Authenticity Identification Of Camellia Oil By Ion Mobility Spectrometry

With the improvement of living standards, consumers are gradually becoming stringent on the quality of edible vegetable oil. Camellia oil, which is also called "Eastern olive oil", has been gradually accepted and favored for the nutritional and healthy property. Up till now, adulteration of camellia oil has becomed a serious factor to damage the interest of consumers, due to the relative high price and the limited detection methods. Therefore, it is of great significance to develop appropriate detection instruments and method for the rapid identification for camellia oil, as well as to accelerate the establishment and modification of national standards of camellia oil, which could help us to be far away from fake camellia oil and ensure food hygiene safety of edible oil and consumers’ rights and interests.In this paper, Ion Mobility Spectrometry(IMS) was applied for the adulteration identification of camellia oil. The main results have been listed as follows:1. The best parameters of IMS method for the identification of camellia oil has been established with the standard operating procedures(SOP) of oil. Alkali refining tests of different kinds of camellia oil were also conducted. The analysis of IMS indicated that the intensity of spectrum in IMS test almost remained to be unchanged with the refining processes, including degumming, deacidifying, bleaching and dewaxing. Heating time and temperature in deodorization process would slightly influence IMS spectra of camellia oil, however the main IMS spectrum still remains stable.2. 79 kinds of camellia oil samples and 615 kinds of other edible vegetable oil samples were collected and detected according to SOP requirements, in order to establish the authenticity model of camellia oil. Obvious differences between camellia oil samples and other edible vegetable oil could be observed. Taking the first-order and the second-order derivative of the data and extracting the feature points, most of other edible vegetable oil could be apart from camellia oil, but it was prone to produce error, because of the data of camellia oil was overlapped with that of other edible vegetable oil. Nevertheless, the correct rate of camellia oil can reach 93.15% after training and identifying with test data through artificial random forest.3. Volatile flavor components have been detected and identified by using boxes sampling- rapid IMS, in order to build up the fingerprint of camellia oil. However, results indicated that there was almost no difference between authentic and fake camellia oil, which means that adulterated and faked camellia oil couldn’t be identified just according to these spectra. So, principal component analysis and inspecting the principal component scores plot were used and in order to separate the authentic camellia oil from the fake one. After training and identifying with test data through artificial neural networks method, the correct rate of authentic camellia oil identification can reach 100%, and the accuracy to identify the fake oil can reach 88.5%.