Corked wine is wine contaminated by TCA, or 2,4,6-trichloroanisole, a chemical compound formed when mould meets chlorine residues in natural cork bark or in cellar timber. Cork taint is harmless to drink, and a sensitive taster detects it at a level of one to two nanograms per litre, while other people need five or ten. Below that threshold a tainted bottle is often silent: it strips out the fruit and the aroma rather than adding the musty odour of wet cardboard, so the wine simply tastes dull.
What to check before you pour it away
- TCA is harmless. A corked bottle is unpleasant to taste, never bad for your health.
- At a low level it removes fruit instead of adding a smell, so the wine reads as dull rather than faulty.
- Heat damage and oxidation are commoner faults online: a pushed closure or a stained capsule is the giveaway.
- A screw cap removes the risk outright, but a tainted winery spoils wine whatever seals it. The taint rate under treated cork is now below one bottle in a hundred.
What corked wine is, and what TCA does to a glass
Cork taint is contamination by a family of compounds called chloroanisoles, of which TCA is by far the most common. The compound is not made by the vine, the grapes or the winemaking: it is picked up after the wine is bottled, from the closure or from the cellar around it, and the source is always outside the liquid. Concentrations are measured in nanograms per litre, which is a few parts per trillion, and that is why no amount of swirling, warming or waiting will drive it out. A tainted wine stays tainted.
The part that catches drinkers out is what trichloroanisole does to the nose. It is not simply an added odour laid over the wine. TCA also suppresses the olfactory receptors themselves, so it mutes the aromas that should be present. A heavily contaminated glass smells of damp cellar and old newspaper. A lightly contaminated one smells of almost nothing, and reads as thin, short and disappointing rather than spoiled. Two people can honestly disagree about the same bottle, because sensitivity to this compound varies enormously between individuals.
What cork taint smells like, and why you may miss it
The vocabulary sommeliers use for the obvious form is consistent across wine regions: wet cardboard, a damp basement, a dog that has come in from the rain, mouldy newspaper. Once you have met it in a glass you do not forget it, and you will find the same aroma on the wet end of the cork if you pull it out and smell it.
The quiet form is the one that matters commercially. At two or three nanograms per litre the wine has no obvious defect. Its fruit is gone, its finish is short, and the impression is that the producer wrote a cheque the wine could not cash. Drinkers blame the vintage, the grape or their own palate. If a wine you know well arrives flat and stripped, and a second glass poured an hour later is no better, cork taint is a reasonable suspicion even without the musty note.
One visual inspection and one precaution cost nothing. A glass stored upside down in a wooden cupboard, or dried with a tea towel, carries a stale smell of its own that a drinker will happily attribute to the wine; pour a mouthful into a second, freshly rinsed glass before you conclude anything, because the condition of your glassware ruins more aromas than most people suspect. Look at the cork too. A wet, crumbling or mould-covered top tells you about storage, not about TCA.
Where TCA comes from
The chemistry runs in two steps. First a chlorophenol has to be present. These come from chlorine used to bleach or to sanitise, from chlorinated cleaning agents used around a winery, and from wood preservatives applied to beams, pallets and roof timbers. Then filamentous fungi living in that damp environment methylate the chlorophenol and turn it into a chloroanisole. Neither step needs much material: a trace of chlorine, water, and a colony of mould are enough. Mould grows naturally wherever cork and damp meet, so its presence in a cellar is normal and nothing to be alarmed by; the chlorine is the avoidable half of the reaction.
Cork bark is a natural sponge for both, which is why the stopper takes the blame. It is not the only route. When every wine in a cellar shows the same character, the culprit is usually 2,4,6-tribromoanisole rather than TCA, formed from brominated flame retardants sprayed on structural timber decades earlier. That version contaminates the air, the barrels and the bottling equipment, and it reaches wine sealed under screw caps just as easily as wine under cork. Outbreaks of this kind have forced cellars in California and elsewhere to strip out the affected timber before they could bottle again.
Three sources account for the majority of tainted bottles coming back through a merchant's door: the cork itself, the cellar air, and the wooden crates a wine has travelled in. Only the first is a cork issue in the narrow sense, which is why changing supplier never solves a building problem.
How much wine is actually affected
Published rates vary more than they should, because studies count different things. Some measure TCA in the liquid by chemical analysis, others count bottles rejected by a tasting panel, and a panel will always find a lower percentage than a machine. The total depends on where the threshold is set. Figures from the 1990s and early 2000s commonly sat between one and five per cent of bottles. Industry data published since the mid 2010s puts the estimated percentage under one for cork treated by modern methods, and a drinker will now go a long time between one tainted bottle and the next. Figures based on laboratory analysis remain the only ones worth comparing across studies.
Two industrial answers explain most of that fall. Amorim screens individual natural corks by gas chromatography under the NDtech name and rejects any that carry detectable TCA. Diam takes the opposite route: it grinds cork into granules, strips them with supercritical carbon dioxide, then reassembles the closure. Both add cost, and both are chosen deliberately by producers who have been burned. A cheap agglomerate stopper on a supermarket wine has had neither treatment, and no manufacturer claims otherwise.
How cork taint is detected and measured
Sensory detection has a threshold and it is not the same for everyone. The often quoted level of one and a half to two nanograms per litre describes a sensitive taster in a neutral white wine; in a heavily oaked red the same quantity can pass unnoticed, because the wood aromas mask it. Some drinkers need five or ten nanograms before anything seems wrong. This is why a tasting panel and a laboratory rarely produce the same result.
Instrumental analysis removes the argument. A sample of wine is prepared and passed through gas chromatography coupled to mass spectrometry, and the technology now resolves fractions of a nanogram per litre. It will identify the presence of TCA in a wine that no taster has flagged, and it says whether the fault was caused by a chloroanisole at all. Cork producers use the same science on the closure itself, soaking a batch and measuring what leaches out, which is how a supplier can put a figure on the risk of cork taint in a delivery rather than a promise.
The practical consequence for a consumer is simple. If you and a friend disagree about a glass, neither of you is necessarily wrong. Sensitivity is individual, and the person who cannot detect it is not being polite.
What a careful winery does about it
Prevention is entirely a matter of housekeeping, and the good producers treat it as such. Chlorine is banned from the building, cleaning is done with peroxide, ozone or steam, and nothing chlorinated is allowed near the bottling line. Wooden pallets are kept out of the cellar, barrels are checked, and incoming closures go through quality control before a single case is filled.
None of this is glamorous and none of it appears on a label, but it is the difference between a winery that has an occasional bad bottle and one that has a systemic problem. Wineries that have lived through a tribromoanisole outbreak are usually the most rigorous of all, because they have seen a whole vintage turn worthless. Ask a producer what they clean with and you learn more about their attention to detail than any tour of the vineyard will tell you. The vineyard is where the wine is made; the taint issue is decided in the building next to it, and the industry has known as much for decades.
Telling the faults apart
Four things go wrong often enough to matter, and only the first is cork taint. This table is the short version of what a merchant runs through before agreeing that a bottle was faulty.
| Fault | What you notice | Does air help? | Returnable |
|---|---|---|---|
| Cork taint | Damp cardboard, or fruit that has simply vanished | No, it builds | Yes |
| Oxidation | Browning colour, bruised apple, sherry character | No | Yes, if the wine is young |
| Heat damage | Pushed stopper, sticky capsule, stewed or jammy fruit | No | Yes |
| Reduction | Struck match, boiled egg, burnt rubber | Yes, usually within minutes | No |
| Brettanomyces | Farmyard, leather, sticking plaster | No, it builds | Rarely, it divides opinion |
Oxidised and cooked wine, the faults you are most likely to meet
Oxidation happens when too much oxygen reaches the wine, through a failing closure, a long spell standing upright, or simply age. A white turns deep gold or brown, a red goes brick at the rim well before its years justify it, and both drift towards bruised apple and dry sherry. The wine is not dangerous, it is finished.
Heat damage is the one that stalks anything ordered online. Sustained temperatures above roughly 25 degrees expand the liquid, force the closure upwards and break the seal; you will find the cork proud of the glass under the capsule, or a sticky trail of seepage down the neck. The fruit that follows is stewed and jammy where it should be fresh. A delivery van left in July sun does more harm than a rough courier ever will, which is why what happens in transit deserves as much attention as your cellar. Once the wine is home, a steady temperature matters far more than a cold one.
Reduction and Brettanomyces, which are not always faults
Reduction is a shortage of oxygen rather than an excess. Sulphur compounds build up under the closure and the wine arrives smelling of struck flint, boiled egg or, at its worst, drains. The distinguishing feature is that it goes away. Pour it, wait ten minutes, swirl it hard, and if the aroma lifts and the fruit comes forward you had a reduced wine and you now have a good one. A stubborn case responds to a vigorous decant. Some producers, particularly in white Burgundy, court a little reduction on purpose.
Brettanomyces is a yeast that produces ethylphenols, and the descriptors run from leather and warm spice at low levels to farmyard and sticking plaster at high ones. Some drinkers read the first as complexity, others as dirt. It is a genuine spoilage yeast, closer to a bacteria problem than to a chemical contamination, and spoilage of this kind is a question of style as much as of hygiene. A wine carrying it is not defective in the way a corked wine is, and a merchant will rarely accept it back.
Can TCA be removed from a wine?
Not commercially, and not in any way that leaves the wine intact. There is one salvage trick worth knowing, and it comes out of the wine chemistry group at UC Davis, where Andrew Waterhouse showed that polyethylene absorbs trichloroanisole out of the liquid.
The method is crude. Take a length of ordinary polyethylene cling film, scrunch the plastic wrap into a loose ball, drop it into a jug with the wine, and stir for five to ten minutes before straining. Enough TCA migrates into the film to bring a lightly tainted wine back under the detection threshold. Two warnings: the wrap has to be polyethylene rather than PVC, and it strips out desirable aromas along with the taint, so what you save is drinkable rather than good. Use it on a weekday wine, never on something you were keeping. In our experience the original character never comes back in full, and a second pass through fresh plastic strips more than it saves.
Filtration and treatment at the winery are possible in principle. No producer wants that on a label, and the loss of quality is rarely worth it, so the practical solution remains prevention rather than repair.
Screw caps, synthetic corks and what they actually solve
A screw cap cannot suffer cork taint, because there is no cork. That is the whole of the argument, and it is a good one for wine meant to be drunk young. It is not a guarantee of purity: a cellar contaminated with tribromoanisole will taint wine whatever seals it, and closure liners have occasionally carried taint of their own.
The trade off has moved on since the debate started. Modern screw cap liners are made in several oxygen transmission rates, so a winemaker can choose how much air reaches the wine over time rather than accepting whatever a given piece of bark happens to allow. Synthetic corks or screw caps both remove the risk of cork taint outright, and technical corks such as Diam offer the same consistency with cork on the outside. Natural cork keeps its place at the top of the market, partly on tradition and partly because long ageing under it is still the best understood path we have. For a consumer the practical rule is short: buy the closure the producer chose, and judge the wine in the glass rather than by what seals it.
What to do with a bottle you believe is faulty
Re-cork it and keep it. A merchant or an importer assessing a claim wants to smell the wine and the closure, and a wine poured away is a claim nobody can settle. Note where it was stored and for how long, and photograph the cork if it has risen.
In a restaurant, say so before the glass is finished. Sommeliers meet cork taint often and will usually replace the wine without discussion. Buying online, contact the merchant with the lot details from the back label. Established sellers carry the loss on a genuinely tainted bottle; that is the accepted cost of selling under cork and it is priced in. What no seller will accept is a wine returned because the buyer did not care for it. Knowing what a region tastes like beforehand avoids that conversation entirely, and our guide to tasting is a better starting point than a returns form.
Questions drinkers ask us about corked wine
Is corked wine dangerous to drink?
No. TCA is unpleasant at concentrations far below anything that could harm you, and there is no known health risk from drinking a tainted wine. The only casualty is the pleasure.
Does cork floating in the glass mean the wine is corked?
It does not. Crumbled cork is a sign of a dry or old closure, or of a difficult extraction, and it says nothing about TCA. Fish the fragments out and judge the wine by its aroma.
How much TCA does it take before anyone notices?
Sensitivity varies widely between people. The most sensitive drinkers pick it up at one to two nanograms per litre, while others need five or ten before anything seems wrong. Two people can honestly disagree about the same glass.
Can a corked wine be sent back after a week?
Most sellers ask you to report a fault within a few days and to keep what is left in the bottle. Cork taint does not fade, so a wine that was tainted on opening is still tainted a week later, but the longer you wait the harder the claim is to support.
Are screw caps better than cork?
They are better at one specific job: no cork means no cork taint. For wine drunk within a few years they are the safer choice. For long ageing the case is less settled, and plenty of serious producers still choose cork.
