Note: since the appearance of the slashdot story, lots of people are commenting and making suggestions. Thanks! I'll reply to them as a group in a day or so.
As I've mention in several prior postings in the blog, the big experiment we wanted to do during the flight was seeing how the Diet Coke & Mentos reaction works in microgravity. If you haven't read those posts, please do so to get an idea of how the experimental apparatus and goals evolved over several weeks. You may also want to check out my Youtube videos for other high-speed videos I've done, including one that shows the reaction at 1200fps on the ground.
Before I go any further, I want to thank everyone at ZeroG and Space Adventures that went the extra mile to make this experiment possible, in particular the science liason, Michelle Peters. And I'd also like to thank the TSA screeners, who arrived on site already totally up to speed on what we wanted to do (they'd even seen my test videos). Whatever you may think about the rules that the TSA enforces (and I agree with Bruce Schneier in that regard), the fact of the matter is that the frontline staff that you deal with have little or no freedom to apply common-sense discretion, and are often placed in situations where they don't have the time, or the background knowledge, to make an informed decision, which means that the default answer is "no". When you couple that with the fact that anyone can be having a horrible day, and some small percentage of people are jerks to begin with (a smaller percentage than most people assume), and multiply by hundreds of thousands of people going through security a day, it's a recipe for horror stories.
But in our case, since the screeners had been pre-briefed, it was easy to demonstrate that everything we wanted to use was well within the TSA rules. The only thing that didn't fly was a tiny ball of modelling clay that we were going to use to mount the mento onto a ziptie with, and the screeners helped brainstorm an acceptable (and better!) mounting method.
As described in a previous post, the hypothesis that we wanted to test was that convection of the soda was an important part of the whole reaction; under normal gravity, bubbles formed around the mentos rise up through the soda, allowing more soda to come in contact with the candy, and thus more bubbles form. However, in microgravity, there's no "up", so any bubbles that form will just stay near the mento, and will in fact keep new cola from reaching it.
So lets get right to the video, and then discuss what we learned.
I'll begin by discussing what went wrong.
Our first problem was that we weren't quite in free-fall, because in order to do the experiment, we had to be strapped into seats next to the emergency exit window (to get the light we needed). If you think about the geometry of the parabolic arcs from the perspective of the actual plane, the true zero-g path forms a parabolic plane, and (I'll try and state this as plainly as possible) the pilot attempts to fly a path that causes the (physical) plane to follow the (geometric) plane.
I say "attempts", because it's humanly impossible for a pilot to do this perfectly. So anything attached to the plane (as we were) is going to experience some small amount of residual gravity, not to mention the effects of any air turbulence.
Even if the pilot does this perfectly, only those parts of the (physical) plane that intersect the (geometric) plane will be in true free fall. Anything above or below that perfect plane (in the perfectly-flown plane!) will experience a slight amount of residual acceleration, because it's not free to orbit the earth in a free fall path, but is rather being dragged along either slightly faster or slower than it really wants to travel. In practice, of course, this tidal effect is tiny compared to the errors caused by the pilot as he or she attempts to fly the true parabolic path.
So because the experiment was attached to us, and we were strapped into the seats, and the seats were attached to the plane, the entire apparatus was subject to some residual accelerations. In order to eliminate them, the Mark II Menticulation apparatus will have to be free-floating, self-contained, and will have to operate without being touched. This will permit it to freely fall in a zero-g parabola independent of the path of the body of the plane (although there will be some residual effects due to things like air currents in the cabin, of course).
As a side-point, it's interesting to note that on these flights, from the standpoint of the passengers, the pilot doesn't need to fly an absolutely perfect parabola; as long as he or she can fly a path that is, on average, close to the perfect path, nobody inside the plane will really notice the difference.
The second problem we had was that we wanted to create a reasonably sized ball of soda, and then put the mento into it. However, the nozzle we used to control the soda flow was too small, and we couldn't control the exit velocity properly (and so we got all the squirts). On the second parabola (not in the video), we tried removing the nozzle, but we couldn't get a bubble of soda to detach from the bottle, and everything got messy - a blob of soda impacted right on the camera window, obscuring a lot of the view.
In order to fix this, we're going to have to come up with a method that deploys a preformed, properly positioned bubble of soda.
Finally, the video was slightly out of focus. We had to manually set the focus on the camera before doing the experiment, and it was difficult to keep the coke and mento at the correct distance, considering all the other things we were having to do.
Even with all these problems, I think we got some interesting video and results. It's not definitive, but it looks like the reaction is slowed down by the lack of gravity, and we also noticed something we weren't looking for; the cola apparently clings to the mento (likely, it's hydrophilic), resulting in the interesting dynamics when I moved the candy. A followup experiment might be to compare the behavior of some water surrounding a mento vs. a ball bearing.
The Mark II Menticulator - Initial Design
I've spent a few hours thinking about an improved experimental apparatus, and I think I've come up with something that, with a little refinement, will do the trick.
The experiment will be housed in a transparent case constructed of acrylic or lexan panels locked together by t-slotted structural framing. Inside will be a rubber balloon of soda trapped between two spoonlike holders, one above, and one below. A hobby r/c servo will be mounted inside the box, with two arms extending from it, such that if it rotates in one direction, a sharp tip will puncture the balloon, hopefully releasing the soda which will remain between the spoons; then when it rotates in the other direction, a mento will be moved to the center of the bubble of soda.
The servo will be directly controlled (no radios onboard!) by a small hobby microprocessor such as an Arduino. The microprocessor could also have some sensors to collect related telemetry (accelerometers and so forth).
Finally, several strips of high-intensity LEDs will provide the needed light for the Exilim EX-F1 camera.