Ng, Jing Sheng
(2021)
Adulteration Detection Of Edible Palm Oil Using Infrared Untargeted Screening And Targeted Fatty Acid Diagnostic Ratio Coupled With Chemometrics.
Masters thesis, Universiti Sains Malaysia..
Abstract
In Malaysia, repeated use of cooking oil in deep frying, especially refined, bleached and deodorised palm oil (RBDPO), is common. The adulteration of RBDPO with used oil has been studied with frying simulation using different kinds of meats, namely beef, fish, mutton, and pork. The oil underwent frying for 25 cycles, at 180 ºC, 10 minutes per cycle. Then, used oil was sampled at the 5th, 10th, 15th, 20th, and 25th frying cycles. Admixture of pure oil with used oil was done at 1%, 5%, and 10% (mass/mass) concentrations. All the pure, admixed, and used oil were evaluated with untargeted screening analysis, near-infrared spectroscopy (NIRs), mid-infrared spectroscopy (MIRs), and targeted analysis, gas chromatography coupled with flame ionisation detector (GC-FID). NIR and MIR spectra and fatty acid diagnostic ratio (DR) composed of fatty acid abundance were obtained. The variation between pure and used oil of the same meat type was visualized with principal component analysis (PCA). Besides, the 5th frying cycle's used oil showed a significant difference to the pure oil in the PCA score plot. The orthogonal partial least square discriminant analysis (OPLS-DA) of the NIR models had successfully identified used and admixed oil from the pure oil. Next, in the MIR model, there was one misclassification in Beef. Lastly, in fatty acid diagnostic ratio models, Beef and Fish had one misclassification of pure oil, respectively.
Actions (login required)
 |
View Item |
