APS News

Physicists Achieve Molecular BEC, Coexistent ''Fermi Sea''

A molecular Bose-Einstein condensate (BEC) has been achieved by Carl Wieman and his colleagues at the University of Colorado. Wieman reported at the APS March Meeting in Indianapolis that his team had observed a quantum superposition of diatomic molecules and disassociated atoms in a trap.

Wolfgang Ketterle of MIT, like Wieman a recipient of the 2001 Nobel Prize in physics for BEC discoveries, spoke at the same session, and reported on some of his latest findings.

Having long used Rb-87 in his BEC experiments, Wieman has as of late been studying Rb-85 which, although it is harder to condense, possesses just the right fine-grained set of quantum energy levels (hyperfine levels) so that the application of a magnetic field can alter the interaction force among the atoms in the trap, even as they reside in the single quantum state which is the hallmark of Bose Einstein condensates. By adjusting the magnetic field to be very close to the point where the interatomic force goes from attractive to repulsive, a "Feshbach resonance" occurs and some of the atoms form molecules.

The atoms and molecules are thought to be coherent, at least locally, and maybe over longer distances too. In this process the condensate appears first to implode and then rebound somewhat like a supernova, even to the extent of sending out jets of particles and leaving behind a remnant. The physics behind this "Bosenova" behavior is still a mystery.

Ketterle reported findings in three areas. First, he has used a sodium-23 BEC to help cool a gas of lithium-6, a fermionic atom. The Pauli-exclusion principle forbids such atoms from falling into the single state available to bosonic atoms such as Na-23, but the Li-6 atoms can, if cooled low enough, occupy all the lowest energy quantum states possible.

This has now been done in the MIT experiment, the first time such a "degenerate Fermi sea" has co-existed with a large BEC. One wants to see how such a fermi gas behaves at nK temperatures and whether the atoms can be coaxed (by manipulating the interaction between them) into forming Cooper pairs, becoming thereby a superfluid.

Ketterle also reported the propagation of a condensate in a magnetic waveguide. First, his group made a large (2 million atoms) BEC, then loaded it into a magnetic trap, and finally loaded it into a microtrap on a printed circuit board.

The micro-journey around the chip was partly smooth and partly bumpy, especially when the cigar-shaped BEC came toward a Y divide. At the divide the condensate wiggled itself into a snake shape. Close to the chip surface, the condensate broke up into several detached segments. Future atom chips will need better control of surface roughness.

Another goal is the generation of pair correlated atoms. Ironically, the atoms in a condensate all share a single quantum state but are not otherwise entangled.

The MIT researchers have created two BEC blobs (let us call them 1 and 2) together with another small "seed" condensate (blob 3). The elastic collision of these blobs produced a fourth blob in a process called four-wave mixing.

In effect, the atoms in blobs 1 and 2 help to amplify blob 3 (a gain of 20, in this case). For each atom added to blob 3, one atom is put into blob 4. This creates two pair-correlated atomic beams.

In some future experiment this pair correlation might be verified directly if one could detect single atoms in the two condensates, which are moving off in opposite directions.

Right now it is difficult to spot single neutral atoms in a BEC. Single-atom detection is likely in helium BECs since the atoms, deliberately put into an excited state in order to confine and cool them in the first place, are easily ionized, making it far easier to detect them.

Chris Westbrook, a member of Alain Aspect's team at Orsay, summarized recent helium work and described a scheme for producing helium molecules within a BEC.

This, he said, might allow an atom-wave equivalent to the current process of down-conversion, by which UV photons can be converted, in a special crystal, into a pair of lower-energy but entangled photons; if one photon has a horizontal polarization, the other must have a vertical polarization. A beam of related atoms could, analogously, be sundered into beams of pair-correlated atoms.

APS encourages the redistribution of the materials included in this newspaper provided that attribution to the source is noted and the materials are not truncated or changed.

Editor: Alan Chodos
Associate Editor: Jennifer Ouellette