Measurement of Cs-Cs elastic scattering at<i>T</i>=30 μK

Monroe, C.; Cornell, Eric A.; Sackett, C. A.; Myatt, C. J.; Wieman, Carl

doi:10.1103/physrevlett.70.414

Cited by 244 publications

(182 citation statements)

References 17 publications

Supporting

Mentioning

171

Contrasting

Order By: Relevance

“…For evaporative cooling to be successful requires, roughly speaking, that the ratio of elastic to inelastic collision rates, K el /K loss , exceed 100 [28]. For the estimated results shown in Figure 6, this condition holds only at energies below ∼ 1mK for fields as low as 10 gauss, and not at all for near-critical fields of ∼ 50 gauss.…”

Section: Applicationsmentioning

confidence: 85%

Magnetic-field effects in ultracold molecular collisions

Volpi

Bohn

2002

Phys. Rev. A

View full text Add to dashboard Cite

We investigate the collisional stability of magnetically trapped ultracold molecules, taking into account the influence of magnetic fields. We compute elastic and spin-state-changing inelastic rate constants for collisions of the prototype molecule 17 O 2 with a 3 He buffer gas as a function of the magnetic field and the translational collision energy. We find that spin-state-changing collisions are suppressed by Wigner's threshold laws as long as the asymptotic Zeeman splitting between incident and final states does not exceed the height of the centrifugal barrier in the exit channel. In addition, we propose a useful one-parameter fitting formula that describes the threshold behavior of the inelastic rates as a function of the field and collision energy. Results show a semi-quantitative agreement of this formula with the full quantum calculations, and suggest useful applications also to different systems. As an example, we predict the low-energy rate constants relevant to evaporative cooling of molecular oxygen.

show abstract

Section: Applicationsmentioning

confidence: 85%

Magnetic-field effects in ultracold molecular collisions

Volpi

Bohn

2002

Phys. Rev. A

View full text Add to dashboard Cite

show abstract

“…Indeed, for an evaporation leading to quantum degeneracy, a factor larger than 50 is sufficient [20]. Nevertheless, the highest phase space density we could attain was only ϳ10 25 for a cloud at 4 mK.…”

mentioning

confidence: 99%

Giant Spin Relaxation of an Ultracold Cesium Gas

et al. 1998

View full text Add to dashboard Cite

We have measured the rate of inelastic collisions in a cloud of doubly polarized ground-state cesium atoms (F m F 4) confined in a magnetic trap for temperatures T between 8 and 70 mK. We find a two-body rate coefficient varying as T 20.63 . At 8mK it reaches 4 3 10 212 cm 3 s 21 which is 3 orders of magnitude larger than predicted, ruling out a Bose-Einstein condensation of Cs in this internal state.[ S0031-9007(98) [3] has attracted a lot of attention. The atomic sample is spin polarized and confined in a magnetic trap. The condensation is reached using the elastic collisions between the trapped atoms to evaporatively cool the gas to a temperature below 1 mK. Although these gaseous assemblies are not truly stable at such a low temperature, the time involved with the relaxation processes, such as two-body spin dipolar relaxation or three-body recombination, is quite long. For typical atomic densities of 10 13 10 14 cm 23 , the lifetime for the condensate is several seconds, leaving enough time for a study of the condensate properties.We show in this Letter that this metastability is not "universal": For a gas of doubly polarized cesium atoms, inelastic collisions limit the atomic density to a value much below the condensation threshold. The case of cesium is of particular interest because the hyperfine splitting DE Ӎ h 3 9.2 GHz of its ground state into two sublevels with angular momenta F 3 and F 4 is the basis of primary time and frequency standards.A cesium atomic gas in its doubly polarized (electron 1 nucleus) ground state ͑F m F 4͒, trapped in a magnetic field minimum, was initially considered to be a very good candidate for Bose-Einstein condensation (BEC) [4]. In zero magnetic field and for densities below 10 13 cm 23 , the dominant loss process is predicted to be a spin dipole relaxation, occurring through the magnetic dipole-dipole interaction. After a collision of two polarized cesium atoms, one or both atoms may emerge in the level F 3. The internal energy (DE or 2DE) is converted into kinetic energy shared between the two atoms. As DE k B 3 0.44 K is much larger than the typical trap depth (ϳk B 3 1 mK), the two atoms escape from the trap after the inelastic collision. The rate coefficient for this process was estimated to be in the range of 10 215 cm 3 s 21 [4], which should not be a limitation for the achievement of BEC with densities similar to other alkali experiments.We report here measurements performed on polarized Cs atoms trapped in a static magnetic trap [5]. The cloud is prepared with adjustable temperature using evaporative cooling. We monitor the decay of atoms out of the trap as a function of time. At 8 mK we find a two-body loss rate of 4 3 10 212 cm 3 s 21 . This spectacular deviation from the theoretical expectation by 3 orders of magnitude may be a consequence of the zero-energy resonance for elastic collisions that has recently been observed in cesium [6].

show abstract

“…[11], and from this data we improve the accuracy of the Rb interaction potential substantially. In contrast to Ref.[10], we do not observe two or three body loss at resonance because we work at much lower densities (10 9 atoms/cm 3 ).We measured the collision rates using the technique of "cross dimensional mixing," as in our earlier work [7]. In this technique, a nonisotropic distribution of energy is created in a magnetically trapped cloud of atoms, and the time for the cloud to reequilibrate by elastic collisions is measured.…”

mentioning

confidence: 99%

“…We measured the collision rates using the technique of "cross dimensional mixing," as in our earlier work [7]. In this technique, a nonisotropic distribution of energy is created in a magnetically trapped cloud of atoms, and the time for the cloud to reequilibrate by elastic collisions is measured.…”

mentioning

confidence: 99%

Resonant Magnetic Field Control of Elastic Scattering in ColdR85b

et al. 1998

Self Cite

View full text Add to dashboard Cite

A magnetic field dependent Feshbach resonance has been observed in the elastic scattering collision rate between atoms in the F = 2, M = −2 state of 85 Rb. Changing the magnetic field by several Gauss caused the collision rate to vary by a factor of 10 4 , and the sign of the scattering length could be reversed. The resonance peak is at 155.2(4) G and its width is 11.6(5) G. From these results we extract much improved values for the three quantities that characterize the interaction potential: the van der Waals coefficient C6, the singlet scattering length aS, and the triplet scattering length aT .Very low-temperature collision phenomena can be quite sensitive to applied electromagnetic fields. Several groups have altered inelastic collision rates in optical traps by applying laser fields [1]. There have also been numerous proposals [2-6] for using laser and static electric or magnetic fields to influence the S-wave scattering length (a), and equivalently, the elastic collision cross sections (σ = 8πa 2 ) between cold atoms. Particularly notable is the prediction by Verhaar and co-workers [2] that as a function of magnetic field there should be Feshbach resonances in collisions between cold (∼ µK) alkali atoms. These resonances were predicted to cause dramatic changes in the cross section, and to even allow the sign of the scattering length to be changed. Such resonances are very interesting collision physics, but they also offer a means to manipulate Bose-Einstein condensates (BEC), the properties of which are primarily determined by the scattering length. Some time ago we searched for such resonances in 133 Cs and 87 Rb, without success [7,8]. This was not surprising because there was enormous uncertainty as to the positions and widths of the predicted resonances. However, recent photoassociation spectroscopy has greatly improved the knowledge of the alkali interatomic interaction potentials, allowing these resonance parameters to be predicted with much less uncertainty [9]. As we show below, the fact that the widths and positions of these resonances are so sensitive to the interatomic potentials makes their measurement a good way to determine these potentials.The improved predictions for the resonance positions facilitated their observation. In the past few months Feshbach resonances have been observed in both 23 Na [10] and 85 Rb [11]. In the sodium work, two magnetic field dependent processes were observed: 1) a change in the expansion rate of BEC due to a change in the scattering length, and 2) an enormous, and as yet unexplained, increase in the loss rate. This loss rate precluded the study of collisions in the interesting field regime near the center of the resonance where the scattering length changes sign; the width of the resonance was not measured. In the rubidium work [11] the resonance was detected as a magnetic field dependence of the photoassociation spectra. The resulting resonance width was measured to be far larger than originally predicted.Here we report the study of a Feshbach resonance in t...

show abstract

Measurement of Cs-Cs elastic scattering atT=30 μK

Cited by 244 publications

References 17 publications

Magnetic-field effects in ultracold molecular collisions

Magnetic-field effects in ultracold molecular collisions

Giant Spin Relaxation of an Ultracold Cesium Gas

Resonant Magnetic Field Control of Elastic Scattering in ColdR85b

Contact Info

Product

Resources

About