09 July 2008

Screw it

[cork]In some quarters, the long-running closure debate rages on. It shouldn’t. It’s almost over.

How can I make this claim? Do I have access to some new research that no one else has seen? No. Here’s the deal: the screwcap is almost universally acknowledged to work flawlessly (or very, very nearly so) over the short term. Most wine is produced for, and consumed in, the short term. Thus, there is no logical reason for any wine made for near-term consumption to be closed with anything but a screwcap. The only resistance remains adherence to tradition (though this must include the wonderful “tradition” of a taint rate that runs between 2 and 8%, depending on the sample) and fear of the market, especially in Europe. Both understandable reasons, despite the science, but the tide is turning even on Old World shores, and more and more short-term wines are being shipped under screwcap, especially for export markets.

Thus, the entirety of the unanswered question rests on the shoulders of long-aging wines. And here, one can forgive any producer for being confused. There’s an ever-growing body of evidence (most of it from Australia, though that’s changing) that white wines age beautifully under screwcap. And, in fact, where one sees adoption of the technology in the Old World, it’s mostly for longer-aging whites: rieslings, white Burgundies, etc. (One wonders if the market-killing premature-oxidation problem among white Burgundies might be solved in a stroke by the adoption of a different closure. It certainly couldn’t hurt, at this point.) For red wines, the anecdotal evidence is thinner, and certainly more long-term studies would be extremely valuable.

But there’s also a pair of unanswered questions: how much oxygen does a wine need to age, and when does it need it? (It’s useful to define our term here: “age” would, to most people, mean “age in a way similar to that of the best cork-finished wines to which we’re accustomed.” If screwcaps allow 10-year wines to go 40 years before they achieve a similar end-state, that’s to the credit of the closure, but less useful to the consumer without a multi-generational cellar from which to draw.) The problem for winemakers is that the research on this point is cloudy and contradictory.

It has long been believed, by some, that corks allow a gradual transfer of oxygen. “Not so,” said cork manufacturers in the past, when they wanted to assure people of the efficacy of their products despite persistently unmanageable cork taint. And “not so,” said the inventors of synthetic corks, who – after research – determined that the “perfect” seal was what they wanted to emulate (though it’s worth noting that all synthetic corks fail over the short- or medium-term, breaking their seal and allowing oxygen to enter the bottle). More recently, a groundbreaking Australian study came up with an answer of “not exactly,” demonstrating that while some corks did indeed prevent the transfer of oxygen, others allowed it at greatly varying rates. And most recently, a study in Bordeaux found that corks do allow oxygen ingress, at a much less variable rate than found in the Australian study.

The Bordeaux study was funded by a cork producer, and thus its conclusions need to be viewed with mistrust until verified by independent study. The safe conclusion is that there’s no actual conclusion, as yet, from the research. But I think we can do better than that by simple thought experiments, while we wait for the research to achieve surety.

We absolutely know one thing about oxygen: a lot of it is very bad for a wine. Open a bottle, pour half of it into a wide-bottomed decanter, and let it sit. How does it taste after a day? A week? A year? Twenty years? That’s a bottled wine in the presence of a lot of oxygen.

And we know one more thing about oxygen: very little of it does very little. “Do you want to open the bottle and let it breathe?” is, as most wine folk know, a useless question: the dime-shaped portion of wine exposed to air in the neck of an uncorked bottle does nothing to change the wine in the times usually under consideration. This is why young wines are sometimes decanted: to accelerate the effect of oxygenation.

Or consider old bottles. Which are, at auction, the most valuable, controlling for vintage, producer, and cellaring conditions? The ones with the least ullage (the space between the bottom of the cork and the top of the wine). It is believed, and supported by nearly all the tasting evidence available to us over the decades, that low-ullage bottles (that is, those that are closest to their original fill) are the best-preserved; there are occasional exceptions, but they’re rare. High-end producers, though not with closures on their mind, support this notion during their periodic forays into bottle reconditioning, wherein an old wine is quickly uncorked, refilled with a quantity of that wine from their cellar or other bottles of the same wine (or, in some cases, a younger wine), and then sealed with a fresh cork. What purpose to this practice other than to reduce the nefarious influence of excess oxygen, and to replace a closure with a high rate of physical failure before that failure becomes catastrophic?

Moreover, if there’s ullage in a bottle, then something is getting out. And not just oxygen. This is physical evidence that some corks are massively permeable (or allow transfer between their surface and the interior of the bottle; an important difference if you’re a cork producer, but a completely unimportant difference if you’re the producer or owner of the bottle in question; either way, the closure has failed to preserve the wine).

[disassembled screwcap]So we know a lot of oxygen is bad, and a tiny bit of oxygen is meaningless over the short term. The questions, then, are: what about a tiny bit of oxygen over the long term, and what about no oxygen at all?

Some wine chemists have long argued that aging is an anaerobic process, that wine doesn’t need oxygen to do most of what it’s going to do. Newer thinking on this question is a little less certain, and suggests that the tiny amount of oxygen trapped in three places: 1) the wine, post-bottling, 2) the headspace (the gap between the wine and the bottom of the closure) [edit: it's worth noting that this oxygen is usually forced out; a process known as sparging], and 3) the cells of the cork itself, might be all the oxygen a wine needs to age. In practice, steps are taken to remove much of this oxygen in question at bottling, but some remains. However, if the cork is shedding oxygen into the bottle, then a cork really doesn’t provide a barrier against oxygen; if oxygen can move across (say) 35% percent of the cork, what’s to stop it from moving across 100% of the cork, other than the vagaries of the cork’s cellular structure? (I don’t want to dismiss this last point too easily. Cork is from a tree, not a lab, and as such will always exhibit variation; the “natural” quality touted by producers is also the principal source of its physical variability.)

But what about no oxygen at all? Can wine age in a hermetically-sealed container? How about one in which an amount of oxygen similar to that of a cork-finished wine is provided? We just don’t know yet, but early evidence suggests not. We do have some evidence that closures that permit absolutely no oxygen transfer (screwcaps with certain types of liners, for example) can lead to reductive wines, the eventual fate of which we don’t yet know. This problem seems to be preventable by changing a few minor aspects of bottling chemistry (that is to say…and with no little irony…modifying techniques that were designed to protect a newly-bottled wine from the effects of excess oxygen). But this can happen to cork-finished wines as well, and that it doesn’t do so 100% of the time is yet more evidence that corks are variable in their ability to move oxygen around.

We also know that, despite the claims of the Bordeaux study, corks can vary a lot in their ability to let stuff in and out, based again on the evidence of variable ullage in old bottles. Whether or not that study was funded by cork producers and tainted (no pun intended) as a result, the simple fact of the matter is that it only covered three, not thirty or more, years of aging. Minor variations in oxygen transfer can and, sometimes, do become major ones over long periods of time. And as I’ve already noted, the effectiveness and qualitative supremacy of screwcaps over the short term is well-established, so it’s really only these long-aging wines about which we’re currently concerned.

So if we don’t know, but suspect, that an absence of oxygen is no good, at least we do know that as little oxygen as possible seems to be ideal…based, as noted before, on the ullage in old bottles. The choice before us is this: a little oxygen at bottling plus a little oxygen over a long period, or a little oxygen at bottling plus no oxygen at all over a long period.

Given this, isn’t the only question the one asked earlier? How much oxygen does a wine need to age, and when does it need it? Once this is known, and given the continued existence of both cork taint and physical cork failure and/or variability, what reason – other than tradition and fear – is there to continue to use “natural” cork? For – and this is a major point – screwcap liners can be selected for permeability. Once we know “the number” (which may be different for different types of wine), we can tailor closures for that result; something that we’re unlikely to be able to do with corks, unless they cease to be a natural product.

There’s no blame to be assigned to a winery who wants to wait for the evidence before making a switch. Were I making a pricey wine with a reputation to preserve, I’d probably do the same, despite the large number of New World producer’s who’ve made the change already, and are betting their futures on the screwcap. But the evidence will eventually come, despite a several-decade interim period…a period in which we will continue to suffer cork taint, physical cork failure, and cork variability. At some point, this question will be decided, and the screwcap – though by that stage it could just as easily be another alternative closure with similar properties – will emerge the scientific victor.

Cork & screwcap photos used under the terms of a Wikimedia Commons license.


Anonymous said...

"How much oxygen does a wine need to age, and when does it need it? Once this is known, and given the continued existence of both cork taint and physical cork failure and/or variability, what reason – other than tradition and fear – is there to continue to use “natural” cork?"
Ah, you're an optimist! I don't know that it will ever be "known". There will be many opinions, and many studies with conflicting results, and committed parties will argue the toss for at least the next 30 years.
Meanwhile, we'll taste wines cellared side-by-side with different closures, and make up our own minds. The minefield of cork inconsistency is known. The measure of the success of the screwcap will be that the wine tastes as good as the best bottle under cork. In fact, I might even swap the curse of 'bottle variation' for all twelve of the screwcapped bottles tasting like the second-best-tasting bottle from the cork-sealed case...

thor iverson said...

Not so much an optimist as someone who has faith that, because it's in the commercial interest of very profitable wineries to have this information, the necessary money will be spent to obtain it. In other words, I have faith in the desire for increased profitability.

Anonymous said...

I agree with about the debate regarding the need for oxygen for the aging of wine. I have argued this point with Michael Pronay: the deposition of sediment, the result of tannin polymerization, is a well-defined oxidative process that needs a fairly strong oxidant such as... oxygen or hydrogen peroxide. The problem, of course, is the complexity of the system and the relative paucity of analytical tools for addressing the question. I wonder, though, if someone couldn't follow the conversion of SO2 (bound and free) to sulfate as a proxy for oxygen ingress and correlate that with wine evolution.
I'm sure that some bright soul will eventually figure the problem out.

Mark Lipton

thor iverson said...

All you need is a sufficiently large grant. ;-)