Simultaneous determination of ethanol and methanol in wines using FTIR and PLS regression

Abstract

Wine is a complex hydroalcoholic solution, with ethanol levels serving as a critical quality parameter. Produced by yeast during the fermentation of hexose sugars, namely glucose and fructose, ethanol not only contributes to the sensory attributes of wine but also plays a vital role in influencing interactions with other components, such as tannins and volatile compounds [1]. Methanol, a naturally occurring component in wine, is primarily produced during alcoholic fermentation through the enzymatic breakdown of pectic polysaccharides in grape skins [2]. While methanol levels in wine are typically low and pose minimal toxicological risk, the International Organisation of Vine and Wine (OIV) has established maximum acceptable limits of 400 mg/L for red wines and 250 mg/L for white and rosé wines. Traditional methods for measuring ethanol and methanol include electronic densimetry and gas chromatography with flame ionization detection (GC-FID). While accurate, these methods are time- and labor-intensive, prompting the adoption of more efficient, non-destructive analytical techniques. FTIR is recognized for its rapid, automated, and non-invasive capabilities, making it a valuable tool for modern wineries and laboratories. This study explores the potential of FTIR spectroscopy, enhanced by chemometric approaches, to accurately quantify simultaneously methanol and ethanol in wine samples. To quantify ethanol and methanol in wine samples standard solutions of ethanol in water, methanol in water, and various combinations of both, were used. Duplicate FTIR measurements took place in transmission mode with the use of a Perkin Elmer LQATM 300 FT-IR wine analyzer. A mathematical model was developed using Partial Least Squares (PLS) regression. For external validation purposes, the predicted ethanol and methanol concentrations of 20 wine samples were compared to electronic densimetry for ethanol determination and GC-FID for methanol determination. The developed FTIR model demonstrated the feasibility of simultaneous ethanol and methanol quantification in wine samples, offering the potential for quality control in wine production. Future improvements should include expanding the dataset with samples from diverse wine types, grapevine cultivars, and winemaking techniques across various wineries. To enhance methanol prediction accuracy, samples with higher and varied methanol levels should also be incorporated [3].

References

[1] Fontoin, H.; Saucier, C.; Teissédre, P. L.; Glories, Y. (2008). Effect of pH, ethanol, and acidity on astringency and bitterness of grape seed tannin oligomers in model wine solution. Food Qual., 19, 286–291.

[2] Hodson, G.; Wilkes, E.; Azevedo, S.; Battaglene, T. (2017). Methanol in wine. Bio Web Conf., 9, 02028.

[3] Thanasi, V.; Caldeira, I.; Santos, L.; Ricardo-da-Silva, J. M.; Catarino, S. (2024). Simultaneous Determination of Ethanol and Methanol in Wines Using FTIR and PLS Regression. Foods, 13, 2975.