Trap-loss collisions of ultracold lithium atoms

Abstract

Accurate measurements are presented of the rate of trap-loss-producing collisions between ultracold magneto-optically trapped Li atoms for a range of trap laser intensities and frequencies. Intensities from near the atomic saturation intensity to well above it are investigated. At low intensities, fine-structure-changing collisions cause trap loss with a rate constant of -10 ' cm /s. At sufficiently high intensity, the trap can be deep enough to effectively freeze out the dominant fine-structure-changing collisions as a loss mechanism, [11].This ability to separate the contributions of RE and FS greatly simplifies the interpretation of the experiment. In this paper, we present measurements of the RE and FS rates in I i for a range of trap laser intensities and detunings. The measured rates are compared with calculations that account for the variation of ET with laser intensity and detuning.The experimental apparatus and procedure are similar to other measurements of collisional loss in a MOT [2,4]. The MOT consists of six near-resonant laser beams along the three orthogonal axes that provide for dissipation of the atomic kinetic energy, and, when combined with an inhomogeneous magnetic field, also produce a restoring force [1].The six beams are produced by a dye laser that is frequency locked relative to a saturated absorption feature of Li in an absorption cell to provide long-term frequency stability and a relative frequency reference. The waist (1/e intensity radius) of the laser beams is 0.64 cm. A pair of anti-Helmholtz configured coils generates an axial magnetic-field gradient of 3 mT/cm and a radial gradient of 1.5 mT/cm. To ensure a near-spherically-symmetric cloud of trapped atoms, the intensities of the four radial laser beams are equal, while the intensity of an axial beam is -60% of the intensity of a radial beam. Transitions from both I' = 1 and 2 ground-state hyperfine levels to the excited states are driven with equal intensity using the upper and lower first-order sidebands produced by a standing-wave electro-optic modulator, modulated at 406.4 MHz. The lower first-order sideband is de-

Heteronuclear ionizing collisions between laser-cooled metastable helium atoms

“…(10) by substituting the values P −3/2 = P −1/2 = P 1/2 = P 3/2 = 1 4 and P −1 = P 0 = P 1 = 1 3 into the expressions for the coefficients (2S+1) b (Table II) and evaluating Eq. (10).…”

“…Since then, numerous investigations, both experimental and theoretical, have been made into the collisional properties of many different homonuclear [2,3,4,5,6,7,8,9,10,11,12,13] and (later) heteronuclear [14,15,16,17,18,19] systems. Collisional studies are in themselves interesting, leading to an in-depth understanding of the various scattering mechanisms present at such low kinetic energies and methods by which we can have some measure of control over elastic and inelastic collisions.…”

Cold inelastic collisions between lithium and cesium in a two-species magneto-optical trap

“…The only process releasing sufficient energy to explain the observations is therefore an inelastic collision changing the Cs hyperfine state. Note, that inelastic Li * -Cs collisions changing the Li excited-state fine structure, which would release h × 10 GHz and which are relevant for trap loss through inelastic Li * -Li collisions [26,27], are excluded by the repulsive quasi-molecular potential (see Sec. 2.3).…”

Trap-depth determination from residual gas collisions

“…A well-studied example of this is the large intensity-dependent variation displayed by the two-body intratrap loss-rate coefficient for atoms trapped in a magnetooptical trap (MOT). This variation results from an interplay of trap depth and the energy imparted to trapped atoms due to hyperfine or fine structure changing collisions, as well as radiative escape [1][2][3][4][5][6][7][8][9][10][11][12]. More recently, inelastic and elastic collision rates in dipole traps have been of interest, particularly for metastable species [13,14].…”

“…However, for dissipative traps that rely on radiation pressure forces, the trap depth is not easily estimated. Estimating the trap depth for a MOT from first principles requires the numerical integration of the optical Bloch equations for a multilevel atom [4][5][6]38,39] and often provides only qualitative agreement with experimental results. Trap depth measurement techniques for a MOT have been developed.…”