Inert gas: carbon dioxide, nitrogen and argon

Carbon dioxide dissolves in cider readily — far more readily than the other two — which is what makes it useful for carbonation and what makes it a nuisance for everything else. Blanketing a still cider with carbon dioxide for months will put some of that gas into solution, and a cider that arrives at the palate with a faint prickle it was not meant to have has usually been sitting under carbon dioxide.

Nitrogen is almost insoluble by comparison. It will displace air from a vessel or a line and change nothing about the cider’s carbonation, which is exactly what is wanted when a still product is being moved or a carbonated one is being pushed at a pressure higher than its own. The same insolubility makes nitrogen a stripping gas: bubbled through a cider it will carry dissolved gases out with it, removing oxygen and, if the operator is not paying attention, the carbonation as well.

Argon sits between the two in solubility and stands apart in density. It is markedly heavier than air, heavier than nitrogen and heavier than carbon dioxide, so a layer of argon laid over an open surface stays where it is put for longer than the alternatives. That is its one advantage and it is a real one for a part-full vessel or an open operation, set against a price that keeps it for small volumes and critical moments rather than for routine purging.

None of the three is a substitute for keeping a vessel full. A blanket is diluted by diffusion, lost through every joint, and displaced whenever the vessel breathes with a change in temperature. It protects for a week reliably and for a year not at all.

The dispense row explains a piece of cellar practice that otherwise looks arbitrary. On a long or a tall beer line, the pressure needed to push the product to the tap is higher than the pressure that would hold the cider at its intended carbonation, so pure carbon dioxide at that pressure would over-carbonate the product over a few days. A mixed gas supplies the push with the nitrogen fraction, which does not dissolve, while the carbon dioxide fraction is set to match the carbonation the cider should have.

Purging is displacing air from an empty space before the cider arrives — a tank, a hose, a bottle, a keg. On a vessel that can hold pressure, the efficient method is pressure cycling: pressurise with gas, vent to atmosphere, repeat, each cycle diluting what remains. On a vessel that cannot, gas is introduced at the bottom and allowed to fill upwards, using the density difference, and the vent is left open at the top. Filling a vessel from the bottom with the cider itself is also a purge, because the rising liquid pushes the air ahead of it rather than mixing with it.

Blanketing is maintaining a layer over a liquid surface that is already there. It is what a floating lid removes the need for. Where it is used it needs a low, continuous or periodically renewed supply and a vessel that is otherwise closed, and it should be understood as a delaying measure rather than a solution.

Sparging is deliberately bubbling gas through the liquid to change what is dissolved in it — nitrogen to strip oxygen out of a cider that has picked some up, or carbon dioxide to carbonate. It is the only one of the three that changes the cider rather than the space above it, and it is the one most easily overdone: a nitrogen sparge run for too long on a lightly carbonated cider will leave it flat.

Small operations buy cylinders. Larger ones take liquid carbon dioxide or nitrogen in a bulk vessel with a vaporiser, which is far cheaper per unit and requires a supply contract and a compound to put it in. A plant with a steady nitrogen demand may generate its own from compressed air with a membrane or pressure-swing adsorption unit, which produces nitrogen at a purity the user selects — and purity matters here, because the residual oxygen in a cheaply generated nitrogen stream is precisely the thing being excluded.

Grade matters as much as source. Gas for food contact is supplied to a food-grade specification with a certificate; industrial gas from the same molecule is not specified for it and may be filled into cylinders that have carried something else. This is a real distinction, not a commercial one, and it is worth confirming with the supplier rather than assuming.

Very large fermentation plants can recover their own carbon dioxide from the ferment, scrub, dry and liquefy it, and use it back in the process. It is a capital project rather than a purchase, and it is one of the few genuinely circular pieces of equipment in the industry.