Corkiness-related off flavor compounds found in wine: SIM chromatogram of 1.0 ng/L 2,4,6-trichloroanisole (TCA).
Corkiness-related off flavor compounds found in wine: SIM chromatogram of 2.4 ng/L 2,4,6-tribromoanisole (TBA).
  Limit of Detection Oder tresholds 
2,4,6-Trichloroanisole (TCA)   0.3-0.5 ng/L  1.4 - 4 ng/L
2,4,6-Tribromoanisole (TBA)   0.5 ng/L  3 - 8 ng/L
2,3,4,6-Tetrachloroanisole (TeCA)   1.1 ng/L  4 - 24 ng/L
2,4,6-Trichlorophenole (TCP)   1.4 ng/L  4000 ng/L
2,4,6-Tribromophenole (TBP)   1.6 ng/L  
Pentachlorophenol (PCA)  0.9 ng/L  4000 ng/L
Limits of detection and odor thresholds of the corkiness-causing compounds determined using SBSE-GC/MS.
Calibration curve for 2,4,6-trichloroanisol (TCA): Limit of detection: 0.39 ng/L; Limit of determination: 0.79 ng/L
Calibration curve for 2,4,6-tribromoanisol (TBA): Limit of detection: 0.50 ng/L; Limit of determination: 1.0 ng/L
  File TCA [ng/L] 
Day 1   1511041  6.1
   1511042  4.9
   1511043  4.7
    1511044  4.5
Day 2   1511045  8.5
    1511046  6.4
    1511047  5.4
Day 3   1511048  7.4
    1511049  6.1
    15110410  5.3
Standard deviations under real laboratory conditions. A 1.5 L sample of water used to wash cork material was homogenized and ten separate aliquots were extracted using SBSE (GERSTEL Twister). GC/MS analyses were performed over three consecutive days. Mean value: 5.9 ng/L; standard deviations = 1.26 ng/L; Minimum value 4.5, maximum value 8.5 ng/L.

Off Flavors in Wine: Corky

Efficient and sensitive determination of TCA and other off-flavors

When your premium wine tastes corky, it is little consolation that this is not caused by the natural cork material used to produce the classic wine stopper. Corkiness points to the presence of 2,4,6-trichloroanisole (TCA), the most well-known malodorous culprit. But other chemical suspects are at large that can equally cause the unpleasant musty, moldy off-odor assault to your nose, and these may not even be coming from the cork stopper. To ensure an efficient, reliable and sensitive determination of all corkiness-related off-flavor compounds, the DLR Mosel in Germany successfully turned to GC/MS combined with Stir Bar Sorptive Extraction (SBSE).

You don’t have to be a sommelier or wine expert to tell the difference between a perfect wine and a corky wine. 2,4,6-trichloroanisole (TCA) has an extremely low odor threshold. As little as a few nanograms per liter of air is enough to detect an unpleasant musty off-flavor. In water and wine it is a similar story with odor thresholds at 0.3 ng/L and 1.4 ng/L respectively, “but that is only a theoretical value”, says Horst Rudy from the Agricultural Service Center (DLR) of the Mosel wine region in Germany. “After all”, the laboratory manager explains, “sensory perception is highly individual and very subjective: While one consumer may sense no problem even at much higher concentrations, another person’s olfactory bulbs sets off the mal-odor alarm at concentration levels as low as 0.5 ng/L”. In addition, several factors could influence sensory perception of the off-flavors; among these are sweetness, alcohol content and grape type. “Whoever wants to identify corky off-flavor compounds and track down their source has no choice but to use gas chromatography combined with mass selective detection (GC/MS)”, Horst Rudy points out.

Barking up the wrong tree – on the origins of Corkiness

The usual suspect as a source of 2,4,6-TCA is the cork stopper made from the bark of the cork oak tree (Quercus suber). TCA is a microbial metabolite, formed by methylation of trichlorophenol (TCP) that may have been applied to the bark as a pesticide. To suspect the cork stopper of introducing TCA to the wine is therefore only logical according to the wine expert, but when wine drinkers started experiencing corkiness in wines with modern polymerbased stoppers experts knew that they had been barking up the wrong tree.

Over the course of the ensuing research projects, it was found that various compounds, mainly halogenated anisols, would give a musty, moldy note to the wine. These compounds could be formed from other chlorinated chemicals that are used for cleaning of wine production equipment or for treating wooden transport pallets or packaging material.

Until the end of the 1980’s, pentachlorophenol (PCP) was used as a fungicide to protect, for example wooden pallets from microbial decay. Among others byproducts, PCP contained 2,3,4,6-tetrachlorophenol (TCP), a compounds that is metabolized microbially to 2,3,4,6-tetrachloroanisol (2,3,4,6-TeCA, TeCA), which also causes corkiness in wine.

In animal tests, PCP was found to be carcinogenic. In Germany, the use of PCP has been prohibited since 1989. PCP was substituted by 2,4,6-tribromophenol (TBP), a combined fungicide and flame retardant, which is often used to protect cardboard packaging, polymer materials, paints and coatings. As it turns out, microorganisms metabolize TBP to 2,4,6-tribromoanisol (2,4,6-TBA), a compound given the sensory attributes musty, earthy, and chemical with a smell of solvent. “TBA is a corkiness causing compound of the first order”, Horst Rudy points out.

Chemical analysis and sensory evaluation – complementary techniques

When the DLR Mosel is asked to determine the cause of a musty and moldy off flavor in a wine, sensory evaluation is only the first step in the process. “While corky off-flavors are typically determined quite reliably”, Horst Rudy says, “TCA concentrations at or below the odor threshold often lead to a subtle and indefinable change in the wine flavor, not perceived as a corky flavor note”. In such cases, chemical analysis is needed in order to prove that the wine is under the influence of TCA. To snoop out the source of the contamination, all aspects of the wine production and bottling process, as well as the entire production site environment, must be carefully investigated. Horst Rudy and his team deploy passive samplers based on Bentonite clay to pick up TCA traces. “Passive samplers are easy to work with and they deliver valuable information such as a distribution profile enabling us to more accurately localize the source”, says Mr. Rudy. As a general rule, the DLR does not restrict its GC/MS investigations to off-flavors that are perceived as corky. Among the targeted compounds are: 2,4,6-trichloroanisole (TCA), 2,4,6-tribromoanisole (TBA), 2,3,4,6-tetrachloroanisole (TeCA), 2,3,4,5,6-pentachloroanisole (PCA), as well as the TCA and TBA precursors 2,4,6-trichlorophenol (TCP) and 2,4,6-tribromophenol (TBP), which are less odor-active. The presence and distribution of TCP and TBP can give valuable information as to the source of an off-flavor. To determine the identity and concentration of odor agents, cork stoppers are extracted for two hours in a 10 % ethanolwater mixture using sonication to speed up the process. The bentonite clay used for passive sampling is extracted in the same way. Subsequently, 100 mL of the Ethanol solution is extracted for one hour by Stir Bar Sorptive Extraction (SBSE) using a GERSTEL® Twister™ (The bentonite should be allowed to precipitate before sampling is performed). The Twister is a glass-coated magnetic stir bar with an outer Polydimethylsiloxane (PDMS) layer. While the Twister stirs the sample, analytes are extracted and concentrated into the PDMS phase. Depending on the application and on the sample volume available, SBSE can be up to 1,000 times more sensitive than SPME due to both the significantly larger PDMS volume available and to the larger sample volumes extracted. Quantification in this work was performed using 2,4,6-trichloroanisole-D5 as internal standard.

Step by step: How to extract odor active compounds
from wine or water using the GERSTEL Twister
Place the Twister in the sample (1). While stirring the sample, the Twister concentrates analytes in its PDMS phase (2). The Twister is removed from the sample (3), dried with a lint-free cloth (4) and placed in the MPS tray (5) for automated thermal desorption in the TDU (6).


Fast and sensitive analysis using the GERSTEL Twister

The multi-stage liquid-liquid extraction previously used by DLR Mosel was highly laborand cost intensive. “Sometimes I spent all day in the laboratory and still only managed to analyze four samples”, states Mr. Rudy. More modern analysis techniques, such as Solid Phase Micro-Extraction (SPME) enabled the DLR to reduce the analysis time significantly, but the limits of detection achieved, for example 2.9 ng/L for TCA, meant that this technique was only of limited use. “We have to reliably determine concentrations of odoractive compounds at their odor threshold levels”, says Horst Rudy, “and for this reason we started using SBSE and the GERSTEL Twister”. SBSE is a fast and accurate extraction technique that enables the DLR laboratories to reach a detection limit of between 0.3 and 0.5 ng/L for TCA as per the DIN 32645 method and the analysis time has been reduced from 6 hours to 1.5 hours per sample and multiple samples can be processed in parallel for improved productivity.

SBSE is extremely easy to perform: Following the extraction step, the Twister is removed from the sample, dried using a lint-free paper cloth and transferred to the autosampler tray. Up to 196 Twisters can be desorbed and analyzed by GC/MS in a single batch using the GERSTEL MPS and TDU directly mounted on a GC/MS system. The work reported in this article was performed using a GC 6890/MSD 5975 (Agilent Technologies).