Cellulose Polymers Shown to Reduce “Brett” Character in Wines

Ethylphenols are compounds that are known to be present in wines and other alcoholic beverages, and often produce “off” flavors such as pharmaceutical or animal-like (e.g. “horse sweat” or leather tones), and are considered to be responsible for “Brett character” in wine.  Ethylphenols are formed in wine during the fermentation process, and involve the transformation of cinnamic, courmaric, and ferulic acids, by Brettanomyces/Dekkera spp yeasts, though sometimes Lactobacillus spp and Pichia guilliermondii facilitate the transformation as well.

There are winemaking techniques that can be employed to reduce this “Brett character” in wine; however, those involve increasing SO2, in addition to other less desirable methods, since the modern preference in wine is to decrease the usage of SO2 in wine and to increase pH.  As a result of these practices, “Brett character” in wine has steadily increased, causing economic damage to many corners of the wine industry.


Several other methods have been examined for reducing ethylphenol content in wine, including reverse osmosis and adsorption, sorption on yeast lees and cell walls, and also molecularly imprinted polymers.  One technique more recently examined for reducing ethylphenol content in wine, and the one in which the study presented to you examines, is the use of carbohydrate cellulose, the most abundant polymer in nature.  Structurally, glucose units are attached together to create a long chain, which allows it the ability to form hydrogen bonds with other chains, creating a cellulose fiber which give it mechanical strength and chemical stability, while still allowing it to hydrogen bond with other molecules.

The brief study presented today aimed to examine the ability of different esterified cellulose polymers to reduce levels of 4-ethylphenol and 4-ethylguaiacol in “Brett-infected” wines.


The four polymers tested for effectiveness in reducing 4-ethylphenol and 4-ethylguaiacol in wine were: 1) cellulose acetate; 2) cellulose acetate propionate; 3) cellulose acetate butyrate; and 4) cellulose propionate.

For all wine samples, the following were measured: alcohol, reducing sugars, total acidity, total dry extract, volatile acidity, color intensity, and hue.  Proanthocyanidins, catechins, proanthocyanidins that are reactive to vanillin, total anthocyanins, and ethylphenols were also measured for each sample.  The polymers’ ability to adsorb ethylphenols was also analyzed.

Experiments were conducted using a sound and volatile phenol-free red wine that was fortified with the standard solutions in order to obtain samples with specific levels of 4-ethylphenol and 4-ethylguaiacol: 0.5, 1.0, and 2.0mg/L (separately).

After the most effective polymer for reducing 4-ethylphenol and 4-ethylguaiacol was found via the aforementioned experimentation, efficiency was further measured in 10 red wines with alcohol levels varying from 12% to 13.8%, residual sugars less than 2.9g/L, total acidity between 5.8 and 4.6g/L, and maximum volatile acidity of 1.49g/L.

Sensory analysis was performed by 12-16 expert wine judges, and was presented in random order.  Overall acceptability and preferences were recorded.


Effectiveness of Individual Polymers

  •       All polymers showed similar abilities to reduce levels of 4-ethylphenol and 4-ethylguaiacol in wine.

o   Most effective: cellulose acetate propionate and cellulose propionate, with average reduction values at 38% and 37%, respectively.

o   Middle performer:  cellulose acetate butyrate, with average reduction values of 32% and 29%.

o   Least effective: cellulose acetate, with average reduction values of 24% for both phenols.

§  Due to these effectiveness results, cellulose acetate propionate was chosen for the remainder of tests in this study.


  •        Level of reduction increased with increasing dose of the polymer, though slowed down around doses of 4g/L to 6g/L.

o   Highest levels of reduction were found at dose of 20g/L, with around 70% reduction in 4-ethylphenol and 4-ethylguaiacol.

o   Based on the results, reduction ability is independent of the initial concentration of the phenol (doesn’t matter how much is present, the polymer will reduce the levels of it).

Contact Time

  •       Reduction over 60 minutes was similar for both 4-ethylphenol and 4-ethylguaiacol, hitting a plateau at about 15 minutes into contact with the polymer.
  •       The initial level of phenol concentration did not affect the percentage reduction by the polymer.

Side Effects

  •       After treatment with the polymer, all wines showed significant decreases in the levels of 4-ethylphenol and 4-ethylguaiacol compared to control wines that were not treated with the polymer.

o   Average reductions of 4-ethylphenol and 4-ethylguaiacol were 31% and 32% (min 25%-27% and max 41%-40%), respectively.

  •       In regards to phenolic composition and color of wines, there were significant reductions in color intensity and total anthocyanins.

o   According to the authors, these reductions are technologically irrelevant, since average reductions were relatively low at 4.9% and 6.6%, respectively, and would not be noticeable to consumers.

  •       Sensory analysis indicated that all wines treated with the polymer were significantly different and more preferable to wines not treated.

Regeneration and Re-Use of Polymer

  •       In economic terms, it’s always more beneficial to have a product that can be used more than once while still retaining its abilities to function.

o   Washing the polymers in a 100mL solution of 99% ethanol or with an aqueous solution (NaOH, pH 12) and stirring constantly at 60rpm for 10 minutes proved to be effective.

o   Washing the polymers allowed the fibers to maintain their ability to reduce levels of 4-ethylphenol and 4-ethylguaiacol, even after 20 consecutive washings, without showing any damage to the polymer.


The results of this brief study are promising in that it appears that using cellulose acetate propionate polymers (or any similar polymer with a propyl group) is effective in decreasing the levels of 4-ethylphenol and 4-ethylguaiacol in wine, which are known to impart negative sensory effects (“Brett character”) on finished wine.  This polymer was able to effectively reduce levels of these phenols in wine, while also allowing the possibility of washing and reusing the polymer, in order to be not only technologically efficient, but also economically efficient.

This study provides a launching pad for further study in engineering similar polymer treatments for reducing “Brett character” in wine, or other situations where 4-ethylphenol and 4-ethylguaiacol are elevated (i.e. in smoke tainted wines).

I’d love to hear what you all think of this brief study.  Please feel free to level your comments below (no html tags that are advertising in nature allowed)!

Source: Larcher, R., Puecher, C., Rohregger, S., Malacarne, M., and Nicolini, G. 2012. 4-Ethylphenol and 4-ethylguaiacol depletion in wine using esterified cellulose. Food Chemistry 132: 2126-2130.

DOI: 10.1016/j.foodchem.2011.12.012

I am not a health professional, nor do I pretend to be. Please consult your doctor before altering your alcohol consumption habits. Do not consume alcohol if you are under the age of 21. Do not drink and drive. Enjoy responsibly!

6 comments for “Cellulose Polymers Shown to Reduce “Brett” Character in Wines

  1. Greg
    September 15, 2012 at 9:41 am

    Bravo! I look forward to this research leading to a commercially available product in the near future!

    • Becca
      September 15, 2012 at 9:42 am

      I bet we’ll see it sooner than later! Thanks for reading and commenting!

  2. Wineknurd
    September 15, 2012 at 9:41 am

    Interesting in that I am doing some filter studies in my current job duties and the cellulose acetate filters I have used showed increase levels of nitrates and nitrites in my filtrate. This isn’t unexpected because nitric acid is used to regenerate the cellulose for acetate, but there needs to be an investigation into what these filters might add to the wines, not just what they remove. Good post Becca! I hope things are going well for you.

    • Becca
      September 15, 2012 at 9:43 am

      I love hearing from those that are actually doing the work (or at least similar work) to the studies I present! You’ll have to keep us updated on any developments in the area 🙂

      I’m doing well! Busy! Started a new job, so my posts haven’t been as frequent! Hoping that will change once I settle in!

  3. DavidZ
    September 15, 2012 at 9:42 am

    Unfortunately, reductions of phenol and anthocyanen of 4.9% and 6.6% are big. Close to the difference between cab and rose. Similar changes were found in this study of lyophilised yeast to absorb brett: http://www.infowine.com/default.asp?scheda=11240. But authors of this study know wine. I think the authors of your study were food chemists, not winemakers and just though oh those are not bad numbers, without asking what the difference really means.

    Combined with wineknurd point about contamination.

    • Becca
      September 15, 2012 at 9:43 am

      Great point, David!

      While sometimes it’s beneficial to have an “outside academic” examine topics slightly outside their field of expertise, it can at times result in “missed connections” or misinterpreted results. Thanks for the link to the other study as well–that site is an excellent resource!

Comments are closed.