A colleague and I were discussing the following fluid mechanics experiment, which most people have probably tried in their kitchen: take a bottle full of water (a 2-litre transparent soda pop bottle works well), invert it, and swirl it around to impart some angular momentum to the fluid. It will quickly form a vortex, which allows the bottle to drain rapidly.
The question that neither of us knows the answer to is this: is there some mechanism by which additional angular momentum is transferred into the system from outside as the water drains, or is it just that the initial angular momentum (which you add by swirling) becomes more concentrated?
Probably equivalently, if I were to cut the bottom off the bottle and then do the same experiment while continually adding water from the tap, would the vortex continue to spin indefinitely, or would the angular momentum in the system eventually be depleted, killing the vortex?*
If angular momentum is transferred to the system from outside the bottle, what is the mechanism for this? I assume that the Coriolis force is not relevant on such a small scale (the fact that a bottle vortex can be formed with either chirality would seem to back this up) so I would imagine that any transfer of angular momentum would be due to the fluid exerting a torque on the bottle it's in. Obviously frictional forces can only exert a torque in the direction that would slow the rotation down, but perhaps there are pressure gradient effects that can exert one in the opposite direction.
* this is obviously a rather easy experiment, and my only excuse for not doing it is that I don't have a pair of scissors handy.