Study Of Oxidation Stability And Discrimination Technology Of Walnut Oil

Walnut oil is a kind of healthy edible oils, which is rich in unsaturated fatty acid and can reduce blood fat and has the ability of anti-oxidation. However, walnut oil is easy to become oxidative deterioration. In addition, some adulterated goods appear frequently as the walnut oil market expands significantly. Therefore, it is important to estabilish some technologies for rapid detection of adulterated walnut oil. In this study, walnut oil was chosen to study its oxidation stability and the detection of adulteration. The main conclusions are as follows:1. A series of chemical reactions happened in walnut oil during heating process. Acid value and peroxide value showed rising trends. The content of saturated fatty acid was increased with the decrease of unsaturated fatty acid. Oleic acid, linoleic acid and linolenic acid were almost destroyed during the heating process. Meanwhile, the content of γ-tocopherol decreased from the initial value of 57.20 mg/kg to zero after oxidation. The results showed that unoxidized walnut oil had good antioxidant ability, the inhibition rate on DPPH·, ·OH and ·O2- had a positive correlation with the concentration of walnut oil. The inhibition rates on DPPH· and ·OH reached 50% when the concentration of walnut oil were 76.21 mg/mL and 86.65 mg/mL, respectively. The average inhibition rates of walnut oil and VC on ·O2- were almost at the same level when the concentrations of walnut oil and VC were 1.0 mg/mL. The inhibition rates of walnut oil on DPPH·, ·OH and ·O2- were decreased after oxidization. The average inhibition rate of walnut oil on ·O2- dropped from the initial value of 55.53% to 15.43% after oxidization of 30 d.2. Fourier transform near infrared spectroscopy(FT-NIR) combined with partial least squares(PLS) was used to rapidly quantify the adulteration of walnut oil. Walnut oils adulterated with different concentrations of rapeseed oil, soybean oil and corn oil were distinguished. FT-NIR spectra of the adulterated mixtures were measured in the range of 12000~4000 cm-1, and the spectra of walnut oils adulterated with rapeseed oil was subjected to Min/Max normalization. Results showed that the relative correlation of Calibration(RC), Root Mean Square Error of Calibration(RMSEC), were 99.97%, 0.532%, 99.96% and 0.562%, respectively. The spectra of walnut oils adulterated with soybean oil was subjected to First Derivative and Vector Normalization, and the RC, RMSEC, RCV and RMSECV were 99.99%, 0.336%, 99.99% and 0.352%, respectively. The spectra of walnut oils adulterated with rapeseed oil and soybean oil were subjected to Second Derivative, and the RC, RMSEC, RCV and RMSECV of rapeseed oil were 99.99%,0.220%, 99.98% and 0.313%, respectively; and the RC, RMSEC, RCV and RMSECV of soybean oil were 99.99%,0.210%, 99.97% and 0.386%, respectively. The spectra of walnut oils adulterated with rapeseed oil, soybean oil and corn oil were subjected to First Derivative and Multiple Scattering Correction,Second Derivative, and First Derivative, and the RC, RMSEC, RCV and RMSECV of rapeseed oil were 99.93%,0.347%, 99.90% and 0.401%, respectively; and the RC, RMSEC, RCV and RMSECV of soybean oil were 99.87%,0.524%, 99.80% and 0.615%, respectively; and the RC, RMSEC, RCV and RMSECV of corn oil were 99.94%,0.371%, 99.89% and 0.451%, respectively. FT-NIR method offered a new approach to rapid quantification of pure and adulterated walnut oil.3. The detection of adulteration in walnut oil was performed using a PEN3 electronic nose system. Walnut oils adulterated with different concentrations of cheaper seed oils such as soybean oil, rapeseed oil and corn oil were distinguished. Electronic nose combined with principal component analysis(PCA) or linear discriminant analysis(LDA) was used in the study. Results showed that the adulteration of walnut oil could be detected effectively, and LDA was more effective than PCA. Walnut oils adulterated with soybean oil at levels of adulteration exceeding 20% could be distinguished easily by PCA, and the distinguishable adulteration level for rapeseed oil or corn oil was beyond 7%. Walnut oils adulterated with soybean oil or rapeseed oil at levels of adulteration exceeding 1% could be distinguished easily by LDA and the distinguishable adulteration level for corn oil was beyond 7%.4. The discrimination of adulteration in walnut oil was performed using electronic tongue system combined with statistical methods such as PCA and LDA. Walnut oils adulterated with different concentrations(0~50%) of rapeseed oil in steps of 10% were distinguished. Results showed that the adulteration of walnut oil could be detected effectively, and the PCA was better than LDA for electronic tongue measurement. Walnut oils adulterated with rapeseed oil at levels of adulteration exceeding 10% could be distinguished by electronic tongue combined with LDA but not PCA.