Alternatives to Casein as Non-Allergenic Fining Agents for White Wine

Fining agents are extremely common in the wine industry, and are used to clarify wine, control browning caused by oxidation, and improve overall stability of the finished wine.  Animal proteins have often been used, however, with recent knowledge that contaminated animal proteins can cause severe health problems in humans; fining agents of vegetable or other origin have been considered.  Aside from possible contamination, allergic reaction risk in humans is another reason why the wine industry is considering other alternatives for fining agents of wine.  One of the most common fining agents of animal origin is casein, which is derived from milk proteins.  Potassium caseinate, which is derived from casein by dissolving it in aqueous potassium hydroxide, is more often used in wine production than casein itself, due to its higher solubility in wine.

These fining agents of animal protein origin are used quite frequently in the wine industry.  For example, in France (as of 2007), 2060 million liters of wine per year were fined with casein, which equates to approximately 41% of French wines.  Similarly, up to 20% of German wines are fined with casein (again, as of 2007).

One potential alternative to casein as a fining agent in wine is polyvinylpolypyrrolidone (PVPP), a synthetic water-soluble polymer, which is known to effectively remove phenols from wine via adsorption.  It is suggested that PVPP could be used either by itself, or combined with lower levels of casein to reduce the oxidation potential of wines.  PVPP works to remove phenols from wine by hydrogen bonding between the carbonyl group on the PVPP molecule and the hydroxyl groups on the phenolic compounds.  Casein fines wine by directly inhibiting the oxidation of polyphenols into quinones by formation of casein-quinone conjugates.  Since PVPP has been shown to effectively remove phenols from wine, it could be a potentially non-allergenic alternative to casein for wine fining.  On the down side, PVPP tends to be much more expensive, though it is at least more effective at lower concentrations.

Another possible alternative to casein as a wine fining agent is commercial pea protein, a non genetically-modified protein that is currently used in vegan wine production.  It is a non-allergenic protein, and studies have found that it does not leave significant residues behind in either red or white wines. 

Studies investigating alternatives to casein as a wine fining agent are not common.  The goal of the study presented today was to evaluate the potential of a selection of allergen-free fining agents (pea protein and PVPP) as alternatives to casein (specifically, potassium caseinate). 


Wine used for this study was a white blend (2007 vintage) of 60% Trajadura and 40% Loureiro from the Vinho Verde region (north of Portugal).  The chemical profile of the wine was as following:  alcohol content was 90.7g/L, titratable acidity was 7.0g/L as tartaric acid, volatile acidity was 0.18g/L as acetic acid, the pH was 3.37, free sulfur dioxide was 44mg/L, and total sulfur dioxide was 128mg/L.

Experiment 1:  Individual fining agents and a commercial formulation of fining agents was added to the experimental wine at laboratory and semi-industrial scales.  Individual fining agents included potassium caseinate at 0.4g/L, pea protein at 0.4g/L, and PVPP at 0.25g/L.  The commercial formulation of fining agents was composed of pea protein and PVPP at a ratio of 0.25g/L pea protein: PVPP.  This experiment was performed to compare the effects of fining agent alternatives to potassium caseinate.

Experiment 2:  This experiment was performed at a laboratory scale using four commercial formulations of fining agents:  1) bentonite and potassium caseinate (0.65g/L); 2) bentonite and pea protein (0.65g/L); 3) bentonite, PVPP, and potassium caseinate (0.55g/L); and 4) bentonite, PVPP, and pea protein (0.55g/L).  This experiment was performed to determine if caseinate could be replaced by pea protein in commercial formulations.

Both experiments were performed in 1000mL flasks for the laboratory scale, and 500L tanks for the semi-industrial scale.  A no fining agent wine was used as the control.  Fining agents were thoroughly mixed into the wine, and allowed to remain in contact with it for 7 days at 20oC.  After this time, samples were centrifuged before analysis.  All samples were run in duplicate.

The following chemical characteristics were measured for each wine treatment sample: phenolic content (both flavonoid and non-flavonoid), polyphenolic content, turbidity, browning potential, color, alcohol content, pH, titratable acidity, volatile acidity, free sulfur dioxide, and total sulfur dioxide.

A sensory analysis was performed by a trained panel with “extensive” wine experience.  Wines were sampled at three different sessions; two for the laboratory scale samples, and one for the semi-industrial scale samples.  Samples were presented in a randomized fashion, and were not identified to the panelist as any particular treatment.  Panelists were asked to score on a 5-point intensity scale 15 different attributes: visual (limpidity, ton and color intensity), aroma (limpidity, intensity, finesse, harmony and vegetable), and taste (flavor limpidity, flavor intensity, body, flavor harmony, persistence, mouth end and vegetable flavor).  A total score was calculated by taking the average of the visual, aroma, and taste scores.   All sessions occurred between 10am and 12noon in individual booths following standardized protocols.



  •       Most of the fining agents significantly decreased the levels of total phenols in the sample wines, with the exception of pea protein/PVPP formulation.
  •       The fining agents had a greater influence of the flavonoid phenols than any other phenol.
  •       The results from the semi-industrial scale were statistically similar to the results from the laboratory scale.
  •       Alternative fining agents had a greater effect on non-flavonoid phenols than traditional fining agents, however all agents significantly reduced all measured types of phenols.


  •       Samples including potassium caseinate (either alone or in a commercial formulation) were most effective at reducing the browning potential of the wine, which would ultimately lead to a more stable wine. 
  •       Wine color was significantly reduced by all fining agents.
  •       Clarification capacity was higher for pea protein, followed by formulations with PVPP or bentonite.
  •       PVPP was not effective in increasing clarity in either experiment.

o   This suggests that perhaps the sedimentation of PVPP powder particles may be difficult by gravity alone.

  •       After adding PVPP, lightness values were unchanged.

o   This suggests no clarifying action on the part of PVPP.

  •       Yellowness values decreased for all fining agents.

o   The greatest decrease in yellowness was for wines fined with potassium caseinate.

  •       Chroma (yellowness and redness) decreased significantly for pea protein, potassium caseinate, and formulations of pea protein and PVPP.
  •       Hue-angle values increased for pea protein, potassium caseinate, PVPP, and formulations with pea protein and PVPP.

o   This suggests that some yellow pigments may have been removed.

  •       Color variation (the geometric mean of lightness, redness, and yellowness) was greatest for potassium caseinate and pea protein, followed by formulations of pea protein with PVPP and pea protein with bentonite.

Sensory Analysis

  •       There were no sensory differences between any of the treatment sample wines.

o   These results suggest the addition of pea protein or other alternatives to casein has no sensory impact on the finished wines.


According to the results of this study, potassium caseinate, pea protein, and PVPP all decreased the concentrations of phenols in the finished wine.  It was also found that only potassium caseinate, the animal protein most often used in winemaking, was effective at reducing browning potential of the wine.  Finally, there were no sensory differences between the wines produced from all of the tested fining agents.

The authors of this study suggest that these results using pea protein as an allergy-free alternative to casein as a wine fining agent is a real possibility, though more research would most certainly need to be completed.  Pea protein was not as effective as casein in regards to reducing oxidation potential; though using pea protein in a formulation with more effective fining agents would be an acceptable alternative to animal protein-based agents alone.

I’d love to hear what you all think about this topic!  Feel free to comment below (any HTML tags will be deleted).

Source: Cosme, F., Capão, I., Filipe-Ribeiro, L., Bennett, R.N., and Mendes-Faia, A. 2012. Evaluating potential alternatives to potassium caseinate for white wine fining: Effects on physiochemical and sensory characteristics. LWT-Food Science and Technology 46: 382-387.

DOI: 10.1016/j.lwt.2011.12.016
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!

5 comments for “Alternatives to Casein as Non-Allergenic Fining Agents for White Wine

  1. David Swaddle
    March 27, 2012 at 9:59 pm

    Great article. What I'm not sure about is the researcher's conclusion. I think that they've fallen into the trap of concluding what they set out to prove, rather than what the data suggest. Oxidation is a major problem in commercial wine shipments, so suggesting that the fining agent with the worst oxidation results is the most suitable seems more than a little bit odd. Or is it just me?

  2. March 27, 2012 at 10:06 pm

    Thanks for your comments, David! I agree, their suggestions that a fining agent with poor oxidation results may be a good alternative to casein (which had the best oxidation results) can be somewhat puzzling. However, what I think they were trying to say is that even though the oxidation results weren't the great, the fact that the pea protein did not harm the sensory characteristics of the wine suggests that we shouldn't rule it out quite yet, and perhaps pea protein in combination with lower levels of casein might be a suitable alternative.

    Looking at their results, if you had to pick which would be the best option as a fining agent, it would be casein, but since the sensory results were good for the alternatives, I think the authors didn't want to toss out the idea all together.

  3. Parda Mancini
    April 12, 2012 at 3:55 am

    I would have liked to have seen a control for sensory evaluation. Compare the taste of the wines against a wine that has "not" been fined. Then come to a conclusion.

  4. April 17, 2012 at 4:12 pm

    Agreed! One of the first things I learned as a researcher was to always have the appropriate controls!

  5. February 14, 2013 at 1:37 am

    Thank you for a wonderful informative article. My wife and I just finished a three day tour of bio and non bio wineries in the region of Trapani. One of the largest producers of a co-operative wine product. We were quite happy to hear when the president told us that the fining agent they used was pea protein. The production size is quite enormous covering 1272ha of land and 700 small growers. Another interesting fact is the use by all the vintners use of microsat which consists of certain micro-organisms which are already present in the soil and help the roots expand and absorb more vitamins and minerals thus resulting in larger fruit production. Thanks again for posting your article.

Comments are closed.