|
A Technical Guide for Wine Producers
|
|
RECENT ARTICLES | WYNBOER HOME
Wood maturation of wine (Part I): Factors that influence the composition of wood during the production process
Wessel du Toit
Key words: Oak, barrels, wood maturation, wood toasting.
INTRODUCTION
The maturation of wine in barrels is a practice that has been customary for many years. The wooden barrels are responsible for certain important reactions that occur in the wine during maturation. These reactions cause changes in the wine's colour, flavour, taste and stability. If the winemaker uses the right choice of barrel for a certain style of wine, these reactions are able to improve the quality of the wine. During the production process of wooden barrels certain chemical changes take place in the wood which may drastically influence the composition and characteristics of the wood. This article discusses some of these changes and how they may influence the wood and eventually the composition of the wine. It is the first in a series of three articles that will focus on wood maturation of wine.
TYPES AND ORIGIN OF WOOD
In the past various types of wood have been used and tested with a view to storing wine. Suitable types of wood should impart flavours and tastes that compliment the quality of the wine, rather than imparting undesirable flavours to the wine. It must also be durable and should not leak. Some of the different types of wood that have been tested include white oak, red oak, chestnut, gum, maple, beech, ash, pine, Cape beech, redwood, acacia and stringy-bark. The oak (Quercus sp) has been found to be the most suitable for this purpose. In France especially a large industry has developed around the production of oak for barrels. About a quarter of France's surface is covered with national forests, of which oak plantings comprise about 30 to 40 per cent. The species that occurs in these French regions are Quercus robur, Quercus pedonculata, Quercus petraea and Quercus sessilis. It is possible to distinguish between some of the species. Q. petraea, for example, has leaves with long petioles and acorns with green stalks (the so-called sitting oak). Q. robur has short petioles and acorns with long stalks (the so-called stalk oak). Q. robur and Q. pedonculata are prevalent in Limousin and Burgundy especially. In Central France and the Vosges regions one finds Q. petraea and Q. sessilis. French oaks are therefore named for their region/forest of origin, e.g. Nevers, Limousin, Alliers, Troncais, etc.
American oak (Q. alba) is also commonly used for the production of wooden barrels. Oak forests occur in Missouri, Ohio, Illinois and Wisconsin. Oregon oak, which is grown in the state Oregon, is also used for this purpose. Other potential sources of oak are being explored nowadays. These include Eastern European and Russian oak forests. Preliminary research indicates that there is little difference between wine that has been matured in this wood and wine that has been matured in French oak.
The origin of the wood has an influence on the qualities of the wood. Quercus robur and Quercus pedonculata have a high polyphenol content and lower aromatic components. Quercus petraea and Quercus sessilis on the other hand have lower levels of extractable ellagitannins and are aromatic. On the whole American wood has lower levels of phenols and higher levels of aromatic flavourants. This is especially so in the case of á-methyl-g-octalactone (the so-called oak or whisky lactone), which imparts the characteristic woody and coconut flavours to wine. Levels of between 0.5 and 16 mg total lactones per gram of dry wood from Limousin have been found and 10.5 mg for wood from Burgundy. For wood from the Vosges 65 mg per gram were obtained and for Troncais 77 mg per gram. American wood gave levels of up to 158 mg per gram. The cis isomer is more aromatic than the trans, so it goes without saying that the wood with more cis oak lactone isomer will impart a stronger wood character to the wine. On the whole American wood also has more cis isomer. Vanillin (vanilla flavours) is also more prominent in American wood (11 vs 6 mg per gram) than in French wood. Eugenol (with a clove character) is slightly higher in French wood, however, especially in central France. Other flavourants, such as vanillin acid, sinapaldehyde, coniferaldehyde etc. are also higher in American wood.
French oak has higher levels of non-volatile components, especially phenolic compounds. Wood from Limousin, in particular, has the highest extract and extractable polyphenols. In wood from this region double the amount of total extract, ellagitannins and catechin tannins was measured than in wood from Central France and American wood. Wood from Limousin also contains more hydrolysable tannins (wood tannins) such as vescalagin and castalagin than wood from the Central and Vosges regions of France. Second fill barrels have shown, however, that wood from Limousin did not impart much more phenolic components to the wine than American wood. The greater measure of extract obtained from wood from e.g. Limousin could contribute to a greater concentration of colour, although research at SU has shown that there are hardly any differences in colour in Shiraz and Pinotage matured in French, Russian and American barrels.
The grain of the wood could also play a role. It is the cambium part of the tree that gives rise to the phloem and xyleme. In due course of time the xyleme hardens and in the process changes to the so-called pith wood, from which the barrel is made. The wood also grows differently depending on the season. Spring wood has bigger ducts than summer wood, the latter being darker and harder. The ducts in the spring wood make it more flexible, but also porous. Tylosis closes off the ducts in winter, but in spring they open again for circulation. Without the tylosis wine would leak from the ends of the dowels. The grain is therefore classified according to the size and regularity of the annual growth rings. These are especially influenced by the growing season of the wood and origin. The regions of France may be classified in the following order from wide to tight grain: Limousin-Vosges-Central France (Alliers). Wide grain wood is richer in dry extract and ellagitannins, while wood with a tighter grain contains more wood lactones and eugenol; these are similarly conveyed to the wine. Wood can also differ in chemical composition as a result of differences in the growth tempo and age of the wood. The oldest pith wood has the least ellagitannins, while rapidly growing summer wood has wide annual rings (wide grain) and therefore more ellagitannins.
It is therefore obvious that origin and grain are an integral part of the wood's composition and eventual characteristics imparted to the wine. It should also be borne in mind, however, that differences may occur in barrels from the same wood. Differences have even been found in barrels from the same tree and it is therefore very important for the winemaker to have a reliable cooper who will eliminate these differences through good coopering practice. The cooper can also subject the wood to different treatments which will change its characteristics, such as wood ageing and toasting.
INFLUENCE OF WOOD AGEING
The cooper has to age the wood to obtain a measure of dessication. It is important to reduce the humidity of the wood from 41% to between 8 and 18%. This makes the wood strong enough so that a barrel may be manufactured from it and also brings about certain beneficial aromatic changes in the wood. Two kinds of ageing may be applied by the cooper, namely natural drying in the open air and artifical drying in an oven. The natural drying process usually takes between two and three years, depending on the thickness of the staves. The artifical drying process usually occurs at 40 - 60 øC in a well-ventilated drying oven for approximately one month. Wood that has been aged in these two ways appears to differ, however. This is especially so as far as the aromatic composition of the wood is concerned. Wood that has been aged the natural way has much higher levels of total lactones (89 µg/g vs 26 µg/g for wood from Central France) than wood that has been aged the artificial way. This was also the case for other important flavour compounds in which the concentrations of vanillin and eugenol were respectively up to fifty and twice as high in wood aged the natural way compared to wood aged the artificial way. On the whole ageing and drying result in a reduction in tannins and bitter glycolysed coumarins that are hydrolised to aglicones. The region in which the drying occurs may also play a role. Wood aged in Australia has been found to contain higher levels of lactones than the same wood that was aged in France (see Fig. 1). During ageing the eugenol concentration is also reduced from 8 to 3 mg/kg. This, together with the reduction in phenols, may also account for the reduction in the "green" character of wood during maturation. There is clearly a need to investigate the ageing of wood under SA conditions.
THE INFLUENCE OF TOASTING
After ageing the cooper toasts the wooden barrels in accordance with the winemaker's specifications. This is also done, however, to enable bending of the wooden dowels in order to assemble the barrel. In France this process traditionally occurs over an open fire and in America by means of steam and a gas flame. Toasting is usually divided into three levels, namely light, medium and heavy toasting. Nowadays one also finds medium plus and heavy plus toasting. The toasting process can drastically alter the wood's composition and charactistics. This happens as a result of hydrothermolysis and pyrolysis of cellulose and phenolic components in the wood. During toasting the concentration of certain phenolic components is reduced. It has been found that ellagitannins and total phenols were reduced as a result of an increase in toasting intensity. Even low levels of heating caused a reduction of 73% and 46% of vescalagin and castalagin levels in the wood surface. However, during toasting there is an increase in furanaldehydes such as furfural, methyl-5-furfural and hydroxy-5-furfural and other components with a carbohydrate origin such as the concentrations of maltol and cyclotene, the highest occurring in medium to heavily toasted wood. These components could contribute to the caramel flavours of wine. The concentration of volatile phenols also increases during toasting, such as guajacol and 4-ethyl-guajacol (smokey characters). Ranging from untoasted to heavily toasted wood, these levels in the wine may increase from 1 to 2 mg/L to 30 mg/L. The concentration of eugenol is usually slightly reduced by light toasting, but increases again with a further increase in toasting intensity. The oak lactone concentration, especially in the form of cis, also increases with toasting, although it looks like it might be reduced again by heavy toasting. At medium toasting levels the vanillin concentration may be up to 50 times more than in the case of untoasted wood, but in heavily toasted wood it is also reduced as a result of thermal breakdown. Toasting consequently makes the flavour profile of the wood more complex. Initially more vanilla and spicy flavours are obtained with low to medium toasting. Medium and medium plus toasting result in more coconut and oak lactone flavours, while heavy toasting imparts more smokey and burnt caramel flavours to the wine.
OTHER FACTORS
The age of the barrel also plays a role. Moreover, the number of fills determines the extraction ability of the wine, with the total number of extractable components being reduced as the number of barrel fills increases. Furfural, guajacol and eugenol in particular are significantly reduced from the first to the second fill of a barrel. Lactone concentrations are not reduced equally quickly. After the third fill there is very little extractable volatile and ellagic acid and gallic acid in the wood. The latter may result in poor colour development and stability of the wine in the barrel. It is possible to extend the life-span of barrels by removing and shaving the dowels. The life-span of the barrel is also extended by initial toasting, especially if the cycles spent by the wine in the barrel do not exceed six months. The risk of microbiological spoilage also increases in older barrels and when they do not contain any wine, barrels should be taken care of properly. This entails the regular burning of sulphur piths in the barrels and proper washing and rinsing of used barrels with warm water or steam before the barrels are filled again with wine. The size of the barrel also plays a role, with more flavourful wood components being more rapidly extracted in the wine when the barrels are smaller as a result of the smaller volume to surface ratio of smaller barrels. This difference is also reduced in older barrels. When using wood shavings and staves this should also be borne in mind, since here the wood surface that comes into contact with the wine on a mass basis is bigger than in a barrel. A Spanish study found that vanillin and furfural in particular were much lower in wine that had been matured in 1000 L barrels than in wine that had been matured in 220 and 500 L barrels.
It is therefore clear that various factors during the production process of barrels may influence the composition of the wood. This will also be reflected in the wine that is being matured in the barrel.
SOURCES:
Baldwin, G. 1993. Austr. Grapegrower Winemaker 355: 26 - 27
Chatonnet, P., Dubourdieu, D. and Boidron, J.N. 1991, Wine Ind J, 2, 73-84.
Chatonnet, P.,1995, Influence des procedes de tonnellerie et des conditions d'elevage sur la composition et la qualité des vins elevés en fûts de chene. These Doctorat. Université des Bordeaux II
Chatonnet P., 1999, Am Chem Soc, p 180-207
Feuillat and Keller, 1997, Am J Enol. Vitic. 47, 502-508.
Graf and Tchelistcheff, 1969, Wines and Vines, 50, 26-30.
Günther and Mosandl, 1986, Liebigs Ann Chem, 2112
Masson, G., Guichard, E., Fourinier, N. and R., Puech, J.L., 1995, Am. J. Enol. Vitic. 46, 424-428
Puech, J.L, Feuillat, F., Mosedale, J.R., 1999, Am J. Enol. Vitic, 50, ,Am J. Enol. Vitic., 50, 469-478.
Sefton, M.A., Francis, I.L., Pocock, K.F., Williams, P.J., 1993, Austr Grapegr Winemaker, 355, 17-25
Sefton, M.A., Francis, I.L., Pocock, K.F., Williams, P.J., 1993, Sci Aliments, 13, 629-643.
Schahinger, 1991, Wine Ind J., 2, 65-89.
Schahinger, 1996, Aust. Grapegr. Winemaker, 7, 17-19.
Schulz, R. 1990. Austr. Grapegr Winemaker 319:19
Swan, J.S., Newton, J., Larnie, E., Sayre, R., 1997, Aust. Grapegr. Winemaker, 7, 17-19.
Towey, J.P., Waterhouse, A.L., 1996, Am. J. Enol. Vitic., 47, 17-20.
Towey, J.P., Waterhouse, A.L., 1996, Am. J. Enol. Vitic., 47, 163-172.
Vivas, N., 1995, J. Coup. Sc and Tech, 1, 9-15
Vivas, N., 1995, J. Coup. Sc and Tech, 1, 33-47
Wilker K.L.. and Gallander, J.F., 1989, Am J. Enol. Vitic., 40, 224-226.
|
|
Wynboer is incorporated in WineLand, magazine of the SA wine producers.
Visit our sister sites:
|
| UP |
COPYRIGHT (C) 2000 WineLand |