I had a discussion several weeks ago with my dad about alcohol evaporation. The conversation started because I had picked up a bottle of WTKS from 2004 that looked like it had suffered from some evaporation compared to a bottle from 07 that I also purchased (although I noticed that the glass bottom on the 07 was much thicker and higher so evaporation was most likely not the case). He said that it would have a lower proof, because alcohol evaporates first.
It got me thinking about the proof changes that occur in barrels, and I brought up that things like George T. Stagg come out at a higher barrel proof than what they were initially bottled at, and we didn't have any good answers, just some theories, as to why the proof would go up, instead of down, during the aging process. As far as whiskey in a bottle, we pretty much decided that the proof would always have to go down if there was evaporation.
So why does the barrel proof of a bourbon sometimes climb upwards? I remember reading that it has a lot to do with temperature and location in the rickhouse (high or low). I didn't, however, find a solid answer as to why the evaporation rates would be effected. Here are some of our theories (and I'm not saying they're all good ones!):
1) Barrels at the top of a rickhouse could have a higher proof because the atmosphere is supersaturated with alcohol molecules and they are as likely to enter as to leave (Letting more water escape and less likely to be replenished). The obvious problem with this is that as I understand it, most rickhouses are fairly well circulated.
2)At a given temperature, alcohol could evaporate at a different (slower) rate than water and the proof would increase. We decided that this was highly unlikely if not possible after some discussion.
3)Oak's ability to wick water vs. alcohol. Oak trees grow up with wicking water in mind, but not alcohol. Therefore, water might be more likely to be absorbed and then allowed to escape a barrel whereas the alcohol (which isn't as readily absorbed) stays within the confines of the barrel.
Maybe all of our theories are wrong- a couple of them seem highly unlikely already (the water wicking properties of oak seem most likely to me compared to the other two anyway). Hopefully somebody out there knows an actual reason for why this happens?