New look for Bose condensates
Jan 7, 2002
Physicists have observed a quantum phase transition in a gas of atoms for the first time. Most phase transitions - such as the melting of ice to form liquid water - are a result of thermal fluctuations. Quantum phase transitions are different in that they are caused by fluctuations allowed by the Heisenberg uncertainty principle and can happen at or near absolute zero (M Greiner et al 2002 Nature 415 39).
Immanuel Bloch and colleagues at the Ludwig Maximilians University in Munich, the Max Planck Institute for Quantum Optics, also in Munich, and at ETH Zurich in Switzerland started by cooling a gas of rubidium atoms until they formed a Bose-Einstein condensate – a novel state of matter in which all the atoms collapse into the same quantum state. The condensate was stored in a magnetic trap and then illuminated by six laser beams. These beams formed a three-dimensional energy landscape known as an optical lattice. The peaks and troughs in the landscape form a perfect cubic lattice.
When the potential depth - the energy difference between the top of a peak and the bottom of a trough - was small, the rubidium atoms were able to move freely between the troughs. This is the normal superfluid state of a Bose condensate. However, when the potential depth was enlarged by increasing the laser intensity, the atoms were no longer able to move freely throughout the condensate: such a state is known as a Mott insulator.
In quantum terms, the transition occurs between a superfluid state in which all the atoms have the same quantum phase and the number of atoms in a trough can fluctuate, and an insulating state in which the phase varies but the number of atoms in a trough is fixed. The transition was reversible and the insulating state could be changed back to the superfluid state by reducing the laser intensity again.
Ultracold gases are much ‘cleaner’ and easier to control than the usual solid-state systems that are used to study quantum phase transitions. The techniques developed by the Munich-Zurich team could also have applications in quantum computation.
About the author
Peter Rodgers is Editor of Physics World