Rapid Detection Of Authenticity And Adulteration Of Walnut Oil By Fourier Transform Infrared Spectroscopy And Fluorescence Spectroscopy

In human daily diet,oils provide the necessary nutrition and energy for the body.Walnut oil is of high nutritional value with a large amount of unsaturated fatty acids,but the adulteration of walnut oil sometimes occurred in recent years because of its high price.In the presented thesis,the rapid detection of authenticity and adulteration of walnut oil by Fourier transform infrared spectroscopy(FTIR)and fluorescence spectroscopy combined with chemometrics was established.The main results obtained are as follows.(1)Gas chromatography(GC)was used to analyze fatty acid composition and contents of five different edible oils.The contents of linolenic acid in walnut oil were 9.0-10.0%,and other oils had little or no linoleic acid.The ratio of linoleic acid to stearic acid in walnut oil was 3.1-4.0,comparing with about 1.0 in peanut oil,1.9-2.3 in soybean oil and 2.3-2.8 in sunflower oil.(2)FTIR combined with principal component analysis(PCA)was used to identify the authenticity of walnut oil.After the first derivative and SNV treatment,the spectra of walnut oils had no overlap with those of other oils and they didn't disturb with each other.Therefore,FTIR-PCA classification model was basically feasible.FTIR combined with PCA and PLS were then used to qualitatively and quantitatively analyze the adulterations.Peanut oils or soybean oils adulterated in walnut oils were analyzed by PCA,and the PCA score plot could be divided into two categories evidently.The PCA classification limits of soybean oil or peanut oil adulterated in walnut oil were both 10%.PLS analysis methods were used to establish a binary system about walnut oil adulteration amount.Soybean oil and peanut oil volume fraction by the PLS calibration model with R2(coefficient of determination)and RMSEC(root mean square error of prediction)value were 0.9963,0.9993 and 0.0038,0.0017 respectively.Prediction models with R2 and RMSECV(root mean square error of cross-validation)value were 0.9956,0.9992 and 0.0041,0.0019 respectively.While the deviation of predicted results obtained were between 1.2299-2.2824,the difference was not significant,and the results illustrate calibration model was feasible.(3)Fluorescence spectroscopy combined with PCA and PLS was used to qualitatively and quantitatively analyze the adulteration of walnut oil.The PCA classification limits of soybean oil mixed with walnut oil adulteration was below 5%.Soybean oil volume fraction by the PLS calibration model with R2 and RMSEC value were 0.9991 and 0.4769.Prediction models with R2 and RMSECV value were 0.9951 and 1.2038.While the deviation of predicted results obtained were between 0.6388-1.7163,the difference was not significant,the results illustrate calibration model was feasible.FTIR and GC can identify the authenticity of walnut oil accurately.But GC is more complicated and need long process of analysis,which does not apply to the online testing.Although the adulteration detection limit of fluorescence spectrometry is lower than FTIR in the identification of the adulterated walnut oils,the concentration of adulteration in real life are generally higher than 10%,so the FTIR and fluorescence spectroscopy in the identification of walnut oil adulteration aspect have no much difference.