and U-PLS-DA models,and the results showed that all three multi-dimensional identification methods could successfully classify the tea origins.In the study of adulteration identification of tea,the total correct identification rate of the training and prediction sets of the U-PLS-DA model were100%,and its overall results were better than those of M-PCA and N-PLS-DA.The results showed that satisfactory discrimination results could be achieved in the identification of tea origin and adulteration level by using the weak selective probe,which provided new ideas for the classification study of other food products.In Chapter 4,the 8-hydroxyquinoline-like weakly selective probe L3 was used to interact with metal ions in kudzu root based on the differences in metal elements contained in kudzu root from different origins.The two-dimensional UV-Visible data and three-dimensional fluorescence data of the samples after interacting with the probe were used in the authenticity identification study of kudzu root in combination with chemometrics,respectively.In comparison,the results obtained by the multi-dimensional pattern recognition methods(M-PCA and U-PLS-DA)were better than those obtained by the PCA and PLS-DA methods based on two-dimensional UV-Visible absorption spectroscopy data for the classification and identification of the authenticity of kudzu root,because the three-dimensional fluorescence data could obtain more information about the characteristics of the kudzu root samples.Experimental results showed that more accurately achieved authenticity identification of kudzu root could be obtained by using the weakly selective probe acting on metal ions to collect three-dimensional fluorescence data combined with multi-dimensional pattern recognition.In summary,the weakly selective fluorescent probes technique combined with chemometric multi-dimensional pattern recognition method for origin tracing and adulteration identification of rice,tea and kudzu root is a rapid and accurate analytical tool to achieve satisfactory classification results,and the technique also provides a new idea for authenticity discrimination of non-fluorescent or weakly fluorescent foods.