Atom laser shows potential
for precision measurements

Institut für Experimentalphysik,
University of Innsbruck, and
Austrian Academy of Sciences,
Innsbruck, Austria

Control over interactions opens the way to precise measurements with atom lasers

The invention of the laser revolutionized physics in many ways. For example, lasers have been used to measure the distance to the moon with 4 cm precision and are a key tool in the elusive search for gravitational waves. The matter-wave analogue to the laser, the Bose-Einstein Condensate (BEC) could also be used for very precise measurements. However, there is one important difference between the photons in a laser beam and the atoms in a BEC; atoms interact, photons don't. This poses a problem - the interactions influence the measurement in several ways. They can obscure the phenomenon one wants to examine. Even if it is possible to observe the phenomenon long enough to measure a precise value, this value might be inaccurate due to effects of the interaction energy.

With the cesium atoms we have in the lab, we can use a trick to eliminate these problems. By means of a broad Feshbach resonance, the strength of interactions between atoms can be controlled, and to a high degree switched off, by tuning a magnetic field to the right value.

Our tool - Bloch oscillations

We use Bloch oscillations to demonstrate our precise control over interactions. These oscillations are a counter-intuitive phenomenon that was predicted in the early days of quantum mechanics. If particles in a periodic lattice potential (for example electrons in a solid or ultracold atoms in an optical lattice) are subject to an external force, they will not be uniformly accelerated but instead oscillate in place.

With ultracold atoms, Bloch oscillations are easy to observe. The lattice potential can be switched off and the atoms are imaged after a time of free expansion, revealing the momentum distribution.

Quasimomentum distribution of our cesium BEC after release from the optical lattice.
The time spent in the lattice is increased from left to right.

The effect of interactions

Interactions between the atoms leads to a dephasing of the Bloch oscillations. Under normal conditions, for example with a rubidium BEC, this quickly leads to a broadening of the momentum distribution and only a few oscillations can be seen before the atoms are spread out over the whole Brillouin zone.
With the precise control we have over interactions, we could make a quantitative measurement of this effect. We measured the width of the momentum distribution for different times and scattering lengths, and we were able to verify theoretical models which predicted a linear time dependency and a square root dependency of the scattering length.

(a) Momentum width as a function of the number of Bloch cycles. The different colors indicate different interaction strengths.
(b) Momentum width as a function of interaction strength. The different colors indicate different number of oscillations. The solid lines are numerical simulations performed in our group.


Interactions minimized - new world record

When we set interactions to zero (black points in figure (a) above), the BEC can oscillate without being disturbed for a very long time. It is possible to observe more than 20000 Bloch oscillations over more than 10 seconds - almost an eternity in the world of quantum mechanics. This is a new world record! The oscillation period could be determined to a precision of about one in ten million.

These results show that it is possible to do precise measurements with atom lasers by switching off interactions. The technique might in the future be used in experiments to determine fundamental constants like the fine structure constant.

Bloch oscillations after different times in the lattice.

Our results were published in the Feb 29 issue of the journal Physical Review Letters: Phys. Rev. Lett. 100, 080404 (2008). Similar results were obtained by our colleagues at LENS in Florence and are published in the same issue.

The CsIII Team

The Innsbruck team (top, from left to right) Elmar Haller, Johann G. Danzl, Russell Hart, Manfred J. Mark
(bottom, from left to right) Mattias Gustavsson, Hanns-Christoph Nagerl



Article: Phys. Rev. Lett. 100, 080404 (2008)

See also: arXiv:0710.5083

German press release: Press release

Photos for download: IQOQI media photos


The experiments are supported by the START-price of the Austrian Science Fund (FWF).

last change: 04-03-08 by MG