Xenon atoms are released from a magneto-optical trap and equilibrate in a one-dimensional, ${\mathrm{\ensuremath{\sigma}}}_{+}$-${\mathrm{\ensuremath{\sigma}}}_{\mathrm{\ensuremath{-}}}$ optical molasses, along which a uniform magnetic field is applied. In the rest frame of the ``corkscrew'' pattern of laser polarization, the magnetic field induces a nonzero average atomic velocity. Introducing a moving molasses through a relative difference in the two laser-beam frequencies allows straightforward, accurate measurements of the molasses-frame velocity distribution over a large range of magnetic fields. The measurements reveal the loss of polarization-gradient cooling in fields higher than \ensuremath{\sim}0.2 mT, in quantitative agreement with theoretical predictions. \textcopyright{} 1996 The American Physical Society.
We investigate the effect of a longitudinal magnetic field on the orientational cooling mechanism that operates both in σ+–σ− optical molasses and in a magneto-optical trap. For an atom with different Landé g factors in its ground and excited states, single-photon scattering and stimulated Raman scattering are resonant at different field-dependent velocities. The sharp polarization-gradient feature in the force-versus-velocity curve may therefore be translated to a velocity at which the Doppler force is dominant. In addition to this, we find that the momentum diffusion coefficient can grow as the field is increased. By numerically solving a semiclassical Fokker–Planck equation, we show that these effects lead to a field-induced inhibition of efficient sub-Doppler cooling for atoms with unequal g factors. Our treatment also enables us to discuss other phenomena that are accessible only if both Doppler and polarization-gradient effects are included in the calculation.
Laser cooling to sub-Doppler temperatures by optical molasses is thought to be inhibited in atoms with unresolved, near-degenerate hyperfine structure in the excited state. We demonstrate that such cooling is possible in one to three dimensions, not only near the standard
D2 line for laser cooling, but over a wide range extending to the
D1 line. Via a combination of Sisyphus cooling followed by adiabatic expansion, we reach temperatures as low as 40 $\ensuremath{\mu}$K, which corresponds to atomic velocities a factor of 2.6 above the limit imposed by a single-photon recoil. Our method requires modest laser power at a frequency within reach of standard frequency-locking methods. It is largely insensitive to laser power, polarization and detuning, magnetic fields, and initial hyperfine populations. Our results suggest that optical molasses should be possible with all alkali-metal species.