The Vision
Because atom traps are typically no deeper than a few millikelvin, atoms must be pre-cooled in order to be trapped. Laser cooling and cryogenic pre-cooling have both been employed. In Bose-Einstein condensation, the energy is further reduced by two orders of magnitude through evaporation, and the atoms achieve temperatures in the nanokelvin regime. At such ultracold temperatures, new types of atom traps and new ways to manipulate atoms become possible, as was dramatically demonstrated by the optical trapping of a Bose-Einstein condensate with just a few mW of far-off resonant infrared laser power (in contrast to watts needed for laser-cooled atoms) [1]. This suggests a bright future for the manipulation of atoms. A vision emerges in which atoms are trapped, guided, focused and made to interfere using a complex pattern of magnetic fields and light fields. The pattern can be precisely controlled and easily modified by microfabrication techniques. This suggests the possibility of a new subfield: surface atom optics with goals such as the “quantum engineering” of new coherent sources and rugged atom interferometers on a chip. Such advances would open new scientific possibilities:

• guided structures that lead to lower-dimensional systems, possibly giving access to new types of phase transitions and low-dimensional atomic physics
• new classes of atomic surface probes based on the proximity of the atoms to the surface
• guided structures that make it possible to study fundamental phenomena such as the Casimir force and quantum reflection
• micro-arrays of atoms that could be valuable for quantum computation

Our proposed program combines three frontier areas of atomic physics, which all heavily depend on technological advances. It is based on a new collaboration of groups which have done pioneering work in novel atom sources (atom lasers, Wolfgang Ketterle), atom optics and interferometry (David Pritchard) and in microfabrication, atom waveguides and atom lithography (Mara Prentiss).

Proposed Research (in 2000)
.
BEC SOURCE DEVELOPMENT
Atom beams extracted from Bose-Einstein condensates are orders of magnitude brighter than beams from conventional atom sources, although all atom lasers so far have operated in a pulsed mode with a small duty cycle. By developing rugged, simple to use, high intensity atom lasers, major advances would become possible in atom optics and atom interferometry, and the way would be open to applications in lithography and metrology. Such tasks are ideally suited to our collaborative effort, where existing know-how can be combined and the testing and debugging of different strategies can be carried out in parallel.

The potential for improving existing BEC sources is high: Laser cooling can provide more than 10E11 slow atoms per second [2] and evaporative cooling can take more than 0.1% of these to BEC. This implies possible fluxes of 10E8 to 10E9 coherent atoms per second, orders of magnitude more than the current value of 10E6/s. The atomic species of choice is rubidium because the diode lasers are convenient, making it easier to duplicate the technology elsewhere. The goal is to pool know-how among the Center PIs and outside collaborators. The design of the new experiments will be documented and made available to the scientific community through a web site as well as technical publications. But the most important avenue for disseminating the know-how will be our Visitors Program.

BEC MANIPULATION
All of the applications we envision require improved methods for manipulating Bose condensates. We propose to combine the rubidium BEC source with a transport mechanism to extract the condensate into a separate “science chamber" with vacuum locks to facilitate insertion of waveguides, substrates, and other objects. Since it generally takes much longer to condense the atoms than to carry out an experiment, the vacuum requirements for the science chamber are less severe. The condensate can be transported using all-optical confinement of a Bose condensate [1], employing an infrared laser trap that forms “optical tweezers" for BEC in which the condensate is confined at the focus of the laser beam. This trap uses far-off resonant laser light, thus eliminating spontaneous scattering as a heating mechanism. Atom transport is accomplished by translating the laser beam focus in an adiabatic fashion [3].

Two-chamber concept. The condensate is produced in a cooling chamber, which has to accommodate several laser beams for optical cooling and probing and the magnets for magnetic trapping. An infrared laser acts as optical tweezers and extracts the condensate into a science chamber where free and flexible access to the condensate is available.

ATOM WAVEGUIDES
Developing single-mode atom waveguides and atom couplers is an essential goal in our program in surface atom physics. A number of geometries for guiding atoms have been suggested and demonstrated with guides constituted of some combination of electric fields, magnetic fields, and laser fields. Single-mode waveguides are essential elements of surface atom optics, for they make it possible to transport coherent atoms and also constitute building blocks for surface atom interferometers.

ATOM INTERFEROMETRY
BECs and atom laser beams are ideal sources of coherent atoms for atom optics and interferometry. Advantages of BEC sources include extreme monochromaticity, the small spreading of the BEC cloud, and the ease by which extremely cold atoms can be manipulated. However, BEC offers more for atom optics than simply a supply of very cold atoms: it can help to realize qualitatively new aspects of atom optics. For example, we have recently demonstrated an “active matter wave interferometer" in which atom waves are amplified phase-coherently by passing them though a condensate illuminated by a laser beam [4]. The creation of non-classical and entangled atomic states can lead to interferometry with noise below the shot noise limit. New forms of multi-path interferometry can further enhance the sensitivity. All these prospects add to the excitement of interferometry with BECs.

Atom interferometers have been employed in a gyroscope that has displayed a sensitivity which is superior to that of the best laser gyroscope [5]. Further advances are possible. We propose to investigate new types of atom couplers, which are key elements of atom interferometers. These are based on coherent coupling between closely spaced potential minima associated with pairs of atom waveguides. Such couplers could make possible double-loop atom interferometers for measuring rotation. Atoms enter a single four-way coupler that splits them into two counter-propagating beams, which propagate along a waveguide loop that returns to the coupler. After traversing the loop, the atoms recombine and propagate out of the two output ports of the device. This same geometry is used in optical laser gyroscopes, and eliminates major sources of systematic errors.

Waveguide interferometer for measuring rotation. A single mode atomic beam (1) is split into two beams (2 and 3) which counter propagate through a waveguide loop. When these two beams enter the coupler again, they are again split, resulting in two output beams (4 and 5) which show interference fringes. The use of counter propagating beams within the interferometer eliminates the effect of external fluctuations.

SURFACE PHYSICS WITH ULTRACOLD ATOMS
Surface atom optics opens the way to fundamental studies of atom-surface interactions and offers a new class of lower dimensional systems for studies in condensed matter physics. The unique feature of atoms that are guided on surfaces is their long dwell time at precisely controlled fixed distance. Atoms which are more than 10 microns from a surface experience a Casimir-Polder interaction energy whose phase shift should be measurable by atom interferometry, thus extending such studies to distances almost an order of magnitude further than in previous measurements of this force. At this long distance the Casimir-Polder interaction has a negligible van der Waals component: it is entirely retarded. At closer distances quantum reflection, which has been observed only in the hydrogen-helium system, will be explored for other systems where theory is not able to predict how the reflection coefficient varies with energy. As the atoms approach even closer to the surface they enter a regime where surface roughness and other surface properties should be of dominant importance, with thermal excitations in the surface possibly causing novel forms of quantum decoherence.

MANY-BODY PHYSICS IN LOWER DIMENSIONS
A scientific frontier waiting to be opened by surface atomic physics is the study of behavior of many-body systems in reduced dimensions. Confinement within a 1D waveguide whose dimensions are comparable with the scattering length can lead to dramatic changes of behavior. For example, bosons can scatter so strongly that they become effectively impenetrable. This would constitute an atomic realization of a Tonks gas whose elementary excitations are counter intuitive: they behave according to Fermi statistics. Another interesting 1D phenomenon, based on quantum point contacts, involves quantization of the flux through a narrow constriction at energies comparable to the excitation of transverse modes in the constriction [6].

Two-dimensional BECs are forbidden in homogeneous two-dimensional systems, but even a weak trapping potential can overcome this limitation. Moreover, quasi-condensates can exist in homogeneous 2D systems. The ability to make, manipulate (for instance by a laser-written spatially dependent phase), and precisely study low dimension systems offers interesting studies for condensed matter physics. Relevant issues include the role of single scattering events (e.g. impurities), blockades in the transport of atoms across a region of the waveguide perturbed by arrays of interacting atoms (atom quantum dots), and searches for new types of phase transitions, e.g. topological phase transitions.

1. D.M. Stamper-Kurn, M.R. Andrews, A.P. Chikkatur, S. Inouye, H.-J. Miesner, J. Stenger, and W. Ketterle, Phys. Rev. Lett. 80, 2027 (1998).
2. M.A. Joffe, W. Ketterle, A. Martin, and D.E. Pritchard, J. Opt. Soc. Am. B 10, 2257 (1993).
3. T.L. Gustavson, A.P. Chikkatur, A.E. Leanhardt, A. Görlitz, S. Gupta, D.E. Pritchard, and W. Ketterle, Phys. Rev. Lett. 88, 020401 (2002).
4. S. Inouye, T. Pfau, S. Gupta, A.P. Chikkatur, A. Görlitz, D.E. Pritchard, and W. Ketterle, Nature 402, 641 (1999).
5. T.L. Gustavson, P. Bouyer, and M.A. Kasevich, Phys. Rev. Lett. 78, 2046 (1997).
6. J.H. Thywissen, R.M. Westervelt, and M. Prentiss, Phys. Rev. Lett. 83, 3762 (1999).