Desorption of oxygen from wine and model wine solutions in a bubble column

Removal of dissolved oxygen through desorption is commonly performed in winemaking, however, industry reports that this process varies for different wines, adding complexity to standard operating procedures. In order to better understand the factors inducing these differences, and therefore allow for better operation industrially, this article examines the mass transfer of six wines and three model wine solutions by evaluating the oxygen desorption volumetric mass transfer coefficient (k(L)a), the Sauter mean bubble diameter (D-32), the interfacial area, and the oxygen mass transfer coefficient (k(L)). The k(L) for different wines was found to vary significantly (between 0.015 and 0.05 mm/s), while there was no significant variation in the D-32 and the interfacial area. Results after adding proteins (yeast extract) to model wine, or reducing protein from real wine (using bentonite) suggest that proteins reduce k(L), and that their presence and concentration contribute to some of the variation in desorption of oxygen. (c) 2022 Elsevier Ltd. All rights reserved.