Experiments with Supersonic Beams as a Source of Cold Atoms

(≈ 10 6 ) which amplifies the current at the anode of the MCP, the integral under the
curve of the detected signal should correspond to the number of particles detected.
Since the gains of the MCP and amplifier are known to an order of magnitude at best,
the best estimate number of particles slowed per shot is 10 3 − 10 5 , where the beam
slowed to 222 m/s has approximately six times more atoms than the beam slowed
to 55.8 m/s. Simulations indicate that the 64 stage slower has an efficiency of a few
percent for the 222 m/s beam to a few tenths of a percent for the 55.8 m/s beam,
where efficiency is defined as the slowed fraction of the metastable beam entering the
coilgun. Most of the metastable atoms are not in the mJ = 2 magnetic sublevel that
is slowed, reducing the efficiency achieved. There is also another metastable state of
neon, 3 P0, which is probably produced by the discharge, and which is unslowed, again
reducing the efficiency.
Rather than keep the phase of the coils constant, a variable phase can be used
as well. Figure 4.31 shows how the slowed peak changes when the phase of a coil is
tuned in proportion to the velocity of the atoms entering the coil. As the beam is
slowed, the coil is switched with the atoms closer to the center of the coil, with phase
angle related to the velocity by the relation φn = φivn ,whereφnis the phase of the
vi
nth coil, vn is the velocity of the synchronous atom entering the nth coil, and φi and
vi are the initial phase and velocity respectively. Letting the phase be adjusted in this
manner leads to a decrease in slowed flux, probably due the atoms being too close
to the center of the coil when the coil is switched, reducing the longitudinal phase
stability. However, if the phase adjustment is applied, with a cutoff maximum phase
allowed, then the slowed beam actually has greater flux. All three slowed peaks in
figure 4.31 have the same target velocity at the end of the slower, 136 m/s, however
the variable phase peaks have a greater slowed flux.
Simulations indicate that the reason the flux increases for variable phases is
107