Several years ago, now, a group at Penn State announced a weird finding in helium at extremely low temperatures and high pressures (which is what you need to make helium solidify): when they made a pendulum out of a cylindrical container with a thin shell of solid He toward the outside edge, twisting about its axis, they saw a small but dramatic change in the oscillation frequency as they cooled the system below a particular temperature. They interpreted this as a "supersolid" phase of helium, with a quantum phase transition taking place that caused the "supersolid" to stop rotating with the cylinder, instead moving without resistance, as a superfluid does when you try to spin it.
This was a dramatic claim, and has not been without controversy. Experiments since then have been ambiguous, with some people claiming to see the same sort of transition, while others don't see anything. The counter-theory is, more or less, that the solid helium doesn't exist as a perfect crystal, but instead has cracks and defects in it, and the observed change in rotation frequency is due to residual (super?)fluid slipping through these gaps, not a solid effect.
Two recent papers might mark steps toward clearing this up.
One, written up in Physics and published in Physical Review Letters, uses nuclear magnetic resonance of a small amount of helium-3 introduced into the helium-4 to look for effects that might be related to the "supersolid" phenomenon. Their results mostly agree with the theory that says there's no supersolid, but there are a couple of features at the temperature where the transition takes place that they can't explain, so maybe. The other, published in Science uses the same oscillating-pendulum technique as the original experiments, but doesn't see the same sort of transition, and claims they can explain their results in terms of other effects.
So, we have one paper that says "almost certainly not," and another saying "probably not, but there's a weird thing happening at that temperature." Which means we can expect this to kick around for another couple of years, at least.
I'd do a more complete write-up of this for ResearchBlogging, but really, the above is about my level of understanding of this phenomenon, and that doesn't seem to justify the extra citation. People with actual knowledge of the field should feel free to explain further in comments, though.
(A few years back a colleague was pushing what he thought was the grand unified theory of the whole thing, claiming to be able to explain everything. I think this arxiv paper may be it, but I thought I had talked to him about it more recently than 2004, so maybe not...)
I think you have a typo here:
"One, written up in Physics and published in Physical Review Letterspublished in Science"
... they made a pendulum out of a cylindrical container with a thin shell of solid He toward the outside edge, twisting about its axis...
Is this part of a suite of standard tests in cryophysics, did they have a specific reason to play this particular game, or had they already tried everything else (playing rock albums backward, exposing it to gamma rays under the full moon, injecting it into soldiers about to deploy to the Middle East, etc)?
did they have a specific reason to play this particular game @2.
It is a perfect example of the creativity one sometimes sees in experimental physics. Look at the process that led to the first YBCO superconductor. No theory told them how long to bake it.
The torsional oscillator experiments done by Kim and Chan at penn state were replicated by several other groups and I am pretty sure that the "unfrozen superfluid" idea had been eliminated. The Non Classical Rotational Inertia is a real phenomena, but the explanation has alluded others. There have been experiments done with direct pressure applied with no flow, as well as looking for supersolid behavior through ultrasonic actuation, both not seeing any evidence of "supersolid". Interesting results came out of the University of Alberta a few years ago (I did my undergraduate thesis in this lab, but did not work on this particular experiment) where they found that the shear modulus of a solid helium crystal increased by about 10% with the same temperature dependence as the NCRI from the torsional oscillator experiments. Any explanations I have heard for any of these results are based on crystal defects, but I don't think anything is settled on the theory side, just that the phenomena is real and that explanations to do with some remaining liquid have been eliminated.
@2: making thigs rotate is pretty standard in cryophysics these days. It started in the seventies. I'm not sure who had the first idea, but anyway, it turned out that rotating cryofluids are a good model for rotating neutron stars. Sudden jumps in pulsar pulses can be explained as joining vortices.
But of course you have to remember that while He-3 and He-4 are chemically similar, physically they are very different beasts. As in fermions vs. bosons.
I gather they did the rotation so they could tell when their material changed phase. If you rotate a can of water back and forth, it sort of sloshes inside. Not all of it changes direction at once. If you freeze the water, it rotates as a unit so you can track the freezing process and tell when the water is frozen by measuring the forces as the can changes rotational direction. There's obviously a lot more to it than that, and even more weirdness when you consider the weird quantum effects of liquid helium, but that's probably what the torsional pendulum is about.
Ideally, they'd just be able to watch the liquid solidify, but at the temperatures they are using, the light necessary for looking would melt any solid.
Of course, what we are actually seeing is a game of kick the can as the can full of liquid, or perhaps solid, helium is kicked down the road in hopes that future research will reveal the phase of the helium inside.