The Physics of Gases in Solution

Buckle up. It's time to talk about the Physics of Champagne.


It is a fact that all gases are soluble in liquids to some degree. The temperature of the liquid and the pressure of a particular gas above the liquid are the primary factors that determine how much of that gas will dissolve: the lower the temperature, and the higher the pressure, the more gas will dissolve in solution.

For the sake of illustrating this point, consider a bottle of Champagne that has just been opened, and from which several glasses have been poured. The bottle is then resealed with a cap. Initially, there is a great deal of CO2 in solution (as evidenced by the bubbles in the wine), and very little in the headspace (most of it escaped when the bottle was opened). Because there is much CO2 in solution and little CO2 in the headspace, CO2 tends to escape from the wine into the headspace.

As this process continues, the pressure of CO2 in the headspace increases and the concentration of CO2 in solution decreases. Left on its own, in a matter of hours a state will occur in which the pressure of gaseous CO2 and the concentration of dissolved CO2 stabilize, neither increasing nor decreasing. This is called equilibrium.

The same process can also work the other way. Imagine a half-full bottle of flat Champagne that has been resealed and repressurized with carbon dioxide using the Perlage System. Because there is little CO2 in solution and a large quantity of high-pressure CO2 gas in the headspace, CO2 will dissolve into solution. This will decrease the pressure in the headspace and increase the concentration of CO2 in solution. Left on its own, in a matter of hours a state will occur in which the concentration of dissolved CO2 stabilizes at some level, neither increasing or decreasing. Again, this is equilibrium.

At equilibrium, however, things are far from static. Even at equilibrium, molecules of gas in solution continually escape the liquid and become molecules of gas in the headspace of the bottle; likewise, molecules of gas in the headspace are continually “trapped” by the liquid and become molecules in solution. In equilibrium, these two processes—escape and capture—happen at the same rate, giving the macroscopic appearance that nothing is happening.

It is not correct to say that the Perlage System (or any other device) “holds” carbon dioxide in solution—rather, it creates a condition whereby CO2 efflux from the liquid is balanced by a comparable CO2 influx into the liquid. Obviously, this cannot be accomplished with air, which is chiefly nitrogen and oxygen. This is why the use of high-pressure CO2 is so important to the Perlage System.

This also explains why it is insufficient to merely cap the bottle. CO2 will escape until equilibrium is reached, and each time the cap is removed, the process starts over again. If the headspace is large, there is a large volume to fill each time, and the amount of CO2 in solution (which is where you want it) drops rapidly.


The solubility of gases in liquids is strongly dependent on temperature. This is especially true with carbon dioxide. As mentioned above, the lower the temperature, the more CO2 the wine will hold. This is because the chemical reactions that bind CO2 in solution create compounds that are very weakly associated, and easily broken apart by slight increases in thermal energy. So, as the wine is cooled, and the molecular thermal energy of the liquid decreases, the more carbon dioxide can be entrained in compounds such as carbonic acid, carbonates, and the like.

This has consequences which are familiar. Consider two identical bottles, one cold, and the other warm. In the colder bottle, more CO2 resides in the liquid; in the warmer, more resides in the headspace. That is why a warmer bottle will make a louder “pop” when opened, and why a colder bottle has more carbonation than a warm one.


There is another way for gases to escape solution, besides molecular diffusion: bubbles. When the concentration of gases in a liquid is far higher than what the ambient temperature and pressure would dictate for equilibrium, large agglomerations of CO2 molecules, or bubbles, can form spontaneously.

It turns out, however, that bubbles cannot simply form on their own without some type of “seed,” called a nucleation site. From a thermodynamic perspective, it simply takes too much energy to form a bubble from nothing at the concentrations of carbon dioxide found in Champagne. A bubble can only form on tiny existing bubble or pocket of gas. These nucleation sites are typically provided by impurities stuck to the glass that the Champagne is served in, usually hollow fibers of cellulose that come from paper or cloth towels, or other impurities stuck to the glass that can trap pockets of air that do not get entirely wetted when wine is poured in the glass. Research has shown that these pockets of gas must be at least four microns in diameter in order to form a bubble.

It is true that bubbles can be suppressed, or “squeezed” back into solution by applying pressure to the surface of the liquid with some other gas besides CO2, but this doesn’t affect molecular diffusion, which is the mechanism by which most carbonation is lost. So the point remains: The equilibrium concentration of each gas in solution is determined by the partial pressure of that particular gas in the headspace. That is why only CO2 can be used to prevent the loss of carbonation—no other gas will work.