Atoms behave very differently at temperatures near absolute zero depending on the value of their intrinsic angular momentum or spin. Bosons have spins with integer values in units of the Planck constant divided by 2π, while fermions have spins of 1/2, 3/2, 5/2 and so on. A molecule made of two fermionic atoms will be a boson because it will have an integer value of spin.

All fermions obey the Pauli exclusion principle, which means that they cannot occupy the same quantum state. However, there are no such restrictions on bosons, so they can all collapse into the same quantum ground state. This process, known as Bose-Einstein condensation (BEC), is at the heart of superconductivity – the flow of electric current without resistance.

Since electrons are fermions they must form Cooper pairs – named after Leon Cooper of the Bardeen-Cooper-Schrieffer (BCS) theory of superconductivity – or strongly bound atom pairs before they can collapse into a Bose condensate. The fermionic condensates produced in these new experiments represent a novel phase – where there is yet no theory – that lies in the crossover between the BCS and BEC regimes.

John Thomas and colleagues at Duke University started by confining a gas of lithium-6 atoms in a magneto-optical trap and then reduced the temperature of the gas by using a technique known as evaporative cooling (J Kinast et al. 2004 Phys. Rev. Lett. 92 150402). When the gas reached 400 nanokelvin they made it vibrate by briefly switching off the optical trap, and then turning it on again.

The gas behaved like a hydrodynamic “jelly” and it continued to oscillate in and out – as one unit – for a long time. According to Thomas, this indicates collective behaviour rather than the independent behaviour shown by individual, non-interacting, atoms. More importantly, the jelly vibrated at precisely the frequency predicted by some theories for a fermionic superfluid.

The Duke team says its experiment provides the most direct ever evidence for superfluidity. Earlier this year, a group at Boulder in the US created a Bose-Einstein condensate from a strongly interacting Fermi gas and studied the pairs of fermions that formed. “That was a good experiment, but it doesn’t establish superfluidity,” said Thomas in a press release. “To have superfluidity you’ve got to observe something like hydrodynamics, like we observed.”

Thomas admits that the work could be criticised because it does not demonstrate a well-defined transition at the point where the gas becomes a superfluid. In contrast, the experiment performed by Rudolf Grimm and colleagues at the University of Innsbruck shows an abrupt change – in the collective excitation frequency – at this crossover point (M Bartenstein et al. 2004 arXiv.org/abs/cond-mat/0403716). The transition is also accompanied by oscillations that last a long time.

“Thomas’ team has reported on interesting experimental findings for which superfluidity could be the only explanation,” Grimm told PhysicsWeb. “However, as for both our experiments, it is difficult to know whether the hydrodynamic behaviour observed is due to simple collisions between atoms in a normal gas phase, or to superfluidity. This is a complicated many-body regime that needs further careful investigation,” he said.