Château Rouquette - an oustanding Bordeaux wine

Our first two harvests, in 1998 and 1999, were made without using barrels. Photos of Bordeaux often show impressive rows of barrels but the majority of the wine produced in Bordeaux does not see a barrel, being bottled direct from a vat. It was only when we moved full time to France in 1999 that we constructed a new barrel chai (see photos below) and had the time to organise the buying of the barrels and the extra work involved.

For our first harvest with barrels in 2000, we were totally guided by our oenologue as to which barrels to buy, as we had no idea how Rouquette would react to the use of oak. The results amazed us! Just a few months after putting the wine in barrels the tannins were much softer and an extra layer of complexity and aromas was added to the wine.

Barrels have two effects on the wine: they give a taste of oak and let in small amounts of air. New barrels give a strong taste of oak but this reduces with age. Barrels can be kept for about 5 years and throughout this period will let in air but the impact of the taste of oak rapidly diminishes with age. The impact of the air intake will remain reasonably constant and softens the tannins.

The skill in choosing barrels is not to let the oak over-power the wine. First growth Bordeauxs normally use 100% new barrels and are kept in barrels for 16-24 months. The wine is sufficiently robust not to be over-whelmed by the oak. The next level down typically has around 50% new barrels – examples include Leoville Barton and Pichon Longueville Comtesse de Lalande. In all these wines the oak is not noticeable and is well integrated. On the other hand, there are wines that taste almost entirely of oak – it overwhelms the wine masking the aroma, flavour, and texture of the wine and the wine ceases to express its unique terroir. There was a period in the ‘90s when many Australian Chardonnays came into this category.

Economics is also a factor, since a new French barrel (containing 225 litres) costs about Euros 530 and adds significantly to the cost of making the wine. The use of oak chips or staves will replicate the taste of the oak, at much less cost than a barrel, but will not allow the gentle aeration that a barrel allows. They are also not allowed for French AOC wine.

We kept one barrel for a third year in 2002, as an experiment. The results were encouraging: the tannins were still significantly softened but the oak much less pronounced. On the basis of this, we decided to use some two year old barrels when we expanded the number of barrels from 60 to 95, for the 2003 harvest. We invested in new equipment to clean the barrels, as there is a risk of brettanomyces developing, which gives an unpleasant odour and taste to the wine. The alternative to using the two year old barrels would have been to buy in some one year old barrels from outside. Then, in 2005 we took the plunge and reduced the new oak to 40% and dropped it again in 2006 to 36%, a level we intend to maintain. The result is a wine where the oak is slightly less pronounced.

If the balance between new and older barrels is the most important decision, the next is where the barrels come from. When we started using barrels we bought about 80% French with some American:

		2000	2001	2002	2003	2004	2005	2006
% American		17%	18%	20%	17%	9%	0%	2%

We discovered that whilst the American barrels integrated well with our wine to start with, after 6 months or so they give a slightly dryish after-taste to the wine. This is a pity as they are about 40% cheaper than French barrels, mainly because the trees that are used are sawn rather than split along the grain, as is the case in France. From 2004 we bought only French barrels until last year, when we bought two American barrels with a Burgundy toast and a barrel from Hungary, as an experiment.

The main quality decision when buying new barrels used to be their geographic origin. It is now recognised that the type of grain is more important, although it can coincide with geographic regions. Fine grain is more porous, contrary to what one would expect, and allows more oxygen intake and produces more aromas with less tannic uptake. The fine rain is caused by the slower growth of the trees. The forests of central France and Allier tend to produce finer grain than the forests of the Vosges and Limosin, whilst American trees have a coarser grain.

The third decision when using barrels is the cooper. We have stuck with the same three firms since 2000, although we have changed the percentages. We use Demptos, Quintessence (owned by US World Cooperage) and Millet. The Demptos barrels suit our wine best but the combination of the three cooperages is better than Demptos alone! The combination adds an extra level of complexity to the wine. We use a medium toast. The chart above (click to enlarge) shows the impact that changes in the toasting of the Quintessence barrels can have on the wine.

It can be seen that we have been continually fine tuning the use of barrels over the last seven years. We think this has benefited the quality of Rouquette. However, this is a never ending quest and we are always looking for new ideas!