Updated: Tuesday, September 2, 7am ET
First of all, I'd like to thank everyone who commented on the video (the original posting is just below this one) and made suggestions for how to improve the experiment. I will try to address as many of the issues you raised in this posting, and amend it as I get more suggestions. I'll also be posting about some tests I've already done and how they've evolved the design in a day or so.
James raises some objections about using balloons as deployment devices and suggests some sort of hard container (as do several other people). Tests I've done indicate there are indeed multiple failure modes with balloons that make them problematic as deployment devices. The big problem with hard containers is getting the fluid out of the container in microgravity; you can't pour stuff, and surface tension and hydrophilic/phobic effects are going to dominate, so getting the liquid to go where you want it is going to be a complex issue. Given that you only get one or two shots during a flight to get it right, and you're under significant time pressure, I want to keep things as simple as possible.
Why don't I just pop a mento into a cola bottle? If you look back through the older posts, you can see that this was my original idea. During the last week before the flight, I simplified the experiment after realizing that if convection was an important factor, nothing much would happen, and it would be hard to see what was going on inside the bottle. In retrospect, doing it this way would have demonstrated a difference between the two environments better than what I eventually did, but it wouldn't have shown what was going on.
Using a baggie of cola: this might actually work, but it might make it hard to see what is going on. A variant that uses a flat "slab" of cola in a specially constructed container might work (Ger later suggests a similar idea), but there are several problems that would have to be worked out; getting the cola into the container from the bottle without loosing too much carbonation, and isolating/releasing/moving the mento in zero G. If those could be dealt with, then it might be a very interesting experiment, in particular if some glitter was added to the cola so the convective flow is more clear. It might work better with a clear diet soda, like Diet Sprite, or perhaps Diet Club Soda. This approach had not occurred to me, and the more I think about it, the more I am coming to like it.
This may well be the way to conclusively demonstrate the difference in convection, because it could also be performed with the same apparatus on the ground. However, it won't look as visually cool as a doing something with a sphere of cola.
Precision measurements: I don't know this if will be feasible given the constraints we have to work under, and our uncertain ability to precisely control the environment. For this reason, we need something that demonstrates a really clear difference between 1G and 0G.
Cperdue suggests an ingenious deployment method (and Nick later provides a variant). The main problem with it is that it would have to be scaled down; the biggest cola bottle we're going to have onboard is 10 or 12oz. Also, the TSA might get a little bit concerned about allowing long pointed sticks aboard. Finally, there's the issue of clearly seeing what's going on.
Matt suggests chemically popping the balloon. If popping the balloon were the only issue, this might be interesting to investigate. There's also the issue of whether Space Adventures will allow a chemical that dissolves rubber onboard.
Rustin suggests I use "aircraft" instead of "plane" in my description of some of the physics of Zero G. Yeah, I could, but plainly I was trying for an extended play on words. I did use typography to differentiate the two usages.
An anonymous coward takes issue with my use of "weightless" and "oG". I didn't use the former term at all, and my opinion of his other quibbling is identical to Winston Churchill's opinion of objections to prepositions at the end of sentences: it is arrant pedantry up with which I will not put. Learn something about frames of reference, go find a lagrange point relative to the entire universe, then come back and argue with me.
Using IR controls: I have come to agree with the very next poster, who contends that servos are overkill. I am going to take my own advice and keep everything as simple as possible. Less failure modes that way.
"As long as your ball of soda is floating without hitting anything..." This is easier said than done. Given that most of the passengers on a flight are Zero G virgins, the inside of the main cabin resembles a 3D peopleball machine in multiball frenzy mode. When someone's paying $5K+ for a ticket to ride (apparently the price just went up), they probably will be a wee bit upset if they get hit in the face by an errant cokeroid. This is why we did our experiment in the seat section, which is empty during the parabolic segments. That area has significantly less airspace (so to speak), and any incompletely collected soda is going to splatter somewhere. This is why all the experiments have to be done in a sealed environment.
The Zero G coaches do release some small bottled-water balls in the main cabin, but these don't cause a lot of mess, and getting rained on is considered part of the fun.
The complex idea suggested by the anonymous non-coward about using a spherical cage that squirts cola into its center would probably not work well. The cola would have to be precisely synchronized, and its momentum has to go somewhere, so I think the mixing would be turbulent enough for drops to be ejected. However, I like the general structure of the device, and will keep it in mind the next time I build a device to punch a hole in the space-time continuum.
Richard mentions that the Mythbusters found that removing the coke from the bottle affects the reaction. My thinking is that the shape of the bottle helps with the convective flow, which was what we were investigating, with limited success, in our original test. Also, pouring the coke out of the bottle obviously causes some of the CO2 to be released, so a device that does not require the soda to be poured into an intermediate container would be preferable.
Valisk kvetches about my coining of the word menticulation. I believe that it is an entirely cromulent word, and that its widespread use will embiggen the English language.
Coating the mento with something the coke dissolves: the problem with this is timing. Keep in mind that you're only in 0g half of the time during the parabolic segment, and for 20-25 seconds at a time.
Alka-seltzer tablet in water: Toby suggested this, and he's not the first; one of the coaches on the flight told me he's always wanted to do this one. However, the situations are not equivalent, because the CO2 is generated by a chemical reaction in alka-seltzer, as opposed to a physical one in DC&M. So as long as water can reach the tablet, the reaction can continue.
The video Toby linked to is just too cool not to embed, so enjoy!