Cork TCA Infection Rates

Wes Hagen suggests a very high ratio of wines under natural cork enclosures are negatively affected:

So if 7% of wines are corked to where almost anyone can tell, (5ppt), we can assume a ten times higher incidence of .5 ppt taint, which could be as high as (gulp!) 70%. I believe the 25%-40% rate I mention is a very low and conservative number.

Full explanation here, which sounds quite convincing and damning:
http://clospepe.com/wes-blog/2014/8/4/842014-can-natural-cork-change-25-40-of-the-wine-it-touches

Wow. He better back up that “math” with some scientific research.

Convincing? Seems like a bunch of non-scientific assumptions and conclusions to me. IMO, TCA is not linear ( i.e. 7% at 5ppt does not equate to 70% at 0.5ppt).

However, since I believe that TCA occurs on some sort of a bell curve, I would tend to agree with him, despite any scientific proof from Wes (or me!), that a very high percentage of wines bottled with cork have some degree of TCA taint, with the vast majority being below most people’s ability to perceive it.

Its confusing to me this argument hasn’t been put to bed. Is the lab work prohibitively expensive? Or is it the more subjective impact of TCA below certain thresholds, which tasters can’t agree on?

My understanding is that the human olfactory system can detect TCA at lower levels than the standard chemical analysis. If there is a way to test for extremely low, non-detectable levels, I assume it is prohibitively expensive.

The reasoning behind the article makes no sense to me. I do believe that there’s probably often an undetectable amount of TCA which does affect wine aromas. Of course, the cork industry tells us TCA basically doesn’t exist anymore, Spectator is full of crap, and we’re killing the cork forests and wildlife if we don’t drink cork-sealed wines. That all makes sense, right? [snort.gif] newhere

Three corked bottles in the last two weeks, so I’m on Wes’ side.

I have been asking the same question for about a decade on wine BBs. The only dollar estimate I have gotten is something like $250 per bottle for a test. I have no idea whether that is accurate or still valid.

Regarding testing, in addition to the cost, how are you going to choose your sample and sample size? The number of corks produced every year must be in the millions. You might need to test a few thousand to have a statistically valid sample. That’s a WAG but consider political polls that sample thousands for a +/-3% error.

Which two or three thousand corks do you choose for testing? There can’t be too many from the same batch or possibly even the same producer, because of the risk of cross-contamination.

Well, you might need a few thousand randomly distributed bottles if you were wanting to find the exact rate of TCA contamination. But if we’re talking of checking Wes’s suggestion, then it’s only a matter of 7-10% vs 40-50% contamination which should be much easier to check. Technically, at such high rates, you should be able to walk into your regular wine shop, pick up 20-30 random bottles (preferably from different regions & different producers), you could even pick up the cheapest ones as one might argue they have the highest chance of being contaminated. And check. This will give you a ballpark figure which will correlate with one side or the other…

Chuck,

Your comment may be right on. Many years ago when I used tree bark to seal my wines in bottle, I asked my cork supplier why they didn’t use dogs to pick out the TCA-affected corks. He said they tried that, but that the dogs found it in nearly all the corks!

Peter Rosback

Sineann (ITB)

Mass Spec can detect down to as low as 1ppt, to my knowledge (maybe better). Human threshold is supposedly in the range of 2-6ppt. The real question is if wine below the detectability threshold are affected in a way that anyone can perceive (i.e., taste). I know there are claims that this is the case, but I haven’t seen a study where a bottle was split in two, then half purposely contaminated with an amount of TCA below the threshold, and a blind tasting done.

I think we paid $150 or so per test from ETS labs. Their price list is online.

I have often heard 2 ppt as a threshold for TCA, I don’t know where that comes from. Some of these numbers come from the cork industry, which of course has a motive to inflate this number (the higher the threshold the better their product, kind of). We received a shipment wine bottled with an agglomerated cork in which every bottle seemed to me and my colleagues slightly corked; we had the wine analyzed and it was at 0.5 ppt TCA. At least some of our customers recognized the wine as corked, too. So 2.0 is way too high, at least for this group of tasters. (And this was for recognition as TCA; the ‘difference threshold’ would of course be even lower.)

Back on eBob years ago, someone wondered why there isn’t a simple test strip for TCA, like the ones you use to monitor chlorine in a swimming pool. The reason is that swimming-pool chlorine concentration is measured in parts per million, whereas TCA can be detected by humans in the range of single-digit parts per trillion (as pointed out above). A trillion is a million million… a TCA test has to be a million times more sensitive than a chlorine test strip.

One more thing: TCA contaminates wine, it does not “infect” it.

That’s interesting. So what is the lower limit of detectability in the lab these days? This app note shows detection down to 0.1ppt under ideal conditions, but it may be out of date.

http://hpst.cz/sites/default/files/uploaded_files/si-02373_-_ultra-trace_analysis_of_246-trichloroanisole_in_white_wine_using_automated_solid_phase_microextraction_spme_and_the_varian_240-ms_ion_trap_mass_spectrometer_and_vresultstm_gcms_software..pdf

I don’t know. Somewhere at 0.5ppt or lower.

Glad to see some good conversation going on here.

My comments were not scripted and I spoke from my heart and 20 years in the business, and 18 years as a professional wine judge.

What I have noticed is that the ‘bottle variation’ I used to notice on my wines under natural cork (Grade 1A Flor, almost $1 a cork) was almost expected. Tasting 12 bottles from a case under natural cork, I would notice each bottle would show a different character. Some had muted fruit, some weren’t as expressive in the mouth.

When I switched to Stelvin and DIAM, this variation ceased to exist. The only thing I changed were the closures.

Do I want to spend $15,000 testing 100 random bottles so I can provide data to a bulletin board? Not really.

Natural cork has failed me as a tool to a craftsman, and I currently reject natural cork as a reasonably safe closure for fine wine.

And thanks for the clarification of contaminated/infected. I do appreciate it.

This was a purposefully provocative video post to get people talking, and I’m glad we are.

Again, this is my opinion and my experience of dealing with and evaluating wines under various closures throughout my career.

I agree with you, Wes.

This to me is the scariest part not the tainted bottle that is easy to pick out

As someone who has switched to 100% twist off closures, I obviously agree with the sentiments expressed by Wes. I do have a couple of other thoughts that have come to me over time:

  1. Per a recent blog by Jim Laube at Wine Spectator, the % of corked wines that they experience is running in the 4% range, while the range some years back was more in the 7-8% range: How Often Did Wine Corks Fail in 2013? | Wine Spectator This is but speculation on my part, but I think that to an extent alternative closures are already providing a valuable service, by forcing cork manufacturers to get their shit together and reduce somewhat the percentage of tainted closures. Having competition is good.

  2. All of the tests that I know of, including the ones that we have done, are on bottles pulled off the line and kept in pristine conditions. I’d love to see numbers run on consistency (or lack thereof) of oxygen ingress, etc. on bottles that have actually been shipped by distributors, banged on wire racking at the retailer, etc. That might make cork more appealing or may make it less appealing, I don’t know. But it would certainly be a more accurate measurement of what the consumer is getting.

Adam Lee
Siduri Wines