October 2008 issue


Send a letter to the Editor

PAGES (60-65) October 2008

Wine Flaws: TCA Jamie Goode

To a visitor from another culture, the ritual of restaurant wine service would probably seem a bit strange. Why is the customer given a taste of wine to assess, but never a bite of food? The answer lies in the inconvenient truth of bottle variation, which can result in small differences among bottles of the same wine—even, in some cases, spoilage of the wine, to the point that it is no longer fit to drink.

The chief culprit in this bottle-to-bottle variation is the cork. As natural products, corks can vary in the amount of oxygen they allow into the bottle, and can even be responsible for contamination of the wine. While oxidation is far from unknown as a fault in cork-sealed wines, it is the musty, damp-cardboard aroma of cork taint that causes the most problems in wine service.


Cork taint is nothing new, but it wasn’t until the 1980s that the chemicals underlying this flaw were first identified. Hans Tanner, a tenacious Swiss researcher, isolated a chemical called 2,4,6-trichloroanisole (TCA) from corked-wine samples, publishing his results in a German-language journal in 1981 and then in a more widely read English-language publication in 1982. The remarkable thing Tanner discovered about TCA is just how little of it is needed to contaminate otherwise-sound wine. Most people can detect TCA at a level of less than 5 nanograms per liter (five parts per trillion)—the equivalent of a few drops in an Olympic-size swimming pool. Although it is easier to spot cork taint in white wines than in heavier reds, this minuscule level of perception is one of the reasons why TCA has proved so difficult to eradicate.

Subsequent research has identified other chlorine-related compounds that can contribute to cork taint. TCA is the most potent agent, with an average detection threshold of 1.4-4 nanograms per liter. The other major player is 2,3,4,6-tetrachloroanisole (TeCA), which has a higher threshold of around 14 nanograms per liter. The compound 2,4,6-tribromoanisole (TBA), which smells similar to TCA, has also been implicated in producing musty aromas. It doesn’t come from corks, but from certain preservatives and fire retardants used in wood treatments. Because such “winery taint” has the potential to affect all the wine made in a facility, there have been several high-profile cases in which wineries have had to withdraw stock and perform expensive renovations in their cellars after TBA contamination. But this is far less common than cork taint; even in large competitions, it is extremely rare to find a “corked” bottle sealed with a synthetic cork or a screwcap, which would be evidence of winery contamination.

In the most obvious manifestation of cork taint, the wine will smell musty and moldy, and the aroma will be quite off-putting. At lower levels, detection becomes more difficult. We’ve all come across bottles that we couldn’t be sure about one way or another. Older Bordeaux often smells a bit “mushroomy”: is that cork taint, or just the normal secondary aromas of aging? And then there are wines that don’t display mustiness, but in which the fruit is strangely subdued—a sort of twilight zone that may be caused by subthreshold levels of TCA interfering with the usual aromatics.

One clue is that cork taint generally doesn’t blow off with aeration. I have often regretted the decision not to send back a particular bottle in a restaurant when I’ve had initial suspicions, because an hour or two later, when it’s too late to return the bottle, I usually find that the bottle is indeed tainted.


The actual incidence of cork taint is a hotly disputed figure for a number of reasons. First, people differ widely in their detection thresholds: one person’s spoiled bottle is another person’s source of pleasure. Second, in major competitions, there’s a tendency to overreport cork taint. If one judge thinks a wine is corked, others will tend to agree, even if the wine seems perfectly fine to them, and a backup bottle will be requested. Whether the backup bottle is better or not, it’s unlikely that the first bottle will be called back. Some competitions have reported levels of cork taint as high as 7%. Perhaps a more reliable estimate is provided by the International Wine Challenge, in which some 10,000 wines from all over the world are assessed each year in London. Any bottles reported as flawed in this competition are sent to a dedicated-faults clinic for verification. For the past three years, the incidence of cork-tainted bottles has been around 3% of all bottles sealed with natural cork. This is still an unacceptably high level, but it is less alarming than some of the figures being suggested elsewhere.

One of the interesting questions about cork taint is whether rates have become progressively worse in recent years. George M. Taber, in his recent book To Cork or Not To Cork (Scribner, 2007), uses interviews and anecdotal evidence to identify two key reasons for an increase in the rate of cork taint in the 1980s and ’90s. First, after the Portuguese revolution of 1974, many of the privately owned cork forests fell into common ownership. As a result, the meticulous, time-honored methods of managing these forests was abandoned in favor of a short-term strategy of maximizing profit. Some cork was harvested before it should have been, leading to a drop in quality of the raw materials used for cork production. In addition, some chemicals used in silviculture may have encouraged the production of taint compounds in the bark. The second factor is that growth in the popularity of bottled wines led to a large increase in demand for cork closures in the 1980s. A third reason could be that we’ve simply gotten a lot better at spotting cork taint. The quality of wines has been rising, and even cheap wines are much cleaner and fruitier than they used to be. In this context, cork taint becomes more obvious and less likely to be tolerated by consumers, regardless of their level of wine knowledge.

Two thorny questions remain: What causes cork taint in the first place, and can it be reduced or even eradicated by improvements in the cork-production process? This subject is obviously of great interest to the cork industry, which, after initially denying it had a problem, is now anxious to do all it can to eliminate cork taint.


The great cork-oak forests, concentrated in Portugal and Spain, but also found in France and north Africa, are beautifully sustainable, natural ecosystems, supporting rural economies that, without them, would have to subsist on arid desert landscapes. Cork bark is a remarkable substance, providing an insulation that allows cork oaks to survive the frequent forest fires in these dry environments.

Once a tree is mature, it can be harvested every nine years. The harvested bark is cut into planks, and after a period of weathering in the open air, it is boiled to clean it and make it more flexible. Then the planks are assessed for quality; those deemed good enough for wine cork are graded and cut into strips of appropriate width. Either a machine punches out the corks or, for more expensive products, the corks are punched out manually, one at a time. After being washed with peroxide, the corks are sorted in a manual or automated process and graded by visual appearance. Top-quality, grade-A corks are beautiful, handmade products that retail for more than $1 each.

Natural cork is also used to make “technical corks.” The most widely used is the Twin Top, in which discs of good-quality cork are glued to either end of an agglomerated cork, made from small pieces of lower-quality cork that are glued together and then extruded in long, cylindrical tubes. The lowest-quality corks are made solely of this agglomerated cork.

Cork is ideally suited to sealing wine bottles because it is fairly inert and extremely elastic, and it can survive for 100 years or longer in contact with wine while maintaining these properties. During the first few months after bottling, some oxygen is released into the wine from within the cork body; thereafter, the cork allows relatively little oxygen transmission. This lessens the risk of post-bottling reduction problems, while allowing the wine to age gracefully over many decades. Cork would be the perfect closure—if not for taint.


TCA is produced by fungi living on the cork bark. If the fungi were present only on the outside of the bark, it would be possible to strip away those sections before processing. But cork bark is riddled with small pores called lenticels, running tangentially to the surface of the bark at regular intervals; these holes facilitate gas exchange, and their inside surfaces can be colonized by fungi. Mold, the growth of large colonies of fungi, smells musty—which may be precisely why our olfactory sense is ultra-sensitive to TCA and its related compounds. An aversion to the smell of mustiness helps us avoid eating rotten food that might make us ill. Would the fungi still create TCA in the absence of obvious chlorine sources, such as pesticides or bleaches? Probably, although science has not produced a definitive answer. To avoid any risk, cork manufacturers have switched from chlorine-containing bleaches to peroxide for washing their corks.

The initial boiling step has the potential to worsen the taint problem. Until recently, the cork was immersed in big pits of boiling water that was rarely changed, so that contamination often spread from bad batches to sound batches. Now, leading cork manufacturer Amorim has installed closed containers for boiling, with the water filtered and cleaned before use. The company is also careful not to leave harvested bark in contact with bare soil during the weathering phase. Quality control is seen as vital in combating cork taint: the Amorim research laboratory analyzes a vast number of samples for TCA with sophisticated and expensive gas-chromatography machines that can detect even the tiniest of concentrations.

But quality-control measures may not be enough. Research is now focusing on methods to get TCA and other contaminants out of cork altogether. In the late 1990s, the French company Oeneo, formerly known as Sabaté, introduced a novel technical closure called Altec—a combination of small particles of cork with synthetic microspheres. The product experienced initial success, but after a while, wineries started noticing that their wines were becoming tainted with what turned out to be low levels of TCA. Basically, when tiny cork particles were blended together, the TCA contamination present in some bits of cork was being averaged across all the closures. Altec turned out to be a disaster, and Oeneo suffered a great deal of negative press, as well as some lawsuits.

But Oeneo’s research team did come up with a novel way of eliminating TCA from cork that actually worked. It relies on a technique used to decaffeinate coffee and extract fragrances: critical-point carbon dioxide. At a certain pressure and temperature, carbon dioxide enters what is known as a supercritical state, where it has the extraction power of a liquid, yet the penetration of a gas. When applied to cork granules, the procedure has proved able to remove all TCA and other contaminants while leaving the cork’s mechanical properties intact. Using this technique, Oeneo has produced the Diam closure—effectively, a taint-free Altec. Amorim has devised its own system, called ROSA, a steam treatment that has been found to remove most, but not all, TCA. Either method would represent a major improvement over the current situation.

The fight against TCA has not yet been won, but the hope is that with all the effort by the cork industry to get its house in order, the real-world incidence of cork-tainted wines will fall—perhaps not to zero, but to the point where encountering a corked bottle becomes a rare occurrence.