Doppler cooling limit

SUB-DOPPUIR COOLING In 1988 the NIST-Gaithersburg group made careful measurements of the temperature of atoms laser cooled in optical molasses, cind found temperatures significantly below the Doppler cooling limit 113). The initial measurements on laser cooled sodium atoms gave temperatures of about 40 pK, about six times lower tham the predicted lower limit of 240 pK. 

The value of the temperature in eqn (5.3) is nowadays referred to as the Doppler temperature or Doppler cooling limit. At a typical value of the natural hnewidth of an allowed transition 2y = 27t X 10 MHz, the temperature To is of the order of 100 pK. Because of the great promise that laser cooling and subsequent laser trapping of atoms held for laser spectroscopy, researchers at the Institute of Spectroscopy in Troitsk, Russia, launched experiments in this field. By the time the first successful experiment was conducted (Andreyev et al. 1981, 1982), the first theoretical work, summarized in a review of the manipulation of atoms by the light pressure force of a resonant laser (Letokhov and Minogin 1981a), had already been completed. 

When the red frequency detuning of the laser radiation is optimal for cooling purposes, that is, when Z = — 7, the temperature T, reaches the minimum value given by eqn (5.3), which usually lies in the millikelvin range. As stated earlier, the temperature Td is usually referred to as the Doppler cooling limit. 

Atoms in an optical trap (Doppler cooling Wineland, et al., 1978 optical molasses Chu, et al., 1986 magneto-optic trap Steane and Foot, 1991 Helmerson, et al., 1992) are confined and cooled to translational temperatures on the order of << 1 mK. Ultracold collisions between such trapped atoms permit the recording of bound<—free spectra with resolution limited only by the translational temperature (1 mK, which corresponds to a frequency resolution of 7 x 10-4 cm-1) (Julienne and Mies, 1989 Lett, et al., 1995 Burnett, et al., 2002). This makes spectroscopically accessible the extremely long-range regions of potential energy curves (R >10A 5Re) and otherwise only indirectly observable weakly bound or repulsive electronic states. 

In this section, various effects that can limit the accuracy of the SCSI-MS technique are considered. In particular, deviations from the idealized case considered in Section 10.3.4, due to finite size motional amplitudes and the effects of the Doppler cooling force will be discussed. 

In accordance with the above temperature-scale estimates (Fig. 5.4), it is convenient to consider consecutively the cooling of atoms first to the Doppler limit Td, then sub-Doppler cooling to Tree, and finally subrecoil cooling. Various mechanisms for cooling atoms in light fields of various configurations are considered below in the same sequence. The reader can find more detailed analysis in the book by Metcalf and van der Straten (1999). 

In experiments on the 3D Doppler cooling of Na atoms (Lett et al. 1988), it was found, by way of direct measurement of the velocity of the cooled atoms leaving the cooling region, that the temperature of the atoms could in actual fact be almost ten times lower the Doppler limit fiK. Moreover, there were observed a number... 

The two methods of demonstrative character considered above should be supplemented with the classical Doppler-cooling method for forbidden transitions with a very small radiative broadening 7. In the first experiments of this kind (Katori et al. 1999) with an ensemble of Sr atoms, a 3D-cooling temperature that was close to a single effective recoil was obtained, that is, the alternative cooling pathway shown on the left of Fig. 5.4 was realized. Moreover, in the case of polychromatic excitation, there is even a possibility of reaching a temperature below the recoil limit (Wallis and Ertmer 1989). 

Lett P D, Watts R N, Westbrook C I, Phillips W D, Gould P L and Metcalf H J 1988 Gbservation of atoms, laser-cooled belowthe Doppler limit Phys.Rev.Lett. 61 169-72... 

Dalibard J and Cohen-Tannoudji C 1989 Laser cooling belowthe Doppler limit by polarization gradients simple theoretical models J.Opt.Soc.Am. B 6 2023-45... 

For a lithium beam slowed down to 10 m/s and cooled down close to the Doppler limit (Av/v 0.1), a capacitor 0.5 m long and for ta = Tmax, then ta = 23.5 ms and Tmin=76.5 ms. With the parameters already mentioned one could measure a charge qn as small as the existing limit on the neutron charge qn in about 16 seconds. 

---