In Reverse Chronological Order: 2008, 2007...

BEC in optical lattices

Diffraction pattern of atoms perturbed by an optical lattice. In this experiment, we suddenly flash on a deep 2D optical lattice right after releasing the atoms in free space. The lattice imprints a variation in the condensate density and wavefunction phase which populates momentum classes at multiples of the "lattice recoil momentum" determined by the wavelength of the light and the mass of the atoms.

Fast Evaporation to Cs BEC: cond-mat/0804.0060

The time-of-flight (TOF) absorption images above are taken at various stages in our evaporative cooling path. The TOF expansion time is 70 ms. During the first 64 ms, atoms are levitated with an external field gradient of 31.5 G/cm, followed by a 6 ms free fall. In the first two pictures, atoms are still thermal with a Gaussian momentum distribution. In the later pictures, bimodality is seen, showing the phase transition into a Bose-Einstein condensate.

We employ a novel evaporative cooling scheme realized by tilting the dipole trap. The idea is shown in (a). Trap depth U decreases with time when a magnetic field gradient is applied to the atoms. In (b), we show the magnetic field gradient B'(t) levitates the atoms against gravity and evaporate hot atoms upward.

Performance of our new trap-tilting based forced evaporation.

(a) phase space density,
(b) collision rate, and
(c) particle number
are shown during the evaporation process.

We adopt two evaporation paths:
an efficient, 4 s path (black dots) and
a fast, 1.8 s path (red circles).

The dashed line in (a) shows simple
exponential increase. Our 4 s path outperforms the dashed line, which, together with the increasing collision rate, demonstrate the first runaway evaporative cooling of cold atoms in an optical trap.






We compare the performance of our new evaporation cooling scheme with theoretical models. For all models, we assume an initial collision rate of 133/s and truncation parameter of 6.2~6.8 and negligible collision loss. This set of parameters best simulates our experiment condition.

The performance of our experiment (black dot) is very close to the tilted trap model we formulate assuming 3 dimensional evaporation. Note that all 1D models cannot explain the fast evaporation speed we observe.

Dipole trapping

Our dipole trap is formed by intersecting two laser beams on the horizontal (x-y) plane; both beams are extracted from a single-mode, single frequency Yb fiber laser operating at the wavelength of 1064 nm, frequency offset by 80 MHz, focused to a beam diameter of 540 micron (620 micron) and intensity of 1.9 W (1.6 W) in the y-(x-) direction.

Optical cooling: Molasses and Degenerate Raman Sideband Cooling

Here are three snapshots of ultracold atoms, the yellowish blobs in the images, when they are released in free space. In free space, two things happen: atoms drop downward due to gravity and also expand due to their finite temperature. Here atoms are about 5 micro-Kelvin, after optical molasses cooling.

Using degenerate Raman-sideband cooling, we achieved a much lower temperature. In this figure, we show that the atoms hardly expand within the free fall time of 25 ms. The temperature of these atoms is 470nK. The fine fringes are due to imperfection of our imaging system, which we have fixed now.

Bose-Einstein Condensation of Atomic Cesium

In the same sense that the laser has become a fundamental tool in nonlinear and quantum optics, Bose-Einstein condensation as a source of coherent atoms serves as the starting point for many studies of interacting atomic quantum gases. Bose-condensation of atomic cesium is particularly useful as a first step in studying strongly interacting gases, molecular condensation, few-body interactions, and condensed-matter analog systems showing quantum coherence. At the same time, cesium remained for many years a challenging atom to condense. Our group was the second to achieve this milestone, following the pioneering work achieved in the group of R. Grimm in Innsbruck. Our apparatus is described below.
Chamber Design

The main trapping chamber, on the left side, contains seven 1 1/3" viewports, eight 2 3/4" viewports (six coated for 670~1064 nm light and two ZnSe for 10.6 um) and two 1.77" viewports in the vertical direction. The upper right part of the chamber is the Zeemann slower and the bottom part the main ion pump and the TSP. A special feature of our design is the recessed top and bottom windows, each 1/2" from the atoms, to permit excellent optical access.

Questions? Feel free to email us.



This photo is taken in the same viewing angle as shown above. Inside the aluminum foil (shimmering mess) on the right side is the atomic source oven at 70 degree C and a differential pumping tube at 0 degree C. On the left side, the chamber is surrounded by optical components for laser cooling and trapping atoms. The thick orange cables on the optical table supply current for generating magnetic fields.

Questions? Feel free to email us.

Interactions between ultracold cesium atoms

Questions? Feel free to email us.

Experiment goals and proposal: Click here.