Model for prediction of the carbonyl value of frying oil from the initial composition

To construct a model for predicting the carbonyl compound content of oil during frying, changes in the unsaturated fatty acid, tocopherol, and carbonyl value (CV) of 10 commercial edible oils during frying at 180 °C were investigated. The initial contents of total unsaturated fatty acid, total tocopherol (TToc), and carbonyl value (CV0) in 10 oils were distributed in the range of 83.49%–95.28%, 16.40–236.05 mg/100 g, and 2.36–6.30 μmol/g, respectively. During frying, the CV of 10 oils increased linearly with heating time at different rates. The effects of intrinsic factors (unsaturated fatty acid and tocopherol) on the CV increase were revealed by using multiple regression analysis. The results indicated that polyunsaturated fatty acid (PUFA) and γ-tocopherol had the greatest effects on the CV increase, respectively. Based on these results, the ratio of the initial content of PUFA to that of TToc was selected as a prediction index and a prediction model, CVt=[4.37(PUFATToc)2−5.64(PUFATToc)+3.36]t+CV0, was established. Using this model, the CV of edible oil during frying at 180 °C can be predicted from the initial composition of the oil, or the frying life of edible oil can be inferred according to the maximum allowable CV (50 μmol/g) stipulated by regulations.