Physicists in Germany have come up with a way to cool a mixture of two different atomic species of fermions to extremely cold temperatures, forming a “degenerate quantum gas”. To do this they invented a new method of cooling that involves adding a third species of atom — a boson — which was later removed. The team intend to use such mixtures to study exotic states of matter such as superconductivity and superfluidity.

Over the past decade physicists have learnt much about the quantum nature of liquids and solids by studying dilute atomic gases at very low temperatures. Early experiments were done with bosons — atoms with integer spin such as helium-4 — which collapse into a single quantum state called a Bose–Einstein condensate (BEC). Then, physicists turned their attention to fermions — atoms with non-integer spin such as helium-3 — which are forbidden by the Pauli exclusion principle from directly forming a BEC. Instead, interactions between fermions can be tuned using lasers and magnetic fields so that they pair-up to form bosons — which then form strange states of matter such as Fermi condensates and superfluids.

Pairing interactions

Physicists are interested in pairing interactions because they also occur in superconducting materials, where the electrons — which are also fermions — form pairs that can flow without resistance. Researchers are particularly keen to look at ultracold gases that contain more than one type of fermion because the interactions in such systems could be made to resemble those thought to be responsible for some poorly-understood high-temperature superconductors.

The first success in creating such Fermi–Fermi mixtures came in 2005, when Wolfgang Ketterle and colleagues at the Massachusetts Institute of Technology were able to create ultracold mixtures of the same atoms in two different spin states — effectively two different fermions.

Now, the study of Fermi–Fermi mixtures has been taken one step further by Matthias Taglieber and colleagues at the Ludwig-Maximilians University in Munich — who have developed a new technique for cooling a mixture of two different fermions, lithium-6 and potassium-40 atoms (Phys Rev Lett 100 010401 ).

Colliding bosons

The main challenge facing the team was how to cool a mixture of fermions to extremely low temperatures. Evaporative cooling, which works well on bosons, does not work on fermions because it relies on particle collisions — and the Pauli exclusion principle prevents identical fermions from getting close enough to collide.

This problem had already been solved for one fermionic gas by mixing it with a bosonic gas such as rubidium-87. Collisions between bosons and fermions cooled the mixture in a process called sympathetic cooling.

Catalytic cooling

While potassium-40 can be cooled efficiently by rubidium-87, the process is much less efficient for lithium-6 because the smaller lithium atoms are much less likely to collide with rubidium and lose kinetic energy. However, when Taglieber and colleagues tried to cool the mixture of three atomic gases they discovered that the lithium atoms were cooled by colliding with potassium atoms — which are less than half the mass of rubidium — in a process they have dubbed “catalytic cooling”.

Once the triple mixture has been cooled, the rubidium can be removed from the trap, leaving mixture lithium and potassium atoms at temperatures below one microKelvin.

While Taglieber and colleagues are the first to create a Fermi-Fermi degenerate quantum gas, two independent groups in Austria and the Netherlands have also made some progress in this direction. The race is now on to use a Fermi–Fermi system to gain insight into fundamental quantum interactions. Ketterle, who won a Nobel Prize for his work with ultracold atoms, described Taglieber’s achievement as an “important step towards studying heteronuclear pairs and possible forms of superfluidity”.

While he would not say exactly what the Munich group are now up to, Taglieber believes that ultracold Fermi–Fermi mixtures could be used to study a number of quantum systems including “heteronuclear” molecules comprising one lithium atom and one potassium atom. Such molecules would interact with each other via a dipole interaction similar to that responsible for some forms of magnetism.

Another possibility, according to Taglieber is the study how fermions pair-up in a superconductor and some superfluids. This pairing is believed to be mediated by other particles — phonons in the case of superconductivity. It could be possible to “tune” the interactions between atoms such that the lighter lithium atoms mediate the pairing interaction between the heavier atoms. Another possibility related to superconductivity is the study of what happens when the numbers of fermions are unbalanced such that not every particle can find a partner to pair up with — something that has already been done with identical atoms that are in different spin states.