Wavepackets in Optical Lattices Investigators: Paul Rudy, Renato Ejnisman and Nicholas P. Bigelow.

Where the laser beams responsible for the atomic cooling intercept, interference fringes are produced. The atoms tend to accumulate in the regions of maxima or minima of the interference fringes, depending on the laser frequency (for frequencies above resonance, the atoms accumulate at light minima and vice-versa). Since the interference pattern is three dimensional, the atoms will sit in regularly spaced wells which form a lattice: the so-called Optical Lattice. It shares many of the properties of crystal lattices, as shown in several recent experiments.

However, optical lattices have an important advantage over crystal lattices: one can control the intensity and frequency of the light to change the parameters of the lattice - something not readily realizable in a crystal lattice. In this project, we take advantage of this feature to change the shape of the wells that form the optical lattice. We compress or expand the wells over three different time scales: the sudden (change is much faster than the atomic oscillation in the well), the adiabatic (change is slow compared to the oscillation) and the intermediate case.

The laser beams form interference fringes creating regularly spaced wells where the atoms sit. By changing the shape of these wells, we can induce oscillations in the atomic distribution that can be observed through fluorescence measurements.

After the well is changed, we can observe the effect on the atoms by observing their fluorescence. For example, for red detuned lattices (frequency of the light tuned below resonance), the atoms accumulate in the light maxima, so that the closer they are to the bottom of the well, the more light they "see" and the more they fluoresce. Hence, by measuring the fluorescence, we can extract the atoms' spatial distribution. In the figures below, we show the expected atomic distribution for a given set of parameters (left) , along with the observed result (right) under those conditions, showing a remarkable agreement.

Solving Heisenberg's equations of motion, we can predict how the atomic distribution will evolve in time after the change in the potential well (left). This prediction is in very good agreement with the experimental atomic distribution (right) that can be extracted from our fluorescence measurements.

This work opens the door to exciting prospects: by choosing an appropriate lattice well variation, one can precisely tailor the final state wavepacket creating a desired final atomic distribution [see I.Averbuch and M.Shapiro, PRA 47, 5086 (1993)].

Latest Project News: