L. Cappellin*, **, L. Ciotti*, M. Hutterli*,
*Tofwerk AG, CH-3645 Thun, Svizzera
**Università degli Studi di Padova, 35131 Padova, Italia
Abstract
Contamination of wine by 2,4,6-trichloroanisole (TCA) causes significant economic losses in the wine industry. Over 50 years of research has identified cork as the main cause of this problem. The threshold of human perception for TCA is extremely low; just 1-2 ng/L makes the wine unpalatable. Contaminations below 0.5 ng /L are generally considered negligible. Preventive screening of corks offers a method to avoid TCA contamination, but the need for a rapid and non-destructive analytical method to quantify trichloroanisole contamination in cork stoppers is of enormous urgency. The analytical and sensorial methods commonly used are unreliable and error-prone . In this article, we propose an innovative analytical approach based on chemical ionization time-of-flight (CI-TOF) mass spectrometry using the “ Vocus ” system. This technique has proven capable of quantifying TCA in a cork in just 2 seconds, without destroying the cork, with a sensitivity below the perception threshold. The method is able to analyze and quantify the TCA in approximately 10 million corks in a year. We demonstrate its applicability in the industrial environment and the correlation with standard methods for the quantification of releasable TCA.
Introduction
So-called cork taint, primarily caused by 2,4,6-trichloroanisole (TCA), represents a significant threat to the wine industry, causing annual losses in excess of $10 billion (Taylor et al., 2000). Despite the complexity of wine and the presence of hundreds of aromatic compounds, even a few ng/L of TCA can compromise the aroma. (Tarasov et al., 2017). The discovery of TCA as responsible for olfactory anomalies dates back to the early 1980s and, despite four decades of research, the complete understanding of its presence in wine remains not entirely clear (Buser et al., 1982; Sefton and Simpson, 2005).
Although TCA is the compound that represents the greatest problem, other cork contaminants can also lead to the contamination of a wine, for example chloro- or bromoanisoles (Chatonnet et al., 2004). The origins of TCA in cork remain unknown, with possible links to chlorophenolic biocides or to natural compounds in wood undergoing chlorination and microbial reactions (Simpson and Sefton, 2007). It has been noted that the incidence of TCA is higher in unmanaged cork oak forests:
this observation would justify a link between environmental conditions and microbiological contamination (Simpson and Sefton , 2007).
The analysis of TCA in corks is challenging due to its strong localization, and current methods are often laborious and destructive (Prescott et al., 2005). The analysis of releasable TCA, according to the ISO 20752:2014 standard, involves the immersion of cork stoppers in a wine simulant and
subsequent quantification using the GCMS technique. These methods are mainly used as standards in the laboratory.
Sensory analysis remains a widely used technique, but these traditional methods are subjective, time-consuming, and error-prone (Cork Stoppers – Sensory Analysis, ISO 22308:2005). The need for rapid and nondestructive methods has led to the exploration of chemical ionization mass spectrometry (CI-MS) as a new approach (Lopez-Hilfiker et al., 2017).
This study introduces the Vocus reactor based on chemical ionization (CI-MS), excluding chromatography and using high-resolution mass spectrometry for real-time separation of TCA signals from those of other volatile compounds emitted from natural corks (Cappellin et al., 2020).
The Vocus instrument has demonstrated the ability to detect TCA concentrations below sensorythresholds in little as 2 seconds while establishing good correlation with releasable TCA determinations (Taylor et al., 2000; ISO 20752:2023). The method was further validated in a
simulated industrial scenario, analyzing 10,100 natural corks in just a few hours.
The importance of the new method lies in its speed, non-destructive nature, and correlation with established standards, offering a possible breakthrough for large-scale TCA screening in the cork industry.
Materials and methods
Cork stopper samples
Samples of natural cork stoppers, with a diameter of 24 mm and a length of 49 mm, were obtained directly from various manufacturers. Cork visual ratings ranged from Flor to II based on international guidance. The caps were not coated, therefore free from the presence of synthetic substances.
Calibrations, tests of the new method, and comparisons with other TCA analysis techniques were conducted on natural corks from different batches. The industrial tests involved a mixed batch made up of 100 natural cork stoppers from different manufacturers and two homogeneous lots, each made up of 5000 natural cork stoppers from the same manufacturer.
Vocus Cork Analyzer (VCA)
The VCA , developed by Tofwerk AG , a technical-scientific company based in Switzerland, includes a Vocus 2R high-resolution chemical ionization mass spectrometer combined with a dynamic
headspace automatic sampler specific for cork stoppers. The Vocus 2R achieves mass resolution up to am/dm = 15,000. The reagent ion source was set at p≈ 2 mbar and generates reagent ions from synthetic air (Alphagaz 1 Air, Air Liquide). The reaction chamber is operated at a pressure of 1,5 mbar and a temperature of 150 °C. Chemical ionization (CI), by charge transfer, produces TCA ions with negligible fragmentation. The automatic cork stopper sampler features individual cavities to avoid
cross-contamination. VCA is measured for 1,2 seconds in each cavity by drawing air at a flow rate of 1 standard liter per minute through a heated PTFE sample line (1/8 in. i.d.) at 120°C.
Simultaneously, synthetic air replaces the gas in the sampled void, with a settling time of approximately 0,8 seconds between cavities, resulting in a total cycle period of Δt = 2 seconds for each cork. Benzene, toluene, and xylene in pure nitrogen are mixed into the sample air stream to monitor the stability of the primary ions. The signal intensity of the spectral peaks at 209,940 Th , 211,937 Th and 213,934 Th expressed in counts per second (cps), and, corresponding to the C7H5Cl3O+ isotopes, are summed and used as the signal for the TCA. The benzene C6H6+ signal is used as an internal standard to correct any sensitivity drift. Conversion to equivalent releasable TCA, expressed in ng/L, is performed by calibration against the standard method.
Releasable TCA analysis according to Standard Methods
The releasable TCA in natural corks was determined following the norms or standards OIV-MAAS315-16 and ISO 20752:2023, but with a small modification. A cork stopper was immersed in a hydroalcoholic solution (12% v/v alcohol content) for 24 hours. Volatile compounds in the headspace were collected using solid-phase micro-extraction, and a GC-MS system detected TCA and TCA-d5.
The method, widely used in the cork industry, is crucial for demonstrating TCA contamination in cork stoppers, even in legal contexts, and will henceforth be referred to as “ISO”.
Results and discussion
Considering the very fast measurement times for individual corks, three groups of natural corks from different manufacturers were evaluated for a total of 10. 100. The results for the smallest set, consisting of 100 natural, are shown in Figure 1. The corks were measured sequentially, and the total measurement time for the entire experiment was a few minutes. This batch included several corks highly contaminated with TCA, others with only slight contamination, and still others showing a TCA signal below the detection threshold.
Figure 2 summarizes the quantification of TCA in two different batches, each consisting of 5000 corks. The total measurement time of the two batches was a few hours. For comparison, if the same experiment had been conducted using the ISO method, it would have required more than four months of continuous analysis.
Furthermore, the VCA method is non-destructive and causes no visual damage or deformation to the cork stoppers. Therefore, samples can still be used and sold after analysis, provided that a reconditioning step is performed to restore the moisture content. This procedure is very common in the cork industry.
The histograms shown in Figure 2 indicate that the first batch is characterized by a lower average TCA content than the second batch. Over 99% of corks were contaminated with <1 ng/L TCA and approximately 50% with <0.5 ng/L TCA, which is currently considered a limit for TCA-free corks in cork industry.
In contrast, the second batch had only a negligible percentage of TCA-free caps, while almost all were contaminated with >1 ng/L TCA.
The VCA technique can be useful in an industrial scenario to select natural cork stoppers based on their level of TCA contamination before they are sold or in a quality control laboratory to quickly assess the incidence of TCA in cork batches.
Comparison with ISO 20752:20 23 and OIV-MA-AS315-16
A set of 671 natural cork stoppers was tested with the new proposed method and, for comparison, with the ISO method (i.e. ISO 20752:2023 and OIV-MA-AS315-16). Figure 3 reports the correlation between Vocus and ISO for these samples. The R 2 value of the linear regression is 0.92, implying a Pearson correlation coefficient of 0.96. The correlation is statistically significant (p < 0.01). To better evaluate such correlation results and have a reference point, a set of corks was measured twice according to the ISO standard, waiting a period of 15-30 days between measurements. The R 2 of the regression line is 0.68 and the Pearson correlation coefficient is therefore 0.82 (p < 0.01). These values suggest that the correlation between Vocus and ISO is strongly limited by the precision of the ISO method. The large uncertainty associated with the ISO method probably originates in the sample preparation phase, i.e. “the soaking phase”. In this phase, each natural cork stopper is soaked for 24
± 2 hours in a wine simulant solution. By eliminating the soaking step, the Vocus Cork Analyzer has the advantage of reducing the uncertainty associated with this step.
Conclusions
In this article we presented a highly sensitive, real-time, non-destructive technique for the quantification of TCA contamination in individual cork stoppers. This approach significantly outperforms existing analytical methods in terms of speed, while also having lower detection limits
for TCA. The new technique also shows excellent correlation with the quantification of releasable TCA using standard methods. The new technique represents a breakthrough for the cork and wine
sector, offering the possibility of ultra-fast (2s) pre-selection of individual natural corks based on the quantification of TCA contamination. The technique can also simultaneously analyze other contaminants such as chloroanisoles , chlorophenols , bromoanisoles, and bromophenols .
For more information contact: vca.info@tofwerk.com
References
Buser, H.R., et al. (1982). Journal of Agricultural and Food Chemistry, 30(2), 359−362.
Cappellin, L. et al. (2020). Analytical Chemistry 2020, 92, 9823−9829
Chatonnet , P., et al. (2004). Journal of Agricultural and Food Chemistry, 52(5), 1255−1262.
Cork Stoppers – Sensory Analysis (ISO 22308:2005).
Lopez-Hilfiker, F., et al. (2017). Ion Molecule Reactor and Setup for Analyzing Complex Mixtures.
WO2019077138A1.”
Prescott, J., et al. (2005). Food Quality and Preference, 16(4), 345−349.
Sefton, M.A., et al. (2005). Australian Journal of Grape and Wine Research, 11(2), 226−240.
Tarasov , A., et al. (2017). Talanta , 175, 82−92.
Taylor, M.K., et al. (2000). Journal of Agricultural and Food Chemistry, 48(6), 2208−2211.
Vestner , J., et al. (2010). Analytica Chimica Acta, 660(1−2), 76−80.
For further information, please, contact us at:
