Technical Q&A with Lennard Zinn - Large molecules and short frames | VeloNews.com
"Alan Hills:
I’ve been told it was true by no less than (our local triathlon legend) Dave Scott. Wading through the web yields some insights on tire pressure loss from tires/tubes inflated with carbon dioxide (CO2) cartridges. Two polymers are used for bike tubes; latex rubber and butyl rubber (isobutylene rubber).
Butyl rubber dominates the market and is used for almost all tubeless tires and bike tubes as its permeability to air is incredibly low — butyl tubes have only 10 percent the leakage rates of natural latex rubber tubes.
Permeation by diffusion predicts gas leakage rates proportional to the inverse of the square root of their molecular weights. Using air as a reference the predicted leakage rates for common gases are: helium 2.7, air 1.0, nitrogen 1.02, oxygen 0.95, argon 0.85, carbon dioxide 0.81.
It turns out however that the leakage rate of CO2 is huge, and the reason is that it is actually soluble in butyl rubber and is thus not constrained to normal permeation loss, it can transfer straight through the bulk rubber resulting in severe tire pressure loss on the order of a single day. CO2 is not likely to be replaced by argon or other gases in refill cartridges, however, because CO2 is much more easily liquefied than other gases and can be contained in a moderate-pressure cartridge in a patch kit. An analogous cartridge holding N2 or argon (non-liquified gas) would be dangerous and would require a thick (and very heavy) steel-walled storage vessel. A reference dealing with CO2 transfer through latex rubber sheds light on the loss process."