Dissolved oxygen, due to atmospheric oxygen transfer through food, lowers the overall nutritional, chemical, and physical qualities of food during storage and marketing, causing food product less acceptable or unacceptable to consumers. Oxygen can react with all food components: lipids, proteins, sugars, and vitamins. To imagine new strategies of protection of food, it is necessary to evaluate oxygen transfer and to couple this transfer to chemical reactions. In this work, ascorbic acid oxidation under aerobic conditions has been studied. The objective is to model ascorbic acid degradation regarding both ascorbic acid and oxygen influence, the last one depending on diffusion rate into food and oxygen consumption during reaction. Ascorbic acid degradation was first studied in water under stirring in which oxygen diffusion could be neglected. Experiments were carried out during six days, under six different oxygen contents varying between 0 (simulating packaging in anaerobic conditions) and 21% (simulating packaging in atmospheric conditions) at 8, 14, 20, 27 and 33 degrees C. Thanks to these experiments, kinetic parameters as reaction orders for ascorbic acid and oxygen, kinetics constant and activation energy have been identified. Oxygen sorption kinetics were measured thanks to luminescence oxygen sensors, placed into of a thin layer of agar gels 1% (w/w) exposed to eight successive increasing oxygen content in the atmosphere (from 0% to 21%). The probes measured oxygen content into material as a function of time. Oxygen diffusivity in agar gel was then identified by minimising the sum of square error between experimental results and those predicted by a mathematical model. Finally, using mass balance equation on oxygen and ascorbic acid contents, oxidation reaction and oxygen diffusion in model food systems were modelled. Validation of the model was carried out on agar gels 1% (w/w) enriched in ascorbic acid (1000 mg/l) in which the ascorbic acid content profile was determined by slicing each cylindrical gels in thin layers of 0.5 mm thick and by determining the ascorbic acid content in each layer. Knowing initial ascorbic acid content, oxygen content in atmosphere and temperature, the model can predict evolution of ascorbic acid content in the product as a function of time. This model could be used as a tool to imagine new strategies for protection of interest compounds in food.