Using Analytical Chemistry To Put an End to Corked Wine

The sound when the cork “pops” out of the bottle is satisfying isn’t it? But what disappointment if at that very first sip the look of pleasure and anticipation turns to disgust when the off flavors of a faulty wine hit your tongue – that’s if it even gets past your nose.

There are seven common faults that can be identified in wines, although a limited degree of some may be appreciated as “pleasant” by some consumers. These faults include oxidation (in excess), reduction, Brett (caused by Brettanomyces yeast), excess sulfur dioxide (SO₂) (naturally produced during wine making and used to stabilize wines), volatile acidity, out of condition (over-age, bad storage) and last but by no means least 2,4,6-trichloroanisole (TCA), more commonly known as cork taint.

In the article we are going to focus on cork taint – a fault not appreciated by anyone in the wine production, retail or consumer chain.

What is cork taint?

Cork is a natural product, stripped bark from the cork oak (Quercus suber). Consequently, as with any food or beverage product, good agricultural practice is necessary to minimize contamination by unwanted microbes during processing, transport and storage. Because of the nature of cork’s source, typical contaminants include soil-borne bacteria, yeasts and most commonly fungi. It is a subset of these fungi that are associated with cork taint. Species most frequently associated with cork taint include isolates from the genera Trichoderma and Fusarium, although other fungal species are also known to be involved.

To defend itself from fungal attackers, the tree produces phenolic compounds. In retaliation the fungi defend themselves by methylating the phenolic compounds which makes them less toxic, resulting in compounds such as anisole. When anisole then comes into contact with chlorine, which is frequently used as an antimicrobial during cork sterilization, it is converted to TCA – et voila! If an affected cork is then used to stopper a bottle of wine, the compound is transferred into the wine as it comes into direct contact.

“Cork taint is a harmless to health, naturally-occurring fault in bottles of wines closed by corks. It results in a very unattractive, moldy, wet cardboard smell that additionally reduces the fruit character of wine (it has nothing to do with pieces of cork breaking off and being seen in the bottle or glass). It can be detected by tasters at very low concentrations” commented David Way, Wine Qualifications Developer at the Wine and Spirit Education Trust (WSET). Whilst the characteristics of cork taint have been described as a fault for decades, it wasn’t until the 1980’s that the compound responsible for taint was actually identified^{1, 2}.

Even as recently as 10 years ago, cork taint was estimated to affect 2-7% of wines produced^{3, 4}; not only disappointing for the consumer but resulting in economic loss for producers and retailers. That’s aside from the sacrilege of potentially losing rare fine wines. However, improvements in cork and winemaking methodology, including minimization of chlorine use in cork processing and more broadly in the winery, have seen these values plummet in recent years. The Portuguese Cork Association (APCOR) estimates that less than 1% of new wines are now affected.

Detecting cork taint

A group of trained wine judges were put to the test and a geometric mean of the minimum detectable concentrations of TCA determined at 4.6 ng/L. Levels in the wines themselves below this should theoretically be undetectable to consumers and therefore offer a maximal threshold for acceptable limits of cork-to-wine transfer.

A host of laboratory-based analytical techniques can be applied to wines to determine if they contain TCA. Headspace solid phase microextraction (SPME) in combination with gas chromatography-mass spectrometry/ selective ion monitoring (GC/MS-SIM) has been shown to work well in detecting TCA in wines. SPME has also been used on wines and cork material to detect TCA at levels beyond olfactory detection. Headspace SPME and heart-cut multidimensional gas chromatography with tandem mass spectrometric detection has also proved effective. However, from a production point of view, every cork is an individual, so short of uncorking and testing every wine before it is sold, these techniques are not helpful in preventing cork taint in the first place.

How can cork taint be prevented?

In terms of whole, natural corks, the primary focus of efforts has been on identifying corks that contain TCA to prevent them reaching the bottling line.

A non-destructive technique called the “dry-soak method” was and is still used in some places to detect individual corks containing TCA. Here, every large format cork is held individually in a sealed glass jar containing 5-10 drops of de-mineralized water. Over 48 hours, the moist environment volatizes the TCA and then the corks are “sniffed” by a human expert panel. A study comparing this technique to chemical analysis by GC-MS found the dry-soak method effective in detecting tainted corks. However, this technique is labor intensive and due to sensory fatigue, only around 200 corks can be sniffed in one sitting, stopping for breaks every 50 corks. It may be effective, but this is therefore certainly not a method suited to a high throughput environment.

Given the sheer volume of corks used in the wine industry there is clearly a need for a quantitative, automated system for accurately detecting and rejecting affected corks.

A number of groups in both academia and industry, fostering collaboration between chemists and the cork industry, have successfully sought non-invasive, sensing device solutions that detect volatile organic compounds (VOCs), including TCA, on individual corks, rapidly.

A multi-center EU-funded project successfully developed an “electronic nose”, consisting of a sensor array, capable of detecting TCA down to 2 ng/L at a rate of 250 corks per hour. Whilst promising, the project has however not produced a solution available to producers. “Whilst they can be useful technology, electronic noses lack the sensitivity required to detect TCA at the levels required and as such don’t provide a solution for detecting affected corks currently” commented Professor Ulrich Fischer, Head of the Institute for Viticulture and Oenology, Service Center for Rural Areas (DLR) Rheinpfalz, Germany.

A partnership was able to create a similar GC-based system with an analytical detection limit of a mere 0.5 ng/L, well below that detectable by a consumer. Each cork is checked individually with the “cork sniffers” currently able to test one cork every 16 seconds, each typically checking around 34,650 corks a week. But the goal is one every 10 seconds!

An alternative based on gas phase spectroscopy, also a result of industry-academia collaboration, analyzes corks one-by-one in an automated system and can be used on washed and unwashed corks. It too has a limit of detection of 0.5 ng/L and takes just 5 seconds to analyze each cork.

“Seems like a huge waste of natural resources” you may be thinking but fear not! Rejected corks can still find a useful life in applications such as flooring and gaskets.

An alternative approach is to prevent the formation of TCA in the first place by sterilizing and reconstituting the cork. “The cork industry has responded with two main approaches – cleaning cork with steam or creating closures with recomposed cork particles that have been cleaned and reconstituted with plastic. These responses are probably what has led to the reduction in corked bottles in recent years and give wine consumers the familiar ritual of opening a bottle with a corkscrew” commented Way.

“Corks fragments that have been sterilized with supercritical CO₂ and “glued” back together have been a huge success in the wine industry. As the compounds used to reform them do not contain plasticizers they have been well received and accepted by many as a good and safe alternative to whole natural corks” commented Fischer. He continued, “Reconstitution also enables manufacturers to control the porosity of the corks to oxygen so they can be tailored to different types of wine. A red Bordeaux for example may benefit from a cork that enables greater oxygenation to develop its character, whilst a Riesling from Germany may benefit from less oxygenation to retain the floral aromatic character of the wine.”

Recently, a Swiss group have developed a membrane that is able to filter TCA out of wines. Fischer continued, “This is an exciting development for wine producers and suppliers. Whilst not economically viable for wines at the cheaper end of the market, it ensures rare and expensive wines are not lost and means suppliers are able to provide the vintages their customers want.”

If corks are a problem, then why do we use them?

Plastic corks and screw cap closures have gained popularity in many parts of the world, particularly in New World wines. However, they are still fallible with compromised seals resulting in leakage and spoiled wines from excessive oxidation.

Synthetic closures avoid the hazards associated with TCA, but they lack the microporous properties of cork that allow minute quantities of oxygen into the bottle. Whilst excessive oxygenation is undesirable, small quantities of oxygen are a necessary part of the bottle ageing process that help to develop the tertiary characteristics of the wine – think savory, meaty notes in a red, or honeyed, dried fruit in a white wine.

Therefore, whilst rubber corks and screw caps may be a good solution to protecting wines bottled for immediate consumption, they are not appropriate for premium wines that are designed to be bottle matured for a number of years.

Non-traditional corks are a very much contentious subject within the industry and the strength of opposition such that in some regions any closure except natural cork is banned under local wine production regulations.

“The main approaches to dealing with cork taint have been either to use other types of closure, especially screw caps, or to ensure that corks do not contain TCA. Screw caps have been widely adopted in Australia and especially New Zealand (among many other countries) and have been accepted, especially for inexpensive and mid-priced wines in some markets, for example, the UK. There has been less take up by wine makers in Europe and less acceptance of alternative closures in some major market, such as France, Italy and the USA (although attitudes are changing in the last named)” concluded Way.

The future of bottle closures

Researchers have been developing new cork-free screw caps that incorporate a breathable liner, enabling micro dosing of oxygen into the bottle. It is still however early days and they have not been widely adopted so it is unlikely you will pick one up off the supermarket shelf anytime soon.

Aside from their physical and chemical properties, studies have shown that many consumers prefer natural corks and it impacts their perception of quality, an important consideration for wine producers and retailers wishing to get the best price for their wines.

Despite the potential shortcomings of natural cork. It is worth considering the “green”, biodegradable and renewable nature of cork over alternatives. Cork oaks grow for around 25 years before the first harvest, benefitting the environment and providing habitat. Once harvested – a skillful job, the bark of the cork oak regrows too, and can be harvested every 8-14 years, offering long-term productivity.

References

1. Tanner H, Zanier C, Buser HR. 2,4,6-Trichloranisol: Eine dominierende komponente des korkgeschmacks, Schweiz. Z. Obst-Weinbau. 1981;117:97-103.

2. Buser H, Zanier C, Tanner H. Identification of 2,4,6-trichloroanisole as a potent compound causing cork taint, J. Agric. Food Chem. 1982;30:359-362.

3. Silva Pereira C, Figueiredo Marques JJ & San Romão MV. Cork taint in wine: Scientific knowledge and public perception — A critical review. Critical Reviews in Microbiology. 2010;26(3):147-162. doi:10.1080/10408410008984174

4. Butzke CE, Evans TJ, Ebeler SE. Detection of cork taint in wine using automated solid-phase micro extraction in combination with GC/MS-SIM. Chemistry of Wine Flavor, 1998;15:208-216. doi:10.1021/bk-1998-0714.ch015