Gassing Regimens

Winemaking Talk - Winemaking Forum

Help Support Winemaking Talk - Winemaking Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.


Supporting Members
Supporting Member
Nov 23, 2019
Reaction score
Gassing Regimens (link to original article)

Gassing Regimens​

By: Conor McCaney, Graduate Assistant, Department of Food Science & Technology
The winemaking process is a dynamic one: from crush, to fermentation, on to post fermentation cellar procedures, aging, and bottling. Each step along the way allows for the potential ingress of oxygen, whether wanted or not. While oxygen is considered by many to be the enemy of wine, this is not always the case. In fact proper use of enological oxygen at crucial steps in the winemaking process is paramount to wine development. That said, many winemakers dutifully aim to eliminate it from the process altogether particularly in partial tank headspace. Proper gassing regimens and selection of the correct gas for a particular application is something that many do not do well and fail to fully understand the principals at play. Managing proper inert gas procedures is tricky. Most protocols are generally arbitrary ones copied from bad information and the proliferation of poor techniques passed on anecdotally from winemaker to winemaker. In general it is a procedure that is often over looked and never given much thought. This usually means the use of a high pressure cylinder (most often nitrogen), and a ¼” or ½” hose that is allowed to run for an arbitrary amount of time, generally 15 to 20 minutes. The results are the improper use of inert gases from the failure to measure gas volumes delivered (using a flowmeter), monitoring results with the use of a dissolved oxygen meter, using an under or oversized delivery system and unsubstantiated cost analysis pertaining to gas type and volume needed.
Typical gas choices are: carbon dioxide, nitrogen, and argon. Most wineries choose to use carbon dioxide and nitrogen because they believe it provides the best cost-benefit in terms of oxygen displacement per unit cost. This is not the case. To understand this, we must first delve into some fundamental principles of gases. In the wine industry, we typically use gas by volume, either in standard cubic feet or molar volume delivered from a standard steel pressurized cylinder in which the gas is compressed. These gas volumes are usually measured at 25°C and 1 atm. If you happen to purchase gas by the pound it is necessary to divide the gas by its molecular weight before you can compare gases to one another. The approximate molecular weights are: 40 g/mole for argon (Ar), 44 g/mole for carbon dioxide (CO2), 28 g/mole for nitrogen (N2), and 29 g/mole for air. One mole of any of these gases measured at standard pressure (1atm) and temperature (25°C) occupies one molar volume, roughly equivalent to 22.4 liters, 5.92 gallons, or 0.8 standard cubic feet. Using the ideal gas law PV = nRT the behavior of gases can be described in which pressure and volume is a fixed proportion in relation to the number of moles of gas at absolute temperature. This indicates that gas molecules take up the same amount of space regardless of their mass when they are at the same temperature and pressure (Avogadro’s Law). Thus one mole of any gas contains the same number of molecules (i.e., 6.02 x 1023). This also indicates that the head space in a tank, barrel, or other container will fluctuate regularly throughout the day in response to temperature and pressure changes. Tanks that are kept outside experience greater temperature changes throughout the day compared to a tank kept inside at a constant temperature. Changes in barometric pressure and temperature can cause the headspace in a tank to pump 3% to 7% of its volume in and out daily. This ultimately means that the headspace in a tank is not a static system and could be constantly changing.
Air is roughly composed of 78% nitrogen, 21% oxygen, and 1% argon, so in essence nitrogen is air without the oxygen. In any gassing procedure it is ideal to reduce the percentage of oxygen in the headspace to below 1% or even below 0.5% to inhibit the growth of aerobic microbes and prevent wine oxidation. The most commonly used gas in winemaking is nitrogen (N2) with a molecular weight (MW) of 28 g/mole making it moderately lighter (less dense) than air at 29 g/mole MW. Graham’s law of diffusion (also known as Graham’s law of effusion) states that the rate of effusion of a gas is inversely proportional to the square root of its molar mass at constant temperature and pressure. This principle is often used to compare the diffusion rates of two gasses such as nitrogen and air. The diffusion rates of nitrogen and air are almost identical meaning that nitrogen does not provide adequate layering, but rather readily mixes with air and does not remain in contact with the wine surface for an extended period of time. This also means that in order to reduce the O2level from 21% to less than 1%, the headspace needs to be flushed with a volume of nitrogen that is five times the volume of the headspace. So if the tank has a 100 gallons of head space it would take 500 gallons of nitrogen to reduce the O2level from 21% to below 1%. The cost of nitrogen is approximately $0.05 per cubic foot (Praxair, Inc). However, because nitrogen requires five times the volume equivalents to reduce the O2percentage from 21% to less than 1%, the cost to gas a barrel (60 gallons) is $2.00, 100 gallons of headspace is $3.34 and 1,000 gallons of headspace is $33.42. This is significantly higher than the cost of using argon for the same O2reduction in the equivalent headspace volumes. This is why headspace gassing with nitrogen requires a substantial effort and time commitment on the part of the winemaking team to be effective. It takes substantially more nitrogen and a greater application time compared to argon to achieve the same reduction in oxygen percentage with a shorter effective shelf life.
In contrast to nitrogen is carbon dioxide (CO2), which is significantly heavier than air at 44 g/mole compared to 29 g/mole and by Graham’s law has a much slower rate of diffusion compared to air. This allows for a more significant displacement of air compared to nitrogen. However, when CO2is delivered from a compressed tank, it is difficult to achieve the desired laminar flow necessary for successful layering. This results in substantial mixing of CO2and air. A more effective alternative for CO2delivery is dry ice (solid CO2) which leads to more efficient layering of CO2and subsequent displacement of air but does not form a permanent layer. However, it should be noted that CO2cannot be considered inert in the same way as nitrogen and argon. Because of Henry’s Law, which states that the solubility of a gas is directly proportional to the partial pressure of the gas above the solution, CO2readily dissolves into wine under standard conditions and its solubility can be increased or decreased with changes in pressure. This dissolution of CO2into the wine causes the pressure in the tank to fluctuate and results in the intake of air from the outside environment through an airlock to replace the lost volume of gaseous CO2. If there is no vacuum release valve on the tank, this could cause the tank to implode. Carbon dioxide dissolved in the wine will also alter the acid, flavor, and textural profile of the final wine. Carbon dioxide is much more effective when deployed early in the winemaking process at juice stage or when the wine is young as there will be substantial time to allow excess dissolved CO2to come out of solution. The use of dry ice to protect grape must is an effective way to protect wine must from excess oxygen exposure, deter fruit flies, and subsequently cool the must.
This leaves argon with a molecular weight of 40 g/mole, making it substantially heavier than air (29 g/mole) and similar in weight to CO2but more inert. A major opposition to the use of argon regularly in wine production is because it is significantly more expensive compared to the other two gases. It is true that when purchasing gas by volume argon is roughly three times as expensive as nitrogen or carbon dioxide. However it is much more effective at displacing air and creating a more permanent blanket that remains in contact with the wine surface longer while also remaining inert compared to CO2. Less volume is also needed to achieve the same desired results. At approximately $0.11 per cubic foot (Praxair, Inc) not including daily tank rental fee, a barrel (60 gallons) can be completely gassed with argon for $0.88, 100 gallons of head space for $1.47, and 1,000 gallons of headspace for $14.71. This cost is relatively insignificant to a winery’s bottom line in terms of the degree of quality preservation that argon can provide.
Continued from post above

When using any of the gases discussed previously, it is important to select the proper pressure gauge, hose diameter, hose length, flowrate, and the use of a t-valve in order to deliver the gas under laminar conditions. The use of a lower velocity, will encourage laminar flow delivery and reduce any chance of turbulence and subsequent mixing with air, thus creating a more layered effect.

It is ideal to keep the flow velocity to 1 meter per sec or less. To determine the velocity divide the volumetric flow rate in cubic meters per second by the cross sectional area in meters of the hose being used. If using cubic feet instead of cubic meters, perform the same calculation but convert the units from cubic meters to cubic feet and meters to feet. Table 1 shows that it is best to use a 1.5” or 2” diameter line with a t-valve to deliver an adequate amount of gas in a reasonable amount of time. This will require the use of an oversized regulator compared to the typical 0.25” regulator used on most compressed gas cylinders.
In essence it is best practice to recommend the use of argon as the headspace gas for the majority of wine production processes. Carbon dioxide and nitrogen have their respective roles but when it comes to headspace gassing argon it the number one choice. In the production of high quality wine, it is imperative to establish proper gassing procedures. This includes the successful training of staff in all aspects of gassing procedures and the selection of the correct gas for the appropriate task. This also requires selecting the correct regulator size, hose diameter and length, the use of T-valves, measuring gas flow using a flowmeter, and finally verifying results with the use of a dissolved oxygen meter to monitor oxygen levels in the tank headspace pre and post gassing. The proper investment of time and resources in this often overlooked area of winemaking can have a profound effect on wine quality and preservation in the long run. It can also reduce long term costs by reducing the amount of gas and time required to achieve the desired reduction in the amount of oxygen present in a tank headspace.

End citation.

We've been using argon for many years. I've been curious about switching to nitrogen. Glad I didn't spend the money.

We use a two hole stopper and two pieces of plastic racking cane to apply a blanket of argon in a carboy. One tube is connected to the tank, the other allows the air to flow out. The tank is barely cracked open to allow a very slow, quiet flow of argon into the the carboy. Seems to work very well. Several carboys with a gallon of headspace were blanketed with argon and a solid bung applied. After three months sensory observation does not indicate that oxidation has occurred. Since we do not have a dissolved oxygen meter, there is no empirical evidence. That meter is in the budget for this year.


How can laminar flow be encouraged in a closed container like a carboy?
How do you use a t-valve to blanket a VCT?
Thanks for posting! I believe you can get bell-shaped diffusers that direct the gas flow parallel to the surface (there is a picture of one here), but you still of course need to fill the headspace - the 'layer' is transient at best. I'm not sure if those bells come in a size small enough to fit in a carboy but they would work in a tank. For a carboy, maybe a rigid tube with the end blocked off and holes near the bottom to direct the gas out?
Great idea! It would be easy enough to plug the end of a racking cane and create some slits in the side near the bottom. Razor saw should work for that.

The gassing bell idea makes sense, too. Another article recommends this. They also recommend using a tee at the end of the hose when blanketing a tank or other large vessel.
There is good information in that article, but there is a lot of misapplied information.

In short, the grad student claims that, in the case of N2, the gases mix essentially IMMEDIATELY. Using this assumption, he claims you need to flush the headspace with 5 volumes of N2 in order to dilute the O2 to an acceptable level. In contrast, he assumes that Ar will not mix AT ALL. Therefore, he claims you can flush the headspace with exactly 1 volume of Ar.

How does he justify this difference? By citing Graham's Law, which relates the rate of diffusion to the square root of the molar mass. And what is the numerical value? The ratio of the diffusivities will be SQRT(28/40) = 0.84. Wowee! N2 diffuses a whopping 16% faster than Ar! Obviously we should treat them completely differently! </s>

I also object to the phrasing of Ar "creating a more permanent blanket that remains in contact with the wine surface longer." More permanent? Is it permanent or not? (A: It is not.) Again, Ar is only 16% slower to diffuse than N2. When I do the calculation, it takes on the order of minutes (depending of vessel volume, etc.) for either gas to mix. Which is long enough that you can use either gas to displace the headspace, but short enough that there is no "blanketing effect" for either gas.

I also object to his selective application of some other laws. He cites Henry's Law to discount the usefulness of CO2, but of course this also applies to Ar. He does touch upon the real reason to prefer Ar over CO2, which is that CO2 is not really inert in wine. But why cite Henry's Law, and talk about imploding tanks, for one gas and not the other? Also, apparently it is bad to introduce CO2 from a pressurized tank, but with Ar, it is just peachy to do so. 🤷‍♂️

However, I do find the discussion about ways to introduce gas carefully to promote laminar flow good and useful.
I also object to the phrasing of Ar "creating a more permanent blanket that remains in contact with the wine surface longer."
A surprising number of folks I've spoken to seem to think that there is a 'blanket' effect even if you only introduce a small amount of argon, but as you say physics doesn't work that way.

Latest posts