3. The harmonization

This phase is the most difficult to follow, because the wine at this time is the most sensitive and the phenomena less reversible. The risks are the development of dryness of the palate and/or the excessive maturity accompanied by loss of freshness and oxidized aromatic expression. Adherence to temperature recommendations is likewise most critical in this phase. These risks are especially high if the wine is weak.

The dose of oxygen usually does not go over 1 ml/liter/month. The effects of the variation of the dose are more crucial than during the structuration phase. A variation in dose between 0.5 ml/l/month and 1.0 ml/l/month can produce a very different reaction in the wine. As the dose is very low during this phase, there is little risk of the accumulation of dissolved oxygen, so the temperature level is not very important. However, high temperature results in fast evolution and therefore increases the risk of oxidation.

The ideal dose is determined by tasting, which looks for maximum aromatic benefit without causing dryness on the palate. As soon as the tannins begin to seem too dry, it is necessary to limit the oxygen or even to stop the micro-oxygenation. If micro-oxygenation is continued at this point, the result will be a wine that lacks in volume, becomes very flat, and the aromas of oxidation may appear irreversibly.

On the other hand, because the notion of hard tannins and dry tannins are very close, one can be confused. If the tannins are hard, it means that the wine needs some oxygen to soften them, and that the harmonization phase must continue. If the tannins are dry, it means the contribution of oxygen must be limited.

According to our definition of these terms, the onset of drying tannins is accompanied by a loss of body and richness, whereas hardening/softening of tannins generally follows this pattern:

• During the structuration, the more oxygen added the harder the tannins will become.
• During harmonization, the tannins automatically soften, but if the dose of oxygen is too high, the tannins will become dry.

One way to test the difference is to place a sample of the wine into a bottle. If the wine hardens in the bottle, it means that the harmonization phase should carry on. If the wine is stable in the bottle during the two weeks to one month after bottling, it means that the process has been sufficient.

The process of harmonizing can be accelerated by a violent addition of oxygen such as through a racking, to impart 1 ml/l all at once. Such operations can lead to a rapid evolution and can be useful when one doesn=t have time to finish the ageing gently.

<<<<<<<<<<<<<<<<<<<

Classic evolution of the wine during the harmonization phase.

Depending on the type of wine desired, you can choose the subject you want to work on: aromas, palate or tannins. You can also choose whether you want to age the wine over a period of, say, 18 months in barrels, or 6 months in the tank.

1. Reduction taste - origin:

• Grape and vintage
• Sulfur from the vineyards
• Lees
• etc . . .

2. Reductive flavor expression - cause and solution:

When deprived of oxygen, a red wine has the tendency to develop some reduction taste. A wine can be reductive without showing reduction taste if a sufficient dose of oxygen is available. As soon as the oxygen is missing (in the bottle, for instance), the reductive taste will develop very quickly.

During the phase of structuration, micro-oxygenation will increase the power of reduction of the wine and decrease the expression of reductive tastes in the same time. During the harmonization, the reductive power disappears completely if the ageing is completed.

If the wine is bottled too soon, or if the tank is without aeration, as long as the reduction power is high, reductive flavors will be apparent.

Once the structuration phase is past, sulfides can be very effectively cleaned with late micro-oxygenation at a rate of 1 to 2.5 ml/L/month over a one to two week period.

1. General features

Micro-Oxygenation is not primarily intended for use with problem wines. It supplements rather than replaces the action of the barrel.

The ideal situation is to work with well-concentrated wines which need only to be stabilized (not remedially salvaged) and refined by prolonged ageing, so as to keep a good balance of structure to roundness.

Other ideal features include:

• Clean vines with a low yield
• Long maceration with a selective extraction, in order to gain the maximum potential of the grapes and avoid the vegetal aspect
• Long ageing with no hurry for bottling
• Temperature-controlled cellar
• No fining before pre-bottling
• Ageing in bottle, if and when the wine has been prepared for that method

2. Importance of the temperature

The temperature of the wine is an essential parameter of micro-oxygenation. Ideally, it should be kept stable at around 15^oC. The colder the wine is the more capable it is of dissolving the oxygen. On the other hand, the colder it is the slower the consumption of oxygen.

As oxygen is added, the combination of these two phenomena leads to the accumulation of dissolved oxygen in the wine. However, there are lesser quantities when the oxygen addition occurs at low temperature. If the oxygen is accumulated as dissolved oxygen in the wine, controlled oxidation is not then occurring, and some of the problems linked to oxygen addition may then appear.

If the temperature rises too high, oxygenation is less beneficial. Except during ML, the addition of oxygen must be not done at high temperature (>18^oC).

3. Importance of the repartition of the oxygen

The effects of oxidation vary widely according to the quantity of oxygen available dissolved at a certain time. For example, bringing 1 ml/l all at once won=t have the same effects as bringing it during one month. By metering in small quantities of oxygen, we want to promote some reactions which occur preferentially, while still avoiding others which are less energetically preferred. This partitioning balance point will be different in each case.

Moreover, the wine ages with or without oxygen. Some sort of reactions of condensation and precipitation (setting) will proceed with time, and will change the reactivity of the wine towards the oxygen. For example, brining 1 ml/l oxygen in November won=t have the same effects as in June.

As an illustration, suppose that treatment is suspended temporarily because the temperature has dropped. One cannot simply double the dose of oxygen afterwards and expect the same effect.

4. Importance of CO₂

CO₂ delays the fining of the wine; among other things, by increasing its density. There is no problem in using the oxygenation technique in a wine with a high content of CO₂. However, it is usually preferred to get rid of it before the operation, simply because to eliminate the CO₂ in the wine during ageing would be a violent process that would adversely influence the aromas gained during that time.

5. Importance of SO₂

Micro-oxygenation normally has no effect on SO₂, as long as it is controlled. On the other hand, as soon as there is an accumulation of dissolved oxygen, there is a precipitation of the sulphur into sulphates, and the free SO₂ drops. The raise of the temperature over 18^oC seems to increase the consumption of the SO₂ as well. Therefore, it is wise to check the free SO₂ regularly.

6. Importance of an experimental control

Because micro-oxygenation is a new technology (a new knob to control an old parameter), the user should still consider it as an experiment. Therefore, everything should be carefully recorded: periods of time when used, temperature, racking, and so on. A control lot or an adequate number of samples of the same wine without micro-oxygenation should also be set aside, to be used as a comparison.

1. During alcoholic fermentation

It is known that 4 ml/l of oxygen is needed at the peak of fermentation by the yeasts. The apparatus is capable of producing such a quantity of oxygen (when set on 120 m/l/month, that is to say 3.5 seconds for one day, at the peak; or, in two to four days, starting the second day of fermentation with a setting of 7 and 14 seconds).

For a red wine during maceration it is necessary to pierce a hole in the pomace with a PVC tube, in order to guide the diffuser into the wine. The tube stays in place so it is easy to remove at the end of the process.

2. During malolactic fermentation (MLF)

Among all the tanks observed while micro-oxygenation was occurring during the MLF, no negative effect has been noticed. There has been no particular increase of VA, nor have there been any taste problems. However, it is true that it=s a bit better to go through the MLF in barrels, which is a kind of oxygenation as well.

It seems desirable to delay the onset of MLF so that its action can be used to help consume aldehyde after pigment complexation is well accomplished.

Moreover, when the oxygenation starts right after the alcoholic fermentation, the stability of the color seems quite good. The low content of SO₂ explains why some high quantities of oxygen brought in at that time will have big effects. A slight production of acetaldehyde just before the MLF explains the fixation of the color.

A controlled addition of oxygen may be recommended throughout the MLF. If the period of time during the MLF is long and without SO₂, one must exercise caution.

3. Ageing of the red wines

As with any ageing regimen, a program must be conceived which incorporates the type of product desired, the time allowed for ageing and the bottling date.

The time required for the structuration phase varies according to the characteristics of the wine. During this phase the wine with micro-oxygenation may taste incorrect if the oxygenation is stopped before completion, and some reduction may appear. The wine must complete this period of time to get a correct balance back.

General guidelines include:

• Stop the micro-oxygenation at least two months before the bottling. If the wine closes itself, the rapid addition of 1 ml/l or a light racking can be done to open and stabilize the wine.
• If the ageing is short, start with small quantities so as not to develop the structuration phase too much. End up with small quantities if the ageing is long.

4. Evaluation of the need in oxygen according to the objective

Every wine has a different need for oxygen in order to express the best of its potential. Tasting is at present the only means to evaluate this need.

Two parameters are important:

• The strength (or weakness) of concentration of character - especially for the phenolic compounds.
• The ripeness of the grapes as opposed to the vegetal character.

<<<<<<<<<<<<<<<<<<<

When learning the technique, it may seem safest to try low levels first and see the result before increasing the dose. Actually, high levels for a short time early on when the wine can absorb them are a safer approach.

5. Some indicators to use

The measure of dissolved oxygen.

• Allows checking if the contribution of the oxygen remains below the consumption. However, this measure will not tell you if the quantity applied is the correct one for the desired wine evolution.

Vegetative aroma of the wine.

• This is an easy way of checking. It has to be done regularly, however. As long as the vegetal characters are still showing, the quantity of oxygen can stay high, if the temperature allows it. If these characters are completely gone, the quantity can be decreased, but the other characters have to be controlled as well.

Reduction flavors.

Oxidation flavors.

(An Ainterview@ compiled from e-mail conversation between Clark Smith of Vinovation and Thierry Lemaire of OenoDev.)

Clark: It is common practice in the U.S. to rack off gross lees promptly after alcoholic fermentation is complete (for example at pressing), but to allow contact of the wine on fine lees for two weeks in order to allow micronutrients helpful to ML bacteria to be evolved from yeasts. Then a second racking is done. Should the treatment commence at this point?

Thierry: The best time to start is when the AF is complete with or without lees, with or without cap. If we could deal with Brett, I=m almost sure that starting even before the end of AF would be interesting. We have to check that seriously.

Clark: If we commence while lees are present, will the turbulence caused by the gas be low enough that the lees will still settle so racking can still be done in two weeks?

Thierry: Yes. As long as the ceramic is not bubbling inside the lees, the movement is not strong enough to keep the solids in suspension. In many cases (especially stronger wines), the wine actually may settle faster.

Clark: Many wineries inoculate with Leuconostoc oenoes for ML during the primary fermentation. For this reason, ML may complete very promptly, even during the primary fermentation. In this case, do you wish for a period of micro-oxygenation to occur before SO₂ is added? If so, how long should one wait to add SO₂?

Thierry: We think that having very prompt ML is perhaps expedient for practical reasons and to feel comfortable that it is complete, but it doesn=t make any sense regarding the maturation aspects. However, if the ML occurs, it can be interesting to delay a little the sulfur addition, but it can result in other problems. In any case the idea is at least to bring fairly big amounts of O₂ ASAP without O₂ accumulation. This can be monitored by SO₂ analysis, as it shouldn=t result in an unusual depletion.

Clark: If one Aguesses@ the correct amount of O₂ to add, are there other ways to tell that too much O₂ is being used, which are useful prior to completion of ML?

Thierry: Prior completion of ML, we still don=t know what Atoo much@ means. What we can say is that there=s no need to accumulate tremendous amounts of aldehyde, to the extent that it results in remaining aldehyde after ML, which doesn=t seem to be a good idea.

Clark: From some old literature on microbullage we were given by Bill Dyer, we got a chart proposing treatments 0.5 to 2.5 ml/L/mo from A - E based in degree of dilution and vegetative character. This seems like a useful document to work from in developing an English language guide. Do you like this approach?

Thierry: The levels to start in red wines much be much more than 0.5 to 2.5 ml/l/mo. If it=s prior to ML, the starting point would be 10 minimum for a weak wine and 60 or more for a strong dark extract. The level has to be then adjusted in function of aldehyde perception. Measurement would be helpful but it=s not very easy to conduct because it=s not very stable. If should be measured in tank.

Clark: I=m confused. Is the 10 - 60 ml/L/mo recommendation only for before SO₂ is added (you keep saying Apre ML@, but I think you mean Apre SO₂" n=est-ce pas?). What is desirable aldehyde perception? Faintly evident, not strong? If they measure it, what is a good level? What do you mean by measurement in tank - headspace GC?

Thierry: The chart is OK as a model. The doses are old fashioned. It has not a lot of sense to speak of dose like that without specifying precisely when it=s delivered. I would widen the range after ML from 0.5 to 10 ml/l/mo. 10 to 60 is OK prior to ML.

I keep saying ML (but it may change one day) because, due to the CO₂ production (maybe) or due to a metabolism of the bacteria (maybe) the ML is a very effective period for consuming the extra aldehyde. It=s safer to play (to Asurf@) the aldehyde before ML than later.

At this stage of our understanding, the desirable aldehyde perception to maintain is so far more related to the winemaker=s personal degree of concern than the actual wine=s need. It=s really important to find a way of measuring the aldehyde production and the aldehyde consumption ability and to put figures behind it. So far, our position is rather simplistic: if we can slightly find it, that means that there=s some available for further reactions. Some people go further and the wine is really aldehydic. The last thing I have to add is that we never had any bad experience of having even a lot of aldehyde in the early age of the wine. Later, it=s another question.

Clark: What is the best cellar temperature for the treatment? If this cannot be controlled, what are your recommendations?

Thierry: The best range is between 59 and 65⁰F (after ML). If it cannot be controlled, then it=s safe to keep the level of oxygenation under the curve we presented. If you don=t have it, I=ll send it as a file (ppt).

Clark: Many people are asking for scientific papers to read concerning this technique. Can you give us some good references?

Thierry: That=s a problematic question because the only >scientific= papers we have are from INRA and are quite old (>95). There=s a bunch of articles treating of Microx which are all practical and not scientific. We are planning for the next vintage to set some simple and precise protocols for writing something. Michel MOUTOUNET will be our consultant. We will set a trial for Australian Magazine in Languedoc and Spain. There=s a trial going on at ASTI. And hopefully, we=ll do something with DAVIS, FRESNO, etc?....

Clark: Most California wine is sent to barrels, and we are at this stage promoting this technique as a preconditioning prior to barreling down. Is it reasonable to say that once the harmonization phase has begun (the wine begins to soften) that this is a good time to barrel?

Thierry: No. The good time to barrel can be much earlier than the harmonization. It=s difficult to answer precisely because it could be the subject of a specific trial. The most important thing is to stabilize as many free anthocyanins as possible at the beginning. Then the further constructions may occur slowly (structuration phase) which can take place in barrels.

Clark: Will August vacation be a problem for supply?

Thierry: August vacations shouldn=t be a problem because we don=t close nor do our suppliers either. The stocks of parts have been felt before. However, the problem might be an acceleration of orders. I=m particularly concerned by some last minute orders in Italy.

Clark: What advice do you have on the correct grade of O₂? Can we use welding grade, or must it be hospital grade?

Thierry: In France we can find Food Grade oxygen. I guess it exists in the USA. If not, hospital should be reasonable. It=s actually the same gas. Only the protocol of cylinder conditioning (rinsing, filtering...) is different.

Clark: Is there an influence of pH on the treatment?

Thierry: The pH is certainly important in the oxidation reactions. The higher the pH, the more phenolate form you find versus phenol. It results in a stronger reactivity of the phenols to give semi quinonic radicals. However there=s no direct consequence in the way we apply microx. The only thing is that the higher is the pH, the more careful we will be, especially regarding the Brettanomyces problems.