LAUSR

BACKGROUND Cow ghee is one of the expensive edible fats in dairy sector. Ghee is often adulterated with low-priced edible oils like soybean oil, due to its high market demand. The existing adulteration detection methods are time-consuming, requires sample preparation and expertise in these fields. The possibility of detecting soybean oil adulteration (from 10 to 100%) in pure cow ghee was investigated in this research. The fingerprint information of volatile compounds was collected using a flash gas chromatography electronic nose (FGCEN) instrument. The classification results were studied using pattern recognition chemometric models such as principal component analysis (PCA), soft independent modelling of class analogy (SIMCA) and discriminant function analysis (DFA). RESULTS The most powerful fingerprint odor of all the samples was acetaldehyde (Z)-4-heptenal, 2-propanol, ethyl propanoate and pentan-2-one as identified from FGCEN analysis. The odor analysis investigation was accomplished with an average analysis time of 90 s. A clear differentiation of all the samples with an excellent classification accuracy of more than 99% was achieved with PCA and DFA chemometric methods. However, results of SIMCA model showed that SIMCA could be used to detect higher concentration levels (30 to 100%) of ghee adulteration only. The validation study shows good agreement between FGCEN and GC-MS methods. CONCLUSION The demonstrated methodology coupled with PCA and DFA methods for adulteration detection in ghee using FGCEN apparatus has been an efficient and convenient technique. The present research explored the capability of the FGCEN instrument to tackle the adulteration problems in ghee. This article is protected by copyright. All rights reserved.