I have always liked table-top experiments. In my opinion, those who can come up with those ideas really, and I mean really, understand what they are doing. Plus, they usually come up with nice videos (it's funny, now that I think about it, it is mostly fluid dynamics people that get nice videos. Here's one for example).
In the newest issue of Nature, Howard Stone from Harvard has a reeeeeeeeeally cool result. First, let me show you the experimental setup. The tank consists of water at the bottom and then oil on top. The oil (a poor electrical conductor) and the water are connected to electrodes of opposite sign that provide a high voltage. Once everything is turned on, they manually put a drop of water at the top using a pipette.
Initially, the water droplet is attracted towards the upper electrode by dielectrophoretic forces but when the drop actually comes in contact with the top electrode it acquires a positive charge and then it is repelled towards the bottom (the drop is very small and gravity doesn't really play a role here). You would think that when the droplet moves low enough to touch the water reservoir at the bottom it will just merge and become part of the reservoir. It should do that normally, but in this case it should be even better because the drop is positively charged and the water at the bottom is negatively charged and, as we all know, opposite sign charges attract. What's the big deal? Sounds easy enough,right? Well not quite, what Stone and friends found was that the behavior of the droplet actually depends on the voltage applied between electrodes.
When the voltage is low enough, what I quickly described above will happen. But when the voltage is high enough the result is different. Initially, the positively charged droplet moves toward the negatively charged water at the bottom but when they come in contact (just a tiny contact as you can see here) the water in the reservoir transfer negative charge to the droplet and now they will repel. This means the droplet moves up, towards the positively charged electrode that now is creating an electrical attraction between the two. Eventually the drop will touch the upper electrode and again a charge transfer happens, charging the drop positively and therefore the whole cycle repeats. You can watch a video of this "bouncing" here.
Pretty cool, uh?