Supercold, wiggling 'jelly' presents evidence of new kind of superfluidity

If fermion atoms can indeed form a superfluid state, as the "friendlier" class of boson atoms are already known to, this would provide scientists with new insights for studying such phenomena as very high temperature superconductivity, said John Thomas, the physics professor who heads the Duke team. Because electrons are also fermions, neutral lithium atoms in a superfluid state could serve as a stand in for a high temperature version of a superconducting system where electricity flows without resistance.

"The really long-term interest would be applications to superconductivity," said Michael Gehm, a Duke postdoctoral researcher who along with Thomas and others wrote a new paper announcing these observations. "The near term interest is to try to understand the physics of what is going on in these systems," Gehm said.

As a result of their special electronic properties, fermions such as lithium-6 cannot interact as closely as the bosonic class of atoms. That has made it more difficult to induce superfluidity in fermions than in socially gregarious bosons, which can co-mingle as closely as atoms possibly can.

But in their paper posted April 13 in the online issue of the journal Physical Review Letters, the Duke team announced promising observations of certain "hydrodynamic" traits of fluids under pressure as the scientists manipulated the temperature of a supercooled "Fermi gas" of fermionic atoms in a tiny trap formed using a single laser beam.

The cigar-shaped gas congealed into a jelly-like state that oscillated as the researchers cooled it significantly below 50 billionths of a degree above absolute zero, the team wrote. Absolute zero, about 273 degrees Centigrade below the freezing point of water, is the temperature at which theoretically all atomic motion stops.

This quivering of this jelly did not degrade in response to these temperature manipulations as an ordinary gas would, said Thomas, who is Duke's Fritz London Distinguished Professor of Physics. Instead it became what Thomas called "a more perfect jelly" that continued oscillating, and did so at a rate that some theoreticians have recently predicted would be characteristic of a fermionic superfluid.

"These observations provide the first evidence for superfluid hydrodynamics in a resonantly interacting Fermi gas," the Duke physicists wrote in their paper, the first author of which was Joseph Kinast, one of Thomas's graduate students. "It is…difficult to see how the observations can be explained without invoking superfluidity," the paper added.

Thomas, however, declined to categorically claim his group has observed fermionic superconductivity, saying theories that would explain all their experimental observations are still incomplete. "We have no definitive way of being absolutely sure until everything is corroborated by theory," he said. "Until you get me a theory that predicts what I observe completely and says 'this is a superfluid' I'm not going to claim it is completely impossible that there's another explanation.

Besides Kinast, Thomas and Gehm, other authors of the Physical Review Letters paper include graduate student Staci Hemmer and postdoctoral researcher Andrey Turlapov. The research was supported by the U.S. Department of Energy, the Army Research Office, the National Science Foundation and NASA.

The Duke group confined a few hundred thousand atoms of gaseous lithium-6 atoms in an "optical bowl" formed by a laser beam, cooling down the atoms to temperatures so low that their normal atomic motion practically ceased.

Under those extreme conditions fermion atoms become what is called a "degenerate" gas, swelling in size and approaching as close to each other as the rules of Nature permit. Fermionic atoms are more stand-offish than bosonic atoms. While members of the boson class of atoms can occupy the same energy states, fermions cannot.

This antisocial trait means that in a degenerate state fermionic atoms are too cold and closely confined to collide. "They're basically told 'the place you want to go is occupied; therefore this collision cannot happen,'" Thomas said. It was in this "collision-less" regime that his group found signs of superfluidity.

While degenerate fermion gases cannot collide, theory says they can be magnetically adjusted to interact by forming molecule-like pairs of atoms. Because these "atomic pairs" act like bosons instead of fermions, they are permitted to vibrate together in a coordinated fashion that Thomas's group observed, he said.

In January, 2004 scientists at a joint laboratory of the National Institute of Standards and Technology and the University of Colorado at Boulder announced the first observations of a "fermionic condensate" composed of pairs of fermions.

"That was a good experiment," Thomas said, "but it doesn't establish superfluidity. To have superfluidity you've got to observe something like hydrodynamics, like what we observed."

Thomas, Hemmer and Gehm were also among the authors of an earlier Nov. 7, 2002, paper in the journal Science's online Science Express that found less-obvious hints of superconductivity. In that research the authors also created a "degenerate" fermion gas under similar conditions. But instead of confining and manipulating the gas they released it.

On release, that gas swelled in free space in a startlingly lopsided way, expanding in one direction but not in the other. While theoreticians have predicted such an "anisotropic expansion" as another possible sign of superfluidity, Thomas and his colleagues realized there was a second possible explanation. There was a chance that the gas could have exhibited that behavior if it lost its collision-less properties on release from confinement.

The new set of experiments was designed to rule out the possibility of the gas's becoming "collisional" by keeping it confined in its trap while slightly adjusting its temperature, Thomas said.

Available theory said that if the gas remained truly collision-less at the temperatures the Duke group observed, the jiggling of jelly it formed would not degrade. If it continued oscillating at a certain characteristic rate, theory postulated that the gas was in the superfluid state. The oscillation rate the Duke group observed is "exactly what it's supposed to be," he said.

Thomas said some critics might challenge the Duke group's interpretations because they did not observe "some really abrupt transition" connected with a switch to the superfluid state. However some theoreticians have proposed this system might be "a gapless superfluid," he said. "That doesn't have to have such an abrupt change."

Superfluidity, which is the flow of a fluid without resistance, has similarities to superconductivity, the flow of electric current without resistance. Both involve the special interaction of pairs of particles, Thomas added. "But in a superconductor those particles are electrons that carry charge; in a system with neutral atoms there is no charge."

Because this superfluid forms at a relatively high temperature for such gaseous atoms, it would behave like an extremely high temperature superconductor, Thomas said. The Duke group's system would thus allow scientists to model what would happen if metals like copper could act like high temperature superconductors with transition temperatures above the melting point.

And because such systems are also "strongly interacting" – meaning their constituent particles affect each other at much greater than normal distances – they could also be used to model matter under especially extreme conditions, Thomas added. Examples are processes within neutron stars, and the natures of quark-gluon plasmas postulated to have formed microseconds after the universe began in a colossal "Big Bang."