Does CO2 leak out of tires?

CO2 tank

Some years ago racing bicyclists began carrying tire-repair kits that used common miniature CO2 cartridges to refill a patched tube in seconds, saving the time otherwise wasted using a manual pump. The kits subsequently became popular among recreational cyclists who weren’t necessarily in a hurry, but simply couldn’t stand not having the latest high-performance gear, or felt that the drudgery of working a manual pump was simply . . . hell.

Soon, however, riders began noticing something: The tires they filled with CO2 seemed to lose pressure much more rapidly than when filled with air. Some claimed a tire filled to 90 psi with CO2 would be down to 40 in a few days. 

Impossible, said the makers of the CO2 kits. A CO2 molecule is larger than either a nitrogen molecule or an oxygen molecule, which together comprise 99 percent of air. How could CO2 leak out faster? Don’t ask us how, said the bicyclists, we just know it happens.

Finally a few people familiar with chemistry looked for an explanation, beginning with the properties of gas permeation by diffusion. All gasses exhibit a permeation rate through butyl rubber—the material used in most bicycle tires—proportional to the inverse of the square root of their molecular weights. Using this formula, you can show that the rate of permeation of CO2 through a butyl tube compared to air should be . . . uh, well, lower. Hmm . . .

Finally someone got it. The answer has nothing to do with molecular weight, although it is still a chemical phenomenon. It turns out that CO2 is actually soluble in butyl rubber—it essentially melts right through the material without having to wait for permeation.

And that brings up a question: Does the same thing happen with the tires on four-wheel-drive vehicles when owners use the popular CO2 tanks for airing up?

Tubeless automotive tires are obviously different in construction than bicycles tubes. For one thing they’re probably ten times as thick, with various reinforcing belts and cords. One source tells me the inner liner of all tubeless tires is “halobutyl” rubber infused with other compounds to limit permeability. So the phenomenon should be at least substantially slower on an automotive tires than on a bicycle tube. Additionally, most uses of CO2 involve topping up a tire that was only partially deflated for trail driving; thus there is probably a significant percentage of plain old air left in the tire. I decided to see if the CO2 effect was a factor in truck tires, and if so to what extent.

I used the front tires on our Land Rover One Ten, since it was due to be parked for a while and I wanted perfectly stable conditions. I deflated both tires completely, then filled one with a compressor, the other with my CO2 tank, to 40 psi (in the process learning a valuable lesson about tire gauges, see here). 

A week later I checked both. The compressor-filled tire remained at 40 psi. The one filled with CO2? Thirty six. Aha. Two weeks later the air-filled tire had dropped about a half pound, while the CO2 tire was a smidgen under 33. That’s about a 20-percent loss in three weeks.

My tentative conclusion is that CO2 does indeed suffuse through automotive tires, albeit at a substantially slower rate than it does from a thin butyl bicycle tube. This means that if you use a CO2 tank for tire repairs and airing up, it would be wise to check the pressure more often than you normally would. 

I say “tentative,” because due diligence requires that I repeat the experiment, but swap tires to confirm that my original CO2 tire does not have a slow leak. Since both tires were close in pressure when I began, I doubt this is the case, but I need to confirm it. So right now both tires are filled again, except with the opposite procedure. I’ll report back in a few weeks.