The present invention relates to a machine and a process for treating wooden barrels used for the maturation and/or refinement of wines, such as for example the so called "barriques".

In particular, the present invention relates to a machine and a process for carrying out the rejuvenating (or refreshing) treatment and the sanitizing treatment of the wooden barrels by throwing granulated dry ice (solid CO₂).

The wine, throughout the course of the maturation process in a barrel, is subjected to oxygenation by osmosis. The wooden fibers of a barrel are however progressively choked by various sediments, for example mildews, tartrates and other organic residues. These sediments gradually build up in the first millimeters of thickness of the inner walls until the oxygen osmotic exchange with the outside is almost completely impeded.

In this case, after some years of use, a rejuvenating (or "refreshing") treatment of the barrel is required, which consists in removing a surface layer of wood from the inner wall of the barrel up to the clean wood, such that the conditions for a correct osmotic exchange can be restored.

This treatment is still carried out by mechanical removal, for example by planing, in order to remove the exhausted wood layer until the new and clean wood is reached. For large size barrels it is operated directly in the inside thereof after removing one of the flat ending walls.

For barrels of smaller size, such as for example the barriques, first of all the rejuvenating treatment consists in disassembling all the barrique staves, and then planing them until obtaining the desired result. The barrel has therefore to be reassembled and possibly subjected to a new toasting process.

It is therefore evident that the rejuvenating treatment is long and labor-intensive, often resulting in costs and times particularly high.

It has to be remarked also that the planing of the staves is a very critical work, which requires extreme care and precision, and therefore it has to be carried out by highly specialized staff (master coopers).

Besides the rejuvenating treatment, barrels and barriques further have to be subjected to the required sanitizing treatments before any new use.

In fact, during the barrel vinification process, different microorganisms develop, whose functions are essentially of converting the sugars in alcohol, reducing the wine acidity and introducing flavors and tastes which improve the wine organoleptic and sensory profile.

The microorganisms entailed in the vinification process come from vineyard, grapes and wine cellar; this natural microflora comprises yeasts, lactic bacteria and acetic bacteria. During such a process particular conditions could occur wherein certain microorganisms can develop and cause some changes to the wine, sometimes rendering it undrinkable and therefore not marketable.

The spoilages can arise with unpleasant scents or tastes, color changes, viscosity changes, or else they can arise with high amounts of substances detrimental for human health, such as for example biogenous amines and ethyl carbamate. Furthermore, in regard to yeasts, at present the wine contamination, in particular of red wine, by yeasts belonging to the Brettanomyces/ Dekkera kind is one of the most relevant problems in the wine field. These yeasts are very resistant, able to develop in presence of even low amount of nutrients, and difficult to be isolated and identified. The problem gets further worse when it also arises on the bottled and finished product. In particular, the presence of Brettanomyces is often found on the inner surfaces of the wooden barrels: this is due to the fact that these containers are difficult to sanitize and wood further contains cellobiose, a sugar which is cleaved by these yeasts in glucose needed for their growth. The new wooden barrels contain a larger amount of cellobiose and therefore represent an even more favorable environment for their development.

The researches performed on these yeasts still have not found an always effective tool in the biologic control of such microorganisms and the results obtained so far are conflicting, also in regard to the Brettanomyces effect on the organoleptic characteristics of wines.

Conventionally, in order to control the development of undesired microorganisms, the sulphurous dioxide (or simply "sulphurous") has long been employed since this compound acts as both antimicrobial and antioxidant.

Nowadays however we are witness to a growing attention towards the food safety and also in wine field the item is becoming very popular. The attention focused a lot on the sulphurous dioxide which can have adverse effects on the human health and can lead to chronic toxicities.

Other compounds and/or additives have therefore been proposed which can be potentially useful, but have several drawbacks. For example, the dimethyl carbonate is a microorganisms inhibitor which does not cause alterations on the wine flavor and color. Studies carried out by different authors have shown that however this compound does not effectively inhibit the lactic e acetic bacteria. On the other hand the lysozyme acts very well on bacteria but it does not have any effect on yeast cells and further causes an alteration of the wine color.

For these reasons, there is nowadays a particular attention to different typologies of barrel treatments; for example, various tests have been performed with ultrasounds, UV radiations or pulsed electrical fields for the sanitizing treatment alone.

On the other hand, the barrel treatment carried out by throwing granulated dry ice (dry ice blasting) in form of pellet or grains aroused higher interest. An example of a machine for carrying out such a treatment is described in the international patent application No. WO2008/005001. The machine includes a frame on which a barrel, previously deprived of one of the two ending walls, is set down with its cylindrical symmetry axis parallel to the resting plane of the frame itself. The barrel loading has to be done by manually rolling the barrel along proper ramps up to the treatment position.

Once loaded on the frame, the barrel is rotated around its axis while a head for throwing the granulated dry ice is linearly moved along a direction parallel to the rotation axis of the barrel. The throwing head can be further rotated and/or translated so as to reach all the inner surfaces to be treated.

Such a machine should allow not only to clean the barrel inside, thus removing the surface sediments, but also to possibly remove a thin wood layer increasing the throwing pressure of the granulated dry ice.

However, the removed debris are not expelled at all and spread out again on the side surface of the barrel during the rotation. Such a machine is less than practical in case an energetic rejuvenating treatment has to be carried out since it entails a further removal of the removed debris. In addition, some debris could fix again to the surface compromising the treatment efficiency.

In the same prior art document the problem of grounding the electrostatic charges is also solved by providing braids made of conducting material which have to contact the barrel, preferably at the metal rings, during its rotation. Such an expedient can also result less than effective, create jams of the conductor braids and spoil the outer surface of the barrel.

Object of the present invention is therefore to provide an improved machine and method to effectively carry out the treatment of a wooden barrel, in particular a barrique, by throwing granulated dry ice.

A further object of the present invention is to provide a machine and a method of the above mentioned type which allow to automatically and effectively carry out both the rejuvenating treatment and the sanitizing treatment of the barrel.

A further object of the present invention is to provide a machine and a method of the above mentioned type which allow the automatic total removal of the debris removed from the inner surface of the barrel.

These objects are achieved according to the invention by virtue of a machine according to claim 1 and a process according to claim 10. Further characteristics and details of the present invention are reported in the corresponding dependent claims. A machine according to the present invention allows to automatically carry out the treatment of a wooden barrel, in particular a barrique, by throwing granulated dry ice.

The machine comprises:

• a supporting frame having means for holding a barrel during the treatment;
• a first column with vertical axis on which the supporting frame is movably mounted;
• a treating head for throwing granulated dry ice;
• a second column, with a vertical axis parallel to that of the first column, on which the treating head is mounted with rotation capacity around the axis of the second column; and
• a device for producing granulated dry ice to be fed to at least one delivery nozzle placed on the treating head.

The so made machine allows to hold the barrel in a vertical and overturned position, so that to promote the full outflow of the debris removed from the inner surface of the barrel. In fact, if on one hand the dry ice does not leave any sediment since it instantaneously sublimates on contact with the barrel surface, on the other hand the debris detached from the barrel have to be immediately removed to prevent them from adhering again to the barrel surface, as it usually occurs for the known machines of the prior art.

The barrel is further held on the supporting frame, preferably made of metallic material, which is in its turn movably mounted on a column, made of metal too. A constant grounding of the barrel is thus ensured, in order to discharge possible electrostatic charges generated during the treatment, without having to devise bulky artifices potentially adverse for the machine and for the barrel under treatment.

In particular, the supporting frame is movable between a loading position on the supporting frame, also corresponding to the unloading position of the barrel from the frame, and one or more treatment positions.

The movement of the supporting frame starting from the loading position is realized, for example, by means of a translation in a direction parallel to the vertical axes of the columns for raising the barrel from the loading position; the frame is therefore rotated with respect to the axis of the first column until the barrel has its own axis coincident with the axis of the second column on which the treating head is installed. The barrel is preferably loaded with the open side facing upwards: in this case, in the rotation pathway that the frame performs around the axis of the first column before reaching the treatment positions, the barrel will be brought in a position in which it can be overturned, that is in a position in which a rotation of 180° of the supporting frame with respect to an axis perpendicular to the axis of the first column, can be performed.

At this point the supporting frame is moved by translation along a direction parallel to the vertical axes of the columns to bring in succession the barrel surfaces at the treating head height. Meanwhile, the treating head rotates around the axis of the column on which it is installed, whereas the delivery nozzle arranged on the treating head can be rotated in its turn around an axis perpendicular to that of the second column, on which the treating head is mounted, and it can be further translated along a direction perpendicular to the axis of the second column so that, by the combination of all these movements, all the surface portions inside the barrel can be reached by the flow of granulated dry ice.

The machine includes preferably a sealable chamber in which the columns are housed, together with the respective supported elements, in particular the supporting frame for the barrel and the treating head. It is suitably provided a machinery which allows the suctioning and filtering of the produced debris which precipitate on the bottom of the chamber during a barrel treatment.

The chamber in which the treatment is carried out is part of a room further comprising a compartment for housing at least one device for producing granulated dry ice to be fed to the delivery nozzle. The same room compartment, or possibly a separated compartment, allows to accommodate a controller for automating the barrel treatment cycle.

The room is preferably sized to allow the transportation thereof by a truck, so that to allow the temporary use in a site, such as for example a wine cellar or a wine producer. In the room at least one compartment can be provided as well for accommodating possible compressed air tanks, as well as at least one compartment wherein a separated device for producing and throwing the granulated dry ice is accommodated. This second device, provided for the manual employ, can be used to treat the ending wall too which is removed from the barrel or barrique subjected to the automated treatment inside the sealing chamber.

The invention further relates to a process for treating a wooden barrel, in particular a barrique, by throwing granulated dry ice. The process according to the invention comprises the steps of:

1. a) removing at least one of the ending walls of the barrel;
2. b) loading the barrel on a supporting frame and holding it in position on the frame, the frame being movably mounted on a first column with vertical axis;
3. c) displacing the supporting frame from the loading position of the barrel to one or more treatment positions, the displacement entailing at least one translation movement of the frame in a direction parallel to the vertical axis of the first column, at least one rotation movement of the frame with respect to the axis of the first column and at least one rotation movement with respect to an axis perpendicular to the axis of the first column;
4. d) treating the barrel by means of granulated dry ice thrown against the barrel walls from a treating head through a delivery nozzle, the treating head being mounted on a second column with a vertical axis parallel to that of the first column; at the same time the treatment entails the translation of the supporting frame along a direction parallel to the vertical axes of the columns, the rotation of the treating head around the vertical axis of the second column, the rotation of the delivery nozzle around an axis perpendicular to that of the second column and the translation of the nozzle itself along a direction perpendicular to the axis of the second column.

The treatment is carried out in a sealed chamber, in which the suction and filtering of the debris produced during the barrel treatment are further carried out.

A first device for producing granulated dry ice feeds the delivery nozzle arranged inside the chamber, whereas a second device for producing and throwing granulated dry ice, arranged outside of the chamber, is manually used to treat the ending wall removed from the barrel subjected to the automatic treatment.

The treatment cycle of the barrel is automated by a controller which controls at least the movements of the supporting frame, the treating head and the delivery nozzle; the controller further controls the first device for producing granulated dry ice in order to determine at least the average grain size, the flow rate and the throw pressure of the granulated dry ice.

In particular, the process according to the present invention can consist in a rejuvenating (or refreshing) treatment: in this case, granulated dry ice is produced with average grain size comprised between 2 and 4 mm and flow rate comprised between 0.8 and 1.2 kg/min; the granulated dry ice is thrown by a flow of compressed air at a pressure comprised between 5 and 7 bar with flow rate comprised between 3 and 5 m³/min. With these parameters a rather strong treatment is obtained which causes the removal of all the sediments from the barrel walls and of the surface layer of exhausted wood, thus allowing to restore a correct oxygenation of the wine, or liquor, which will be placed in the barrel for maturation or refinement.

The process according to the present invention can consist in a sanitizing treatment: granulated dry ice is produced with average grain size comprised between 0.5 and 1.5 mm and flow rate comprised between 0.4 and 0.8 kg/min; the granulated dry ice is thrown by a flow of compressed air at a pressure comprised between 7 and 10 bar with flow rate comprised between 4 and 6 m³/min. With these parameters a treatment is obtained more delicate than the rejuvenating treatment: the treatment does not substantially alter the surface layer of the barrel but allows to eliminate and remove possible microorganisms developed inside the barrel during previous uses before the barrel itself is filled again.

It has been found that a machine according to the present invention is able to assure the reproducibility of each treatment, regardless from the exploiting conditions of barrels subjected to treatment. Another advantage provided by the present invention is further given by the fact that optimal results can be obtained also by treatment cycles limited to few tens of minutes.

Further characteristics and advantages of the present invention will be more evident in the following description, for illustrative and not limitative purposes, referring to the attached drawings, wherein:

• Figure 1 is a section view of a chamber in which the treatment in a machine according to an embodiment of the present invention is carried out;
• Figure 2 is a top plan view of a complete room, in which the treatment chamber illustrated in Figure 1 is embedded;
• Figures 3 to 7 show the movements imparted to the barrel in the chamber in which the treatment is carried out to bring it from the loading position to the treatment positions;
• Figure 8 shows the movements imparted to the barrel during the treatment cycle, as well as the movements of the various components of the machine according to the embodiment depicted in the previous figures; and
• Figure 9 shows a typical barrique on which the areas are pointed out in which samplings of possible microorganisms have been carried out, before and after the treatment, during the tests carried out to verify the efficiency of the sanitizing process of the barrels according to the present invention.

In Figures 1 and 2 a machine is schematically depicted for treating a wooden barrel having shape and sizes typical of a barrique, that is to say a wooden barrel with capacity not higher than 350 liters and, more in particular, with capacity generally comprised between 220 and 250 liters.

The machine is enclosed in a room 10 and comprises a chamber 15 provided with a door 16 for sealing the chamber during the treatment. The room 10 is such sized to allow the loading and transportation thereof on the rear load body or on the floor of a truck.

Inside the chamber 15 a first column 20 is arranged, with vertical axis 20a, on which a supporting frame 30 is movably mounted to hold a barrel during the treatment. The frame 30 comprises a loading base 31 realized with profiles which end at the front with a slightly open fork pair 32 to facilitate the loading of the barrel on the frame 30; on the opposite side a retaining base 33 is provided composed by profiles joined together to form an hexagon. The two bases 31 and 33 are connected to each other by tension wires 34 stressed under tension by pneumatic cylinders 35 operable by control.

The frame 30 is rotatable around an axis 30a perpendicular to the axis 20a of the column 20 and is translatable in a direction parallel to the axis 20a, as indicated by the double arrow T1 in Figure 1.

In the chamber 15 a second column 40 is further arranged with axis 40a parallel to the axis 20a of the first column 20. On the column 40 a treating head 42 is mounted which includes a delivery nozzle 45. The treating head 42 is rotatable around the axis 40a of the column 40, as indicated by the double arrow T2 in Figure 2; the delivery nozzle 45 is further rotatable for about 90° around an axis 45a (Fig. 2), as indicated by the double arrow U2 in Figure 1, and it is further translatable (double arrow U1) along a direction perpendicular to the axis 40a of the column 40. On the treating head 40 also a sensor (not shown) is preferably provided able to provide a signal representative of the distance between the delivery nozzle 45 and the surface of the barrel subjected to treatment.

Inside the chamber 15 a machinery 50 is further arranged for the suction and filtering of the debris produced during the treatment of a barrel. The machinery 50 comprises for example a suctioning fan 51 and a couple of filtering cartridges 52. During the treatment, the air is picked up from the outside through a suctioning port 55 (Figure 1) and discharged downstream of the filtering cartridges through discharging ports or suitable piping (not shown). The suctioning port 55 is closed by a trapdoor 56 which can be conveniently controlled for the opening and closing by operating for example a pneumatic cylinder (not shown).

The room 10 further includes different compartments, indicated by the references 5, 6 and 7, all accessible through closable doors. For example, the compartment 5 can house switchboards and compressed air/electrical power connections and the compartment 7 can house possible tanks 77 to store a compressed air stock.

The compartment 6 houses a first device 60 for producing granulated dry ice to be fed to the delivery nozzle 45 through a flow of compressed air, the latter provided for example by the tanks 77 or the connections present in the compartment 5. A second device 61 for producing and throwing the granulated dry ice is also arranged in the compartment 6 and is intended for the manual employ to treat, for example, the ending wall which is removed from the barrel before the latter is subjected to the treatment inside the chamber 15.

In the compartment 6 a controller 70 is further housed which allows to automate the treatment cycle of a barrel in the chamber 15. For example, the controller 70 can control the movements of the supporting frame 30, of the treating head 42 and the delivery nozzle 45, as well as the opening and closing of the trapdoor 56. The controller 70 further controls the operation of the device 60 for producing granulated dry ice to be fed to the delivery nozzle 45 to determine the treatment parameters, for example the average grain size of granulated dry ice, the flow rate of the grains, as well as the pressure and flow rate of compressed air for throwing the grains.

On the partition portion between the compartment 6 and the chamber 15 a window 66 can be arranged as well sealed by a clear panel and facing towards the treating head 42 to allow a visual inspection inside the chamber 15 during the barrel treatment.

In Figures from 3 to 7 various steps of handling a barrel B to bring it from the loading position up to the treatment area are illustrated.

In Figure 3 a barrel 1 is depicted as loaded on the supporting frame 30 after an ending wall of the barrel has been removed; the barrel is loaded on the frame 30 with the closed end which is rested on the loading base 31, whereas the open end looks upwards and is facing to the retaining base 33. The axis 1a of the barrel 1 is substantially parallel to the axis 20a of the column 20. In this position, before starting the handling of the barrel 1, the pneumatic cylinders 35 are operated to apply a tensile force on the tension wires 34 and firmly lock in position the barrel in the frame. The tension wires 34 and the cylinders 35 can possibly be coated with soft material to prevent the outer surface of the barrel from being damaged.

At this point, once the barrel has been loaded on the supporting frame 30, the door 16 can be closed to start the treatment cycle. A safety switch (not shown) enables the cycle start only when the door 16 is in a closed condition.

From the position of Figure 3, the supporting frame is raised (arrow TU) until reaching the position of Figure 4. From this position, the frame 30 on which the barrel 1 is loaded is rotated of about 90° (arrow R1) around the axis 20a of column 20 until reaching the position shown in Figure 5. The barrel 1 on the frame 30 is still oriented in the same loading position, that is to say with its open end facing upwards, whereby it has to be rotated of 180° around the rotation axis 30a of the frame 30 (arrow R2) to overturn and bring it to the position of Figure 6.

A further rotation R3 of the frame 30 around the axis 20a of the column 20 brings the barrel in the position of Figure 7, in which the axis 1a of the barrel 1 is aligned with the axis 40a of the column 40. From this position, the frame 30 is translated downwards (arrow D1) until the barrel is brought at the treatment area.

During the treatment, the frame 30 is translated one or more times (double arrow UD) in a direction parallel to the axis 20a of the column 20 and the treating head 42 is rotated around the axis 40a of the column 40; at the same time, the delivery nozzle 45 is also rotated around the axis 45a and translated along a direction U1 (Figure 2) perpendicular to the axis 40a.

During the treatment, the trapdoor 56 is opened to allow the correct operation of the machinery for suctioning and filtering the debris removed from the barrel. While the automated treatment cycle is in progress, the ending wall preventively removed from the barrel 1 can be manually treated by using the device 61 for producing and throwing granulated dry ice.

Further characteristics of the present invention are disclosed in the following examples related to experimental tests of treatments carried out by a machine according to the present invention.

The same rejuvenating treatment by dry ice blasting has been carried out on two barriques used for the refinement of red wines; the first barrique has been used without any intermediate treatment for three consecutive years, whereas the second barrique has been used without any treatment for five consecutive years.

Both the barriques had an inner surface with sediments stratified at different filling levels.

Before subjecting the barriques to the treatment inside the machine, a preliminary treatment test has been carried out on an ending wall removed from each barrique to verify the efficiency thereof and determining the optimal treatment parameters.

The treatment of the removed ending walls has been manually carried out by using a portable device for producing and throwing the granulated dry ice. The object was to reveal again the surface of new wood while verifying the grain production parameters and throwing parameters thereof.

It has been visually verified that the best results were obtained with a rather strong treatment. The granulated dry ice production has been set with average grain size of 3 mm and flow rate of about 1kg/min; the grains were thrown by means of a flow of compressed air at the pressure of 6 bar and flow rate of about 4 m³/min.

Therefore the same parameters have been set in the controller of a machine implemented according to the present invention to carry out the rejuvenating treatment on the two barriques. For both the barriques the treatment planned by the controller had an overall duration of about 8 minutes, except the loading and unloading times.

At the end of the treatment, the wood inside both barriques resulted safe and clean.

For the test two barriques have been used, preventively used for the refinement of different red wines from a wine producer.

As shown in Figure 9, for each of the two barriques two areas have been identified, that is to say:

• area D, corresponding to an area composed by staves, on the inner surface of the barrel in a position opposite to the bunghole;
• area F corresponding to the center area of the bottom.

In Table 1 below the two barriques have been denoted by numbers 1 and 2. F1 (barrique 1, Bottom) D1 (barrique 1, Stave) F2 (barrique 2, Bottom) D2 (barrique 2, Stave)

The sampling has been carried out by using a 10 x 10 cm sterile mask so that to sample an area of 100 cm². The areas have been sampled before and after the dry ice blasting treatment by using sterile cotton pads scraped and rotated against the surface in three directions, that is to say horizontal, vertical and diagonal.

Each pad has then been dipped in 10 ml of peptonate sterile solution and stored in ice until the next isolation.

Each 10 ml assay has then been subjected to two different isolation techniques:

• for the pre-dry ice blasting samplings a serial dilution has been carried out in peptonate water up to 10^-1 dilutions and subsequent seeding on specific plate count agar;
• for the post-dry ice blasting samplings, assuming a low load, an isolation on a filtering membrane of 0.20 µm by filtering 2.0 ml of sample has been carried out. For the microorganisms culturing the following mediums have been used:
• WL + Ampicillin for the isolation of yeasts and mildews;
• MRS + Actidione for the isolation of lactic bacteria.

With respect to the rejuvenating treatment disclosed in Example 1, it has been assumed that the sanitizing treatment can also be carried out less aggressively. In fact the object is to eliminate and/or neutralize the microorganisms present inside the barrel, regardless from restoring the original conditions which allow the oxygenation by osmosis of the wine under maturation.

Therefore the parameters set in the controller of the same machine already used in Example 1 have been the following:

• production of granulated dry ice with average grain size comprised between 0.5 and 1 mm and flow rate of about 0.6 kg/min;
• throw of the grains by a flow of compressed air at the pressure set to 8 bar with flow rate of about 5 m³/min.

For both barriques the treatment planned by the controller had an overall duration of about 8 minutes, except the loading and unloading times.

The samplings carried out on the barriques before the treatment pointed out the growth of mildews alone (fungi). Yeast species have not been detected, nor lactic bacteria have been detected in the samplings carried out before and after the treatment with dry ice blasting.

The treatment effect results apparent in Table 2 below, with a decrease comprised between 90 and 98% of fungi after the treatment. In barrique 1, at Staves area (D1), fungi cells have not been detected before and after the treatment. (fungi: CFU/100 cm²)* reference sample Sampling before the treatment Sampling after the treatment abatement % F1 (barrique 1, Bottom) 100 5 95% D1 (barrique 1, Stave) 0 0 - F2 (barrique 2, Bottom) 800 15 98% D2 (barrique 2, Stave) 100 10 90% *Fungi load expressed in CFU (colony forming units) on the wood surfaces before and after the treatment

The appearance of the wood surface of the barriques after the treatment is however clearer than the starting appearance. In fact, despite using granulated dry ice having a grain size lower than that used in Example 1, and thus having a kinetic energy lower than the previous, the efficiency of employing dry ice has been verified, also with small size particles, for "weakening" the sediments present in the barriques due to the immediate sublimation of grains when they impact the inner surface of a barrique.

Taking into account the results of tests carried out in the Example 2, wherein neither yeasts nor contaminants bacteria ascribable to species present in the wine, particularly to the Brettanomyces strain or lactic bacteria, have been detected, some experimental models have been setup to simulate a barrel wood pollution by the microorganisms.

The models consisted of Durmast oak wood sections, having suitable sizes, taken from a used barrique and inserted in respective sterile plastic trays. Each tray has been filled with wine preventively contaminated with the microorganisms at a predetermined load.

For every microorganisms 3 wooden models have been setup that have been obtained from a bottom of the used barrique (F1A, F2A, F3A), 3 wooden models obtained from the opposite bottom (F1B, F2B, F3B), 3 models of a first stave A (D1A, D2A, D3A) and 3 models of a second stave B (D1B, D2B, D3B), in this way simulating the pollution assay on 6 barriques, three for the bacteria and three for the Brettanomyces species.

As regard to the Brettanomyces bruxellensis-contaminated models, the ISE 372 strain belonging to the CRA (Centro di Ricerca per l'Enologia di Asti) collection, preventively multiplied in YEPG (Yeast peptone glucose) liquid medium, has been used. The wine inoculum has been carried out at the load of 1 x 10⁶ cells per mL, a concentration typically present in contaminated wines in barrique.

For the models which simulate the bacteria contamination the Lactobacillus brevis species has been used, multiplied on MRS, at the load of 2 x 10⁶ cell/mL, a concentration coherent with the wine cellar bacteria development.

The models have been incubated at room temperature for 5 days in order to allow the microorganism adaptation to the wine and implantation thereof between the wood pores. Afterwards, the wine has been eliminated and the pre-dry ice blasting sampling has been carried out. After the dry ice blasting treatment the models have been inserted in sterile bags and brought back to the laboratory to carry out the post-dry ice blasting sampling.

The models treatment has been manually carried out using a portable device for producing and throwing the granulated dry ice. The parameters settings were the same as the settings in the machine used in Example 2 for the sanitizing treatment of the barriques, that is to say by setting the production of granulated dry ice with average grain size comprised between 0.5 and 1 mm and flow rate of about 0.6 kg/min and setting the throwing of granulated dry ice by a flow of compressed air at the pressure of 8 bars with flow rate of about 5 m³/min.

For the counting of Brettanomyces and lactic bacteria two different methods have been employed.

For the test with Brettanomyces contact plates of YEPD medium added with Rose Bengal have been used. The Rose Bengal is an additive which limits the mildew growth allowing the count and identification of yeasts. On the model with wood sampled from the bottom of the barrique two samplings have been carried out with the plates. These samplings have been repeated before and after the treatment with dry ice blasting, taking care of sampling the same area.

The barrique models contaminated with wine containing lactic bacteria have been sampled by using sterile pads, as previously described in Example 2, taking a surface of 1 dm². The pads have then been placed in 10 ml of sterile peptonate solution. This solution has then been plated, by spread plate method, on MRS medium containing actidione, a substance which inhibits the development of yeasts and mildews to allow the selective growth of lactic bacteria. After the dry ice blasting treatment, the surfaces of the models have been sampled with sterile pads, and thus the diluting solution of peptonate water has then been filtered using a 0.2 µm sterile filter; this operation allows to concentrate possible cells present in solution. The filters have been placed on bacteria-selective MRS medium, as described for the step before the dry ice blasting treatment.

The samplings carried out before the dry ice blasting treatment pointed out the presence and the implantation of the Brettanomyces strain on the surface of the used barrique model. All the contact plates show a carpet of Brettanomyces colonies with a load higher than 10000 CFUs (Colony Forming Units) per dm². In addition to the Brettanomyces species few mildews CFUs have been detected whose development has been monitored by the presence of the Rose Bengal stain.

The samplings carried out after the dry ice blasting treatment pointed out the presence of a limited number of colonies, ranging from a minimum of 0 to a maximum of about 200 CFUs per plate. The percent abatement ranges between values over 97.8% up to 100%, taking the pre-dry ice blasting values of 10000 CFU into account. In the post-treatment samplings no mildews colonies have been detected.

The set of load values detected for the Brettanomyces species are shown in Table 3 below. Brettanomyces model CFU/ 100 cm²* reference sample Sampling before the treatment Sampling after the treatment abatement % F1A (barrique 1, Bottom, Area A) >10000 0 100.00% F1B (barrique 1, Bottom, Area B) >10000 0 100.00% F2A (barrique 2, Bottom, Area A) >10000 25 >99.75% F2B (barrique 2, Bottom, Area B) >10000 21 >99.79% F3A (barrique 3, Bottom, Area A) >10000 30 >99.70% F3B (barrique 3, Bottom, Area B) >10000 38 >99.62% D1A (barrique 1, Stave, Area A) >10000 105 >98.95% D1B (barrique 1, Stave, Area B) >10000 42 >99.58% D2A (barrique 2, Stave, Area A) >10000 68 >99.32% D2B (barrique 2, Stave, Area B) >10000 114 >98.86% D3A (barrique 3, Stave, Area A) >10000 211 >97.89% D3B (barrique 3, Stave, Area B) >10000 127 >98.73% * Brettanomyces yeast loads expressed in CFUs (Colony Forming Units) on the surface of the woods before and after the dry ice blasting treatment.

After a 72 hours incubation at 30°C a high cell growth has been observed with a load over 10000 CFUs on all plates obtained from the samplings before the dry ice blasting treatment; this demonstrates that also lactic bacteria can be nested in barrique wood.

After the dry ice blasting treatment a very low cell growth has been observed, also upon the concentration carried out with the sterile filtering membrane system; the CFU number ranges from 0 to 20 per dm².

By taking into account the starting cell number of 10000 (which can be considered a surely underestimated amount), it can therefore be concluded that the dry ice blasting treatment allows to obtain an abatement of the bacterial load higher than 99%. The data referring to all carried out samplings are shown in Table 4 below. Bacteria model CFU 100 cm²* reference sample Sampling before the treatment Sampling after the treatment abatement % F1A (barrique 1, Bottom, Area A) >10000 6 >99.94% F1B (barrique 1, Bottom, Area B) >10000 10 >99.90% F2A (barrique 2, Bottom, Area A) >10000 22 >99.78% F2B (barrique 2, Bottom, Area B) >10000 20 >99.80% F3A (barrique 3, Bottom, Area A) >10000 0 100.00% F3B (barrique 3, Bottom, Area B) >10000 0 100.00% D1A (barrique 1, Stave, Area A) >10000 0 100.00% D1B (barrique 1, Stave, Area B) >10000 0 100.00% D2A (barrique 2, Stave, Area A) >10000 0 100.00% D2B (barrique 2, Stave, Area B) >10000 0 100.00% D3A (barrique 3, Stave, Area A) >10000 0 100.00% D3B (barrique 3, Stave, Area B) >10000 0 100.00% * Lactobacillus brevis bacteria load expressed in CFUs (Colony Forming Units) on the surface of the woods before and after the dry ice blasting treatment.

Various modifications can be made to the embodiments herein illustrated for illustration purposes only. For example, the barrel could be loaded as already overturned on the supporting frame, that is to say with the open side facing downwards, and therefore already oriented in the correct position for the treatment, thus limiting the number of movements to carry out in order to move the barrel from the loading position to the treatment area.