A unifying approach to lipid oxidation in emulsions: Modelling and experimental validation.

Lipid oxidation is a longstanding topic within the field of food technology, and is strongly related to loss of product quality and consumer acceptance. Both for bulk oils and emulsions, the chemical phenomena involved in lipid oxidation have been extensively researched, and various reaction pathways have been identified. They are different in bulk oil compared to oil-in-water (O/W) emulsions in which the oil-water interface plays a prominent role. Most probably because of the complexity of the reaction scheme in combination with mass transfer effects, there is no model that describes lipid oxidation in emulsions in a unified fashion, and that is the aim that we have set ourselves to achieve. We use lipid oxidation data previously obtained in O/W emulsions made with 5 different emulsifiers (2 surfactants, and 3 proteins), in well-mixed systems where the oxygen-to-oxidizable lipid ratio is strictly controlled. We use data pertaining to headspace oxygen concentration, and to primary and secondary lipid oxidation products to develop a model based on reaction kinetics, including not only the classical reaction scheme (starting from an unsaturated lipid, LH) but also radical initiation from hydroperoxides, which is thought to be an effect that is overlooked in the classical description of the initiation step. We were able to describe the course of the reactions in these emulsions using the same reaction rate constants for all emulsions, with the exception of the two related to radical-based initiation. In Tween 20- and Tween 80-stabilized emulsions, initiation stems most probably solely from decomposition of hydroperoxides; this implies that lipid oxidation in these emulsions is co-determined by the initial ("pre-existing") hydroperoxide concentration. In protein-stabilized emulsions, on the other hand, lipid radical initiation is probably linked to reactions involving proteins (co-oxidation reactions), whereas initiation through decomposition of hydroperoxides seems less important, if at all. From this, we can conclude that the difference between both types of emulsions with regard to lipid oxidation mechanisms is related to differences in radical initiation. The developed model can serve as a unified basis for understanding lipid oxidation in emulsions, through which additional effects beyond the bare reaction kinetics, such as mass transfer effects, can be identified and used to e.g., quantify antioxidant effects, which is part of follow-up research.