Issue Date: March 28, 2005
NMR Method Detects Spoiled Wine In Unopened Bottles
The last thing a wine collector wants to hear is that the bottle of wine for which he or she just paid a small fortune is nothing more than fancy vinegar. Unfortunately, determining whether wine has spoiled usually requires opening the bottle, which drastically reduces, if not destroys, its value.
A solution to this problem is now offered by Matthew P. Augustine, an associate professor of chemistry at the University of California, Davis, and graduate student Daniel Sobieski. In San Diego, they described an NMR spectrometry method with which they can detect spoilage in wine without opening the bottle. Using a special NMR probe, they place the entire bottle directly in the magnet.
Wine commonly spoils in one of two ways. The ethanol can be oxidized to either acetic acid or acetaldehyde. Wine is considered spoiled if it contains acetic acid at 1.4 g per L or acetaldehyde at 300 mg per L. The second type of spoilage--so-called cork taint--is caused by mold in the cork generating the compound 2,4,6-trichloroanisole (TCA). About 5 to 10% of all bottles of wine are spoiled by one or the other of these mechanisms. The NMR method detects only spoilage caused by oxidation.
One of the challenges in detecting spoilage is that wine is about 88% water and 12% ethanol. Wine that has gone bad contains about 0.2% acetic acid or 0.02% acetaldehyde. Augustine and Sobieski have demonstrated that they can easily detect those low levels in spoiled wines.
What limits the sensitivity of detection is time, Augustine said. The sensitivity depends on how much time is available to average the signals from the contaminants, he explained. At first, the analysis was time intensive, requiring as much as an hour to properly adjust the magnet to get a good magnetic field and to calibrate the radio-frequency pulses. Eventually, Sobieski trimmed the analysis time to less than five minutes per bottle.
The difference in the chemical shifts of the methyl protons in acetic acid and ethanol is about 1 ppm, which is enough to distinguish between the two compounds easily. Because the instrument that Augustine and Sobieski use can't distinguish the methyl protons from acetic acid and acetaldehyde, they use the aldehyde proton, which has a chemical shift of about 11 ppm, to detect acetaldehyde.
Augustine has shown that the truism that appearances can be deceiving even applies to wines. Two bottles of wine he had tested make the point: one with a cork that appeared to be cracked and leaking and the other with a cork that appeared to be in good shape. When the wines were tested using NMR, however, the apparently good cork was on a bottle that actually contained spoiled wine. He suspects that the sugars in the leaking wine crystallized and resealed the bottle with the damaged cork. "You can't look at something on the outside and tell something about the inside," he said.
Cork taint is harder to detect than oxidative spoilage because humans are sensitive to as little as 10 ng per L of TCA, according to Augustine. "There's no way that liquid-state NMR is going to be able to see 10 ng per L," he said. To detect cork taint, they are developing a method that would use a vacuum to pull a flow through the cork and sublime TCA, which will then be detected by gas chromatography/mass spectrometry.
Augustine is setting up a company called WineScanner, in Morristown, N.J., to test wines in the auction market. Based on the amount of wine produced annually and the fraction that tends to be cellared for an average of 20 years, Augustine thinks the market for testing could be large.
"If you can see acetic acid in a full bottle, you have a method for screening bottles and telling people who want to buy a bottle of $25,000 Thomas Jefferson wine whether they're buying vinegar," Augustine said.
- Chemical & Engineering News
- ISSN 0009-2347
- Copyright © American Chemical Society