Methodology for measuring peroxide value in vegetable oils

NIR spectroscopy was examined as a possible alternative to the conventional titration method for determining oxidation levels in corn oils (CO). Additionally, the potential of using a single calibration to measure oxidation in oils from multiple sources was evaluated. Calibration models were developed to measure peroxide values in both soybean oil (SBO) and CO's, using partial least squares (PLS) regression and forward stepwise multiple linear (FSML) regression, from NIR transmission spectra. PV can be successfully measured in both SBO and CO's using a single calibration. The most successful calibration was based on PLS regression of first derivative spectra. When this calibration was applied to a validation sample set containing equal numbers of CO and SBO's, with PV ranging from 0–20 meq/kg, a correlation coefficient (r) of 0.99 between titration and NIR values was obtained, with a standard error of prediction (SEP) equal to 0.72 meq/kg. For both types of oil, changes occurred in the 2068 nm region of the NIR spectra as oxidation levels increased. These changes appear to be associated with the formation of hydroperoxides during oxidation of the oils. This study also compared four peroxide analytical methods using SBO. The methods included the official AOCS titration, the newly developed NIR method, the PeroxySafe™ kit, and a ferrous xylenol orange (FOX) method. Statistical analysis of the data showed that the r and standard deviation of differences (SDD) between the titration and NIR methods were r = 0.991, SDD = 0.72 meq/kg; between titration and the PeroxySafe™ kit were r = 0.993, SDD = 0.56 meq/kg, and between the titration and FOX method were r = 0.975, SDD = 2.3 meq/kg. The high correlations between the titration, NIR and PeroxySafe™ kit data indicated that these methods were equivalent. Solid phase microextraction/headspace-gas chromatography (SPME/HS-GC) was evaluated as a tool for determining the rate of oxidation in SBO by measuring the production of hexanal. Different analytical conditions were evaluated. Samples of the headspace were taken from 20 mL vials incubated 30 min at 50°C using a carboxen-polydimethysiloxane (Car-PDMS) fiber and 5 min adsorption time. Volatiles were then desorbed into a GC using a 2 min thermal desorption. The study showed that SPME/HS-GC was a simple method for hexanal determination, and satisfactory repeatability was obtained by the method (n = 11, CV% = 6.31).