SO₂ consumption in white wine oxidation: approaches to low input vinifications based on rapid electrochemical analyses and predictive enology

Abstract

Oxidative stability is a critical factor in maintaining wine quality during its shelf-life. SO₂ is commonly added to wine due to its strong antioxidant activity, although there is a general push to reduce SO₂ use in vinification. Reducing the reliance on SO₂ while maintaining oxidative stability is a pressing challenge for winemakers, emphasizing the need for predictive tools to optimize wine oxidation management.

In this study, the relationship between O₂ and SO₂ consumption of 71 Lugana white wines was studied Samples underwent controlled oxidative stress (oxygen consumed ~ 5ppm) to monitor oxygen and SO₂ consumption, while control anoxic samples were stored in the same temperature controlled room at 20°C. Samples were characterized for SO2 content before and after oxygen consumption, ammonia, primary amino nitrogen (PAN), polyphenols, ascorbic acid, and catechins. Cyclic voltammetry was employed to obtain information on the redox-active compounds present in the wines.

Oxygen consumption rate followed first-order kinetics, with half-lives ranging from 2.1 to 18.1 days.SO₂ consumption ranged from 1.4 to 18 mg/L in the oxygenated samples and from 0.6 mg/L to 14.9 mg/L in the anoxic samples. Final free SO2 concentrations showed a strong correlation with their initial values in both oxygenated and anoxic samples. On average, oxygenated samples consumed 4.6 mg/L more free SO2 compared to anoxic samples, with a pseudo-stoichiometric coefficient of 0.92 mg of free SO₂ consumed per mg of O₂. Notably, variability among individual samples was substantial, with ratios ranging from 0.62 to 1.95 mg free SO₂/mg O2.

The amount of free SO2 consumed in oxygenated samples was significantly inversely correlated with the half-life, suggesting that, under equal oxygen availability, wines with faster oxygen consumption tend to consume less free SO₂. Electrochemical profiles also revealed significant variability among the redox-active compounds of the different wines, particularly in the voltametric regions located around 420mV 820mV.

In consideration of the large variability in SO2 consumption levels of the different wines and the multivariate nature of SO2 consumption chemistry, different modelling approaches were explored to identify opportunities for the development of predictive tools for SO2 stability.

The results of these investigations will be presented, with a focus on a model that predicts the final concentration of SO2 in wine. This model, developed using data from voltammograms and other analytical parameters readily accessible to winemakers, demonstrated good performance on the training set (RMSE = 2.94 mg/L) and was confirmed on the test set (RMSE = 2.55 mg/L).

By combining rapid electrochemical analysis and predictive modelling, more rational use of SO2 appears possible, contributing to more efficient and sustainable wine management practices.