Read the probe before the coil
Phillips and Metcalf reported deceleration and velocity bunching of a sodium beam in March 1982. A separate weak analyzing laser scanned across the emerging atoms; fluorescence selected the velocities resonant with that probe. Timed records compared atoms exposed to the slowing light with a later unexposed population.
The probe had to distinguish a velocity change from optical pumping between ground hyperfine states. A missing fluorescence group alone could mean atoms had become dark. The paper compared those populations and located a new slower group. Its reported partial deceleration should not be replaced by later achievements of nearly stopped or trapped atoms.
Read the evidence: W. D. Phillips and H. Metcalf (1982), Laser Deceleration of an Atomic Beam, PRL 48, 596–599J. V. Prodan, W. D. Phillips and H. Metcalf (1982), Laser Production of a Very Slow Monoenergetic Atomic Beam, PRL 49, 1149 — abstract consulted
Choose the signs before shaping the field
Take the atoms along the positive beam axis and the laser in the opposite direction. Define the positive wave number by its magnitude, and the laser detuning relative to the zero-field atomic angular frequency. The atom sees a positive Doppler contribution. Choose a transition whose angular frequency increases with positive axial magnetic field, as represented below.
The effective magnetic moment is the excited-minus-ground energy slope. As the atom slows, its Doppler contribution decreases; decreasing the field can compensate. This is the sign convention of a decreasing-field slower, not a rule that every Zeeman slower must use that field direction or profile. The detuning and all frequency shifts here use angular-frequency units.
Read the evidence: W. D. Phillips (1998), Nobel Lecture: Laser cooling and trapping of neutral atoms, RMP 70, 721–741, especially pp. 723–726NIST, Context of a Discovery: Bill Phillips — institutional account of the Zeeman slower
A cycling transition is part of the apparatus
The sodium experiment used circularly polarized light and a strong axial bias field to favor a repeating transition between maximum-projection ground and excited states. Absorption increases the magnetic quantum number by one; the selected excited state returns to the cycling ground state. Polarization errors and unwanted excitations can interrupt that sequence.
For our calculation, replace this structure with one closed, steady-state two-level cycle. Each absorbed counterpropagating photon supplies backward momentum; spontaneous emission has zero mean axial recoil under a symmetric emission pattern. We omit recoil diffusion, transverse divergence, collisions, magnetic-gradient forces and light depletion. This mean-force example does not solve sodium’s full hyperfine dynamics.
Read the evidence: W. D. Phillips (1998), Nobel Lecture: Laser cooling and trapping of neutral atoms, RMP 70, 721–741, especially pp. 723–726
Integrate a force, then audit its work
Our invented beam starts at 600 metres per second and aims for 100. Using sodium mass, 589.16-nanometre light, a decay rate corresponding to 9.8 megahertz and saturation parameter two, choose a field demanding forty percent of the fully saturated maximum deceleration. Integrating the finite-linewidth scattering force over about 0.482 metres reaches the chosen exit speed.
The calculation accumulates about 16,972 scattering cycles. Independently, the momentum change divided by one photon momentum gives the same count, and integrated force times displacement matches the kinetic-energy decrease. These checks audit the mean trajectory and its units. They do not measure photon-number fluctuations, a final temperature or experimental slowing efficiency.
Read the evidence: W. D. Phillips (1998), Nobel Lecture: Laser cooling and trapping of neutral atoms, RMP 70, 721–741, especially pp. 723–726
Following has a force budget
At this finite intensity, the largest available deceleration is only two thirds of the fully saturated bound. Make the profile demand that entire bound over a shorter distance: the integrated atom exits near 586 metres per second, far above the requested 100. A magnetic resonance condition cannot supply missing radiation pressure.
Capture is limited too. In the original illustrative profile, a 700-metre-per-second atom begins outside the useful resonance range and exits near 700, while a 500-metre-per-second atom encounters the slowing trajectory later and reaches about 100. Those examples demonstrate conditional capture; they do not establish that every slower atom, polarization or transverse trajectory will be accepted.
Read the evidence: W. D. Phillips and H. Metcalf (1982), Laser Deceleration of an Atomic Beam, PRL 48, 596–599NIST, Context of a Discovery: Bill Phillips — institutional account of the Zeeman slower
A slower beam still needs a thermometer and a trap
Changing the mean velocity and narrowing its spread are different claims. Optical pumping and finite collection can alter the accepted distribution as well. A directed beam need not have a single equilibrium temperature; longitudinal and transverse motion deserve separate records. Random spontaneous recoils can widen transverse motion while the longitudinal group slows.
The existing cooling Lab compares damping, diffusion and magneto-optical capture within its declared model. It does not reproduce the Zeeman slower, probe scan or historical coil. Read this beam experiment alongside the ballistic thermometer article: one asks how atoms remain coupled to a slowing force, the other how motion becomes a defensible temperature estimate.
Read the evidence: W. D. Phillips (1998), Nobel Lecture: Laser cooling and trapping of neutral atoms, RMP 70, 721–741, especially pp. 723–726NIST, Context of a Discovery: Bill Phillips — institutional account of the Zeeman slower