Realization of a Magnetically Guided Atomic Beam in the Collisional Regime

We describe the realization of a magnetically guided beam of cold rubidium atoms, with a flux of $7\times {10}^9\mathrm{a}\mathrm{t}\mathrm{o}\mathrm{m}\mathrm{s}/\mathrm{s}$, a temperature of $400\mu \mathrm{K}$, and a mean velocity of $1\mathrm{m}/\mathrm{s}$. The rate of elastic collisions within the beam is sufficient to ensure thermalization. We show that the evaporation induced by a radio-frequency wave leads to appreciable cooling and an increase in the phase space density. We discuss the perspectives to reach the quantum degenerate regime using evaporative cooling.

Figure 1

Top view of the setup. The atoms emerging from the oven are slowed down by radiation pressure, captured in the magneto-optical trap, and launched into the magnetic guide.

Figure 2

(a) Absorption signal at the end of the guide for 1, 2, 4, and 24 launches. The time origin corresponds to the launching of the first packet. (b) Number of atoms (area of the absorption signal) as a function of the number of launches.

Figure 3

Experiment with two antennas separated by a distance $d$: remaining fraction after the second antenna, with the first antenna set at a fixed value ${\nu }_1$. (a) Antenna one off, flux $\varphi$, $b_1^{\prime }=600\mathrm{G}/\mathrm{c}\mathrm{m}$, and $d=3\mathrm{m}$. Using (2) as a fit function (solid line), we find $T_1=384(\pm 13)\mu \mathrm{K}$. (b) ${\nu }_1=24\mathrm{M}\mathrm{H}z$ with the same conditions as for (a). The fit gives $T_2=281(\pm 8)\mu \mathrm{K}$, indicating a significant cooling. (c) ${\nu }_1=8\mathrm{M}\mathrm{H}\mathrm{z}$ and reduced number of collisions between antennas: flux $\varphi /3$, $b_1^{\prime }=400\mathrm{G}/\mathrm{c}\mathrm{m}$, and $d=0.85\mathrm{m}$. The dotted line is the prediction for the collisionless regime. The solid line is the result expected if thermalization had occurred.

Figure 4

Ratio $T_2/T_1$ between the temperature $T_2$ of the beam when the first antenna is set at ${\nu }_1$ and the temperature $T_1$ of the beam in absence of evaporation, as a function of ${\eta }_1=h{\nu }_1/(k_B{T}_1)$. The solid line is the theoretical prediction for a linear potential (no adjustable parameter). The open square corresponds to the data of Fig. 3b. The squares (circles) have been obtained with $I=240\mathrm{A}$ ($400\mathrm{A}$).