The present invention refers to a method which permits implementation of supercritical fluids treatment directly to natural cork stoppers without losing their shape and maintaining their sealing properties, using a separating device. This method eliminates or reduces a quantity of cork contaminants, namely 2,4,6-trichloroanisole which is the main compound responsible for off - aromas occurring in wine.

Traditionally, cork has been used as stoppers for wine bottle sealing. The most important application of cork is as stoppers in the preservation of wines. The natural cork stopper has unique physical properties such as elasticity, hydrophobicity and impermeability to gases and liquids. To date, there are no other natural or artificial products that possess such desirable characteristics. Besides an efficient sealing, these properties ensure adequate wine maturation or aging., and therefore natural cork stoppers are frequently used for storage of wines for long periods of time (Fischer, 1997; Insa, 2006). However, this application is being challenged due to the so-called "mouldy taint" problem, which leads to the rejection of the product by consumers, thus provoking considerable losses in the wine industry. Soleas et al. (2002) showed that about 6.1% of wine bottles were affected by the "mouldy taint" problem. Another study (Quercus (1996)) revealed that between 0.1 and 10% of European wine bottles are contaminated.

The problem of sensory deviations caused by cork stoppers, which frequently occurs in wines, has been related with the presence of chloroanisoles, especially 2,4,6-tricloroailisole (TCA), but also with 2,3,4,6-tetracloroanisole (TeCA) and pentachloroanisole (PCA) (Insa, 2006). TeCA and PCA are more related with contaminants from the cellar environment. For the first time in 1982 Tanner et al. identified TCA as the main compound responsible for the "mouldy taint" problem.

Even though other aromatic compounds have been related to the same problem (Simpson, 2004; Pena-Neira 2000; Chatonnet, 2004), TCA is generally considered the main indicator (Hervé, 2000). According to the European project "Quercus", the presence of TCA was detected in at least 80% of wine with undesirable sensory deviation. Additionally, TCA is a compound with a sensory detection limit at very small concentrations, in the order of a few nanograms per litre of wine (Amon, 1989; Tanner, 1982).

The presence of TCA in wines is associated with cork stoppers and its appearance in stoppers has been the subject of numerous studies (Silva Pereira, 2000). Direct precursors of chloroanisoles are chlorophenols, which are converted into chloroanisoles via a methylation reaction provoked by microorganisms, especially fungus at certain conditions of temperature and pressure (Insa, 2006). In this way, chlorophenols constitute a potential source of TCA and their appearance in the cork oak can occur through the use of certain insecticides and fungicidal products, as well as from certain materials used in storage, transport and packing (Czaplicka, 2004, Alvarez-Rodríguez, 2002). Although the cork industry has developed preventive procedures in order to avoid the presence of these compounds, residual chloroanisoles continue to be detected in cork stoppers.

TCA removal is a difficult process due to a combination of various factors such as:

1. 1. impermeable characteristics of cork;
2. 2. need to reach values of concentration in the order of ppt, an extraordinarily low value, due to the low concentration threshold of perception;
3. 3. great chemical stability of the compound.

Since the identification of TCA as the main compound responsible for undesirable taste of wine, processes of treatment and minimization have been developed (Gil, 2006).

Industrially, cork stoppers are cleansed using hydrogen peroxide or peracetic acid. The aim of this process is to wash and disinfect the stoppers. There are also other methods such as implementation of microwaves (Jager, 1999). Washing/disinfection is followed by stabilization of the humidity level, optimizing the performance of the cork as a sealant and simultaneously reducing microbiological contamination. However, these methods are not sufficient for volatile compounds (Gil, 2006).

Other processes for removal of cork stopper contaminates have been proposed. The most prominent are sterilization of cork using gamma-radiation (PT103006), ozone treatment (Vlachos, 2007), an ultra-sound method (Penn, C, 2004), steam extraction (WO03041927), photodegradation (Vlachos, 2008) and extraction with supercritical CO2 (FR19990012003, published as US2007/0017550 and WO 01/23155).

Supercritical CO2 is recognized as a very efficient solvent for extraction of cork contaminates, namely TCA. This fact was reported originally in 1997 by Chouchi et al. The authors reported that "supercritical extraction of cork contaminants which provoke undesirable wine taste is efficient and much faster than conventional method".

In 2000 Taylor et al. confirmed that yields of supercritical extraction of TCA have similar values compared to Soxhlet extraction, being however much more efficient in terms of time.

The patent FR19990012003 illustrates a process for extraction of cork contaminants using supercritical fluids, namely supercritical CO2. In the examples presented in the description of the invention, it is clear that the cork treated by supercritical extraction is always in granulated form or as planks and not in the form of cork stoppers. As a matter of fact, there is no known application directly to the natural cork stoppers.

The publication of Eduard Lack in 2006 at the 3rd Chemical Engineering and High Pressure International Meeting is especially relevant. The author, representing NATEX, the Austrian company responsible for the design and construction of the industrial facilities for the patented process of supercritical extraction, claims that cork stoppers do not return to their original form during the depressurization step. Therefore, the author has concluded that it is impossible to implement supercritical extraction to natural cork stoppers. The author also states that treatment with supercritical fluids should be applied directly to granulated cork and stoppers should be produced afterwards.

This invention can be illustrated using the attached drawings. These drawings show the devices and processes used to accomplish this invention:

• Figure 1: It shows a model of a separating device with attached shelves. Figure 1A is a front view of the said separating device, which is composed of one central rod (1) and three side rods (2) with a base composed of hole-bored metallic sheet, to which the shelves that support the cork stoppers to be treated are attached. The shelves are organized one above the other and attached to the above-mentioned rods up to a number limited by the height of the rods. Figure 1B is a plan view of the separating device, having a cover lid with a hook to allow pulling up the whole structure.
• Figure 2: A model of shelf used in the separating device is shown. Figure 2A is a perspective view of the said shelf, composed of a perforated plate (5), having attacking holes (4), a side ring in a flat plate (6) with a rod attached on the top (7). Figure 2B represents a plan view of the said shelf.
• Figure 3: It shows a scheme of the process in which the solvent stored in a cylinder is compressed by a liquid pump (8) and introduced into the high-pressure pressurized vessel (9) containing the cork stoppers to be treated according to this invention. During the process, the solvent is recycled after decompression, TCA separation (10) and passing through an adsorption filter (11). At the end of the process, the solvent is substituted with nitrogen stored in a cylinder.

To date, although the known supercritical extraction process is very efficient for cork contamination removal, it has only been possible to apply it to granulated cork. It is not possible to directly treat cork stoppers using this extraction technology. The present invention permits the application of supercritical extraction in the treatment of cork stoppers.

The present invention refers to a new method, which employs a separating device, permitting direct supercritical fluid extraction from natural cork stoppers without damaging their shape and maintaining their sealing properties.

Taking into consideration all the known facts and publications to date, the success obtained by the process using the separating device is totally unexpected. This process will solve a technical problem that until now is not resolved.

The novelty of the present invention consists of using a separating device which permits direct treatment of natural cork stoppers without damaging their form and maintaining their sealing properties using supercritical fluids.

The present invention is a supercritical extraction method applied directly to natural cork stoppers with the objective of eliminating/reducing contaminant compounds, namely 2,4,6-trichloroanisole (TCA), using supercritical fluids in a high-pressure vessel comprising a separating device. It allows shrinking and expanding of the cork stoppers with minimal contact between them, avoiding significant distortion of their original shape and maintaining their sealing properties.

Preferentially, in the high-pressure vessel (9) the supercritical fluid or mixture of supercritical fluids, with or without addition of co-solvent, is directly contacted with natural cork stoppers.

In one embodiment according to Figure 3, the method can be carried out through the following steps:

• a) Introduction of a compressed fluid or mixture of compressed fluids through a compressor or liquid pump (8) into a sealed high-pressure vessel (9) immersed in a temperature-controlled bath, containing cork stoppers distributed within the shelves of the separating device. The method can optionally use an inert and non-toxic co-solvent which can be introduced by a compressor/liquid pump (8) or placed in the high-pressure vessel (9) prior to extraction;
• b) Circulation of the compressed fluid, mixture of compressed fluids or mixture of compressed fluids and co-solvent through the high-pressure vessel (9) under predetermined conditions of pressure, temperature and time, with predetermined velocity - preferentially low;
• c) Removal of compressed fluid, mixture of compressed fluids, or mixture of compressed fluids and co-solvent, by decompression and, optionally, by substitution with a permanent gas, like nitrogen.

Preferentially, the supercritical extraction step is carried out in a temperature range from 30°C to 120°C and at pressures between 6 MPa and 40 MPa. This step can take 1 to 24 hours, preferentially between 4 to 24 hours.

Any fluid or mixture of fluids in the supercritical state or in the sub-critical liquid state can be used to carry out this invention. Preferentially, the compressed fluid or mixture of these fluids should be very volatile or be in gas state at atmospheric conditions, in order to facilitate its removal by expansion and/or evaporation after treatment. For safety reasons, the compressed fluid, or a mixture of these fluids, should be non-toxic and not flammable, as well as recyclable. Preferentially, the fluid used is carbon dioxide.

Co-solvents can be added, preferentially inert and non-toxic compounds, in order to increase solubility of cork stopper contaminants in compressed fluids.

According to Figures 1 and 2, in one specific embodiment, the separating device for carrying out the invention consists of:

• A metallic structure comprising four rods, a central one (1) and three side ones (2) with a base made of perforated metallic sheet, where the shelves (3) that support the cork stoppers to be treated are attached;
• Shelves (3) made of perforated sheets (5), with four attaching holes (4) where the above-mentioned rods are secured, a side lid made of metallic sheet (6) with a rod soldered at the top (7); the shelves are attached to each other up to a maximum number limited by the height of the rods;
• A cover lid made of metallic sheet, where a hook is attached to allow pulling up the whole structure.

Preferably, the metallic structure is made of stainless steel.

In an even more specific embodiment, the four said rods are metallic and 12 mm thick. The shelves have 3 cm of useful height while the perforated metallic sheet is 2 mm thick. The holes have a diameter of 2 mm and the distance between them is 5 mm.

The second aim of the method of the present invention is maintaining the original form and sealing properties of the cork stoppers after treatment.

One of the ways to carry out the invention is given in Figure 3. The solvent is stored in a cylinder and introduced into a high-pressure vessel (9) up to the desired pressure through a compressor (8). The vessel contains the cork stoppers to be treated using the separating device according to the method of the present invention. In the first phase, the fluid in the high-pressure vessel is retained at desired operational conditions of temperature ranging from 30°C to 120°C and pressure ranging from 6 MPa to 40 MPa during a predefined time period ranging from 1 to 24 hours, typically between 4 and 8 hours. Next, the fluid circulates at predetermined velocity, preferentially low. The TCA is collected in the separator (10) while CO₂ is decompressed and filtered (11), being reused after new compression. Finally, a substitution of the solvent with nitrogen stored in a cylinder is carried out, in order to avoid temperature decrease.

The invention is additionally illustrated by the following non-restrictive examples.

A high-pressure vessel with a capacity of 170 cm³ and a manual liquid pump were used. One cork stopper and 10% mass of distilled water were placed in the vessel. The system was heated up to 40°C and pressurized with CO₂ up to 100 bar. The cork stopper was left in contact with CO₂ during two periods of time, of two and three hours respectively. The extraction cycle was carried out in the following way:

• Phase 1 - Pressurisation up to 100 bar and retention of CO₂ in the vessel for 2 hours;
• Phase 2 - Slow circulation of CO₂ through the vessel with flow of approximately 0.8 g/min;
• Phase 3 - New retention of supercritical fluid in the vessel for 3 hours :
• Phase 4 - New slow circulation of CO₂ through the vessel with flow of approximately 0.8 g/min;
• Phase 5 - Slow depressurization of the system.

The total used mass of CO₂ was 100 g/g of cork. The quantity of TCA in the natural cork stoppers was measured before and after the treatment with supercritical CO₂ and the results are shown in Table 1. Sample [TCA] _initial (ng/l) [TCA] final (ng/l) % of Extraction S45 5.2 <1.0 >90 S59 17 1.4 92 S55 215 5.4 98

Experiments were carried out with the immersion of cork stoppers in CO₂ inside a high-pressure extractor. CO₂ was fed to the extractor up to the desired working pressure, while temperature was kept constant by a thermostat.

The experiments were carried out in the following conditions:

• temperature: 40°C
• pressure: 100 bar
• time of immersion: 1 hour
• decompression time: 15 minutes
• cork stoppers placed at random inside the extractor.

After decompression, the cork stoppers were found to have collapsed to the bottom of the extractor, where they accumulated in highly contorted shapes. As a result, the cork stoppers showed highly irreversible deformations.

The same experiment was repeated, in the same working conditions of pressure and temperature and immersion/decompression. However, the cork stoppers were separated by a metallic grid with holes (to allow homogeneous flow of CO₂ along the extractor).

Very small deformation of the cork stoppers was observed, with small increases of about 2 mm in height and 1 mm in diameter. Only cork stoppers in the base of the extractor showed slightly greater deformation, but much less than in the initial experiments.

These deformations in the base of the extractor are caused by compression due to the weight of the stoppers above arid by the joint forces due to the volume expansion of the stoppers upon carbon dioxide decompression.

Experiments were also carried out where the cork stoppers were placed above the grids, but with freedom to expand in height (but not sideways). The useable volume inside the extractor is therefore decreased.

The increases in height were now slightly smaller (maximum of 1.5 mm) than in the previous examples. The increases in diameter were slightly higher, due to a smaller free space between stoppers in the packing in comparison with the situation where the cork stoppers are simply piled upon the grids.

There was no indication of shape effects resulting directly from decompression. It was therefore confirmed that the slight increase of dimensions of the cork stoppers upon decompression only leads to problems of irreversible deformation if they are freely and randomly placed inside the extracting vessel. These problems are soluble in an extractor where each cork stopper is given the freedom to expand upon decompression without interference from other stoppers.

The compression/decompression cycles consisted of:

• pressurization of the extractor up to 100 bar
• allowing CO₂ to diffuse into the cork for 3 hours
• slow depressurization after 3 hours - depressurization time of 1 hour
• cork stoppers separated by grids in order to be allowed freedom of expansion.

Evaluation of the compression/relaxation behaviour was carried out following the Portuguese Standard NP 2803 and the dimensional recovery of the diameter 15 minutes after compression (Table 2). Cylinders Samples Compression (10 mn) (daN) Relaxation (12 mm) after 1 min (daN) Recovery (%) Natural cork cylinders 45×15 mm Control 63.7 9.8 93.2 After SCF* 55.5 9.7 92.5 Cork stoppers Samples Compression (16 mm) (daN) Relaxation (19 mm) after 1 min (daN) Recovery (%) Natural cork stoppers of 45×24 mm calibre Maximum Acceptable Value 80 Minimum Acceptable value 60 10 95 After SCF 65.2 11.8 92.9 * SCF = supercritical fluid extraction cycle

An adequate behaviour of the cork stoppers subjected to supercritical extraction is therefore verified, as much in relative terms - by comparison with non-processed samples in the case of cylinders with diameters smaller than 24 mm - as in absolute terms - by comparison with reference values for cork stoppers.

• [1] C. Fischer and U. Fischer, "Analysis of cork taint in wine and cork material at olfactory subthreshold levels by solid phase microextraction," journal of Agricultural and Food Chemistry, vol. 45, 1997.
• [2] S. Insa, V. Salvado, and E. Antico, "Assays on the simultaneous determination and elimination of chloroanisoles and chlorophenols from contaminated cork samples," If Journal of Chromatography A, vol. 1122, pp. 215-221, 2006.
• [3] G. J. Soleas, J. Yan, T. Seaver, and D. M. Goldberg, "Method for the gas chromatographic assay with mass selective detection of trichloro compounds in corks and wines applied to elucidate the potential cause of cork taint," Journal of Agricultural and Food Chemistry, vol. 50, pp. 1032-1039, 2002
• [4] QUERCUS-Qualitative Experiments to Determine the Components Responsible and Eliminate the Causes of Undesirable Sensory Characteristics in Drinks Stoppered With Cork, European Union and C.E. Liège, Contract No. AIR1-CT92-0372, 1996.
• [5] R. F. Simpson, D. L. Capone, and M. A. Sefton, "Isolation and identification of 2-methoxy-3,5-dimethylpyrazine, a potent musty compound from wine corks," Journal of Agricultural and Food Chemistry, vol. 52, pp. 5425-5430, 2004.
• [6] A. Pena-Neira, B. F. de Simon, M. C. Garcia-Vallejo, T. Hernandez, E. Cadahia, and J. A. Suarez, "Presence of cork-taint responsible compounds in wines and their cork stoppers," European Food Research and Technology, vol. 211, pp. 257-261, 2000.
• [7] P. Chatonnet, S. Bonnet, S. Boutou, and M. D. Labadie, "Identification and responsibility of 2,4,6-tribromoanisole in musty, corked odors in winge," Journal of Agricultural and Food Chemistry, vol. 52, pp. 1255-1262, 2004.
• [8] http://www.corkqc.com/currentresearch/CorkTaint/ETS%20CQC-SPME.pdf
• [9] J. M. Amon, J. M. Vandeeper, R. F. Simpson, "Compounds responsible for cork taint in wine", Australian & New Zealand Wine Industry Journal, vol.4, pp 62-69, 1989.
• [10] C. S. Pereira, J. J. F. Marques, and M. V. San Romao, "Cork taint in wine: Scientific knowledge and public perception - A critical review," Critical Reviews in Microbiology, vol. 26, pp. 147-162, 2000.
• [11] M. Czaplicka, "Sources and transformations of chlorophenols in the natural environment," Science of the Total Environment, vol. 322, pp. 21-39, 2004.
• [12] M. L. Alvarez-Rodriguez, L. Lopez-Ocana, J. M. Lopez-Coronado, E. Rodriguez, M. J. Martinez, G. Larriba, and J. J. R. Coque, "Cork taint of wines: Role of the filamentous fungi isolated from cork in the formation of 2,4,6-trichloroanisole by O methylation of 2,4,6-trichlorophenol," Applied and Environmental Microbiology, vol. 68, pp. 5860-5869, Dec 2002.
• [13] J. Jaeger, 1999. Cork Sterilization and decontamination process employing microwaves. W09937334.
• [14] C. Pereira, L. Gil, O Problema do Odor a Mofo nas Rolhas de Cortiça e Processos para a sua Reduçãc/Eliminação, Silva Lusitana 14(1): 101 - 111, 2006.
• [15] L. Gil, C. Pereira, 2003. Processo para a eliminação/redução de compostos com gosto/odor a mofo em material para contacto com alimentos e bebidas, e em alimentos e bebidas. PT103006.
• [16] P. Vlachos, A. Kampioti, M. Kornaros, and G. Lyberatos, "Development and evaluation of alternative processes for sterilization and deodorization of cork barks and natural cork stoppers," European Food Research and Technology, vol. 225, pp. 653-663, 2007.
• [17] Penn, C., Newest cork-cleaning treatment comes to market. Wine Business Monthly. September 2004: 25 - 26
• [18] J. Bordado, J. Marques, M. Gonçalves, I. Allegro, A. Mesquita, R. Filipe, 2003. New process for treating cork stoppers or planks for the reduction of strange aromas, namely 2,4,6-trichloroanisole. WO03041927.
• [19] P. Vlachos, E. Stathatos, G. Lyberatos, and P. Lianos, "Gas-phase photocatalytic degradation of 2,4,6-trichloroanisole in the presence of a nanocrystalline Titania film. Applications to the treatment of cork stoppers," Catalysis Communications, vol. 9, pp. 1987-1990, Jun 2008.
• [20] G. Lumia, C. Perre, J. Aracil, 1999. Method for treating and extracting cork organic compounds, with a dense fluid under pressure. FR19990012003.
• [21] D.Chouchi, C. Maricato, M. Nunes da Ponte, A. Pires, V. San Romao, "SFE of trichloroanisole from cork", Proceedings of the 4th International Symposium on Supercritical Fluids, May 11-14, Sendai, Japan, vol A, pp 379-382, 1997.
• [22] M. K. Taylor, T. M. Young, C. E. Butzke, and S. E. Ebeler, "Supercritical fluid extraction of 2,4,6-trichloroanisole from cork stoppers," Journal of Agricultural and Food Chemistry, vol. 48, pp. 2208-2211, Jun 2000.
• [23] E. Lack, "Industrial cleaning of cork with supercritical CO2", 3rd International Meeting on High Pressure Chemical Engineering, 10-12 May, Earlanger, Germany, 2006.