"However, we would expect the centrifuge to gain mass (and hence generate an increasing gravitational field) as fluid spins up to relativistic speeds. Hello warp drive! Oh, and you’re welcome."
Congrats, E=mc², you've collected a large amount of energy in a small area and created a gravitational field. You've now reached the space travel and anti-gravity technology of a small rock.
My understanding of a warp drive (Alcubierre drive) is that it contracts space in front of the ship while expanding space behind it. I don't understand how increasing a gravitational field locally is meant to achieve any of that.
We can do something like that already with current flowing in a circle in a superconductor. Resistance corresponds to friction. In the worst case you get a quench and all of the stored angular momentum results in a torque that breaks your machine.
Until, of course, it gets nowhere near the speed of light because the limiting factor to spinning something fast isn't friction, it's the required centripetal force to keep it in one piece.
And assuming you could solve that by making the spinning thing out of unobtanium, you're still not doing anything to break conservation of momentum.
Put a plate in the center of a turntable and pour some water into it. Now rotate the turntable, slowly accelerating. What happens to the water?
ah ah ah ah nope.
It doesn't sound right to me either, but I have a hard time explaining to myself why it wouldn't maintain it's velocity.
I searched for "rotating superfluid helium" and found http://physicsworld.com/cws/article/news/2018/feb/07/unexpec... , which is a summary of https://journals.aps.org/prb/abstract/10.1103/PhysRevB.97.01... . It says that there is friction, perhaps caused by "quasiparticles that become trapped within the cores of vortexes. As the vortexes accelerate, the quasiparticles gain energy, which they can then dissipate to their surroundings in the form of friction."
