Advanced evaluation method for oxidative deterioration of edible oils using multivariate analysis incorporating weak chemiluminescence data

 Evaluating oxidative deterioration in edible oils is critical for ensuring their quality and safety during storage. Traditional methods, such as the Rancimat test, are reliable but time-intensive. This study introduces a novel evaluation method that combines ingredient data from publicly available food databases with chemiluminescence (CL) measurements to accurately predict induction periods (IP), even in oxidatively degraded oils.
 Four commercially available edible oils were subjected to accelerated oxidative degradation at 60°C under atmospheric conditions to simulate various stages of deterioration. The ingredient data of each oil were obtained from Standard Tables of Food Composition in Japan as public database. As an indicator of oxidative degradation, the induction period (IP) measured by the Rancimat method was used. Chemiluminescence measurements revealed two distinct peaks in oxidatively degraded oils: the first reflecting pre-existing oxidation products and the second representing oxidation occurring during the measurement. To model these behaviors, a sequential- parallel reaction mechanism was employed, allowing the derivation of kinetic parameters (k1–k6). These parameters were integrated into multivariate regression models alongside ingredient data, such as fatty acid profiles and tocopherol content, obtained from the public database.
 Regression models combining CL-derived parameters with compositional data achieved high prediction accuracy for IP, with adjusted R² values up to 0.89. The sequential-parallel model captured the complex dynamics of oxidative reactions, enabling the prediction of IP with precision unattainable by compositional data alone. This integration of universal compositional data and sample-specific oxidative dynamics offers a reproducible and robust method for evaluating oxidative deterioration.
 This approach supports rapid and detailed assessment of oil stability, providing a practical tool for quality control and research in the food industry. By leveraging the strengths of both static and dynamic data, it promotes innovation and efficiency in the evaluation of oxidative stability, addressing the growing need for advanced analytical methods.