Theoretically, plastic can give or take TCA
Natural cork takes a lot of the blame for cork taint, prompting me to ask Andrew L. Waterhouse whether it is possible for a screwcap to transfer trichloroanisole (TCA) to a bottle of wine.
“This is theoretically possible for any plastic material that comes in contact with wine,” he said, “but this is also true for natural cork, as both materials bind to TCA and can release it later.”
However, he added, “The converse is also true: The closure—screwcap lining, synthetic or natural cork—could absorb TCA out of a wine contaminated in production.”
In theory, that means that the TCA level could be lessened in the wine, he said, although he wasn’t aware of any direct study assessing such an occurrence.
Andrew L. Waterhouse has a long background in chemistry, with a bachelor’s degree from the University of Notre Dame and a doctorate from the University of California, Berkeley. But his exploration of wine chemistry began when he joined the Department of Enology and Viticulture faculty at the University of California, Davis, in 1991, and he’s gone on to become internationally recognized in the field.
Waterhouse has published a number of articles related to his research of phenolic compounds, how they relate to the taste of wine and their possible health effects for wine consumers. As part of his research into the effects of oxidation on wine chemistry and quality, he is conducting experiments to determine the oxygen-transfer rates for various types of wine closures.
In addition to being a professor of enology at Davis, Waterhouse is an honorary professor at the University of Auckland in New Zealand, an associate editor of the American Journal of Enology and Viticulture
and is on the editorial board for the Journal of the Science of Food and Agriculture
Wines & Vines:
How did you become interested in researching the oxygen-transfer rates for various types of closures?
Some years ago I was consulting on a wine company quality problem that appeared to be inconsistent pinking. In those discussions, the internal insurance program manager told me that she had no way to anticipate problems. She said, “Someone just calls and says this or that wine is bad, and I have to cover the losses.” She had no means to find out if a particular wine or type of wine is more or less fragile, or liable to losses. I thought that was a particularly problematic way to run a business and have a long-term goal of providing wine companies with some means to evaluate shelf life.
We are interested in wine oxidation chemistry, and oxidation is key to aging. Since the OTR controls the amount of oxygen available to the wine, that variable is most important to the aging process.
Please describe the research you’re conducting that involves bottles of Sauvignon Blanc and the results you’ve seen so far.
To test closure consistency, we bottled a Sauvignon Blanc wine in March 2012 with natural and synthetic corks and screwcaps. Cade in Napa Valley provided the wine. We are analyzing the color of every bottle by looking through them using a spectrophotometer to precisely measure it. Colleagues in Australia have shown that color darkening is related to wine oxidation, and we plan to sort out the bottles with the largest and smallest oxidation for each closure. Those will be opened and tested for chemical changes as well as sensory differences. We would like to know if wine consumers can taste the difference between closures of a specific type in order to assess closure consistency.
We had hoped to bottle the wine with minimal SO2, but that wasn’t possible because the bottling was also for commercial purposes. So the wine was bottled with the standard level of SO2 and is well protected from oxidation. We were hoping to see results in a year, but we have seen no changes in any of the bottles. I hope we start seeing some of the bottles darken during this next year.
You are also conducting a study with empty bottles. Please describe that research and the results you’ve seen so far.
We test the closures by filling the bottles with nitrogen gas to remove all the oxygen. The bottles are then sealed with the closure, and we start measuring the oxygen level almost immediately. We follow the amount closely for several days, then taper off in the frequency of measurement.
We measure the oxygen level inside the closed bottles by using small dots that change their luminescence depending on the amount of oxygen present. We focus a blue light on a dot right through the glass of a wine bottle, and the Nomasense device flashes the blue light hundreds of times per second, and between flashes (it) measures the red luminescence. The rate of decay of that red light is related to the oxygen level around the dot. The advantage for this experiment is that the same bottles can be tested repeatedly. Older technology would have required opening a bottle for each measurement.
We are still running these tests, but it seems that at bottling, there are differences between closures in all cases. The initial oxygen levels vary between 50 and 100 parts per billion. We attribute that to the mechanics of bottling. After that, it seems that the oxygen accumulation is quite consistent for screwcaps and synthetics, but naturals show different slopes. In other words, a few of our dozen naturals show oxygen-transfer rates that are significantly higher than the rest of the batch. It seems that there’s a natural variability in natural corks in terms of the amount of oxygen they transmit across the cork.
Considering the initial oxygen levels you’re finding in these tests, do you think that some closures may be getting a bum rap for oxygen transfer when, in fact, some of the fault may be wit h the bottling process?
That can certainly happen. There’s experimental data showing that different levels of oxygen at bottling can make a wine taste different six months later. If you have a bottling line that’s not properly maintained or managed, you could in theory get bottle-to-bottle variation. It’s very hard to do an experiment on that. It’s a big machine that’s complicated. How do you set it up to run badly? It’s not an experiment for a chemist.
But it’s also becoming acutely clear that for certain wines you really need to select the closure carefully. Some wines really are demanding; some wines are much more flexible, particularly over the short term.
Do you think there’s an optimal closure for each type of wine? When do you think there might be enough information available to make recommendations to winemakers?
There is certainly an optimum closure for a particular wine, and that depends on the wine’s composition at bottling, the flavor profile desired at opening as well as the time frame for consumption, as wines evolve with time. In some cases, the time frame is highly variable, so this can be complex, but in many cases the wine can be expected to be consumed relatively quickly after purchase, so it is very feasible to make these estimates, presuming the winemaker knows how to predict an aging trajectory. In some cases, such as Riesling, the wine can be very responsive to oxygen, while with Cabernet Sauvignon the wine seems impervious to oxygen, at least over a limited time frame. As the time to consumption becomes short, the effect of closure shrinks, so that for fast turnaround wines, the oxygen-transfer properties of the closure are not so important.
You’re using CT scanning to assess the permeability of natural cork. Do you have any results yet? What application could that have in an industry setting?
We hope to relate oxygen-transfer rates with internal structure, but those analyses by our collaborator in the medical school have not been done yet, as we just recently obtained the oxygen level data. This work is really in its infancy, and I am not an expert in CT scanning. I am not willing to project any applications at this point.
A resident of the Santa Cruz Mountains, Laurie Daniel has been a journalist for more than 25 years. She has been writing about wine for publications for nearly 15 years and has been a
Wines & Vines contributor since 2006.