Document Sample
Motivation Powered By Docstoc
					Interaction of a BEC with Dipole

     Mirco Siercke, Chris Ellenor, Matt Partlow,
  Fan Wang, Jan Henneberger, Aephraim Steinberg
    Department of Physics, University of Toronto
• Our first experiments will study the time a tunneling
  particle spends in a dipole barrier
• Later experiments will further probe the interaction of
  coherent atoms with detuned laser light
    – The creation of non-classical momentum components during
      collisions with barriers, and even the complete suppression of
      central momentum components.
    – Long range laser induced dipole-dipole interactions (LIDDI)
•   Possible experiments in quantum hydrodynamics
                                      push beam
       Upper MOT

                            40G max
                            5.4 kHz
 Lower MOT                                      probe

                                                     Quadrupole coils
        Ø=10cm, 295 turns
        max gradient:
        470 G/cm@30A                               Inner Diameter 1.7cm

                                                               Not Drawn:
                                                               coils along 3 axes
                                             TOP coils
Optical pumping beam
              Experimental Details
• Loading of lower MOT with push beam: 7s
• MOT beams
     – 3.2 mW/cm2
     – 2 cm diameter
•   Trap lifetime: 100s
•   Heating without RF shield: 300nK/s
•   Atom loss at 52nK with RF shield: < 13%/s
•   Trap frequencies – 48, 68, 96 Hz (compressed)
•   20 micron resolution, 12-bit CCD array
•   Field turnoff 1ms (TOP field), 100s (quadrupole field)
•   Single loop RF coil driven by Agilent 33250A arbitrary waveform
                        Experiment Stages
•   Upper MOT
     –   400 million atoms, 10-9 Torr
•   Lower MOT
     –   Push beam loading for 7seconds
     –   1 billion atoms, < 10-11 Torr
•   Molasses
     –   2ms at 20MHz detuning 3ms at at 28MHz detuning and 3ms at 36MHz detuning
     –   Optically pump on the F=2 to F=3 transition for 4ms
•   TOP Trap
     –   initial loading parameters: 40 G TOP field, 71 G/cm gradient
     –   Compensate for gravitational sag with an additional field in the vertical direction
     –   Load 300 million atoms at 80K
     –   Collision rate 5 Hz, Phase Space Density 2.810-7
•   Quadrupole compression
     –   Compress from 71 G/cm (weak) to 155 G/cm in 0.5s
     –   Wait 15s for evaporation, compress to 235 G/cm in 0.5s, wait 15 more seconds
•   Lowering of the Bias Field
     –   Effects further compression and brings in circle of death for initial evaporation
     –   Lowered from 40G to 20G in 14.5s
     –   48 K, 40 million atoms, Collision rate of 22 Hz, Phase Space Density 3.410-6
•   RF ramp
     –   Ramp from 24MHz to ~14MHz in 54s
     –   Optimal ramp calculated, and recalculated when cloud deviated from prediction
The Lower Chamber
Expansion after RF Evaporation
                  Absorption Images

      8ms Expansion of a 52nK cloud   20ms Expansion of a 52nK cloud
                    Next steps
• Further compression of trap during RF ramp by
  lowering bias field
   – Raise transition temperature
• Imaging
   – Improve resolution to better than 10m to resolve
     aspect ratios we expect to be on the order of 80%, and
     expanded condensates we expect to be 20-40 m
   – Improve absorption imaging
   – Implement phase contrast imaging
• Stabilize atom number
       How Long Does a Tunneling
       Particle Spend in the Barrier?

After almost 70 years of discussion, no consensus has yet emerged on the answer
to this simple question. This question is not only of fundamental, but also of
technological interest. BEC provides an excellent tool to study this issue
experimentally. Dipole tunneling barriers can be created with a size comparable
to the DeBroglie wavelength of the atoms, allowing for significant tunneling
probabilities of atoms which can then be imaged relatively easily. The internal
(hyperfine, Zeeman) structure of these atoms also offers possibilities for the
study of interaction time.
          A Proposed Geometry
• A dipole beam traps
  atoms in a separate                               trapping
  well, and acts as a
  barrier as atoms                        dipole
  tunnel into the                         beam

  magnetic trap
• Raman beams are               Raman
  overlapped with the           beam(s)

  barrier and weakly
  couple atoms into a   atoms

  different hyperfine
                           Other Things to Look at…
   •      Büttiker and Landauer imagine a barrier
          whose height is modulated at some frequency
          . The frequency is raised to a critical                                                                            C

          frequency c at which modulation in
          transmission is no longer seen, and this
          frequency serves to define a traversal time
          known as the Büttiker-Landauer time.
                                                                                         Büttiker and Landauer, PRL 49, 1739 (1982)

                                                                             •   A probe beam which could in principle be
                                                                                 used to image a tunneling atom could suffice
                                                                                 to enhance transmission probability without
                                                                                 necessarily attempting to perform the imaging
                                                                             •   Work by Bardou suggests that significantly
                                                                                 enhanced transmission may be achieved by
                                                                                 applying a small momentum transfer to a
                                                                                 particle interacting with a steep potential
                                                                                           D. Boosé and F. Bardou, Europhys. Lett. 53, 1 (2001)

A. M. Steinberg, Journal of the Korean Physical Society 35 (3), 122 (1999)
Collisional Transitory
 Enhancement of the
  High Momentum
  Components of a
Quantum Wave Packet
• Collisions are usually considered only in the asymptotic
  regime, but the full quantum mechanical treatment of a
  collision with a potential barrier reveals the transitory
  population of classically forbidden momentum states.
• By quickly switching off the barrier during the collision,
  we will observe these states in the free expansion of our
                             S. Brouard and J. G. Muga, Phys. Rev. Lett. 81, 2621 (1998)
             Transient Interference of
            Transmission and Incidence
   • Extending the previous work,
     a similar effect has been
     described where interference
     completely suppresses the
     central momentum of the wave
   • The interference and resultant
     momentum distribution is a
     result of the barrier shape,
     over which we have splendid
A. L. Perez Prieto, S. Brouard and J. G. Muga, Phys. Rev. A 64, 012710 (2001)
                 Laser Induced Dipole-Dipole
                     Interactions (LIDDI)
        • LIDDI is a long range (~3) interaction induced between
          atoms by an incident, propagating field
        • A product of forward photon scattering
        • Possible roton dip?

Borrowed from a talk by Duncan O’Dell at: http://www.quacs.u-
                                                                DHJ O’Dell, S Giovanazzi and G Kurizki, PRL 90 (2003) 110402
        Conclusions / Future
• We are VERY close!
  – T=52nK, PSD=2.8???
• Currently developing beams for dipole
  barriers and Raman probes
• Always interested in potential

And You
And if all else fails…
Upper Chamber

Lower Chamber