this isn't quite as silly as it may sound, though not exactly in the application you are thinking of specifically. i explored with this very idea extensively at one point, looking for alternate ways to combat condensation at polar temperatures (see -30C or so) when comuter cooling. if we understand that condensation ONLY occurs when warm air hits a cold surface (that happens to be below the dew point - whatever it happens to be at that moment depending on humidity) because the air has water already in it, two alternate ideas were put forth. one idea was to seal the computer in an air-tight enclosure (such as a freezer or whatnot) and displace the air inside it with an inert gas like argon which would (in theory) have no water in it. so by replacing the air with a substance that didn't hold as much water or which the dew point was a different thing all together due to the properties of argon, the end result would be, in effect, a complete lack of condensation and thus, ice. the idea was eventually rejected out of hand due to it being more complex and fiscally cumbersome to test and impliment then i was willing to commit myself to, though i believe my data to be accurate and the idea not without merit - albeit perhaps in another arena.
the other idea put forth was to build a dessicant scrubber using fans to force the air inside through <a href=http://www.drierite.com/>Drierite</a> or Silica Gel filters and thus remove the water from the air and drop the dewpoint to around -100F. so as long as we didn't drop the cooling below -100F (which would have been below the dew point) there wouldn't be any condensation. again the idea was to seal the computer in an airtight container (like a freezer for example), and then turn the scrubbers on for an hour or so to pre-dry the air, and then turn the freezer on, and then turn the computer on.
the idea works wonderfully, and is vastly easier to impliment than gas emersion. though i no longer cool my PC in this fashion (partly because i've become a Macintosh/Unix convert partly because the price of processors have absolutely plummeted making this kind of cooling for me too much of a hassle, and partly because i don't believe the solution to be very elegant or user friendly even though functional), i still keep a dessicant scrubber in both my fridge and freezer at home where i keep my food and so forth and have an ice free/moisture free refrigerator/freezer in my kitchen.
but back to your thread, at no point was gas considered as a cooling mechanism - i looked at is as strictly as a method for combating condensation.
anyhow, theres my 2 cents
peaz