Study On Component Analysis And Adulteration Identification Of High-quality Edible Oil By NMR Combined With Pattern Recognition Method

Adulteration of edible oils affect the health of consumers and is a great challenge to food safety.As two high-quality edible oils,olive oil and camellia oil have been favored by people because of their high nutritional value.This has also led to the high prices of olive oil and camellia oil,which have become important targets for adulteration.Low-field nuclear magnetic resonance(LF-NMR)technology combined with pattern recognition method provides a brand-new idea for the rapid adulteration identification and quality control of edible oil.In this thesis,LF-NMR combined with chromatographic techniques(GC)were used to obtain the lateral relaxation spectrum and fatty acid composition of olive oil(OL)adulterated with the different ratios of soybean oil(SO)or corn oil(CO).The results showed that,the multi-component relaxation time(T21 and T22)were influenced by triacylglycerol components,while the change of their peak area ratio(S21 and S22)were attributed to the variety of MUFAs/PUFAs.Then,the OPLS-DA and PCA-LDA models based on pattern recognition were established for the classification and identification of adulterated OL,when the adulteration rate was higher than 30%,the 100%discrimination rate can be obtained in the OPLS-DA model of OL adulterated with SO or CO;the correct classification rate of PCA-LDA was 98.8%when the adulteration range was 30-100%.The research confirmed that LF-NMR combined with pattern recognition analysis method can be used as a fast,convenient and effective screening technique for the identification of adulteration of olive oil,and this technique was expected to be extended to the identification of adulteration of other edible oils.Based on the identification analysis of olive oil,LF-NMR combined with HF-NMR was used to complete a comparative study of camellia oil known as"Oriental olive oil" and olive oil in the difference of nutritional composition,starting from their composition and relaxation and modeling to identify adulteration of camellia oil.The experiment in HF-NMR showed that the content of oleic acid and linolenic acid in CA;the iodine value in the evaluation index of fatty acids in CA;the content of high-quality biological active substance(squalene)in CA were much higher than that in OL,proving the higher nutritional value in camellia oil.Using the LF-NMR combined with pattern recognition,the identification model of CA realized a complete distinction between the CA and the adulterated CA by OPLS-DA when the adulteration ratio is at least 30%.Combined with the conclusions above,as a rapid,convenient and effective screening technique,LF-NMR combined with pattern recognition should be widely applied in the discrimination of high-quality edible oil adulteration.