Bose 2.2 User Manual Page 89

  • Download
  • Add to my manuals
  • Print
  • Page
    / 155
  • Table of contents
  • BOOKMARKS
  • Rated. / 5. Based on customer reviews
Page view 88
evaporation by continuously lowering the power of the ODT laser beams
83
. Before reaching
degeneracy, we switch on the strong magnetic field to a strength
B
evap
= 602
G
, well
above the Feshbach resonance (FR) located at
B
0
= 589
.
1
G
. At the scattering length
a
(
B
evap
) = 90
a
0
, we finish the evaporation and create the BEC in a trap with frequencies
ω
x,y,z
= 2π · (440 ± 14, 330 ± 10, 290 ± 10) Hz.
We then linearly decrease the magnetic field to
B
=
B
lat
, to approach the FR with
a
(
B
lat
) = 60
a
0
. As a next step, we load the BEC into the lattice by increasing the power
of the lattice laser
84
in an s-shaped ramp to a variable value which defines the lattice depth
U
lat
. While doing so, the lattice introduces an additional transverse (radial) confinement.
In order to keep the radial (
x, y
) trap frequencies constant during the ramp (and to avoid
excitations in the BEC), we decrease the power in the ODT1 laser beam in an inverted
s-shaped ramp
85
.
In a second linear ramp of the magnetic field strength to the final value
B
f
, we then
bring the system close to the stability threshold, defined by the critical scattering length
a
crit
=
a
(
B
crit
). We hold the condensate in this configuration for
t
hold
= 2
ms
, such that
the system can equilibrate. After this holding time, we switch off all trapping potentials
to perform a 6
ms
time-of-flight and finally take an absorption image of the expanded cloud.
Loading of the BEC into the lattice
To probe the stability of the condensate in the lattice, we have to prepare the system in its
ground-state, i.e. in the lowest energy band of the lattice. This implies that any excitation
to higher bands has to be avoided during the loading of the lattice potential. We thus
obtain a criterion for the speed of the lattice ramp [170]. In general, the adiabaticity
criterion is given by [196]
dU/dt
(∆
E
)
2
/~
with
E
the spacing between the first
and the second excited band and with
U
(
t
) the time-dependent lattice depth during the
ramp. At the center of the Brillouin zone, i.e. at quasi-momentum
q
= 0, the energy gap
E
= 4
E
R
is independent of the lattice depth, such that the adiabaticity criterion for
the lattice ramp in our experimental sequence is given by
dU
dt
16 E
2
R
~
. (5.3)
For our parameters (, the lattice must be ramped up much slower than 0
.
3
E
R
/ s
to fulfill
the criterion given by Eq.
(5.3)
. Experimentally, we ramp up the lattice using an s-shaped
83
During the sequence, we not only lower the power of the horizontal ODT beam, but we also decrease
the light level of the vertical beam (“ODT2”). After reaching the BEC, the power in ODT2 is not
changed anymore until it is switched off to perform the time-of-flight.
84
The lattice laser beam is switched on during the production process of the BEC to heat the optical
components in the beam path. During this time the beam is blocked by a shutter in front of the vacuum
chamber. With this technique, we minimize the movement of the focus of the laser beam during the
lattice ramp that is induced by thermal effects in the lenses.
85
For our trap frequencies, and for the given parameters of the lattice laser beams (waists
w
lat,1
=
w
lat,2
=
72
m
), we can provide constant radial trap frequencies up to a lattice depth of around 200
E
R
. This is
sufficient for all our experiments.
89
Page view 88
1 2 ... 84 85 86 87 88 89 90 91 92 93 94 ... 154 155

Comments to this Manuals

No comments