Editor's Note: This article is part of a semi-regular series of research briefs sponsored by the Department of Viticulture and Enology at UC Davis. For more summaries of research sponsored by UC Davis Department of Viticulture and Enology, see http://wineserver.ucdavis.edu/trellissummary_categories.php.

If you are a winemakerconcerned with the price of new barrels, you have probably wondered what type of wood provides the best oak aroma for your wine. These authors wondered, too, and published the results of their trial in "The use of oak chips during the ageing of a red wine in stainless steel tanks or used barrels: effect of the contact time and size of the oak chips on aroma compounds" in the Australian Journal of Grape and Wine Research, 14:63-70, 2008. Since the title says it all, let's move right into the trial lay-out and the results.

Trial Lay-out

The authors used a 2004 Monastrell (Mourvedre) wine from D.O. Jumilla, Spain, which had completed alcoholic and malolactic fermentations. They put this wine in stainless steel tanks (1,000 liters) by itself, in the presence of cubes (2.4 g/L), in the presence of shavings (2.6 g/L) or in the presence of powder (1.3 g/L). Rates were calculated to simulate the surface/volume of a barrel. They also put the same wine in used barrels, once again, by itself, with cubes, with shavings or with powder. These used barrels--American oak--had been used seven times and had followed a sulfur burning sanitation regime. Finally, the authors put the wine in new barrels--medium toast--from the same cooperage as the used barrels. All of the chips used--cubes, shavings and powder--were also medium toast.

To assess the influence of contact time, the different treatments were allowed to stay in the wine for either three, six or nine months. Because wine normally undergoes a period of anaerobic ageing in the bottle after oak ageing, the authors mimicked that by waiting for another six months after bottling before taking any measurements. All treatments were conducted in triplicate (three tanks, three used barrels and three new barrels for each type of chip and contact time). At the end of the experiment, the authors studied how container type, chip shape and contact time length had affected the amount of oak compounds released into the wines.

Background

But first, let's review the main categories of compounds that wood can bestow to a wine. There are five categories of interest to us: lactones, furfurals, vanillin, guaiacols and ethylphenols.

The overall, oversimplified picture would be something like this. At the beginning, raw oak contributes mainly lactones. As the wood is toasted, the lactones diminish and furfural and vanillin increase. At even higher toast levels, these compounds are gradually replaced by guaiacol. In the meantime, if yeast is present--and chances are there will be some embedded in the wood--their microbial activity will contribute ethylphenols. [Even though the current authors do not study it, there is a sixth class of wood compounds--eugenol and isoeugenol--present in raw wood; however, they increase with toasting, which imparts a spicy, clove aroma.] The authors then studied the impact of container, chip shape and contact time on each of these compounds.

Furfurals

Furfural and methylfurfural contribute a pleasant aroma to the wine (sweet, butterscotch, light caramel, light almond). They form during the toasting process as a result of the degradation by heat of the carbohydrates in the wood (cellulose and hemicellulose)--a process called Maillard reaction.

After just three months, furfural--and methylfurfural--were high in the wines to which oak chips had been added. Used barrels plus chips showed higher levels than tanks plus chips. Cubes were more effective than shavings or powder in increasing furfural. Interestingly, the authors found that furfural levels decreased after six and nine months of contact, rather than the other way around. This is because furfurals, which are aldehydes, tend to be converted in time into their respective alcohols. Therefore, for longer periods of time, conversion exceeds extraction, and furfural levels go down.

But the highest furfural levels of all the treatments compared were those of the new barrels. Even though furfural levels tended to decrease after bottling in all treatments, the wines that had been aged in new barrels were still the ones showing the largest concentrations.

Guaiacols

Guaiacol and methylguaiacol impart smoky, char and spicy aromas to the wine; guaiacol is more smoky whereas methylguaiacol is more spicy. They are formed by the degradation of wood lignin due to very high temperatures--a process called pyrolysis.

The authors found very low levels of these compounds in both tanks and barrels. This agrees with reports that guaiacol formation is low when toasting temperatures are below 446°F (230°C).

Lactones

Lactones in either of their two isomers--cis and trans--are the main compounds released by wood from a sensory point of view. They impart a fresh oak and coconut aroma to the wine. Cis-lactone has more of a wood character; trans-lactone is more herbaceous.

In tanks without chips, no lactones were detected--proof of its oak origin. With chips, the authors found more lactones when cubes or shavings were used than when they used powder. This also confirms that lactones were in the wood before toasting; otherwise, more would have been found in the smaller formats.

One of the most remarkable features about lactones was that their extraction was very fast, with levels after three months being similar to those with six and nine months. This effect was exacerbated by the use of chips. Whereas with chips, lactones were very high at the beginning and later decreased when wine was aged in new barrels, lactones continued to increase. In other words, the release of oak lactones is much faster with chips than with barrels. This makes sense since more time is expected for a wine to penetrate a stave barrel--and then for the compounds to permeate back into the wine--than to penetrate a chip.

Ethylphenols

The animal, leather, horsy, sweaty aromas that we are so familiar with are attributed to 4-ethylphenols (and sometimes also 4-ethylguaiacol). These compounds are not naturally present in the wood. As we know, they are formed by Brettanomyces and other yeasts when they decarboxylate the phenolic acids present in the wood (coumaric acid and ferulic acid). Therefore, the amount of these compounds in wines aged in wood is going to depend on the presence of their precursors--and on the amount of yeast. As a reminder, yeast have been found as deep as 8 mm (3 inches) within the wood staves of a used barrel.

The authors found the very lowest levels of ethylphenols after three months, both in the wine that had been in tanks--with or without chips--and in the wine that had been in barrels--used or new, with or without chips. In contrast, the highest levels of ethylphenols were found after six and nine months of ageing. In other words, time had more impact than type of container. For longer periods of ageing, the authors found higher levels of ethylphenols with cubes than with the other types of chips.

Surprisingly, ethylphenol levels were high even in the control wine (tank without chips) after six months. The authors believe that a previous Brettanomyces contamination from the winery led to this increase in ethylphenols. Thus, Brett--and ethylphenols--were likely present in this Mourvedre wine from the start. The authors commented how this is an increasing problem in wineries that are adopting the trend of not stabilizing their wines, something that makes them more nutritionally attractive to Brettanomyces.

Overall, ethylphenol levels were slightly higher in those wines that had been aged in barrels compared with those that had been aged in tanks. After six months of ageing in the bottle, the authors detected no changes in ethylphenols. Apparently, it had no chance to increase in the anaerobic conditions of the bottle.

Vanillin

Vanillin, the main aroma compound in natural vanilla beans, is also present in raw oak. Vanillin increases with medium toast levels but decreases with very high toast. Vanillin can also be transformed into non-aromatic vanillyl alcohols by yeast.

Similarly to ethylphenols, the authors found vanillin in the control wines, meaning vanillin can form by mechanisms other than wood extraction.

The highest vanillin levels were found after three months of contact with chips, particularly when cubes were used. Because finer pieces of wood are more combustible, higher vanillin is expected to form. However, when size is too small (<5mm), losses due to evaporation have been reported.

For longer contact times, vanillin levels in wine treated with chips decreased and in wine kept in new barrels increased. As we mentioned before, as wine matures, vanillin changes into non-aromatic compounds. Unlike what happened with chips, from which vanillin was released and quickly converted, when the wine was in new barrels, prolonged extraction was able to exceed conversion and the result was vanillin accumulation. As was also the case with lactones, new barrels provide a slower, more sustained release of vanillin.

Conclusions

Using sophisticated analyses, the authors tried to organize the large amount of data developed from the various container/oak additive combinations tested. They were able to reach the following conclusions:

1. The type of container and the contact time were the parameters leading to the most significant differences in the aromatic profile of the wines. The shape of the chips was much less important.

2. Wines with new barrels could be totally separated (by discriminant analysis) from those with chips added. Similarly, wines aged for three months were very different from those aged for six or nine months.

3. Chips released aroma compounds into the wine very rapidly while new barrels kept extracting aroma compounds for longer periods. Used barrels with added chips had a behavior somewhere in between.

Based on the above results, the authors consider that oak chips are a good cost-effective way of ageing wine in a short time while reusing used barrels, but that "overall wine quality is better in wines matured in new barrels." wbm