Toward evaporative cooling of potassium atoms
S.
Inouye1,2,
1
T. Kishimoto1, J. Kobayashi2, K. Aikawa1, K. Noda1, T. Arae1, and M. Ueda2,3
Institute of Engineering Innovation, The University of Tokyo, Bunkyo-ku, Tokyo 113-8656, Japan
2 ERATO, JST, Kawaguchi, Saitama 332-0012, Japan
3 Department of Physics, Tokyo Institute of Technology, Meguro, Tokyo 152-8851, Japan
Motivation
Results
Our goal is to produce ultracold polar moleculer gas. This will be a new
“playground” for physicists, since molecules interact with dipole-dipole
interaction, which is long range and unisotrpic. Applications include
“a toolbox” for lattice-spin models[1], quantum computation[2], and ultracold
chemistry.
We are currently focusing on cooling 41K
(polar molecules in a lattice)
“Feshbach molecules” is an important intermediate step for efficient
production of ultracold polar molecules from ultracold atoms
STIRAP
(stimulated
Raman adiabatic passage)
magnetic field
sweep or
rf-association
Mixture of degenerate
atomic gases
Feshbach molecules
Deeply bound
molecules
Our plan is to prepare fermionic mixture made of 6Li and 40K
• minimized three-body loss at the resonance
• alkali-alkali mixture ensures spin-exchange interaction needed for F.R.
We plan symapthetic cooling of 6Li and 40K using bosonic potassium
isotope (41K) as a coolant. This simplifies the laser and vacuum setup.
(LENS group has Bose condensed 41K via sympathetic cooling with 87Rb.
See G. Moduguno et al., Science 294, 1320 (2001))
Feshbach resonance
[1] A. Micheli et al., Nature
Physics, 2 341 (2006)
[2] D. DeMille, PRL 88,
067901 (2002)
• Homemade dispensers
• Two glass cells connected
by stainless-steel chambers
• Multi-stage diff. pumping
• Atom transfer by a push beam
(No magnetic guiding).
• Larger MOT beam diameters
(~38mm)
MOT in the 1st cell
level diagram of 41K
CMOT and Doppler cooling are quite effective for 41K
The number of atoms is less sensitive to experimental conditions. Controlling
temperature and density requires more attention, since the small excited state
hyper-fine splitting prevents straight forward optimization using internal state
control. Following is our current status of our experimental parameters.
Temp.
(μK)
Density
(1010cm-3)
Number
(109)
(R-MOT)
(2500)
(0.33)
MOT
5000
1.6
after CMOT
100
5.1
suppressed three-body loss
for fermionic mixture
Experimental Setup
Double-MOT system
Atom transfer between the cells is reasonable
Loading rates: 2.9 x 109 atoms/s (1st MOT, w/o push)
0.56 x 109 atoms/s (2nd MOT)
Transfer efficiency = ~ 19 %
(*The efficiency is sensitive to MOT beam diameters.
It determines the amount of transverse cooling during
the push, and “stopping power” of the 2nd MOT beams)
PSD
(10-7)
(Note)
(1.4)
0.005
(FL optimized)
1.6
0.009
1.0
10.1
(experimental time sequence)
Conclusions
External Cavity LD
+ Five Slave LDs
+ Four tapered amps
• Enough laser power
(~ (4x) 300mW after each fiber)
for multi-isotope trapping
(four tapered amps)
Ioffe-Pritchard
magnetic trap
• Clover-leaf configuration
• Square tubing
Radial: 380Hz @ 200A
Axial: 15.1Hz @ 150A
(ωi/2π, for 41K)
• 41K is easy to collect (~109 atoms), but the temperature tends to
be high, and the density tends to be low. This is partly due to the
small excited state hyperfine splitting, which makes excited
states un-resolvable during MOT.
5mm
41K after CMOT,
0.6 ms TOF
• CMOT and Doppler cooling is quite effective for 41K. It is possible to reduce the temperature and increase the density by CMOT
-like procedure, reaching the phase-space density of 10-6 with
more than 109 atoms.
Outlook
2nd MOT
(schematics)
1st MOT
Dispensers
Investigate the properties of 41K
• See if 41K is appropriate as a coolant
Photoassociation of double-species MOT
• Start setting up energy and state sensitive detection of molecules
made by photoassociation in a double species MOT
(Front from the left:Kobayashi,
Aikawa, Arae, Noda. Back:
Inouye, Kishimoto)