Enology – Fermentation Management Session

James Nelson

A Current Evaluation of Wine Fermentation Models

James Nelson*
*University of California, Davis, 595 Hilgard Ln, Davis, CA, 95616 (jjnel@ucdavis.edu)

The field of wine fermentation modeling can be traced back to the late 1970’s, when alcoholic fermentation models were first extended to address initial sugar concentrations of 200 to 240 g/L, ethanol concentrations of 100 to 110 g/L, a mixture of glucose and fructose and the decline in cell viability. Today, wine fermentation models, in combination with sensors, are being used to provide early diagnosis and prediction of abnormal fermentations at commercial wineries. In this work, multiple wine fermentation models were compared for their mathematical descriptions of cell growth, substrate consumption, product formation, temperature variation, cell death and the calculation of density from solution composition and solute properties. The models were also evaluated on a red and white commercial fermentation data set across a wide range of winemaking conditions. Only one model appeared suitable for application to commercial fermentations. The limitations of current models and suggestions for areas of advancement in the field are presented.

Funding Support: T.J. Rodgers University Fellowship in Electrical and Computer Engineering; Rossi Endowment in Viticulture & Enology

Federico Casassa | Isabelle LoMonaco | Marcel Velasco | Dimos Papageorgas

Effect of Cap Management Frequency on the Phenolic, Chromatic, and Sensory Composition of Cabernet Sauvignon Wines

Federico Casassa,* Isabelle LoMonaco, Marcel Velasco, and Dimos Papageorgas
*Wine and Viticulture Department, Cal Poly San Luis Obispo, 1 Grand Avenue,  San Luis Obispo, CA, 93407 (lcasassa@calpoly.edu)

Cabernet Sauvignon (clone 7) from the Paso Robles AVA was processed with a contrasting array of punch-down frequencies, ranging from zero to four punch-downs/day, over two vintages (2019 and 2020). In one vintage, fruit was harvested at two contrasting maturity levels. The only variable was the frequency of punch-down, with temperature and other variables kept constant within treatments. Wines were followed for up to three years of bottle aging for basic and phenolic chemistry, and the late-harvest wines of 2020 were subjected to sensory analysis. There were no effects of punch-down frequency on alcohol, pH, titratable acidity, or lactic acid. This is likely because all the wines, irrespective of punch-down frequency, were submitted to very similar temperature regimes. There was a rather unremarkable effect of punch-down frequency on the basic phenolic composition of the wines, including anthocyanins, tannins, and polymeric pigments. These results counter empirical winemaking observations, whereby more frequent physical mixing should lead to higher phenolic extraction. It is possible that more frequent mixing may lead to phenolic losses due to chemical and/or enzymatic oxidations. Such losses may be minimized under conditions of no mixing, akin to wines made with no cap management. In fermentors of relatively small volume (60 L) and at high fermentation temperature, convection movements may allow for sufficient phenolic extraction. This may extend to fermenters up to 1.5 tons; however, reduction aromas may concomitantly appear as well under such conditions. Sensory results indicated that all four wines, irrespective of punch-down frequency, were perceived as drying. While other subqualities of astringency changed little as a function of the treatments, the perception of bitterness increased concomitantly with the frequency of punch-downs, with 3 punch-down wines showing the greatest bitterness perception.

Funding Support: Agricultural Research Institute (ARI)

Aude Watrelot | David Carter

Effects of Including Different Forms of Grape Stems on Chemical Parameters of Red Wines

Aude Watrelot* and David Carter
*Iowa State University, department of Food Science and Human Nutrition, 536 Farm House Lane, Ames, IA, 50011 (watrelot@iastate.edu)

There is increased interest in using grape stems in red winemaking, but their actual impact on wine quality is poorly understood. Grape stems are either not harvested by mechanical harvest, or are discarded after manual harvest and the destemming process. However, grapes stems are rich in phenolic compounds, which makes them a potential good source of tannins to improve red wine quality. In 2023, two cold-hardy grape cultivars, Marquette and Frontenac, were crushed, destemmed, and processed with different stem additions. Stems were collected and either kept fresh (FS), oven-dried (DS), or oven-dried and ground (DGS) prior to being added to the musts. Those conditions were compared to a control treatment to which no stem was added. Red wines were made in triplicate for each condition following a standard red winemaking procedure, using yeast strain ICV D254 and co-inoculated with VP41 malolactic bacteria. Basic chemistry, color parameters, and phenolic compounds of the resulting red wines were analyzed by UV-vis spectrophotometer and high-performance liqud chromatography diode array detection at bottling. The pH was higher in Frontenac than in Marquette wines, but pH and titratable acidity were not affected by the treatments. The color intensity was greatest in control Frontenac wines and least in FS Frontenac wines. This could be explained by the ability of anthocyanins to bind with cell wall material of fresh stems. However, no effect of stem inclusion on color intensity was observed in Marquette wines, which may be due to their different anthocyanins profile. The FS and DS Marquette and Frontenac wines contained significantly more tannins than the control wines. Surprisingly, the concentrations of phenolics and tannins were not improved by DGS, suggesting a lower impact on the surface area-to-juice ratio on tannin extraction.

Funding Support: This project was supported by the U.S Department of Agriculture’s (USDA) Agricultural Marketing Service through grant 23SCBPIA1187.

Yiliang Cheng | Aude Watrelot

Isothermal Titration Calorimetry to Describe Polyphenol-Macromolecule Interactions

Yiliang Cheng and Aude Watrelot*
*Iowa State University, 2567 Food Science Building, 
536 Farm House  Lane, Ames, IA, 50011 (watrelot@iastate.edu)

This study builds on our previous findings that grape flesh-derived macromolecules, mainly polysaccharides and proteins, significantly affect retention of tannins and anthocyanins during winemaking. We specifically examined how anthocyanins in mono- and diglucoside forms influence these interactions. Employing isothermal titration calorimetry (ITC), we evaluated the binding affinity and thermodynamic properties between poly(L-proline) (PLP, at 0.7 mM) and pectins (at 30 mM  galacturonic acid eq.) with commercial grape tannin (BioTan) and procyanidin C1 (both at 3 mM epicatechin eq.). To assess whether anthocyanins alter the binding interaction of tannins with these macromolecules, we conducted a pre-incubation of the macromolecules with malvidin-3-glucoside (M3G) and malvidin-3,5-diglucoside (M35DG), each at a concentration of 3 mM prior to titrating with tannins.
Tannins showed a strong affinity for PLP, while no interaction was observed with pectin, even at greater concentrations. Tannins did not interact with pure anthocyanin solutions under the experimental conditions. Interestingly, pre-incubation of PLP with M35DG reduced the binding affinity of  BioTan to PLP, while M3G did not significantly alter this affinity. The nature of the tannin-PLP interaction, whether enthalpy- or entropy-driven, varied based on the type of tannin.

The different effects of malvidin mono- and diglucoside on the strength of interaction between tannins and proteins help clarify the effects of different grape varieties on the sensory attributes and color stability of red wine. This could be used to manage the texture and mouthfeel of wine and to adjust winemaking processes such as fining, filtration, or blending. Because we observed haze formation during the ITC experiments, future research will focus on investigating the aggregation patterns between polyphenols and macromolecules using dynamic light scattering.

Funding Support: No funding source.