Modelling the oxygen dissolution rate during oenological fermentation

Discrete oxygen additions are commonplace during winemaking to avoid fermentation problems and improve wine quality. Since oxygen dissolution under oenological conditions is not fully understood, these additions are carried out empirically, resulting in unpredictable concentrations and heterogeneous oxygen distributions within the fermentation tank. In this research, we develop a model to predict the evolution of the oxygen dissolution rate in the must during alcoholic fermentations. The model includes two partial differential equations that describe oxygen spatial distribution during the fermentation in both liquid and gaseous phases. These equations incorporate the effects of recirculation and mixing. Also, a set of ordinary differential equations describing the kinetics of relevant fermentation variables is included. The model is validated by fermenting a Sauvignon blanc must in a bubble column with five micro-bubblers for oxygen injection. Additions are made every ten points of decrease in the fermenting musts' density. Model simulations and experimental results confirm that CO2 is responsible for the observed reduction in the oxygen dissolution rate during the exponential growth phase, and that viable yeast cells significantly reduce this rate compared with an abiotic bubbling CO2 system. The model developed will help to design more effective oxygen addition systems through micro-bubbling for wine fermentation tanks.