How six laser beams slowed Chu’s atoms

A cloud that lingered after the source stopped

In 1985, Steven Chu and his colleagues at Bell Laboratories sent precooled sodium atoms into the crossing point of six laser beams. The atoms glowed as they scattered the light. After the incoming pulse had passed, the glow remained. Light was slowing their motion in all three directions, making a patch of empty space behave a little like thick syrup.

The team called it optical molasses. The name captured what made the experiment striking: atoms that would otherwise fly through the region stayed around long enough to be watched. Chu, Leo Hollberg, John Bjorkholm, Alex Cable and Arthur Ashkin could follow the fading fluorescence and ask how the atoms eventually left.

Read the evidence: S. Chu, L. Hollberg, J. E. Bjorkholm, A. Cable and A. Ashkin (1985), Three-dimensional viscous confinement and cooling of atoms by resonance radiation pressure, PRL 55, 48–51

An atom meets more light from the front

An atom absorbs light most readily near a particular frequency, much as a radio responds to a tuned station. Set two opposing beams slightly below that frequency. An atom moving toward one beam sees its light shifted upward in frequency, closer to the atomic resonance. It scatters more photons from that beam than from the one behind it.

Each absorbed photon gives a small push against the motion. For slow atoms the imbalance acts like friction: move faster and the opposing push grows. Three pairs of beams provide this braking along three perpendicular directions. The atoms still receive random kicks when they emit photons, so their motion settles into a balance between slowing and heating.

The same argument works wherever the atom sits inside the overlap region. Move a stationary atom away from the center and the opposing beams still balance. Molasses resists motion without supplying a spring that pulls an atom back to one particular point.

Read the evidence: S. Chu, L. Hollberg, J. E. Bjorkholm, A. Cable and A. Ashkin (1985), Three-dimensional viscous confinement and cooling of atoms by resonance radiation pressure, PRL 55, 48–51 S. Chu (1998), Nobel Lecture: The manipulation of neutral particles, Reviews of Modern Physics 70, 685–706 — optical molasses and the retrospective thermometry discussion

A slow random walk eventually reaches the edge

Picture a person taking many small steps in a room. Shorter steps delay the moment they reach a wall. Any step can also bring a walker back; across many walkers, the spread from their starting point grows with time. The sodium atoms likewise kept wandering under the random recoil of scattered photons. Their typical speed could stay nearly constant while the cloud spread across a larger region.

The laser beams had finite width. An atom that wandered beyond their overlap stopped contributing to the retained cloud. The 1985 fluorescence record lost about half its atoms in roughly a tenth of a second. A fast initial spike belonged to another group passing through early; the slower decay followed the atoms that had lingered.

Turning the light off supplied a second test

The team also extinguished the six beams for a chosen interval, then turned them back on. The returning fluorescence counted atoms still within reach of the light. During the dark interval the atoms flew freely, making this a different experiment from watching gradual loss while the molasses stayed on.

Interpreting those losses required the size of the illuminated region and the shape of the cloud at release. A broad starting cloud has less distance to travel before leaving; a directed drift carries the whole cloud sideways. Comparing the light-on and light-off procedures helped separate these effects from the random motion that sets the temperature.

Read the evidence: S. Chu, L. Hollberg, J. E. Bjorkholm, A. Cable and A. Ashkin (1985), Three-dimensional viscous confinement and cooling of atoms by resonance radiation pressure, PRL 55, 48–51

Chu later revisited the temperature estimate

The original paper reported a temperature near 240 millionths of a kelvin, or 240 microkelvin, and described agreement with the expected cooling. In his later Nobel lecture, Chu recalled that an early analysis had produced a lower temperature. A correction for the spatial distribution moved it toward the theoretical prediction. He criticized how strongly that expectation had guided the analysis.

His account makes the correction itself part of the story. The cloud’s geometry really mattered, and an independent thermometer could test whether it had been handled well. Later measurements would reveal cooling below the simple Doppler prediction. In 1985, the immediate achievement was already remarkable: six beams had turned a fast passage through space into a cloud whose slow escape could be measured.