From MOT to BEC using a single crossed-wire pair

Abstract

We demonstrate a new magneto-optical trap (MOT) configuration using a simple pair of crossed wires rotated at 45 deg and an appropriate bias field to generate a MOT of >10^8 atoms. The same pair of wires, with slightly adjusted control parameters, is then used to magnetically trap the atoms and cool them via forced evaporative cooling into a Bose-Einstein condensate (BEC) with >10^4 atoms. We present the theoretical framework for generating a quadrupole field using a pair of crossed wires with arbitrary rotation angle, along with the atom chip design and fabrication. Finally, we describe the experimental protocols required for BEC production using only a single crossed-wire atom chip.

Figures (5)

• Figure 1: Representation of the side and top view of the atom chip experiment showing the relative locations of the DBC atom chip, silicon membrane, current-carrying wires, lasers, the atomic flux from a 2D MOT that is not shown, and the 3D MOT. The representation of the side view cross section of the wires on the atom chip is schematic and is intended to show the offset of the MOT from the intersection of the wires. The red cloud represents the approximate location of the MOT relative to the chip. The indicators LHC and RHC represent left-handed and right-handed circular polarization of the laser beams, respectively. In the side view, note the curvature of the membrane due to atmospheric pressure on the outside. The relative thickness of the membrane and the DBC is to scale. In the top view, the atom chip is shown with all of the traces. Only the wires used in the paper are colored; the other wires are shown only in outline. In the zoomed in view, only the wires used for making a MOT are shown. The dashed outline represents the trace that is on the bottom layer of Cu, while the solid outline represents the trace on the top layer. The red cloud represents the approximate $x,~y$ location of the MOT relative to the crossed wires. The vector $\mathbf{r}$ starts in the middle of the DBC stack, and the center of the cross shows the offset of the quadrupole zero from the crossing of the wires.
• Figure 2: Aspect parameter and experimentally measured MOT number as a function of rotation angle. As the rotation angle approaches $45^\circ$, the aspect parameter approaches 1, which is unsuitable for MOT production. Due to this interdependence between angle and aspect parameter, the maximum MOT number is found at an angle of $20^\circ$.
• Figure 3: Currents and bias fields required to produce a MOT at different distances from the chip ($r_z$). The angle for these calculations is set to $20^{\circ}$, but the results are similar for other angles. a) The currents needed in the crossed wires with the distance between wire centers ($T$) set to 828 $\mu$m and 0 $\mu$m. The actual experiment has a chip where $T=828$$\mu$m. The peak required currents range from about 20 A at a distance of -5 mm, to about 2 A at a distance of 1 mm. Such currents are quite feasible for a DBC atom chip. b) Required bias fields are $\mathord{<} 6$ Gauss which is easy to achieve with Helmholtz coils. c) The $(r_x, r_y)$ displacement needed to achieve a $20^{\circ}$ rotated gradient as a function of trap height. As the height decreases, so does the $(r_x, r_y)$ displacement, which allows for easy mode-matching to a magnetic trap upon transfer closer to the atom chip.
• Figure 4: Pictures of atom chip trace cuts. a) A top-down view of the atom chip showing the polynomial wire pattern. b) A top-down view of the atom chip showing the waveguide wire pattern c) A side view of a sectioned atom chip showing the relative size of the wires to the AlN and copper layers, as well as the profile of the laser cut copper traces. The laser cut gap tapers as it moves from the edge of the Cu down to the layer of AlN at the bottom. The copper regions on the chip are considered isolated with the resistance between copper traces $\mathord{>}1\,$M$\Omega$.
• Figure 5: Optical density of the atomic cloud after 10 ms time of flight. The main plot shows the density profile of an x-axis slice through the center of the cloud. The inset shows a picture of the full atomic cloud. A clear double-cloud structure is visible and evidence of partial condensation. From this cloud, we calculate a condensate fraction of 0.25.