WO2020160644A1

WO2020160644A1 - Apparatus and method for preserving the aroma of a fermentable beverage

Info

Publication number: WO2020160644A1
Application number: PCT/CA2020/050115
Authority: WO
Inventors: Richard L. Jones
Original assignee: Aromaloc Inc.
Priority date: 2019-02-06
Filing date: 2020-01-31
Publication date: 2020-08-13
Also published as: EP3921403A4; AR118018A1; EP3921403A1; AU2020218378A1; CL2021002055A1

Abstract

A fermentation apparatus and method for preserving the aroma of a fermentable beverage is provided. The fermentation apparatus comprises a flow passage fluidly connected to the headspace located above a fermentable beverage in a fermentation container. A rubbery membrane in the flow passage receives a gaseous headspace fluid comprising carbon dioxide gas and a first concentration of an aromatic fluid. The rubbery membrane strips a portion of the aromatic fluid from the gaseous headspace mixture thereby reducing a concentration of the aromatic fluid in the headspace mixture in the upstream portion of the flow passage. A first portion of the gaseous headspace mixture from the fermentation apparatus is discharged, the first portion having a second concentration of aromatic fluid less than the first concentration. Carbon dioxide is diffused from a second portion of the gaseous headspace mixture through the rubbery membrane, the diffused carbon dioxide gas entraining the aromatic fluid at the permeate side of the rubbery membrane. The flow passage directs the permeated modified fluid back to the headspace to at least partially retain the aromatic fluid in the fermentable beverage in the fermentation container.

Description

TITLE: APPARATUS AND METHOD FOR PRESERVING THE AROMA OF A

FERMENTABLE BEVERAGE

CROSS-REFERENCES TO RELATED APPLICATIONS

[0001 ] This application claims priority to U.S. Patent Application No. 16/268,990, filed on February 6, 2019, which is a continuation-in-part of United States Application Serial No. 14/801 ,947, filed July 17, 2015, which is a continuation-in-part of PCT Application Serial No. PCT/CA2014/000359, filed April 17, 2014. PCT Application Serial No. PCT/CA2014/000359 claims priority to United States Provisional Application Serial No. 61/813,529, filed April 18, 2013. The entire contents of United States Application Serial No. 14/801 ,947, PCT Application Serial No. PCT/CA2014/000359 and United States Provisional Application Serial No. 61/813,529 are hereby incorporated by reference.

FIELD

[0002] The described embodiments relate to a fermentation apparatus and method for preserving the aroma of a fermentable beverage.

INTRODUCTION

[0003] There are numerous fermented beverages, including, for example, wine, beer, spirits, ciders, and champagnes. As an example, wine fermentation typically begins with pressed grape juice or crushed grapes including both the juice and grape skins. The juice and skins contain aroma compounds that are unique to the type of grape and the locale where the grapes were grown. When the juice is inoculated with yeast to initiate fermentation, the yeast cells consume the sugar in the grape juice and in the process produce ethyl alcohol and carbon dioxide plus a wide variety of yeast-derived aroma compounds and waste by-products. The aroma compounds can desirably affect the smell and/or taste of the wine and the enjoyment associated with drinking the fermented wine. [0004] As part of the typical fermentation process, carbon dioxide gas is vented to the atmosphere during fermentation to avoid increasing pressure in the fermentation container. Some exceptions to this involve the production of carbonated beverages where a small supplemental fermentation is conducted inside an already bottled beverage. In typical fermentation, the carbon dioxide gas takes with it some of the desirable aroma compounds. In turn, the consumable fermented beverage is deprived of at least some desirable aroma compounds that would have otherwise enhanced the taste and/or smell of the fermented beverage. In some cases, desirable aroma compounds escape with carbon dioxide by diffusing into carbon dioxide bubbles. In some cases, desirable aroma compounds diffuse directly into the headspace from the surface of a fermenting beverage.

[0005] The enjoyment associated with drinking fermented beverages, such as wine or beer, for example, may be affected by aroma compounds present in the beverage. Some of these aroma compounds are representative of the base products, such as the variety of grape used in wine making, for example. Important aromas are also derived from yeast metabolism during fermentation of the beverage. Some of these aroma compounds have a low gas/liquid partition coefficient which makes them less volatile and helps keep them contained in the beverage during fermentation where they remain, for the most part, in the finished beverage. Other aroma compounds have higher gas/liquid partition coefficients. These aroma compounds are relatively more volatile and will diffuse into the headspace above the fermenting beverage, which exposes them to loss during fermentation when carbon dioxide gas, produced during fermentation, is released from the system and takes with it some of the volatile aroma compounds. This decreases the concentration of the aroma compounds in the finished, consumable beverage. This loss of aroma in the escaping carbon dioxide gas is sometimes referred to as“carbon dioxide stripping”. [0006] Aroma compounds with higher gas/liquid partition coefficients are important since, in order to detect aroma, the aroma compounds are typically volatile. Notably, more volatile aroma compounds are relatively more susceptible to carbon dioxide stripping. Esters, for example, are a group of aroma compounds important in giving wine its fruity flavor, but tend to have higher gas/liquid partition coefficients and are relatively susceptible to carbon dioxide stripping. Desirable aroma compounds may include, but are not limited to, ethyl hexanoate, ethyl butyrate, ethyl octanoate, isoamylacetate and hexylacetate.

[0007] The movement of volatile aroma compounds from the beverage to the headspace above the beverage depends on the concentration difference between the two compartments. The higher the concentration difference, the greater the movement of aroma from the fermenting beverage to the headspace, and the greater the“carbon dioxide stripping”.

[0008] Fermenting beverages can be complicated due to the constantly changing levels of sugar and ethanol during the fermentation process, along with the presence of proteins, all of which affect the transfer of aroma to the headspace, and subsequent loss of aroma through carbon dioxide stripping. Depletion of sugar and the production of ethanol both decrease the gas/liquid partition coefficient thus helping to preserve what is left of the aromatic esters in the fermenting beverage by limiting carbon dioxide stripping.

[0009] The loss of aroma from carbon dioxide stripping may also be temperature dependent. Higher temperatures during fermentation can deplete certain esters due to their lower production at high temperature and also their accelerated loss due to a higher gas/liquid partition coefficient and higher rates of carbon dioxide production by the more active yeast metabolism. Therefore, production of fruity wines is best done at lower fermentation temperatures, but this lengthens the wine making process. Low fermentation temperatures can also lead to unreliable and stuck fermentations when using certain yeasts. Many of the most desirable aroma compounds have a high gas/liquid partition coefficient and, in some cases, more than 70% of these aroma compounds can be lost by carbon dioxide stripping when fermentation is conducted at a relatively warm temperature.

[0010] Winemakers, for example, have attempted to preserve the fruity aroma and flavor of white wine by fermenting grape juice at cool temperatures. This decreases the gas/liquid partition coefficient, thus helping to keep the aroma compounds dissolved in the wine while at the same time a cooler temperature slows yeast metabolism which decreases the rate of carbon dioxide production. The end result is decreased loss of aroma by carbon dioxide stripping and more aroma retention in the wine. However, aroma loss typically still does occur and producing the wine takes longer.

[0011 ] For red wine production, for example, color extraction from the red grape skins is important as is extraction of tannins and other compounds from the skins. Red wine is generally fermented at warmer temperatures to achieve these goals, but the resultant higher temperatures enhance the loss of varietal and yeast-derived aromas due to the temperature-dependent increase in gas/liquid partition coefficients and increased rate of carbon dioxide production due to more vigorous yeast metabolism. So, regardless of the type of wine produced, there is often carbon dioxide stripping of desirable aroma compounds.

[0012] The issue of aroma loss during fermentation is an important one and a variety of methods have been advanced for preserving some of the aroma in wine or beer. However, these methods often require tampering with the beverage or capturing and isolating aroma compounds during fermentation for reintroduction to the beverage after fermentation. Some of these existing methods require the use of complex refrigeration or vacuum equipment for treating or isolating the aroma compounds. It would be desirable to have an apparatus and method for selectively reducing, removing, absorbing, or the like (collectively“scrubbing”) carbon dioxide while preserving aroma compounds during fermentation of a beverage; preferably, without tampering directly with the fermenting beverage, or capturing aroma compounds and then returning them later directly into the fermented liquid beverage. It would also be desirable to have an apparatus and method whereby aroma compounds are not isolated and captured but rather are allowed to recirculate back to the headspace from which they originated.

[0013] There exists a need for an improved apparatus and method for preserving the aroma of a fermentable beverage.

SUMMARY

[0014] The following is intended to introduce the reader to the more detailed discussion to follow. The summary is not intended to limit or define the claims.

[0015] According to one broad aspect of this disclosure, a fermentation apparatus is provided for preserving the aroma of a fermentable beverage. The fermentation apparatus comprises a closure engageable with a fermentation container containing a fermentable beverage. The fermentation container has at least one port. The fermentation apparatus comprises a flow passage coupled to the closure. The flow passage is fluidly connectable to the at least one port of the fermentation container when the closure engages the fermentation container to fluidly connect a headspace located above the fermentable beverage in the fermentation container with the flow passage. The fermentation apparatus comprises a carbon dioxide scrubber in the flow passage. The carbon dioxide scrubber receives from the at least one port a headspace fluid mixture made up of at least carbon dioxide gas and an aromatic fluid originating from the fermenting beverage. When the headspace fluid mixture contacts the carbon dioxide scrubber, the carbon dioxide scrubber permits at least a portion of the carbon dioxide gas to exit the flow passage and retains at least a portion of the aromatic fluid in the flow passage to thereby retain a modified fluid in the flow passage. The modified fluid has a lower carbon dioxide gas concentration and a higher aromatic fluid concentration than the headspace fluid mixture. The flow passage directs the modified fluid to the at least one port in the fermentation container to direct the modified fluid through the reentry port of the fermentation container to reenter the headspace to at least partially retain the aromatic fluid in the fermentable beverage in the fermentation container.

[0016] In some embodiments, the at least one port comprises an exit port and a reentry port. In these embodiments, the flow passage is fluidly connectable to the exit port of the fermentation container when the closure engages the fermentation container to fluidly connect the headspace located above the fermentable beverage in the fermentation container with the flow passage. In these embodiments, the carbon dioxide scrubber receives from the exit port the headspace fluid mixture comprising at least carbon dioxide gas and the aromatic fluid originating from the fermenting beverage. In these embodiments, the flow passage directs the modified fluid mixture to the reentry port of the fermentation container to direct the modified fluid mixture through the reentry port of the fermentation container to reenter the headspace to at least partially retain the aromatic fluid in the fermentable beverage in the fermentation container.

[0017] In some embodiments, the carbon dioxide scrubber comprises a carbon dioxide selective membrane. The carbon dioxide selective membrane may permit the carbon dioxide gas to pass therethough and out of the flow passage and to retain the aromatic fluid in the flow passage [0018] In some embodiments, the fermentation apparatus comprises a sweep gas directed across an exterior surface of the carbon dioxide selective membrane located external to the flow passage to lower the carbon dioxide partial pressure on an exterior side of the carbon dioxide selective membrane relative to a flow passage side of the carbon dioxide selective membrane, to facilitate the passage of the carbon dioxide gas through the carbon dioxide selective membrane and out of the flow passage.

[0019] In some embodiments, the fermentation apparatus comprises a negative pressure generator located external to the flow passage to direct the sweep gas across the exterior surface of the carbon dioxide selective membrane and permit decreased pressure on the exterior surface of the membrane.

[0020] In some embodiments, the carbon dioxide selective membrane includes at least one of: a planar gas selective membrane, a plurality of stacked planar membranes, a tubular hollow membrane and a bundle of tubular hollow membranes.

[0021 ] In some embodiments, the carbon dioxide selective membrane is a fixed-site carrier membrane.

[0022] In some embodiments, the carbon dioxide scrubber comprises a carbon dioxide absorber containing a carbon dioxide absorbing material. In these embodiments, the carbon dioxide absorbing material absorbs and removes the carbon dioxide gas from the flow passage.

[0023] In some embodiments, the flow passage comprises a release valve in fluid communication with the headspace. In these embodiments, the release valve is openable to vent a portion of at least one of the headspace fluid mixture and the modified fluid to the external atmosphere [0024] In some embodiments, the fermentation apparatus comprises a fluid pump in fluid communication with the headspace and the carbon dioxide scrubber to transport the headspace fluid mixture and the modified fluid through the flow passage.

[0025] In some embodiments, the fermentation apparatus comprises an expansion chamber having a flexible wall and containing an expansion gas therein, the expansion chamber being in fluid communication with the flow passage. An interior volume defined by the flexible wall may be adjustable to accommodate pressure fluctuations in the flow passage.

[0026] According to another broad aspect of this disclosure, a method is provided for preserving the aroma of a fermentable beverage. The method comprises fermenting the beverage in a fermentation container to produce a headspace fluid mixture comprising at least carbon dioxide gas and an aromatic fluid in a headspace located above the beverage contained in the fermentation container. The method comprises permitting the headspace fluid mixture to exit the container into a flow passage. The method comprises permitting the headspace fluid mixture to flow through the flow passage and into contact with a carbon dioxide scrubber to separate the carbon dioxide gas in the headspace fluid mixture from the aromatic fluid in the headspace mixture by permitting at least a portion of the carbon dioxide gas to exit the flow passage and retaining at least a portion of the aromatic fluid in the flow passage to thereby retain a modified fluid in the flow passage. The modified fluid has a lower carbon dioxide concentration and a higher aromatic fluid concentration than the headspace fluid mixture. The method further comprises permitting the modified fluid remaining in the flow passage after contacting the carbon dioxide scrubber to reenter the headspace to at least partially retain the aromatic fluid in the fermentable beverage in the fermentation container. [0027] In some embodiments, a fluid pump is in fluid communication with the headspace and the carbon dioxide scrubber to transport the headspace fluid mixture and the modified fluid through the flow passage.

[0028] In some embodiments, the carbon dioxide scrubber removes from the flow passage substantially all of the carbon dioxide gas that contacts the carbon dioxide scrubber. In some embodiments, the carbon dioxide scrubber retains in the flow passage substantially all of the aromatic fluid that contacts the carbon dioxide scrubber.

[0029] In some embodiments, the carbon dioxide scrubber is capable of removing the carbon dioxide gas from the flow passage at a removal rate at least as high as the maximum production rate at which the carbon dioxide gas is produced in the headspace by the fermentation of the beverage.

[0030] In some embodiments, permitting the headspace fluid mixture to flow through the flow passage into contact with the carbon dioxide scrubber occurs during the entire duration of the fermentation.

[0031 ] In some embodiments, permitting the headspace fluid mixture to flow through the flow passage into contact with the carbon dioxide scrubber occurs during less than the entire duration of the fermentation.

[0032] In some embodiments, the headspace fluid mixture comprises aromatic fluid and waste by-products produced by the fermentation of the beverage. In these embodiments, the headspace fluid mixture may flow into contact with the carbon dioxide scrubber during a first portion of the fermentation. The headspace fluid mixture may be permitted to escape from the flow passage to the external atmosphere via an airlock during a second portion of the fermentation, thereby allowing the aromatic fluid, the waste by-products and the carbon dioxide gas produced during the second portion of the fermentation to escape to the external atmosphere. The second portion of the fermentation occurs after the first portion of the fermentation. [0033] In some embodiments, the modified fluid includes a portion of the carbon dioxide gas and the portion of the aromatic fluid in the headspace mixture retained in the flow passage after contacting the carbon dioxide scrubber. A portion of the modified fluid may be permitted to exit the flow passage via a release valve in the flow passage, and the modified fluid remaining in the flow passage may be permitted to reenter the headspace.

[0034] In some embodiments, the carbon dioxide scrubber comprises a carbon dioxide selective membrane. In these embodiments, when the headspace fluid mixture contacts the carbon dioxide selective membrane, the carbon dioxide selective membrane permits the carbon dioxide gas in the headspace fluid mixture to pass therethrough and out of the flow passage and retains the aromatic fluid in the headspace fluid mixture in the flow passage.

[0035] In some embodiments, the carbon dioxide selective membrane is a diffusive membrane that permits the carbon dioxide gas to diffuse therethrough and out of the flow passage.

[0036] In some embodiments, a sweep gas is directed across an exterior surface of the carbon dioxide selective membrane located external to the flow passage to facilitate the passage of the carbon dioxide gas through the carbon dioxide selective membrane and out of the flow passage.

[0037] In some embodiments, the sweep gas includes one of: air, inert gas, and a combination of inert gas and air.

[0038] In some embodiments, the carbon dioxide scrubber comprises a carbon dioxide absorber containing a carbon dioxide absorbing material. In these embodiments, when the headspace fluid mixture contacts the carbon dioxide absorbing material, the carbon dioxide absorbing material absorbs the carbon dioxide gas from the headspace fluid mixture. [0039] In some embodiments, the carbon dioxide gas that exits the flow passage via the carbon dioxide scrubber is released to the external atmosphere.

[0040] In some embodiments, the carbon dioxide gas that exits the flow passage via the carbon dioxide scrubber is transferred to a carbon dioxide storage vessel fluidly connected to the carbon dioxide scrubber.

[0041 ] According to one broad aspect of this disclosure, a fermentation apparatus is provided for preserving the aroma of a fermentable beverage. The fermentation apparatus comprises a closure engageable with a fermentation container containing a fermentable beverage. The fermentation container has an exit port and a reentry port. The fermentation apparatus comprises a flow passage coupled to the closure. The flow passage is fluidly connectable to the exit port and the reentry port of the fermentation container when the closure engages the fermentation container to fluidly connect a headspace located above the fermentable beverage in the fermentation container with the flow passage. The fermentation apparatus comprises a carbon dioxide scrubber in the flow passage. The carbon dioxide scrubber receives from the exit port a headspace fluid mixture made up of at least carbon dioxide gas and an aromatic fluid originating from the fermenting beverage. When the headspace fluid mixture contacts the carbon dioxide scrubber, more of the carbon dioxide gas than the aromatic fluid in the headspace fluid mixture exits the flow passage via the carbon dioxide scrubber, to retain a modified fluid mixture in the flow passage. The flow passage directs the modified fluid mixture to the reentry port of the fermentation container to direct the modified fluid mixture through the reentry port of the fermentation container to reenter the headspace to at least partially retain the aromatic fluid in the fermentable beverage in the fermentation container. [0042] According to another broad aspect of this disclosure, a method is provided for preserving the aroma of a fermentable beverage. The method comprises fermenting the beverage in a fermentation container to produce a headspace fluid mixture comprising at least carbon dioxide gas and an aromatic fluid in a headspace located above the beverage contained in the fermentation container. The method comprises permitting the headspace fluid mixture to exit the container into a flow passage. The method comprises permitting the headspace fluid mixture to flow through the flow passage and into contact with a carbon dioxide scrubber to separate the carbon dioxide gas in the headspace fluid mixture from the aromatic fluid in the headspace mixture by permitting more of the carbon dioxide gas than the aromatic fluid to exit the flow passage via the carbon dioxide scrubber to retain a modified fluid mixture in the flow passage. The method comprises permitting the modified fluid mixture remaining in the flow passage after contacting the carbon dioxide scrubber to reenter the headspace to at least partially retain the aromatic fluid in the fermentable beverage in the fermentation container.

DRAWINGS

[0043] Reference is made in the description of various embodiments to the accompanying drawings, in which:

[0044] FIG. 1 is cut-away side view of a fermentation apparatus in accordance with an embodiment of the invention;

[0045] FIG. 2 is a cut-away side view of a fermentation apparatus in accordance with another embodiment comprising a carbon dioxide absorber.

[0046] FIG. 3 is a cut-away side view of a fermentation apparatus in accordance with yet another embodiment comprising a carbon dioxide selective membrane; [0047] FIG. 4 is a cut-away side view of a fermentation apparatus in accordance with a further embodiment comprising a carbon dioxide selective membrane and a sweep gas flow via a negative pressure system;

[0048] FIG. 5 is a cut-away side view of a fermentation apparatus in accordance with a further embodiment comprising a carbon dioxide selective membrane and a sweep gas flow via a positive pressure system;

[0049] FIG. 6 is a side elevation view of a fermentation apparatus in accordance with a further embodiment in which the beverage container has a single port;

[0050] FIG. 7 is a side elevation view of the fermentation apparatus of FIG. 3 operating with a headspace fluid mixture comprising aromatic fluid, carbon dioxide gas and undesirable waste by-products;

[0051 ] FIG. 8 is a side elevation view of a modified fermentation container operating with a headspace fluid mixture comprising aromatic fluid, carbon dioxide gas and undesirable waste by-products;

[0052] FIG. 9 is a side elevation view of a fermentation apparatus in accordance with a first experimental embodiment;

[0053] FIG. 10 is a side elevation view of a fermentation apparatus in accordance with a second experimental embodiment;

[0054] FIG. 11 is a graphical representation of the carbon dioxide gas removal/production rate and specific gravity of a fermented juice vs. fermentation time, as associated with the second experimental embodiment of FIG.10;

[0055] FIG. 12 is a graphical representation of the attributes of the experimental wine associated with the second experimental embodiment of FIG.10; [0056] FIG. 13 is a side elevation view of a fermentation apparatus in accordance with a fourth experimental embodiment;

[0057] FIG. 14 is an example of a scorecard used to score attributes of the experimental wine associated with the fourth experimental embodiment of FIG. 13;

[0058] FIG. 15 is a graphical representation of the attributes of the experimental wine associated with the fourth experimental embodiment of FIG.13;

[0059] FIG. 16 is a graphical representation of the percent change in aroma compounds of an experimental wine associated with the fourth experimental embodiment of FIG. 13 compared to a control wine;

[0060] FIG. 17 is a graphical representation of the percent change in aroma compounds of another experimental wine associated with the fourth experimental embodiment of FIG. 13 compared to a control wine;

[0061 ] FIG. 18 is a schematic illustration of an apparatus for preserving aroma in a fermenting beverage in accordance with another embodiment; and

[0062] FIG. 19 is a graph illustrating carbon dioxide gas production from a fermenting beverage over the course of an experiment.

DESCRIPTION OF VARIOUS EMBODIMENTS

[0063] Although the apparatus and methods may relate to the fermentation of wine, the apparatuses and methods described herein may be used during the fermentation of other beverages. In some cases, it is desirable to selectively remove, reduce, absorb, or the like (collectively“scrub”) one or more fluids while preserving one or more other fluids, such as, when fermenting beer, fermenting the base liquid used for spirit distillation, fermenting ciders, initially fermenting champagnes and other fermented beverages. [0064] Reference is made to FIG. 1 , which shows an exemplary first embodiment of fermentation apparatus 100 for preserving the aroma of fermentable beverage 102. The fermentation apparatus 100 includes closure 104 engageable with fermentation container 106 holding fermentable beverage 102. Fermentation container 106 has at least one port. As exemplified in FIG. 1 , the at least one port comprises exit port 108 and reentry port 110. Flow passage 112 is coupled to closure 104. Flow passage 112 fluidly connects to the at least one port (exemplified in FIG. 1 as exit port 108 and reentry port 110) of fermentation container 106 when closure 104 engages fermentation container 106. When closure 104 engages fermentation container 106, headspace 114 located above fermentable beverage 102 in fermentation container 106 is fluidly connected with flow passage 112. In some cases, exit port 108 and reentry port 110 are located in closure 104 above headspace 114 when closure 104 is engaged with fermentation container 106.

[0065] In some embodiments, closure 104 is a lid that fluidly seals container 106. The lid may be snap-fit, friction-fit, fit with a pneumatic seal or threaded onto container 106, for example. In some embodiments, closure 104 is removably engageable with container 106. Alternatively, closure 104 may be engageable with fermentation container 106 by being integrally formed therewith. In some cases, closure 104 may be fixedly connected to container 106.

[0066] As illustrated in FIG. 1 , flow passage 112 has carbon dioxide scrubber 116. Carbon dioxide scrubber 116 receives from the at least one fluid port (exemplified in FIG. 1 as exit port 108) headspace fluid mixture 150 comprising, at least, carbon dioxide gas 152 and aromatic fluid 154 originating from fermentable beverage 102. Aromatic fluid 154 comprises desirable aroma compounds that enhance the taste and/or smell of fermentable beverage 102 when it is consumed after fermentation. Exemplary aromatic fluids 154 (which enhance beverage taste and/or smell) include, but are not limited to, ethyl hexanoate, ethyl buryrate, ethyl octanoate, isoamylacetate and hexylacetate.

[0067] When headspace fluid mixture 150 contacts carbon dioxide scrubber 116, carbon dioxide scrubber 116 permits at least a portion of carbon dioxide gas 152 to exit flow passage 112 and retains at least a portion of aromatic fluid 154 in flow passage 112 to thereby retain modified fluid 156 in flow passage 112. At least a portion of carbon dioxide gas 152 is “scrubbed from” flow passage 112, via carbon dioxide scrubber 116, to retain modified fluid 156 in flow passage 112. Modified fluid 156 has a lower carbon dioxide gas 152 concentration and a higher aromatic fluid 154 concentration than headspace fluid mixture 150.

[0068] Carbon dioxide gas 152 concentration is typically expressed in units of volume of carbon dioxide gas to volume of total fluid mixture. The concentration of carbon dioxide gas 152 in headspace fluid mixture 150 may be expressed as volume of carbon dioxide gas 152 per volume of headspace fluid mixture 150 (e.g. parts per million or percentage volume). The concentration of carbon dioxide gas 150 in modified fluid 156 may be expressed as volume of carbon dioxide gas 152 per volume of modified fluid 156 (e.g. parts per million or percentage volume).

[0069] Aromatic fluid 154 may be in liquid, gas or vapour form, or some combination thereof. In some cases, aromatic fluid 154 is in the form of liquid droplets. In some cases, aromatic fluid 154 is in the form of vaporized droplets. If present, these vaporized droplets may be in isolation or entrained in gaseous aromatic fluid. In some cases, aromatic fluid 154 contains the same aromatic compounds. In other cases, aromatic fluid 154 contains more than one type of aromatic compound. The concentration of aromatic fluid 154 in headspace fluid mixture 150 may be expressed as mass of aromatic fluid 154 per unit volume of headspace fluid mixture 150 (e.g. milligrams per litre). The concentration of aromatic fluid 154 in modified fluid 156 may be expressed as mass of aromatic fluid 154 per volume of modified fluid 156 (e.g. milligrams per litre).

[0070] Flow passage 112 directs modified fluid 156 to the at least one port (exemplified as reentry port 110 in FIG. 1 ) of fermentation container 106 to direct modified fluid 156 through the at least one port of fermentation container 106 (reentry port 110, FIG. 1 ). In this manner, modified fluid 156 reenters headspace 114 of the fermentation container 106 to at least partially retain aromatic fluid 154 in fermentable beverage 102.

[0071 ] Typically, when fermentable beverage 102 starts to ferment, the concentration of aromatic fluid 154 in the fermentable beverage 102 (which may be expressed in mass of aromatic fluid 154 per volume of fermentable beverage 102) is relatively low. When fermentable beverage 102 starts to ferment, the concentration of this aromatic fluid 154 in headspace 114 (which may be expressed in mass of aromatic fluid 154 per volume of headspace fluid mixture 150) is also relatively low. As fermentation progresses, aromatic fluid 154 production increases rapidly in fermentable beverage 102 and aromatic fluid 154 migrates into headspace 114 (along with carbon dioxide gas 152). At this stage, the concentration of aromatic fluid 154 in beverage 102 is higher than in headspace 114, so aromatic fluid 154 moves from beverage 102 to headspace 114. When modified fluid 156 reenters headspace 114 (after headspace fluid mixture 150 contacts carbon dioxide scrubber 116), it has a relatively high proportion of aromatic fluid 154 (as compared to headspace fluid mixture 150). Due to the relatively higher concentration of aromatic fluid 154 in the headspace at this point (by way of aromatic fluid 154 reintroduced into the headspace), aromatic fluid 154 in fermentable beverage 102 is more inhibited from leaving the fermentable beverage 102 than if carbon dioxide scrubber 116 were not utilized. By maintaining the relative concentration of aromatic fluid 154 in headspace 114, less aromatic fluid 154 in fermentable beverage 102 is required to leave fermentable beverage 102 as the system strives to equilibrate the concentration of aromatic fluid 154 between beverage 102 and headspace 114. In this manner, the carbon dioxide scrubber 116 at least partially retains aromatic fluid 154 in fermentable beverage 102.

[0072] As shown in FIG. 1 , flow passage 112 may be a conduit, such as plastic piping, for example. In alternative embodiments, the flow passage may be an opening in closure 104. In these alternative embodiments, carbon dioxide scrubber 116 may extend across the flow passage (i.e. extending across an aperture in closure 104), to separate headspace 114 from the external atmosphere when closure 104 is engaged with fermentation container 106.

[0073] Continuing to refer to FIG. 1 , another embodiment of this disclosure provides a method for preserving the aroma of fermentable beverage 102. The method comprises fermenting beverage 102 in fermentation container 106 to produce headspace fluid mixture 150 comprising at least carbon dioxide gas 152 and aromatic fluid 154 in headspace 114 located above fermentable beverage 102 contained in fermentation container 106. The method further comprises permitting the headspace fluid mixture 150 to exit container 106 into flow passage 112. Fleadspace fluid mixture 150 flows through flow passage 112 and into contact with carbon dioxide scrubber 116 to separate carbon dioxide gas 152 in headspace fluid mixture 150 from aromatic fluid 154 in headspace fluid mixture 150 by permitting at least a portion of carbon dioxide gas 152 to exit flow passage 112 and retaining at least a portion of aromatic fluid 154 in flow passage 112 to thereby retain a modified fluid 156 in flow passage 112. Carbon dioxide scrubber 116 retains a modified fluid 156 in the flow passage 112 by“scrubbing” carbon dioxide gas 152 from flow passage 112. Modified fluid 156 has a lower carbon dioxide gas 152 concentration and a higher aromatic fluid 154 concentration than headspace fluid mixture 150. The method comprises permitting modified fluid 156 to remain in flow passage 112 after contacting carbon dioxide scrubber 116 to reenter headspace 114 to at least partially retain aromatic fluid 154 in fermentable beverage 102 in fermentation container 106.

[0074] Returning aromatic fluid 154 to the headspace 114 above the fermentable beverage 102 in the fermentation container 106 helps to maintain the concentration of aromatic fluid 154 in the headspace 114 above the fermenting beverage 102. This helps to decrease the movement of aromatic fluids that enhance the smell and/or taste of fermentable beverage 102 from the beverage 102 into headspace 114 during fermentation.

[0075] Since carbon dioxide gas 152 is generated to some extent throughout the fermentation process, scrubbing the carbon dioxide gas 152 from the headspace fluid mixture 150 via the carbon dioxide scrubber 116 may allow further carbon dioxide gas 152 to be generated by fermenting beverage 102 without excess buildup of carbon dioxide gas 152 in the fermentation apparatus. Excess buildup of carbon dioxide gas 152 in the fermentation apparatus is undesirable as it could increase the pressure in the fermentation apparatus. In extreme cases, this increase in pressure could lead to explosive failure of the fermentation apparatus.

[0076] In some cases, carbon dioxide scrubber 116 removes from the flow passage 112 substantially all of the carbon dioxide gas 152 that contacts the carbon dioxide scrubber 116. In some cases, carbon dioxide scrubber 116 retains in flow passage 112 substantially all of aromatic fluid 154 that contacts carbon dioxide scrubber 116.

[0077] It will be appreciated that although FIG. 1 and other figures that will be discussed below do not explicitly indicate carbon dioxide gas 152 reentering headspace 114, this may occur in some embodiments. For example, in some cases, carbon dioxide scrubber 116 may not remove all of carbon dioxide gas 152 in headspace fluid mixture 150 from flow passage 1 12.

[0078] FIG. 2 exemplifies a second embodiment of a fermentation apparatus 200. It will be appreciated that for simplicity and clarity of illustration, elements of fermentation apparatus 200 corresponding or analogous to elements of fermentation apparatus 100 are labeled with the same reference numerals as for fermentation apparatus 100. For brevity, the description of similar or corresponding elements is not repeated.

[0079] As exemplified in FIG. 2, carbon dioxide scrubber 116 may be, or may include, carbon dioxide absorber 118 containing carbon dioxide absorbing material 120. Carbon dioxide absorbing material 120 absorbs and retains more of carbon dioxide gas 152 than aromatic fluid 154. In this manner, more of carbon dioxide gas 152 than aromatic fluid 154 may be removed (i.e. scrubbed) from headspace fluid mixture 150 in flow passage 112. Carbon dioxide absorbing material 120 may be soda lime, for example. In other embodiments, carbon dioxide absorbing material 120 may be, for example, other agents commonly used for chemical carbon dioxide absorption, such as monoethanolamine (MEA), methyldiethanolamine (MDEA) or other amine solvents known by those skilled in the art. The potential for contamination of the beverage by recirculating the headspace gas through these chemical carbon dioxide absorbers is a consideration. This is also a consideration for physical carbon dioxide absorption using solvents such as Selexol® or Rectisol®, for example. In some embodiments, such chemical and physical carbon dioxide absorbers may be recycled in order to enhance their usefulness as carbon dioxide absorbents. [0080] In some embodiments, carbon dioxide capture molecules may be used as physical adsorbents in conjunction with molecular sieves and metal organic frameworks. Preferably, such carbon dioxide capture molecules are used on relatively larger scale fermentation processes and in relatively low humidity environments.

[0081 ] Continuing to refer to FIG. 2, in some embodiments, when headspace fluid mixture 150 contacts carbon dioxide absorbing material 120, carbon dioxide absorbing material 120 absorbs carbon dioxide gas 152 from headspace fluid mixture 150 and removes carbon dioxide gas 152 from flow passage 112. In some embodiments, the fermentation apparatus further includes fluid pump 122, as exemplified in FIG. 2. Fluid pump 122 may be, for example, an electric gas pump, a gas blower, or a fan. More specifically, fluid pump 122 may be a centrifugal blower, a regenerative blower, a positive displacement blower, a diaphragm pump or an axial fan, for example. In some cases, the fermentation apparatus may include a plurality of fluid pumps. In some cases, fluid pump 122 is in fluid communication with headspace 114 and carbon dioxide scrubber 116. In some embodiments, fluid pump 122 transports headspace fluid mixture 150 and modified fluid 156 through flow passage 112. Fluid pump 122 may increase pressure in flow passage 112. Alternatively, different fluid pumps may be used to transport headspace fluid mixture 150 and modified fluid 156, respectively, through flow passage 112. While the fluid pump 122 is shown upstream of carbon dioxide scrubber 116 in FIG. 2, it will be appreciated that fluid pump 122 may be located in different positions in the fermentation apparatus, such as, for example, downstream of carbon dioxide scrubber 116. In some embodiments, the methods described herein include transporting headspace fluid mixture 150 and modified fluid 156 through flow passage 112 using fluid pump 122. [0082] Continuing to refer to FIG. 2, in some embodiments, flow passage 112 includes a release valve 124 in fluid communication with headspace 114. Release valve 124 may be openable to vent a portion of at least one of headspace fluid mixture 150 and modified fluid 156 to the external atmosphere. Release valve 124 may be located downstream of the carbon dioxide scrubber 116 and upstream of the reentry port 110 of fermentation container 106. Release valve 124 may vent a portion of modified fluid 156 to the external atmosphere when the pressure in the fermentation apparatus exceeds a predetermined value. Release valve 124 may be included in the flow passage 112 when the amount or rate of carbon dioxide gas 152 produced by fermenting beverage 102 exceeds the maximum amount or rate of carbon dioxide gas 152 that can be removed from flow passage 112 by carbon dioxide scrubber 116. This release of carbon dioxide gas 152 (containing some aromatic fluid 154) may be used to reduce the pressure within the fermentation apparatus to acceptable levels. While the release valve 124 is shown in FIG. 2 downstream of carbon dioxide scrubber 116 and upstream of the reentry port 110, it will be appreciated that release valve 124 may be located in other positions within the fermentation apparatus. For example, release valve 124 may be located in fermentation container 106 or closure 104. It will be understood that a portion of aromatic fluid 154 may be lost if the release valve 124 vents the modified fluid 156. Flowever, the quality of the fermentable beverage 102 can still be improved by opening release valve 124 for only a portion of the fermentation of beverage 102 and retaining a portion of aromatic fluid 154 in flow passage 112.

[0083] Continuing to refer to FIG. 2, in some embodiments, a portion of carbon dioxide gas 152 and at least a portion of aromatic fluid 154 in headspace fluid mixture 150 may be retained in flow passage 112 and included in modified fluid 156 after headspace fluid mixture 150 contacts carbon dioxide scrubber 116. In these embodiments, a portion of modified fluid 156 may be allowed to exit flow passage 112 via release valve 124, while the modified fluid 156 remaining in flow passage 112 may be permitted to reenter headspace 114. In this manner, a portion of carbon dioxide gas 152 retained in flow passage 112 (in modified fluid 156) may be allowed to vent to the external atmosphere to avoid a detrimental build-up of carbon dioxide gas pressure in the fermentation apparatus. Another portion of modified fluid 156 remaining in flow passage 112 may be permitted to reenter headspace 114 to increase the concentration of aromatic fluid 154 in headspace 114.

[0084] As exemplified in FIG. 2, in some embodiments, the fermentation apparatus includes an expansion chamber 126 having flexible wall 128 and containing expansion gas 158 therein. Expansion gas 158 may be an inert gas, such as argon, nitrogen or helium, for example. Expansion gas 158 may also be air or a combination of air and inert gas. In some cases, expansion chamber 126 is in fluid communication with flow passage 112. An interior volume defined by flexible wall 128 may be adjustable to accommodate pressure fluctuations in flow passage 112. The methods described herein may include accommodating pressure fluctuations in the fermentation apparatus through the use of expansion chamber 126. It will be understood that while expansion chamber 126 may contain expansion gas 158 therein, expansion gas 158 may be replaced by headspace fluid mixture 150 or modified fluid 156 as the fermentation of beverage 102 progresses. Expansion chamber 126 may be located at a point other than as exemplified in FIG. 2 that is in fluid communication with flow passage 112 or headspace 114. Release valve 124 may serve as a port to add or remove gas from the headspace circuit, including expansion chamber 126.

[0085] FIG. 3 exemplifies a third embodiment of a fermentation apparatus 300. It will be appreciated that for simplicity and clarity of illustration, elements of fermentation apparatus 300 corresponding or analogous to elements of fermentation apparatus 100 and/or 200 are labeled with the same reference numerals as for fermentation apparatuses 100 and/or 200. For brevity, the description of similar or corresponding elements is not repeated.

[0086] As exemplified in FIG. 3, carbon dioxide scrubber 116 may be, or may include, carbon dioxide selective membrane 130. In some embodiments, carbon dioxide selective membrane 130 permits carbon dioxide gas 152 in headspace fluid mixture 150 to pass therethrough and out of flow passage 112. In some cases, carbon dioxide selective membrane 130 retains aromatic fluid 154 in headspace fluid mixture 150 within flow passage 112. In some cases, carbon dioxide selective membrane 130 substantially retains aromatic fluid 154 in headspace fluid mixture 150 within flow passage 112.

[0087] Continuing to refer to FIG.3, in some embodiments, a pressure greater than atmospheric pressure may exist on interior surface 132 of the carbon dioxide selective membrane 130. Increasing the pressure on interior surface 132 encourages carbon dioxide gas 152 to pass through carbon dioxide selective membrane 130 and out of flow passage 112. In some cases, the pressure on interior surface 132 of carbon dioxide selective membrane 130 is generated by fluid pump 122. In some embodiments, carbon dioxide selective membrane 130 may be of sufficient permeability and have a sufficient surface area to accommodate the maximum rate of production of carbon dioxide gas 152 produced by the fermentation of the beverage 102. In some embodiments, carbon dioxide selective membrane 130 is not capable of eliminating the maximum rate of carbon dioxide production, but the partial removal of carbon dioxide gas 152 from flow passage 112 can still enhance the aroma and/or taste of fermentable beverage 102. In these embodiments, carbon dioxide gas 152 produced in excess of the carbon dioxide removal rate of carbon dioxide selective membrane 130 may exit flow passage 112 through relief valve 124 to prevent over pressurization of the system.

[0088] Referring to FIG. 3, the methods described herein may include contacting headspace fluid mixture 150 with carbon dioxide selective membrane 130, carbon dioxide selective membrane 130 permitting carbon dioxide gas 152 in headspace fluid mixture 150 to pass through carbon dioxide selective membrane 130 and out of the flow passage 112, and retaining aromatic fluid 154 in headspace fluid mixture 150 in flow passage 112. In some cases, carbon dioxide selective membrane 130 substantially retains aromatic fluid 154 in flow passage 112.

[0089] In some cases, all of carbon dioxide gas 152 in headspace fluid mixture 150 may not pass through carbon dioxide selective membrane 130, and some of aromatic fluid 154 may pass through carbon dioxide selective membrane 130 and be lost. However, in some cases, by permitting at least a portion of carbon dioxide gas 152 to pass through carbon dioxide selective membrane 130 and retaining at least a portion of aromatic fluid 154 in flow passage 112 and thereby retaining modified fluid 156 in flow passage 112, carbon dioxide selective membrane 130 can improve the quality of the fermented beverage 102 by increasing the amount of aromatic fluid compounds in the final beverage after fermentation.

[0090] Continuing to refer to FIG. 3, in some embodiments, the carbon dioxide selective membrane 130 includes at least one of: a planar membrane, a plurality of stacked planar membranes, a tubular hollow membrane and a bundle of tubular hollow membranes.

[0091 ] Continuing to refer to FIG. 3, in some embodiments, carbon dioxide selective membrane 130 is a diffusive membrane that permits carbon dioxide gas 152 to diffuse therethrough and out of the flow passage 112. In other embodiments, selective membrane 130 is a chemically activated membrane that utilizes chemical reactions between a membrane surface material and a contacting fluid to convey the fluid through the membrane. Non-limiting examples of selective membrane 130 include a fixed-site-carrier membrane, microporous organic polymer membrane, mixed matrix membrane, carbon molecular sieve membrane, inorganic membrane or a polypropylene hollow fiber membrane. Some, like a fixed- site-carrier membrane, for example, are more efficient under the conditions of fermentation which include atmospheric pressure, temperature near room temperature and the presence of water vapour.

[0092] FIG. 4 exemplifies a fourth embodiment of a fermentation apparatus 400. It will be appreciated that for simplicity and clarity of illustration, elements of fermentation apparatus 400 corresponding or analogous to elements of fermentation apparatus 100, 200 and/or 300 are labeled with the same reference numerals as for fermentation apparatus 100, 200 and/or 300. For brevity, the description of similar or corresponding elements is not repeated.

[0093] As exemplified in FIG. 4, in some embodiments, sweep gas 160 is directed across exterior surface 134 of carbon dioxide selective membrane 130. In some cases, sweep gas 160 is a sweep gas that contains little or no carbon dioxide gas. Examples of sweep gas 160 include, but are not limited to, air having a low carbon dioxide content, an inert gas, a plurality of inert gases, or a combination thereof. In some cases, the air having a low carbon dioxide content has a carbon dioxide concentration of equal to or less than about 0.04%. In some cases, the air having a low carbon dioxide content has a carbon dioxide concentration of approximately 0.04%. More specifically, sweep gas 160 may be nitrogen, helium, hydrogen or argon. In some cases, exterior surface 134 of carbon dioxide selective membrane 130 is located external to flow passage 112. In some cases, sweep gas 160 lowers the carbon dioxide partial pressure on exterior side 162 of carbon dioxide selective membrane 130 relative to flow passage side 164 of the carbon dioxide selective membrane 130. In these cases, this facilitates the passage of carbon dioxide gas 152 through carbon dioxide selective membrane 130 and out of flow passage 112.

[0094] The methods described herein may include directing sweep gas 160 across exterior surface 134 of carbon dioxide selective membrane 130 to facilitate the passage of carbon dioxide gas 152 through carbon dioxide selective membrane 130 and out of flow passage 112.

[0095] Continuing to refer to FIG.4, in some embodiments, negative pressure generator 136 is located external to flow passage 112. A suitable negative pressure generator may include, but is not necessarily limited to, a centrifugal blower, regenerative blower, positive displacement blower or an axial fan. In some embodiments, a plurality of negative pressure generators is used. In some cases, negative pressure generator 136 directs sweep gas 160 across exterior surface 134 of carbon dioxide selective membrane 130. In some embodiments, a lower pressure exists on exterior surface 134 of carbon dioxide selective membrane 130 than on interior surface 132. This lower pressure increases the carbon dioxide partial pressure difference across the membrane and may be generated by negative pressure generator 136. In some embodiments, positive pressure on the flow passage side 164 from fluid pump 122, or a combination of both negative pressure generator 136 and positive pressure from fluid pump 122 increase the carbon dioxide partial pressure difference across the membrane. This pressure differential across carbon dioxide selective membrane 130 encourages carbon dioxide gas 152 in headspace fluid mixture 150 to pass through carbon dioxide selective membrane 130 and out of flow passage 112. Furthermore, in some cases, negative pressure generator 136 permits decreased total pressure on exterior side 162 of selective membrane 130 to thereby facilitate the passage of carbon dioxide gas 152 in headspace fluid mixture 150 through carbon dioxide selective membrane 130 and out of flow passage 112. In some cases, fluid pump 122 permits increased total pressure on flow passage side 164 of selective membrane 130 to thereby facilitate the passage of carbon dioxide gas 152 in headspace fluid mixture through carbon dioxide selective membrane 130 and out of flow passage 1 12.

[0096] Some embodiments of the methods described herein include releasing carbon dioxide gas 152 that exits flow passage 112 via carbon dioxide scrubber 116 to the external atmosphere. Referring to FIG. 4, some embodiments of the methods described herein include transferring all (or part) of carbon dioxide gas 152 that exits flow passage 112 via carbon dioxide scrubber 116 to a carbon dioxide storage vessel 166 that is fluidly connected to carbon dioxide scrubber 116. Carbon dioxide storage vessel 166 may be connected to compressor unit 167 which condenses the carbon dioxide that travels into carbon dioxide storage vessel 166, and thereby allows more carbon dioxide gas to be stored in carbon dioxide storage vessel 166. It may be desirable to store carbon dioxide gas 152 in a storage vessel 166 where environmental regulations restrict the release of carbon dioxide gas 152 to the environment, or where carbon dioxide gas 152 can be utilized in other processes. These reasons for storing the carbon dioxide gas 152 are merely illustrative and are not meant to limit the methods described herein.

[0097] FIG. 5 exemplifies a fifth embodiment of a fermentation apparatus 500. It will be appreciated that for simplicity and clarity of illustration, elements of fermentation apparatus 500 corresponding or analogous to elements of fermentation apparatus 100, 200, 300 and/or 400 are labeled with the same reference numerals as for fermentation apparatus 100, 200, 300 and/or 400. For brevity, the description of similar or corresponding elements is not repeated. [0098] Continuing to refer to FIG. 5, in some embodiments, positive pressure generator 138 may be located external to flow passage 112. Positive pressure generator 138 may direct sweep gas 160 across exterior surface 134 of carbon dioxide selective membrane 130. Although the use of positive pressure, in itself, may slightly elevate the total pressure on exterior side 162 of selective membrane 130, in some cases, the flow of sweep gas 160 generated by positive pressure generator 138 may facilitate a large carbon dioxide partial pressure difference between the exterior side 162 and the flow passage side 164, to facilitate carbon dioxide gas passage through selective membrane 130.

[0099] FIGS. 4 and 5 exemplify sweep gas 160 moving across exterior surface 134 of selective membrane 130 and headspace fluid mixture 150 moving across interior surface 132 of selective membrane 130 and travelling in substantially the same flow direction to one another. In some alternative embodiments, sweep gas 160 and headspace fluid mixture 150 flow in substantially opposite directions to one another across opposing surfaces of selective membrane 130.

[00100] FIG. 6 exemplifies a sixth embodiment of a fermentation apparatus 600. It will be appreciated that for simplicity and clarity of illustration, elements of fermentation apparatus 500 corresponding or analogous to elements of fermentation apparatus 100, 200, 300, 400 and/or 500 are labeled with the same reference numerals as for fermentation apparatus 100, 200 300, 400 and/or 500. For brevity, the description of similar or corresponding elements is not repeated.

[00101 ] Fermentation apparatus 600 shown in FIG. 6 is similar in some respects to fermentation apparatus 100 shown in FIG. 1. Fermentation container 106 of fermentation apparatus 100 has exit port 108 and reentry port 110, which are separate ports from one another. Unlike fermentation apparatus 100, fermentation apparatus 600 has a combination port 169 (FIG. 6), which functions as both exit port 108 and reentry port 110 (FIG. 1 ). In some cases, as shown in apparatus 600, fluid pump 122 facilitates flow through flow passage 112 and the dual operation of combination port 169. As exemplified in FIG. 6, fermentation apparatus 600 includes an expansion chamber 126 having flexible wall 128 and containing expansion gas 158 therein.

[00102] It will be appreciated that each aromatic fluid has different production, loss and retention curves depending on such factors as its gas/liquid partition coefficient, the temperature, and the level of carbon dioxide production. Preservation of aromatic fluid may be enhanced in the final beverage by reducing carbon dioxide stripping with carbon dioxide filtration or carbon dioxide absorption and headspace recirculation, as described above in accordance with embodiments of this disclosure.

[00103] In some wine fermentation processes, most of the yeast- derived aromatic fluid is produced and lost early in fermentation. Consider the example of fermenting a 1000L of grape juice with 24°Brix (24 g of sucrose per 100g of solution), specific gravity of 1.100 (using a hydrometer corrected to15.6° Celsius, with water as the reference solution), and an initial sugar content of 264g/L. When fermenting 1000L of this juice the carbon dioxide gas flow from the start of initial yeast growth would be expected to gradually increase up to a peak carbon dioxide gas flow of about 750 L/hour at about 1 ½ days after the start of the yeast growth. After peaking, the carbon dioxide gas flow rate would be expected to decrease, and about four days from the start of yeast growth, there may be a carbon dioxide flow rate of about 225 L/Hr. Production of yeast-derived aroma compounds begins at about the time of peak carbon dioxide gas flow rate. In this example, the production of yeast-derived aroma compounds would be expected to reach a peak two to four days after peak carbon dioxide production. In this example, total fermentation time would be expected to be more than two weeks. The final concentration of individual aroma compounds contained in a wine at the end of fermentation varies widely and depends on a wine’s individual production rates and an aroma compound’s susceptibility to being stripped by the continuing carbon dioxide flow rate. Since a higher carbon dioxide gas flow rate represents a higher alcohol production rate (i.e. fermentation rate), it is evident that, in at least some cases, a large proportion of the alcohol, carbon dioxide gas and aromatic fluid production occurs relatively early in the overall fermentation process. In this example, it is expected that approximately 40% of aroma compound loss would occur between days one and four, and an additional approximately 20% of aroma compound loss would occur between day four and the end of fermentation (for a total aroma compound loss of 60% over the entire duration of fermentation).

[00104] Consequently, in some embodiments disclosed herein the methods for preserving aroma may be conducted for only a portion of the fermentation process, such as the first portion of fermentation only. In some cases, the first portion is less than one week. In some cases, the first portion is approximately three days in duration. Thereafter, carbon dioxide gas may be allowed to escape from flow passage 112 without the use of a carbon dioxide scrubber 116 (FIGS. 1 to 6 exemplify carbon dioxide scrubber 116). The decrease in sugar concentration and the increase in ethanol typically both decrease partition coefficients. This typically decreases the loss of aroma in the latter part of fermentation.

[00105] In some cases of wine fermentation, reducing carbon dioxide stripping for the first week of fermentation, through the use of a carbon dioxide scrubber 116 as described above and illustrated in FIGS. 1 to 6, may retain approximately 80% of the maximum desirable aroma and have a positive impact on the quality of the final fermented wine. It will be appreciated that in other cases of wine fermentation, employing a carbon dioxide scrubber 116 as described above and illustrated in FIGS. 1 to 6 for the first week of fermentation, may reduce the carbon dioxide stripping to that point so that less than 80% of the maximum desirable aroma is retained, yet still have a positive impact on the quality of the final fermented wine.

[00106] In addition to aroma preservation, there may be other advantages to carbon dioxide removal during fermentation, such as allowing higher temperatures for white wine fermentation, which may shorten the time required to make white wine. In addition, in some cases, higher temperatures make the yeast fermentation more reliable. Additional advantages of the described methods and apparatus may become apparent upon reading this description.

[00107] Referring back now to FIG. 1 , the methods described herein may include carbon dioxide scrubber 116 being capable of removing carbon dioxide gas 152 from flow passage 112 at a removal rate at least as high as the maximum production rate at which carbon dioxide gas 152 is produced in the headspace 114 by the fermentation of beverage 102.

[00108] Continuing to refer to FIG. 1 , some embodiments of the methods described herein include allowing headspace fluid mixture 150 to flow through flow passage 112 into contact with carbon dioxide scrubber 116 during the entire duration of the fermentation. Some alternative embodiments of the methods described herein include allowing headspace fluid mixture 150 to flow through flow passage 112 into contact with carbon dioxide scrubber 116 during less than the entire duration of the fermentation. Carbon dioxide scrubber 116 may be used for any part of, or for the entire duration of, fermentation. Carbon dioxide scrubbers incapable of eliminating the maximum rate of carbon dioxide production may still be used with the beneficial effect of preserving some of the aroma compounds in a fermenting beverage. [00109] FIG. 7 exemplifies fermentation apparatus 300 (the apparatus shown in FIG. 3) operated in the presence of waste by-products 170. Referring to FIG. 7, in some of the methods described herein, fermentation of beverage 102 produces undesirable waste by-products 170 in headspace 114, in addition to aromatic fluid 154. Exemplary aromatic fluids 154 (which enhance taste and/or smell) include, but are not limited to, ethyl hexanoate, ethyl buryrate, ethyl octanoate, isoamylacetate and hexylacetate. By contrast, exemplary undesirable waste by-products 170 (which produce an unpleasant taste and/or smell) include, but are not limited to: hydrogen sulfide, acetic acid, hexanoic acid and methyl mercaptan. In these cases, headspace fluid mixture 150 contains both aromatic fluid 154 and undesirable waste by-products 170. In some cases, depending on the type of carbon dioxide scrubber 116 employed and the type of waste by-products 170 present, waste by-products 170 may be removed from flow passage along with carbon dioxide gas 152, as shown in FIG. 7. In some cases, substantially all of waste by-products 170 flow through selective membrane 130 and are substantially removed from flow passage 112, as exemplified in FIG. 7. In alternative embodiments, waste by-products 170 may be partially retained in flow passage 112 along with aromatic fluid 154, with only a portion of waste by-products exiting flow passage 112 via selective membrane 130.

[00110] FIG. 8 exemplifies modified fermentation container 172. Modified fermentation container 172 permits headspace fluid mixture 150 to escape from flow passage 112 via airlock 168. As exemplified in FIG. 8, airlock 168 covers exit port 108 and stopper 174 fluidly seals reentry port 110. In cases wherein the beverage container has only one port (see combination port 169 in FIG. 6, for example), airlock 168 covers the one port and stopper 174 is not necessarily required. In other embodiments, each port may be covered by a corresponding airlock 168. Airlock 168 is a conventional airlock that allows internal gases to escape a container while preventing the inflow of external gases to a container. When airlock 168 covers and partially seals a port, it prevents external oxygen from entering modified fermentation container 172, but permits aromatic fluid 154, carbon dioxide gas 152 and waste by-products 170 in headspace fluid mixture 150 to escape from flow passage 112 via airlock 168.

[00111 ] FIG. 7 exemplifies substantially all waste by-products 170 passing through selective membrane 130 and out of flow passage 112. However, in some cases, some waste by-products 170 may be retained in flow passage 112. Unlike fermentation container 106 shown in FIG. 7, modified fermentation container 172 (exemplified in FIG. 8) replaces flow passage 112 with airlock 168 and stopper 174. Airlock 168 of modified fermentation container 172 (FIG. 8) allows headspace fluid mixture 150 to escape from the flow passage 112 to the external atmosphere (or into a separate vessel) and permits desirable aromatic fluid 154, undesirable waste by-products 170 and carbon dioxide gas 152 produced during fermentation, to escape. Advantageously, the undesirable waste by-products 170 are allowed to escape from flow passage 112 via airlock 168 and are not reintroduced into fermentable beverage 102.

[00112] In some embodiments, a fermentation apparatus (ex. fermentation apparatus 300, FIG. 7) and a modified beverage container (e.g. modified fermentation container 172) may be used in combination to allow a sufficient amount of desirable aromatic fluids to be retained in the flow passage during a first portion of the fermentation process (as shown in FIG. 7), and a sufficient amount of undesirable waste by-products to be permitted to escape from the flow passage during a second (i.e. later) portion of the fermentation process (as shown in FIG. 8). In this manner, a fermented beverage may be obtained having a sufficiently high concentration of desirable aromatic compounds and a sufficiently low concentration of undesirable waste by-product compounds. In some of the methods described herein, the first portion of fermentation may, for example, be the first week of fermentation (in which a significant portion of desirable aromatic compounds may be preserved, in some cases). In some of the methods described herein, the first portion of fermentation may be more or less than the first week of the fermentation process.

[00113] In some embodiments, the fermentation apparatus has a carbon dioxide absorber (see carbon dioxide absorber 118 of fermentation apparatus 200 in FIG. 2, for example). In some cases, substantially all waste by-products may be removed from flow passage 112 by carbon dioxide absorber 116. In other cases, some waste by-products may not be fully absorbed by carbon dioxide absorbing material 120, and may be retained in flow passage 112. In the case of the latter, a fermentation apparatus (e.g. 200, FIG. 2) may be used for a first portion of the fermentation process, and a modified beverage container (e.g. 172, FIG. 7) may be used for a second portion of the fermentation process. It will be appreciated that the second portion follows the first portion of the fermentation process.

EXAMPLARY EXPERIMENTAL IMPLEMENTATIONS

Experiment #1

[00114] A first experimental embodiment is discussed below with reference to FIG. 9 and exemplary fermentation apparatus 700.

[00115] After crushing and pressing the juice from Pinot Gris grapes, fermentation was started with ICV-D47™ yeast (sold by Lallemand™ Inc.,

Montreal, Quebec, Canada), which is a yeast known for producing wines with typically well-regarded aroma. A portion of the juice, identified as the control juice, was placed in a fermentation container. Fermentation of the control wine was conducted in a sealed container with an airlock to allow carbon dioxide gas to freely escape while limiting oxygen exposure. It was noticed that wine juice aromas were coming out of the airlock and the carbon dioxide gas was carrying away the very aromas desired in the finished control wine.

[00116] Referring to FIG. 9, approximately 48 hours after fermenting the juice in the control container (as outlined in the paragraph above), a portion of the control juice (one litre) was drained into a two liter glass jug 704 (see experimental juice 702). This one litre of juice was the experimental juice to be fermented into an experimental wine. Approximately 0.8 litres of soda lime carbon dioxide absorber 706 was poured into a one liter plastic container 708 fitted with a rubber stopper 710. A tube 712 carried carbon dioxide gas and aromatic fluid via a sealed aquarium pump 714 from headspace 716 above juice 702 to the bottom of the soda lime carbon dioxide absorber 706 where it then traveled upwards through carbon dioxide absorber 706 and back into headspace 716 of two liter glass jug 704 containing fermenting juice 702. The absorption of carbon dioxide gas via carbon dioxide absorber 706 was started just after the peak carbon dioxide production of fermenting juice 702 was achieved, and continued until the end of the its fermentation. The completeness of carbon dioxide absorption by the soda lime in carbon dioxide absorber 706 was evidenced by the lack of pressure increase inside glass jug 704 and in the plastic container 708 during fermentation. After the fermentation was complete, the control wine and the one liter experimental wine were tasted and the aroma was judged to be more intense and have a more fruity smell and taste in the one liter experimental sample that had its carbon dioxide gas removed and some of its aroma preserved, using the above described fermentation apparatus 700. This taste difference was noted despite the fact that the juice treated with the above described fermentation apparatus 700 had already undergone some fermentation with carbon dioxide aroma stripping before the experimental juice 802 was set up as illustrated FIG. 9.

Experiment #2 [00117] A second experimental embodiment is discussed below with reference to FIG. 10 and exemplary fermentation apparatus 800.

[00118] Pinot Gris grapes were picked and pressed to yield 52 litres of juice. This pressed juice had the following initial properties: Brix of 22.2° (22.2 g of sucrose per 100g of solution), specific gravity of 1.100 using a hydrometer corrected to 15.6°Celsius with water as the reference solution, titratable acidity of 6.5 grams/litre, and pH of 3.30. Potassium metabisulfite, in a quantity of 2.2 grams, was added to the juice to obtain a molecular sulfur dioxide content of 0.80 parts per million.

[00119] The pressed juice was allowed to settle for 30 hours under a controlled, cool climate in which the surrounding temperature was held at approximately 10° Celsius.

[00120] The settled juice was then racked into a clean container into which the following ingredients were added to treat the juice: 14.7g of OptiWHITE™ (sold by Lallemand™ Inc., Montreal, Quebec, Canada), 1.5ml of Antifoam FG9030™ (sold by Vinquiry™, Windsor, California, U.S.A.), 12.3g of Fermaid-K™ (sold by Lallemand™ Inc., Montreal, Quebec, Canada), and 1.2g of hydrated BA11™ yeast (sold by Lallemand™ Inc., Montreal, Quebec, Canada). The juice was inoculated with the BA11™ yeast. The treated juice was then thoroughly stirred and transferred into two clean fermentation containers, with 24 litres of juice and approximately 6 litres of headspace in each container. One container was filed with the 24 litres of juice 802 and labelled“experimental” (see experimental container 804 of fermentation apparatus 800 in FIG. 10). The other fermentation container was labelled“control” and was fitted with a common airlock to allow carbon dioxide gas to freely escape to the atmosphere from the control container. The control container and experimental container were both flushed with argon gas prior to fermentation, to remove oxygen from the system. [00121 ] FIG.10 illustrates headspace 806 above juice 802. Headspace 806 contained aromatic fluid 812 and carbon dioxide gas 814 during fermentation. An airtight pump 808 circulated headspace fluid mixture 810 (including aromatic fluid 812 and carbon dioxide gas 814) from headspace 806 through a sealed carbon dioxide absorber 816. Carbon dioxide absorber 816 contained four litres of carbon dioxide absorbing material 818 in the form of Spherasorb™ soda lime (sold by Trudell Medical Marketing™ Ltd., Surrey, BC, Canada). Pump 808 also circulated a modified fluid 820 from carbon dioxide absorber 816 through flow passage 822 and back into headspace 806.

[00122] Prior to fermentation of juice 802, headspace 806, flow passage 822, carbon dioxide absorber 816 and expansion chamber 824 were flushed with argon to remove oxygen from the system. Fermentation apparatus 800 included expansion chamber 824 containing argon gas to provide the argon gas to the system and accommodate pressure fluctuations in flow passage 822. Fermentation apparatus 800 also included flow meter 826 to measure the flow of headspace fluid mixture 810 from headspace 806 to carbon dioxide absorber 816. The experimental apparatus also contained water trap 828 for collecting excess moisture created during the soda lime absorption of carbon dioxide gas 814.

[00123] The control and experimental containers sat side by side on a flat surface and were exposed to the same temperature range (approximately 15-17° Celsius) during fermentation.

[00124] FIG. 11 shows the fermentation dynamics for experimental juice 802 during fermentation. The specific gravity of experimental juice 802 was monitored (using a hydrometer corrected to 15.6°Celsius with water as the reference solution) during the fermentation process. During fermentation, sugar was consumed and alcohol was produced. Since sugar is denser than alcohol, the specific gravity of juice 802 decreased as fermentation progressed, as shown by the specific gravity curve in FIG. 11. The specific gravity did not fluctuate until approximately one day after yeast inoculation, thereby suggesting that alcohol production began about 1 day after yeast inoculation. After about 1 day, carbon dioxide gas production also began, thereby further evidencing the start of alcohol production.

[00125] Referring to FIGS. 10 and 11 , the rate of removal of carbon dioxide gas 814 from headspace fluid mixture 810 by carbon dioxide absorber 816 was then calculated. Since the carbon dioxide absorber 816 was found to remove all carbon dioxide gas 814 produced during the fermentation experiment, the rate of removal of carbon dioxide was equal to the rate of production of carbon dioxide gas 814 by fermenting experimental juice 802.

[00126] Therefore, the rate of removal and production of carbon dioxide (CO2) gas was calculated using the following formula:

L

C0₂ Removal Rate and Production Rate hour

L of juice

Headspace gas flow (Entry C0₂ concentration - Exit C0₂ concentration) x 60

Juice Volume

wherein

Headspace gas flow = The rate of flow (L/minute) of headspace fluid mixture

810, measured at flow meter 826;

Entry CO2 concentration = Fractional concentration of carbon dioxide gas 814 in the headspace fluid mixture 810, measured at three- way stopcock 832 using a Viasensor™ carbon dioxide analyzer (model G110, AFC International™ Inc., DeMotte, Indiana, USA);

Exit CO2 concentration = Fractional concentration of carbon dioxide gas 814 carbon dioxide gas in modified fluid 820, measured at three-way stopcock 834, which at all times of full operability of carbon dioxide absorber 816 was 0.00; and

Juice Volume = The amount of juice 802 in experimental container

804, which remained constant at 24 litres during fermentation.

[00127] The above formula provided the amount of carbon dioxide gas 814 removed per hour during the fermentation process (per 1 L of experimental juice 802), as shown graphically vs. fermentation time in FIG. 11. Since carbon dioxide absorber 816 removed all carbon dioxide gas 814 from headspace fluid mixture 810, this formula also represented the amount of carbon dioxide gas 814 produced during fermentation, and thereby, indicated the periods of more active fermentation (and higher aromatic fluid 812 and carbon dioxide gas 814 production).

[00128] Referring now to FIG.11 , since fermentation rate can be gauged by the rate of production of carbon dioxide gas 814, the maximum fermentation rate occurred about 50 hours after yeast inoculation. The measured peak rate of carbon dioxide gas production was 0.93 (L/hour)/L of juice, measured at a temperature of 17° Celsius and 711 mmFIg of atmospheric pressure. By applying the standard correction for these ambient conditions to standard conditions, this translated to 0.81 (L/hour)/L of juice under conditions of standard temperature and pressure (i.e. 760 mmFIg and 0°C). Following the peak, the rate of production of carbon dioxide gas 814 decreased rapidly and then more slowly until day seven after yeast inoculation. At about day 7, the capacity of the soda lime to absorb carbon dioxide gas 814 was exhausted. At that time, pump 808 was deactivated, and the circulation of fluid out of and back into headspace 806 (through flow passage 822 and carbon dioxide absorber 816) was terminated. [00129] After fluid circulation was terminated, experimental container 804 was fitted with a conventional airlock that allowed the remaining carbon dioxide gas 814 to freely escape to the external atmosphere. The airlock was attached to exit port 836 and reentry port 838 was fluidly sealed.

[00130] On day 12, the airlock and seal were removed and circulation of the headspace fluid mixture 810 through carbon dioxide absorber 816 was reestablished for two hours to obtain carbon dioxide gas 814 production readings as described above using flow meter 826 and gas samples from three-way stopcock 832 and three-way stopcock 834 to calculate a carbon dioxide production rate (per litre of juice) at that time.

[00131 ] The fermented experimental and control juices (in the form of wine at that this point) were racked off the lees on day 35 when the wines’ specific gravity were 0.996 and the wines were relatively clear. The wines were cold stabilized and then fined using bentonite clay, specifically, Bentonite Vitiben™ (sold by Bosagrape Winery Supplies™, Burnaby, BC, Canada). After 84 days from yeast inoculation, the wines were finely filtered using a Buon Vino Super Jet™ pump with #2 filter pads (sold by Buon Manufacturing™ Inc., Cambridge, ON, Canada) and bottled on day 85.

[00132] During the 6 days of headspace recirculation and carbon dioxide gas absorption for experimental wine 802, approximately 70% of carbon dioxide gas 814 produced during the entire fermentation was removed by the soda lime.

[00133] To assess whether any organoleptic differences existed between the control and experimental wines, a taste test was conducted by an expert. The expert was given one 750 millilitre bottle of each of the control and experimental wine to taste periodically over a two day period. The two bottles were labeled A and B and the expert was blinded as to which was control and which was experimental. The expert was not aware of the nature of the experiment. The assessment was documented using a numerical scoring system of sixteen attributes, each having a range of values between 1 and 10 (based on Winespider.com™ rankings). Higher numbers for each attribute represented more desirable properties. The experts’ taste test ratings are expressed as numerical data and qualitative descriptors in Table 1 below.

Table 1 : Attribute Ratings for Control and Experimental Wines

[00134] The respective scores of the sixteen attributes are expressed on a radar, or spider web, chart in FIG. 12.

[00135] The above-described fermentation process was designed to produce a light-bodied white wine and many of the scores provided in Table 1 and FIG. 12 reflect that style. Table 1 and FIG. 12 show differences between the control and experimental wines, even though the control wine’s complexity and aftertaste were rated as commercial quality. Most notably, the experimental wine had more desirable aromatic scent, a more intense flavor, more varietal character (aroma originating from the grape variety, Pinot Gris, in this case), its palate was more intense and complex, and it had longer residual taste. The experimental wine was deemed generally superior to the control wine.

Experiment #3

[00136] A fixed-site carrier membrane that functions at a relatively low pressure differential was used for Experiment #3. Fixed-site carrier membranes are described, for example, in United States Patent Publication No. 20080078290, which is hereby incorporated by reference in its entirety. This membrane is highly selective to carbon dioxide gas relative to larger desirable aromatic fluid compounds. This membrane is able to function at room temperature and in the presence of water vapour. The membrane transports carbon dioxide gas by simple diffusion from the feed side to the permeate side of the membrane. In addition, the membrane surface on the feed side of the membrane is coated with polyvinylamine. When the carbon dioxide gas contacts the feed side of the membrane, the polyvinylamine reacts with the carbon dioxide gas to produce bicarbonate on the feed side of the membrane. The bicarbonate then migrates to the permeate side of the membrane. On the permeate side, the bicarbonate then dissociates into carbon dioxide gas, thereby efficiently transferring the carbon dioxide gas from the feed side to the permeate side of the membrane.

[00137] The above-mentioned fixed-site carrier membrane was used in the fermentation apparatus of FIG. 10, to replace carbon dioxide absorber 816 and soda lime absorbing material 818 illustrated in FIG. 10. In this configuration, 27 litres of Gamay grape juice was placed in container 804. This juice was fermented and the fixed-site carrier membrane was used for the first 10 hours of active fermentation, until the carbon dioxide production rate exceeded the scrubbing rate of the fixed-site carrier membrane under accelerated carbon dioxide production rates. The membrane successfully removed 716 millilitres/hour of carbon dioxide gas from headspace fluid mixture 810. Although this was less than 1 % of the peak carbon dioxide gas production rate, it was notable that the membrane successfully removed carbon dioxide gas from the flow passage. The surface area of the feed side of the flat, fixed-site carrier membrane was only 18cm x 18cm. It is generally understood that the degree to which a membrane is able to remove carbon dioxide depends on its surface area at the interface of its feed side with feed gas (i.e. headspace fluid mixture 810, FIG. 10). It is generally understood that a larger surface area will increase the extent of material separation. It was expected that with a greater surface area, the fixed-site carrier membrane would have removed a much larger amount of carbon dioxide gas from the system while retaining aromatic fluid in the flow passage, thereby having a significant effect on the aromatic properties of the fermented juice.

Experiment #4

[00138] A fourth experimental embodiment is discussed below with reference to FIG. 13 and exemplary fermentation apparatus 900.

[00139] Pinot Gris and Gamay grapes were picked, crushed and pressed to yield 49 litres and 57 litres of unsettled juice, respectively. Both pressed juices were allowed to settle for 38 hours under a controlled, cool climate in which the surrounding temperature was held at approximately 10° Celsius. The settled juices were then racked to clean containers.

[00140] The 44 litres of settled Pinot Gris juice had the following initial properties: Brix of 21.8°, specific gravity of 1.100, titratable acidity of 6.5 grams/litre and pH of 3.30. Potassium metabisulfite, in a quantity of 1.94 grams, was added to the Pinot Gris juice to obtain a molecular sulfur dioxide content of 0.80 parts per million.

[00141 ] The 44 litres of settled Pinot Gris juice also had the following ingredients added: 13.2g of OptiWFIITE™ (sold by Lallemand™ Inc.,

Montreal, Quebec, Canada), 1.3ml of Antifoam FG9030™ (sold by Vinquiry™, Windsor, California, U.S.A.), 5.5g of Fermaid-K™ (sold by Lallemand™ Inc., Montreal, Quebec, Canada), and 11. Og BA11™ yeast previously hydrated in 260ml water containing 13.2g Go-Ferm (sold by Lallemand™ Inc., Montreal, Quebec, Canada). The treated Pinot Gris juice was then thoroughly stirred and transferred into two clean fermentation containers, with 21 litres of juice and approximately 6 litres of headspace in each container. One container was filed with the 21 litres of juice 902 (Figure 13) and labelled“experimental”. The other fermentation container was also filled with 21 litres of juice with approximately 6 litres of headspace and labelled“control”. The fermentation containers for both the control and experimental juice were initially fitted with a common airlock to allow carbon dioxide gas to freely escape to the atmosphere from the container’s headspace. The headspace of both the control and experimental containers was flushed with carbon dioxide gas prior to fermentation, to remove oxygen.

[00142] The skins and juice of the crushed Gamay grapes were allowed to remain in contact for 18 hours to impart a light red color (rose) to the Gamay juice prior to pressing. The 53 litres of pressed and settled Gamay Rose juice had the following initial properties: Brix of 21.1°, specific gravity of 1.094, titratable acidity of 7.8 grams per litre and a pH of 3.30. Potassium metabisulfite, in the quantity of 2.33 grams, was added to the Gamay juice to obtain a molecular sulfur dioxide content of 0.80 parts per million.

[00143] The 53 litres of settled Gamay Rose juice also had the following ingredients added: 15.9g of OptiWHITE™ (sold by Lallemand™ Inc., Montreal, Quebec, Canada), 1.6ml of Antifoam FG9030™ (sold by Vinquiry™, Windsor, California, U.S.A.), 6.7g of Fermaid-K™ (sold by Lallemand™ Inc., Montreal, Quebec, Canada), and 13.3g BA11™ yeast previously hydrated in 320ml water containing 15.9g Go-Ferm (sold by Lallemand™ Inc., Montreal, Quebec, Canada). The treated Gamay Rose juice was then thoroughly stirred and transferred into two clean fermentation containers, with 23 litres of juice and approximately 5 litres of headspace in each container. One container was labelled“experimental”, and the other fermentation container was labelled“control”. Both fermentation containers were initially fitted with a common airlock to allow carbon dioxide gas to freely escape to the atmosphere. The headspace of both the control and experimental containers was flushed with carbon dioxide gas prior to fermentation, to remove oxygen.

[00144] FIG.13 illustrates the experimental fermentation apparatus 900 used to ferment Pinot Gris juice and Gamay Rose juice (juice is indicated at 902) transferred to the fermentation containers 904 labelled experimental. Headspace 906 contained aromatic fluid 912 and carbon dioxide gas 914 during fermentation.

[00145] At the first sign of active fermentation of juice 902, the air lock initially fitted to the experimental container was removed and replaced with the carbon dioxide separation membrane system 915 and flow passage 922, which were connected to ports 936 and 938 respectively. Headspace 906, flow passage 922, carbon dioxide separation membrane system 915 and expansion chamber 924 were flushed with carbon dioxide gas to remove oxygen.

[00146] Airtight pump 908 circulated headspace fluid mixture 910 (including aromatic fluid 912 and carbon dioxide 914) from headspace 906 through outlet 936 to the sealed carbon dioxide separation membrane system 915. Carbon dioxide separation membrane system 915 consisted of membrane 916 with feed side 918 and permeate side 917. Membrane 916 was a fixed site carrier membrane supplied by the Department of Chemical Engineering, Norwegian University of Science and Technology, Trondheim, Norway. Membrane holder 915 consisted of two units connected in series, each containing two flat sheet membranes with a total of 0.27 m² membrane area. Pump 908 also circulated a modified fluid in retentate stream 920 of carbon dioxide separation membrane system 915 through flow passage 922 and back into headspace 906 via inlet port 938.

[00147] Fermentation apparatus 900 included expansion chamber 924 initially containing carbon dioxide gas to accommodate pressure fluctuations in flow passage 922. Fermentation apparatus 900 also included flow meter 926 to monitor the flow of headspace fluid mixture 910 from headspace 906 to feed side 918 of carbon dioxide separation membrane 916. Pump 909 sucked room air sweep gas 911 through humidifier 913, then through permeate side 917 of carbon dioxide separation membrane system 915 and then through flow meter 927. Room air 911 carried away carbon dioxide that diffused through membrane 916 and vented it to atmosphere 921.

[00148] Knowing total permeate flow from flowmeter 927 and carbon dioxide concentration at permeate sample port 933, the carbon dioxide flow across membrane 916 could be calculated according to equation:

Carbon dioxide flow across membrane 916 = (Total permeate flow 927)*(fractional concentration of carbon dioxide from sample port 933)

[00149] The carbon dioxide measured at sample port 933, measured as the fraction of carbon dioxide in the permeate stream (FpC02), causes the flowing gas to be more dense which affects the accuracy of flow meters calibrated for room air. The indicated flow in flow meter 927 was corrected for the effects of carbon dioxide using the following equation:

Corrected flow at flowmeter 927 = (SQRT(1/(((1 -Fp CO2)*0.766) + ((1 - FpCO2)*0.233 + (FpC02*1 53))))*(lndicated flow at flowmeter 927).

[00150] Carbon dioxide was measured at port 933 using a Viasensor™ carbon dioxide analyzer (model G110, AFC International™ Inc., DeMotte, Indiana, USA) and calculated carbon dioxide flow was used to follow the kinetics of fermentation.

[00151 ] Membrane scrubbing of carbon dioxide from flow passage 922 was conducted in the same way for both Pinot Gris and Gamay Rose. Membrane carbon dioxide scrubbing commenced when carbon dioxide first appeared in the headspace 906 (detected by bubbling in the airlock). For Pinot Gris, the membrane 916 was used for only the first 22 hours of active fermentation until just after the peak carbon dioxide production occurred. For Gamay Rose membrane carbon dioxide scrubbing took place from the first indications of carbon dioxide production until at least 90% of the fermentation had taken place. At the termination of membrane carbon dioxide scrubbing, a simple airlock was installed and any carbon dioxide and aroma in headspace 906 was allowed to escape to atmosphere.

Pinot Gris Fermentation:

[00152] Carbon dioxide was first detected by bubbling in the airlock 32 hours after yeast inoculation. Fleadspace recirculation past membrane 916 was initiated at that time. Membrane carbon dioxide separation and headspace recirculation continued until the peak carbon dioxide production was detected using the equation above for carbon dioxide flow across membrane 916. This occurred at 54 hours post yeast inoculation (22 hours of headspace recirculation). At this point an airlock was installed on the experimental container which allowed headspace 906 to escape to atmosphere for the remainder of fermentation. The entire fermentation required 32 days to complete.

Gamay Rose Fermentation:

[00153] Active fermentation, indicated by carbon dioxide bubbling in the airlock, was first observed 25 hours after yeast inoculation. Headspace recirculation and membrane carbon dioxide separation was started at that time and continued until 204 hours (8.5 days) post-yeast inoculation. At this time the bulk of fermentation was complete and the airlock was put back on the container to allow any remaining carbon dioxide production to escape to the atmosphere. The entire fermentation required 25 days.

[00154] For both the Pinot Gris and the Gamay Rose, the fermented experimental and control wines were allowed to settle for about two weeks before racking the wine off the lees. The wines were cold stabilized at 0° Celsius for 2 weeks and then fined using bentonite clay, specifically, Bentonite Vitiben™ (sold by Bosagrape Winery Supplies™, Burnaby, BC, Canada). After 97 and 88 days, respectively, the Pinot Gris and Gamay Rose wines were filtered using a Buon Vino Super Jet™ pump with #2 filter pads (sold by Buon Manufacturing™ Inc., Cambridge, ON, Canada) and bottled.

[00155] Ninety days after bottling the wines were tasted and scored. The tasting panel consistent of seven winemakers with an average of more than 15 years commercial winemaking experience. There were also two chefs with extensive knowledge of wine quality and one winery owner with extensive wine tasting experience. The Pinot Gris and Gamay Rose were blinded as to whether they were control or experimental but the tasting panel did know the varietal of grape from which the wine was made. According to a code prepared by random number generation, the control was labeled either“A” or“B”. Samples were poured into wine glasses and the samples were placed on colored placemats labeled“A” in red and“B” in blue.

[00156] The tasters were instructed to sample wine A and then compare it to wine B by marking their score on a horizontal 10 cm long visual analog scale (VAS) (shown in Figure 14). Marks toward the left-hand end of the line were less positive than marks toward the right-hand end. There were seven attributes rated in this way for each sample. Following the tasting, the code was broken and the scores for each of the four wines were obtained by measuring the distance in centimeters from the left-hand end of the VAS.

[00157] The objective was to determine if control and experimental wines were different so the difference between the two ratings for each attribute was determined. This difference was translated into % change and plotted on a radar graph (Figure 15) based on the mean scores from the 10 tasters. The control wine was given a zero score for each attribute. Negative values indicate the experimental wine was judged inferior for that attribute and positive values for the experimental wine indicate a more favorable score for that attribute. The tasters rated all seven attributes with higher scores for the membrane-treated experimental wines. The largest positive changes were observed for“fruity aroma, complexity, fruity taste and overall rating”. These are the types of changes expected if the membrane-treated wine contained higher levels of aroma compounds.

[00158] The aroma compounds were also analyzed at the Mass Spectrometry Facility Lab in the Wine Research Centre at the University of British Columbia in Vancouver, BC. Samples of 300 ml wine were transported by automobile from Naramata, BC to Vancouver (450 km). Prior to sample analysis, an aliquot of each sample (5ml) of wine was pipetted into a 20 ml headspace vial (Agilent, US) that contained 1.5g Sodium Chloride, crimped with a 3mm tan PTFE/ white silicon rubber septa (Canada Life Science, Canada).

[00159] Metabolites profiling was done by headspace solid phase micro extraction (HS-SPME) method using carboxen-polydimethylsiloxane- divinylbenzene, CAR/PDMS/DVB fiber (Supelco, USA) on an Agilent 7890/5975C gas chromatograph quadrupole mass spectrometer in electron ionization mode fitted with an Agilent J&W DB-WAX 30m x 0.25mm ID, 0.25um thickness fused silica column (Agilent, USA). All samples for GCMS analyses were injected in duplicate, one injection per vial and data were analyzed using Enhanced Chemstation software (Agilent, USA). The spectra were matched against Wiley9Nist08 mass spectrum libraries (Wiley, US).

[00160] Each sample was incubated and agitated for 5 minutes at 30°C, 500rpm prior to SPME fiber adsorption for another 5 minutes followed by fiber injection and desorption in the GC injection port for additional 3 minutes. The GC oven program temperature started at 40°C hold for 4 min then increased to 150°C at 3°C/min and further increased to a final temperature of 230°C at 25°/min and hold for 10 minutes. The injector temperature was 250°C and the injection was done with a split ratio of 3:1.

Helium was used as the carrier gas with a flow of 1 ml/min. The mass spectrometer was operated in scan mode (33-450)

[00161 ] Figure 16 shows the results for all measured peak areas in the Pinot Gris wine. The plot in Figure 16 shows the percent change for each measured peak in the experimental Pinot Gris wine as compared with the peaks for the control Pinot Gris wine. Table 2 shows the aroma compounds measured for the plot shown in Figure 16.

[00162] Peak areas for 18 of 49 measured compounds changed less than 10% from control peak areas. There were 3 compounds (2 ethyl esters and 1 acetate ester) that decreased in the experimental wine compared to the control wine. However, 28 of 49 aroma compounds increased by more than 10% in the experimental membrane-treated Pinot Gris wine compared to the control wine. The average change in peak area for all measured aroma compounds was +22.3%.

[00163] Figure 17 shows the results for all measured peak areas in Gamay Rose wine. The plot in Figure 17 shows the percent change for each measured peak in the experimental Gamay Rose wine as compared with the peaks for the control Gamay Rose wine. Table 3 shows the aroma compounds measured for the plot shown in Figure 17.

[00164] The pattern for the Gamay Rose peaks was similar to that obtained for the Pinot Gris wine. 15 of 50 measured aroma compounds changed by less than 10%, while 7 compounds (4 ethyl esters, 1 alcohol, 1 acetate ester and 1 alkylated phenol) decreased more than 10%. However, peak area for 28 of 50 measured aroma compounds increased more than 10% in the experimental Gamay Rose wine compared to the control wine. The average change in peak area was +66.9%, although much of this increase was due to the large increases in diethyl succinate, ethyl 9- decenoate and phenylethyl alcohol after membrane treatment.

[00165] In summary, the results of Experiment #4 indicate that fermenting wine while scrubbing carbon dioxide from the recirculated headspace with a fixed-site carrier membrane and allowing the carbon dioxide to pass out to the atmosphere retains wine aroma compound concentrations at higher levels than using conventional fermentation with an airlock or open top container. The recirculation of aroma compounds in flow passage 922 and headspace 906 retards flow of volatile compounds from the wine and produces wine with increased aroma and better taste.

[00166] The Pinot Gris results are notable because membrane carbon dioxide scrubbing was applied only during the first day of fermentation. The increased aroma compound concentration in the finished Pinot Gris suggests it can be important to retain these compounds during the early phase of fermentation, while the later phase of fermentation may in some instances be less important in preventing aroma compound loss. This is reinforced by the Gamay Rose results. Membrane treatment for the entire fermentation of the Gamay Rose increased aroma concentrations but also caused greater loss of volatile aroma compounds than was observed for the Pinot Gris. Without being limited by theory, this may be because of diffusion across the membrane 916. Therefore, long duration use of the membrane during fermentation may in some instances not be desirable for maximizing aroma in the finished wine. This makes the commercial utility of membrane carbon dioxide scrubbing more appealing.

Rubbery Membrane

[00167] Embodiments described below relate to the use of a rubbery membrane in an apparatus for preserving the aroma of a fermentable beverage in a fermentation apparatus. These embodiments address undesirable loss of aroma across the glassy membrane observed in experiment #4. The rubbery membrane may divide a flow passage between an upstream portion and a downstream portion, whereby a gaseous headspace mixture that enters the flow passage must pass through the rubbery membrane from a feed side to a permeate side in order to re-enter the headspace of the fermentation container. This contrasts with some methods and apparatus described above, in which gas (including aromatic fluids in vapor form) that passed through the membrane to the permeate side are removed from the system. [00168] The rubbery membrane has high permeability to both carbon dioxide gas and aromatic fluids. In use, the rubbery membrane may strip aromatic fluids from the gaseous headspace mixture located in the upstream portion of the flow passage. A portion of the stripped gaseous headspace mixture is drawn across the rubbery membrane to the permeate side by a gas and aromatic fluid pressure differential. The aromatic fluids absorbed in the rubbery membrane are drawn to the permeate side by a chemical potential difference (e.g. difference in partial pressure of aromatic fluids between the feed and permeate sides of the membrane), and by entrainment with the permeated carbon dioxide gas. At the permeate side of the membrane, a gaseous mixture of the carbon dioxide gas and aromatic fluids forms and moves downstream to re-enter the headspace. This gaseous mixture has a higher concentration of aromatic fluids than the gaseous headspace mixture that first entered the flow passage. Accordingly, the concentration of aromatic fluids in the headspace is increased by circulating the gaseous headspace mixture through the flow passage.

[00169] The rate at which aromatic fluids escape the fermenting beverage into the headspace has an inverse correlation to the concentration (e.g. partial pressure or volume density) of aromatic fluids present in the gaseous headspace mixture. Accordingly, the rate at which aromatic fluids escape the fermenting beverage can be reduced by increasing the concentration of aromatic fluids in the gaseous headspace mixture. Embodiments disclosed herein circulate the gaseous headspace mixture through a flow passage in order to increase the concentration of aromatic fluids in the gaseous headspace mixture. This slows the loss of aromatic fluids from the fermenting beverage. This approach is very different than prior techniques, which make no attempt to slow the loss of aromatic fluids from the fermenting beverage into the headspace gas, and instead condense the escaped aromatic fluid out of the headspace gas into a liquid, and reintroduce the liquid aromatic fluid directly into the beverage, after fermentation is complete. Such prior techniques require large condensers, which are expensive to purchase and operate. For many companies, the expense of such prior techniques makes them commercially infeasible.

[00170] Aromatic fluids in fermenting beverages, such as wine, include hydrophobic fluids characterized by an increased logP(octanol/water) partition coefficient (e.g. logP(o/w) greater than about 2), and hydrophilic fluids characterized by a lower logP(o/w) partition coefficient (e.g. logP(o/w) less than about 2). Aromatic fluids such as ethanol and fusel alcohols are hydrophilic and commonly cause a bitter taste in wine. Aromatic fluids, such as esters that are prized for their beneficial effect on wine quality, are generally hydrophobic.

[00171 ] Thus, for many fermenting beverages, including wine, it is desirable to preferentially retain hydrophobic aromatic fluids over hydrophilic aromatic fluids in the fermenting beverage. Due to their hydrophobicity, the hydrophobic aromatic fluids escape more readily into the gaseous headspace mixture. Embodiments described below may preferentially increase the concentration of hydrophobic aromatic fluids in the gaseous headspace mixture to slow the loss of the hydrophobic aromatic fluids from the fermenting beverage. This helps preserve the concentration of hydrophobic aromatic fluids in the fermenting beverage, which improves the flavor and quality of the final beverage.

[00172] In some embodiments, a rubbery membrane such as for example a silicone rubber membrane (e.g. polydimethylsiloxane (PDMS) membrane) is used. The rubbery membrane, e.g. particularly PDMS, is highly permeable to both carbon dioxide gas and aromatic fluids. In particular, the rubbery membrane may be hydrophobic such that it has much higher permeability to hydrophobic aromatic fluids than to hydrophilic aromatic fluids. Hydrophobic membranes other than PDMS may also be used to achieve this same effect.

[00173] Reference is now made to FIG. 18, which shows a fermentation apparatus 1000 in accordance with an embodiment. As shown, apparatus 1000 may include a fermentation container 1002, a flow passage 1004, a rubbery membrane 1008, and a negative pressure generator (e.g. vacuum pump) 1012. The fermentation container 1002 may include a headspace 1016 above a fermenting beverage 1020, and exit and re-entry ports 1024, 1028, respectively, in fluid communication with the headspace 1016. Exit and re-entry ports 1024, 1028 may be located anywhere on fermentation container 1002 in fluid communication with the headspace 1016 (e.g. at an elevation above fermenting beverage 1020). For example, one or both of exit and re-entry ports 1024, 1028 may be located on a container sidewall 1032 or an openable (e.g. removable) closure 1036 as shown.

[00174] As shown, fermenting beverage 1020 includes aromatic compounds 1040 (also referred to as aromatic fluid 1040). Aromatic compounds 1040 may escape (e.g. evaporate) from fermenting beverage 1020 along with carbon dioxide gas 1042 into a gaseous headspace mixture 1044 in container headspace 1016. The rate at which the aromatic compounds 1040 escape from fermenting beverage 1020 into the gaseous headspace mixture 1044 is negatively correlated to the concentration of the aromatic compounds 1040 in the gaseous headspace mixture 1044 and positively correlated to the hydrophilicity of the aromatic compounds 1040.

[00175] As used herein and in the claims, an element may be described as“gaseous” where it is a gas, a vapor, a colloidal suspension of particles in a gas or vapor, or combinations thereof.

[00176] Still referring to FIG. 18, the gaseous headspace mixture 1044 exits in headspace 1016 and enters flow passage 1004 through exit port 1024. Flow passage 1004 fluidly connects exit port 1024 to re-entry port 1028. For example, flow passage 1004 may extend from exit port 1024 to re-entry port 1028. As shown, rubbery membrane 1008 is located in the flow passage 1004 and divides flow passage 1004 between an upstream portion 1048 and a downstream portion 1052. In the illustrated example, fluids (e.g. carbon dioxide gas and aromatic fluids) must pass through rubbery membrane 1008 in order to move from upstream portion 1048 to downstream portion 1052. For example, rubbery membrane 1008 may have a feed side 1056 defining a terminal end of upstream portion 1048, and a permeate side 1060 defining the beginning of downstream portion 1052.

[00177] As gaseous headspace mixture 1044 travels within upstream portion 1048 and makes contact with membrane feed side 1056, rubbery membrane 1008 strips the gaseous headspace mixture 1044 of aromatic fluid 1040. For example, the aromatic fluid 1040 may dissolve into the rubbery membrane 1008. As discussed above, rubbery membrane 1008 may be preferentially permeable to hydrophobic aromatic fluids. For example, upon making contact with feed side 1056, a greater proportion of the hydrophobic aromatic fluids therein may dissolve into rubbery membrane 1008 than the proportion of hydrophilic aromatic fluids therein that dissolve into rubbery membrane 1008. The aromatic fluids 1040 removed by the rubbery membrane 1008 may therefore have a greater ratio of hydrophobic aromatic fluids to hydrophilic aromatic fluids than was in the gaseous headspace mixture 1044 before making contact with the rubbery membrane 1008. In some embodiments, membrane 1008 may strip at least 50% (e.g. 50% to 100%) of the hydrophilic compounds of the aromatic fluid 1040 in the gaseous headspace mixture 1044 within upstream portion 1048.

[00178] The aromatic fluids 1040 dissolved in the rubbery membrane 1008 diffuse from the feed side 1056 to the permeate side 1060. The rate at which the aromatic fluids 1040 move to the permeate side 1060 is positively correlated to a chemical potential difference (i.e. difference in partial pressure of aromatic fluids 1040) between that existing within membrane 1008 and the permeate side 1060. Due to the high solubility of hydrophobic aroma compounds in PDMS and a vacuum at permeate side 1060 of membrane 1008, flux of the aromatic fluids across membrane 1008 is high resulting in permeate aroma concentration at re-entry port 1028 being higher than aroma concentration in the feed gas mixture. For example, the permeate aroma concentration at re-entry port 1028 may be at least 2 times (e.g. 2 to 5,000 times) the aroma concentration in the feed gas mixture. This is known as the membrane enrichment factor.

[00179] A portion of the carbon dioxide gas 1042 in the gaseous headspace mixture 1044 that makes contact with the membrane 1008 also diffuses across the rubbery membrane 1008 from the feed side 1056 to the permeate side 1060. The rate at which the carbon dioxide 1042 moves through rubbery membrane 1008 to the permeate side 1060 is positively correlated to the carbon dioxide gas partial pressure differential between feed side 1056 and the permeate side 1060. A lower gas pressure (vacuum) on the permeate side 1060 causes the carbon dioxide gas 1042 to move to the permeate side 1060.

[00180] As shown, a negative pressure generator 1012 may be positioned downstream of rubbery membrane 1008 to reduce the gas pressure on permeate side 1060 below the gas pressure on feed side 1056. This causes the carbon dioxide gas 1042 to move downstream through membrane 1008.

[00181 ] Advantageously, the diffusion of carbon dioxide gas 1042 downstream through rubbery membrane 1008 also accelerates the movement of aromatic fluids 1040 through rubbery membrane 1008. By removing the aromatic fluids 1040 from the permeate side 1060, the partial pressure of aromatic fluids 1040 at the permeate side 1060 is reduced (compared to the absorbed aroma concentration at feed side 1056 of membrane 1008), which increases the chemical potential difference across rubbery membrane 1008, and accelerates the rate at which the aromatic fluids 1040 diffuse to the permeate side 1060. Although permeated carbon dioxide dilutes aromatic fluids on permeate side 1060 of membrane 1008 the re-pressurized gaseous mixture 1064 downstream from vacuum pump 1012 has a higher aroma concentration than headspace gaseous mixture 1044.

[00182] The concentrated gaseous mixture 1064 is composed of the carbon dioxide gas 1042 and aromatic fluid 1040 that diffused to the permeate side 1060 of rubbery membrane 1008.

[00183] When a sweep gas is used to dilute aroma on the permeate side of membrane 1008 it increases the chemical potential difference for enhanced aroma diffusion. The higher the sweep flow the greater is the aroma flux across membrane 1008. In some embodiments (not shown) additional sweep, over and above that provided by permeated carbon dioxide, can be obtained by shunting a portion of retentate discharge 1068 to permeate side 1060 of membrane 1008.

[00184] The concentrated gaseous mixture 1064 travels through downstream portion 1052 and exits flow passage 1004 into container headspace 1016 through re-entry port 1028. In container headspace 1016, the concentrated gaseous mixture 1064 mixes with the headspace gaseous mixture 1044, which increases the concentration of aromatic fluids 1040 in headspace 1016. Gaseous headspace mixture 1044 is continuously circulated through flow passage 1004 so that the concentration of aromatic fluids 1040 (particularly hydrophobic aromatic fluids) in headspace gaseous mixture 1044 continues to rise. The rising concentration of aromatic fluids 1040 progressively slows the escape of aromatic fluids 1040 from fermenting beverage 1020, so that the final beverage at the end of the fermenting process will have an improve flavor profile.

[00185] As noted above, only a portion of carbon dioxide gas 1042 of the gaseous headspace mixture 1044 that contacts feed side 1056 will diffuse across membrane 1008. As shown, flow passage 1004 may include an exhaust outlet 1068 at retentate port 1068 which is fluidly connected to upstream portion 1048. Gaseous mixture 1072, which is composed of carbon dioxide gas 1042 and residual aromatic fluids 1040 that were not stripped by rubbery membrane 1008 may exit apparatus 1000 through exhaust outlet 1068. For example, gaseous mixture 1072 may exit through exhaust outlet 1068 to atmosphere.

[00186] In some embodiments, exhaust outlet 1068 may include a mass flow sensor 1076 to obtain the carbon dioxide flow rate and provide a useful indication of fermentation kinetics. The gas flowing out of port 1072 is primarily carbon dioxide and substantially equals the amount not permeating membrane 1008. Therefore, mass flow sensor 1076 provides a measure of the carbon dioxide production of fermenting beverage 1020.

[00187] In one aspect, the illustrated system may operate successfully with a negative pressure generator 1012 (e.g. vacuum pump) of low strength, purchase cost, and operating cost. For example, negative pressure generator 1012 may generate a gas pressure at permeate side 1060 of between 200-500 mbar absolute. This mild vacuum is sufficient to cause high diffusion rates of hydrophobic aroma compounds because apparatus 1000 diffuses carbon dioxide gas 1042 across rubbery membrane 1008 to dilute the aromatic fluid 1040 at the permeate side 1060 of membrane 1008 and works synergistically with the mild vacuum to increase a chemical potential difference of the aromatic fluid 1040 across membrane 1008. Both the mild vacuum and the sweeping flow of permeated carbon dioxide increase the rate at which the aromatic fluid 1040 diffuses to the permeate side 1060. In contrast, a much stronger vacuum of 1 -20 mbar absolute is normally employed in pervaporation and vapor permeation systems in order to increase the aroma chemical potential difference across the membrane. The expense of high powered vacuum pumps is one factor why commercial techniques for preserving aromatic fluids are uncommon.

[00188] In some embodiments, negative pressure generator 1012 may generate a gas pressure at permeate side 1060 of 0.1-1000 mbar absolute. A stronger vacuum of between 0.1 -200 mbar absolute may increase the carbon dioxide flow rate across membrane 1008 and increase the chemical potential gradient for aroma compound diffusion, although the equipment to generate this vacuum may be more expensive. A weaker vacuum of between 500-1000 mbar absolute may reduce the carbon dioxide flow rate across membrane 1008, and possibly the effectiveness of the system, but may be implemented with less expensive equipment.

[00189] Moreover, whereas prior systems may require a pressure switch to start and stop pumps to mitigate the development of excess headspace pressure or headspace vacuum, such equipment is not required for apparatus 1000 because an exhaust outlet 1068 is provided to discharge gaseous mixture 1072 from the system. This too contributes to the reduced cost of apparatus 1000.

Example #5

[00190] The flow passage was rearranged as illustrated in FIG. 18. Feed gas from the headspace flowed passively to a membrane module containing thousands of small hollow fibers made of polydimethylsiloxane (PDMS). PDMS is a rubbery polymer, commonly known as silicone, which is highly permeable to both carbon dioxide and to organic molecules such as aroma compounds, especially those with high logP(o/w). In this membrane the feed gas flows through the lumen side of the hollow fibers where aroma sorption occurs into the PDMS with subsequent diffusion of both aroma and carbon dioxide across the membrane to the permeate portion of the flow passage on the shell side of the hollow fibers. Any carbon dioxide which does not diffuse across the membrane is retained in the retentate part of the flow passage and is discharged to the atmosphere. The permeate gas is pumped back to the headspace to increase headspace aroma concentration and achieve the goal of increasing aroma in the finished beverage. The retentate flow that is discharged to atmosphere is substantially equal to the amount of carbon dioxide being produced at that time during fermentation. This allows information about the fermentation kinetics to be determined by monitoring the discharging retentate flow.

[00191 ] To increase aroma sorption and diffusion across the membrane and to ensure a high carbon dioxide diffusion into the permeate portion of the flow passage a vacuum was applied to the permeate side of the membrane. However, contrary to usual practices during vapor permeation, the amount of vacuum applied was mild (only 200 to 500 mbar, absolute). This lower vacuum is sufficient to ensure diffusion of most of the aroma because the relatively high volume of carbon dioxide crossing the membrane provides a sweep effect of lowering the permeate aroma concentration and increasing the chemical potential difference which helps drive aroma diffusion. This is a unique feature of this system and it is possible because the flow passage contains substantially 100% carbon dioxide during fermentation (the aromatic fluids amount to only tiny fractions of 1 %). The end result is the recompressed permeate portion of the flow passage downstream from the vacuum pump that re-enters the headspace contains carbon dioxide and aroma and the aroma concentration is higher than the headspace concentration. When the permeate gas re-enters the headspace, the increased aroma concentration retards loss of aroma from the liquid beverage resulting in a more aromatic finished product. [00192] Eleven separate experiments were done on grape juice volumes ranging from 17 liters to 1800 liters and each membrane-treated wine had a control comprised of the same juice as the membrane-treated wine. The experiment described below had sensory tasting results qualitatively similar to the other experiments.

Procedure

[00193] Forty liters of Pinot Gris juice was well mixed with hydrated DV10 yeast. Then 18 L of the yeast-containing mixed juice was added to each of two 23 L glass carboys. One carboy was fitted with an airlock and allowed to ferment to dryness. This was the control. Another carboy was also filled with 18 L of the same juice but a stopper was inserted in the top through which two ½” OD aroma barrier tubes protruded into the headspace above the liquid in the carboy. One tube supplied headspace gas to a PDMS hollow fiber membrane. This was the feed portion of the flow passage which flowed passively through the lumen side of the hollow fibers. The other tube was connected to the shell side of the membrane module which returned permeate gas from this portion of the flow passage back to the headspace. Located near the shell side outlet was a small vacuum pump which created a vacuum of 500 mbar (absolute) on the shell side of the hollow fibers to help create a chemical potential gradient for aroma diffusion and to decrease permeate carbon dioxide partial pressure to facilitate carbon dioxide diffusion across the membrane. The permeated carbon dioxide acted as a carrier gas to transport the diffused aroma back to the headspace downstream from the vacuum pump, plus it provided a diluting effect on the permeated aroma to further increase the chemical potential difference over and above that provided by the vacuum to enhance aroma diffusion across the membrane. Carbon dioxide and aroma which did not diffuse across the membrane from the feed supply exited the membrane module at the retentate port. Nearly all of the retentate gas consisted of carbon dioxide and the volume was equal to the carbon dioxide production in the fermenting juice. Therefore, monitoring the retentate flow with a mass flow meter provided information about fermentation kinetics. The fermentation kinetics for experiment 5 are illustrated in FIG. 19. The headspace pressure rose to 1” water column at the peak of fermentation and then steadily decreased as carbon dioxide production fell. This mild headspace pressure supplied the force needed to push headspace gas through the feed and retentate parts of the flow passage. The vacuum pump was turned on at peak carbon dioxide production and it was turned off when carbon dioxide production was very low toward the end of fermentation. When the vacuum pump was on, aroma was being preserved in the headspace. Experiment 5 lasted for 260 hours with the vacuum pump running continuously during this time.

[00194] Upon completion of the fermentation, the wine was racked into clean 11 L glass carboys and allowed to clarify over a two week period.

Following the clarification period, both the control and membrane-treated wines were bottled and presented to a panel of expert wine tasters in a blinded fashion to determine how much the experimental wine differed from the control wine for each of seven characteristics: fruity aroma, fruity taste, complexity, smoothness, aftertaste and overall rating. Table 1 , below presents the results as the percent the experimental wine differed from the control wine. For all 11 experiments, the“overall rating” increased between 17% and 75% for the membrane-treated wines.

[00195]

Overall Rating

+16.8

[00196] Therefore, experiment 5 showed that using a hydrophobic membrane to allow aroma compound diffusion into the permeate stream and then returning the permeate stream back to the headspace consistently resulted in better tasting wine.

[00197] While the present embodiments as herein shown and described in detail are representative of the subject matter which is broadly contemplated by the present invention, the scope of the claims should not be limited by the preferred embodiments set forth in the examples, but should be given the broadest interpretation consistent with the description as a whole. Moreover, it is not necessary for an apparatus or method to address each and every problem sought to be solved by the present invention, for it is to be encompassed by the present claims.

Claims

CLAIMS:

1. A method for preserving the aroma of a fermentable beverage in a fermentation apparatus, the method comprising: fermenting the beverage in a fermentation container thereby producing a gaseous headspace mixture, the gaseous headspace mixture including carbon dioxide gas and aromatic fluid in a headspace located above the beverage contained in the fermentation container, the gaseous headspace mixture having a first concentration of aromatic fluid; moving the gaseous headspace mixture into an upstream portion of a flow passage into contact with a feed side of a rubbery membrane, the rubbery membrane stripping at least a portion of the aromatic fluid from the gaseous headspace mixture thereby reducing a concentration of the aromatic fluid in the gaseous headspace mixture in the upstream portion of the flow passage; discharging a first portion of the gaseous headspace mixture from the fermentation apparatus, the first portion having a second concentration of aromatic fluid less than the first concentration; diffusing carbon dioxide gas from a second portion of the gaseous headspace mixture through the rubbery membrane, the diffused carbon dioxide gas entraining the aromatic fluid at the permeate side of the rubbery membrane; and moving a concentrated gaseous mixture comprising (i) the diffused carbon dioxide gas and (ii) the aromatic fluid entrained at the permeate side of the rubbery membrane, through a downstream portion of the flow passage out of the flow passage and into the headspace of the fermentation container, the concentrated gaseous mixture having a third concentration of aromatic fluid greater than the first concentration.

2. The method of claim 1 , wherein: said diffusing carbon dioxide gas comprises generating a pressure differential across the rubbery membrane, wherein a downstream gas pressure at the permeate side is less than an upstream gas pressure at the feed side.

3. The method of claim 2, wherein: said generating a pressure differential comprises generating a vacuum pressure of between 200 to 500 mbar absolute to the permeate side.

4. The method of claim 3, wherein: said generating a pressure differential comprises operating a vacuum pump located downstream of the rubbery membrane.

5. The method of claim 1 , wherein: the membrane is a silicone rubber membrane.

6. The method of claim 1 , wherein: the membrane is hydrophobic.

7. The method of claim 1 , wherein: the aromatic fluid includes hydrophobic compounds and hydrophilic compounds, and the membrane has greater permeability to hydrophobic compounds than to hydrophilic compounds.

8. The method of claim 1 , wherein: said fermenting the beverage generates carbon dioxide gas at a first volumetric rate, said discharging comprises discharging carbon dioxide gas from the fermentation apparatus at a second volumetric rate, and the second volumetric rate is substantially the same as the first volumetric rate.

9. The method of claim 1 , wherein: said stripping comprises dissolving the portion of the aromatic fluid from the gaseous headspace mixture into the rubbery membrane.

10. The method of claim 1 , wherein: the aromatic fluid includes hydrophobic compounds and hydrophilic compounds; and said stripping removes at least 50% of the hydrophilic compounds from the aromatic fluid of the gaseous headspace mixture in the upstream portion of the flow passage.

11. The method of claim 1 , wherein: said discharging comprises measuring the first portion of the gaseous headspace mixture with a mass flow sensor.

12. A fermentation apparatus for preserving the aroma of a fermentable beverage, the apparatus comprising: a fermentation container, the fermentation container including a gas exit port and a gas re-entry port, the gas re-entry port located downstream of the gas exit port; a flow passage fluidly connecting the gas exit port to the gas re-entry port, the flow passage including an upstream portion connected to the gas exit port and a downstream portion connected to the gas re-entry port; a rubbery membrane located in the flow passage, the rubbery membrane separating the upstream portion from the downstream portion, the upstream portion terminating at a feed side of the membrane, and the downstream portion beginning at a permeate side of the membrane; an outlet fluidly connected to the upstream portion of the flow passage; and a negative pressure generator that when activated reduces a downstream gas pressure on a permeate side of the membrane to below an upstream gas pressure on a feed side of the membrane.

13. The fermentation apparatus of claim 12, wherein: the membrane is a silicone rubber membrane.

14. The method of claim 12, wherein: the membrane is hydrophobic.

15. A fermentation apparatus for preserving the aroma of a fermentable beverage, the apparatus comprising: a closure engageable with a fermentation container, the closure including a gas exit port and a gas re-entry port, the gas re-entry port located downstream of the gas exit port; a flow passage fluidly connecting the gas exit port to the gas re-entry port, the flow passage including an upstream portion connected to the gas exit port and a downstream portion connected to the gas re-entry port; a rubbery membrane located in the flow passage, the rubbery membrane separating the upstream portion from the downstream portion, the upstream portion terminating at a feed side of the membrane, and the downstream portion beginning at a permeate side of the membrane; a negative pressure generator that when activated reduces a downstream gas pressure on a permeate side of the membrane to below an upstream gas pressure on a feed side of the membrane.

16. The fermentation apparatus of claim 15, wherein: the membrane is a silicone rubber membrane.

17. The method of claim 15, wherein: the membrane is hydrophobic.

18. The method of claim 15, wherein: the membrane has greater permeability to hydrophobic aromatic compounds than to hydrophilic aromatic compounds.

19. The method of claim 15, wherein: the membrane is permeable to carbon dioxide gas.

20. The fermentation apparatus of claim 15, wherein: the negative pressure generator when activated reduces the downstream gas pressure to between 200 and 500 mbar absolute.