Is there any fundamental--show stopper--reasons this idea could not work (in a Hard SciFi context respecting the laws of physics)?
No. It's a perfectly fine idea.
Which materials are suitable?
If you can use a superfluid, then a lot of engineering problems get a lot easier. The whole point of superfluids is that they flow with zero viscosity, so once you get it rotating in the pipes, it'll just keep going, with no need for magnetic confinement or replacement of frictional losses. On the other hand, you now need a really good refrigeration system, and there is the question of how you get it started moving in the first place. There are certainly ways to do it; it's just not obvious what the most practical engineering solution would be.
In an SF context, you could always make up some new exotic superfluid material to use for this purpose. If we're sticking with known, real materials, however, the superfluid option is going to require helium (probably helium 4, since it's much more common and transitions to the superfluid state at a higher temperature than He3). Superfluid helium is also a Bose-Einstein Condensate, so there's you top-two choices already taken care of. The fact that it is a BEC is, however, not really relevant; yes, its BEC nature is part of the explanation for why it becomes superfluid, but your engineers won't need to care about that. All that matters to them is that it is, in fact, superfluid. Bose-Einstein Condensates do not generically lend themselves to this application.
Using hot plasma is a bad idea. It's incredibly difficult to control, and you'll want something with a relatively high density to provide the maximum counter-force.
Copper would work, making use of its diamagnetic properties to "levitate", pushing against permanent-magnetic bearings in the outer wall. It's far from the best solid material you could use for this purpose, however. You wouldn't want to use just a continuous band of copper; it would need to be broken up to minimize induction of eddy currents, which will sap energy and momentum from the system. Rings of bismuth and/or pyrolytic carbon would be much more effective, as they are much stronger diamagnets. Unlike the superfluid option, which can just use regular ol' static pipes to circulate the fluid in, using solid rings like this would require some kind of active support; the main load-bearing elements could be simple permanent magnets, but you'll also need an electromagnet, or air cushion bearings, or some other control system with extremely fast response times to make sure they remain in their tracks. If that fails, your pressure vessel gets rather violently torn apart.
The upside of this sort of system is that, while there are some significant system control issues to solve, the basics of how to build and supply power to the system, and rev it up and down, are much more straightforward, and the materials are cheap.