What Do We Know About Modeling Red Wine Fermentations?
June 14, 2019
David Block presented this research at the last Innovation+Quality (IQ) conference. The lead author, Konrad V. Miller, is a Ph.D. candidate at UC Davis and has produced a number of papers around the subject of producing a red wine reactor model for red wine fermentations. This research may end up being really important. That’s easy for me to say after a couple of related papers ended up being on the cover of the issues of AJEV and AJGWR in which they appeared.
This work constitutes a step in the progression from understanding what happened in a given fermentation to understanding what will happen in a fermentation for a given lot of grapes to being able to design a fermentation regime to get the results one wants. Put another way, if the model really matches enough real-world results, it can be used to predict the results of the next fermentation. From there the model becomes a tool for achieving a particular desired outcome.
Call it “Block’s Black Box” or “Miller’s Magnificent Model” but if successfully taken to its logical conclusion, the work could change the way we approach winemaking.
Red wine fermentations have long eluded accurate simulation because of their inhomogeneous nature. In this work, a three-dimensional, time-dependent, reactor engineering model for jacketed red wine fermentations was utilized to explore the effect of fermentor volume (500; 50,000; and 500,000 L), aspect ratio (height:diameter of 1:1 and 3:1), temperature set point (15, 25, and 35°C), and initial yeast assimilable nitrogen (YAN) concentration (100, 225, and 350 mg/L) on fermentation dynamics. The model simulated N-limited, ethanol-inhibited, and temperature-dependent Monod fermentation kinetics; mass transfer of sugar, yeast, N, and ethanol; evaporative, convective, and conductive heat transfer; and the motion of the bulk fluid beneath the cap. Fermentor surface area-to-volume ratio, temperature set point, and initial YAN were all found to significantly affect fermentation performance in simulated fermentations. Heat transfer by conduction into the cap was nondimensionalized and analyzed as well. Finally, the formation of temperature gradients in the cap between cap-management cycles was visualized from simulations using the model.