Laser Cooling: How Light Chills Atoms Near Absolute Zero

Laser cooling is a way of using light to slow atoms down, and since the temperature of a gas is a measure of how fast its atoms move, slowing them means cooling them. With it, physicists chill gases to millionths of a kelvin. This post continues our explainer What Happens at Absolute Zero?.

Why cold means slow

The Nobel committee’s 1997 press release puts it in numbers. At room temperature the molecules of air move at about 4,000 km/h. Even at −270 °C speeds are about 400 km/h. Only at about one millionth of a kelvin (1 µK) do free hydrogen atoms move at less than 1 km/h, about 25 cm/s (NobelPrize.org, 1997 press release^[1]). Normally a gas cooled that far would condense into a liquid or freeze, but in a vacuum the density can be kept low enough to stop that happening.

How light can slow an atom

Light carries momentum. When an atom absorbs a photon it gets a small kick in the photon’s direction, and when it re-emits the photon the recoil goes in a random direction. Averaged over many photons, the absorbed kicks add up and the random ones cancel. Laser cooling of neutral atoms was proposed in 1975 by T. W. Hänsch and A. L. Schawlow, and the trick is to arrange that photons are absorbed only when they meet the atom head-on (NobelPrize.org, 2001 popular background^[2]).

That arrangement uses the Doppler effect, the same effect that makes a train whistle sound higher as it approaches. An atom only absorbs light of just the right frequency. Tune the laser slightly below that frequency (towards red) and an atom moving toward the beam sees the light shifted up into resonance, so it absorbs and slows. An atom moving away sees the light shifted further off and mostly ignores it (NobelPrize.org, 1997 press release^[1]).

Optical molasses

Around 1985, Steven Chu and co-workers at Bell Laboratories in Holmdel, New Jersey, crossed six laser beams in opposed pairs along three directions at right angles. Sodium atoms, first stopped by an opposing laser beam, drifted into the crossing point. Whichever way an atom tried to move, it met photons that pushed it back. To the naked eye the result looked like a glowing cloud the size of a pea, holding about a million chilled atoms. Moving through it was like moving through thick liquid, so it was named optical molasses (NobelPrize.org, 1997 press release^[1]).

With the lasers switched off, the atoms measured about 240 µK, corresponding to a sodium speed of about 30 cm/s, in good agreement with a theoretical floor called the Doppler limit. Our Molassa tee is built on that limit.

The Doppler limit formula

Each kick of recoil also heats the atoms a little, so Doppler cooling stops where cooling and recoil heating balance. Daniel Steck’s widely used reference data for sodium gives the lowest temperature expected for a simple two-level atom in optical molasses (Steck, Sodium D Line Data^[3]):

T_D = ħΓ / (2k_B)

Here ħ is the reduced Planck constant, 1.054 571 817 × 10⁻³⁴ J s (NIST, ħ^[4]); k_B is the Boltzmann constant, 1.380 649 × 10⁻²³ J/K (NIST, k^[5]); and Γ is the decay rate of the atom’s excited state, 61.542 × 10⁶ s⁻¹ for the sodium D2 line.

Worked example

ħΓ = 1.054 571 817 × 10⁻³⁴ × 61.542 × 10⁶ ≈ 6.490 × 10⁻²⁷ J. Divide by 2k_B = 2.761 × 10⁻²³ J/K and T_D ≈ 2.350 × 10⁻⁴ K, which is 235 µK, matching Steck’s listed 235.03 µK. For rubidium-87 the same formula gives 145.57 µK (Steck, Rubidium 87 D Line Data^[6]).

To turn a temperature into a typical speed along one direction, use

v = √(k_BT / m)

For a sodium atom (mass about 3.82 × 10⁻²⁶ kg, from Steck’s table) at 240 µK, v ≈ √(3.31 × 10⁻²⁷ / 3.82 × 10⁻²⁶) ≈ 0.29 m/s, about 30 cm/s, which agrees with the Nobel committee’s figure.

Breaking the Doppler limit

Molasses cools atoms but does not hold them; gravity pulls them out in about a second. In 1987 a magneto-optical trap added two magnetic coils to the six beams to hold the atoms in place. Then came a surprise. William D. Phillips measured temperatures as low as 40 µK in 1988, six times below the Doppler limit (NobelPrize.org, 1997 press release^[1]). The paper reporting it is titled “Observation of Atoms Laser Cooled below the Doppler Limit” (Lett et al., 1988^[7]). The limit had been worked out for a simplified two-level atom. Claude Cohen-Tannoudji’s group in Paris explained the real result using the sodium atom’s more complex energy levels, a mechanism likened to Sisyphus endlessly rolling a stone uphill. Chu, Cohen-Tannoudji and Phillips shared the 1997 Nobel Prize “for development of methods to cool and trap atoms with laser light” (NobelPrize.org, Physics 1997^[8]).

From laser cooling to a new state of matter

Laser cooling alone was not quite enough to reach a Bose–Einstein condensate. The last step was evaporative cooling: let the fastest atoms escape so the rest are colder, the same way a cup of coffee cools (NobelPrize.org, 2001 popular background^[2]). The 2001 laureates first reached that state in 1995; Eric Cornell and Carl Wieman then produced a pure condensate of about 2,000 rubidium atoms at 20 nK, and Wolfgang Ketterle made larger ones with sodium (NobelPrize.org, 2001 press release^[9]). Our Condensa sticker is that state of matter, explained in full in Bose–Einstein Condensates Explained. For a cold character to wear, see Cryo Fault.

Quick FAQ

Can lasers really make things colder?

Yes, for dilute gases of atoms in a vacuum, when the laser is tuned just below the atom’s resonance so it mostly pushes against the atom’s motion.

What is the Doppler limit?

The lowest temperature simple Doppler cooling can reach, ħΓ/(2k_B), about 235 µK for sodium.

Can laser cooling reach absolute zero?

No. It reaches micro- and, with further steps, nanokelvin temperatures, but never exactly zero.

References

1. NobelPrize.org, Press release: The 1997 Nobel Prize in Physics. https://www.nobelprize.org/prizes/physics/1997/press-release/
2. NobelPrize.org, Popular science background, The Nobel Prize in Physics 2001. https://www.nobelprize.org/prizes/physics/2001/popular-information/
3. D. A. Steck, “Sodium D Line Data” (revision 2.3.4, 2025), steck.us/alkalidata. https://steck.us/alkalidata/sodiumnumbers.pdf
4. NIST CODATA, “reduced Planck constant” ħ. https://physics.nist.gov/cgi-bin/cuu/Value?hbar
5. NIST CODATA 2022 recommended values: Boltzmann constant (exact). https://physics.nist.gov/cgi-bin/cuu/Value?k
6. D. A. Steck, “Rubidium 87 D Line Data” (revision 2.3.4, 2025), steck.us/alkalidata. https://steck.us/alkalidata/rubidium87numbers.pdf
7. P. D. Lett, R. N. Watts, C. I. Westbrook, W. D. Phillips, P. L. Gould & H. J. Metcalf (1988), “Observation of Atoms Laser Cooled below the Doppler Limit,” Phys. Rev. Lett. 61, 169–172 (DOI 10.1103/PhysRevLett.61.169; metadata verified via Crossref). https://doi.org/10.1103/PhysRevLett.61.169
8. NobelPrize.org, “The Nobel Prize in Physics 1997” (S. Chu, C. Cohen-Tannoudji & W. D. Phillips). https://www.nobelprize.org/prizes/physics/1997/summary/
9. NobelPrize.org, Press release: The 2001 Nobel Prize in Physics (Cornell, Ketterle, Wieman). https://www.nobelprize.org/prizes/physics/2001/press-release/

Written by Pixelated Physics. Every factual claim is linked to the source we checked; points that are still debated are labelled open.