In the process of exploring different ramifications of the laser-focused
atomic deposition process, we have undertaken some studies
[EPG Pub#660] that utilize the deposition of
chromium atoms as a tool for investigation of the spatial distributions of
atoms in a light field.
Motivated by the possibility of increasing the density of lines in a
deposition, we have studied the behavior of chromium atoms depositing through
an optical field made of counterpropagating beams with orthogonal polarizations
("lin x lin" configuration). With this polarization
configuration, the light field can be thought of as being made up of two
standing waves, one of right-hand-circular polarization and the other left.
These standing waves are offset from each other by Lambda/4, resulting in an
average intensity that is uniform, but a polarization that is spatially
varying.
Upon performing the experiment, we discovered that the deposition pattern
had a clear Lambda/8 component. The explanation of this high spatial frequency
component has led us to a detailed analysis of the behavior of Cr atoms in a
lin x lin field and of the critical role played by Raman-induced
avoided crossings in the adiabatic optical potentials.

AFM image of laser-focused atomic deposition of Cr in a lin x lin field,
showing clear Lambda/8 spatial frequency component.
A basic understanding of the phenomenon can be had by examining the
adiabatic potentials for a chromium atom in an intense laser field. There are
seven of these potentials because there are seven magnetic sublevels in the
ground state of Cr. However, the potentials do not correspond to the
individual magnetic states because the states are coupled by Raman couplings
that arise from both left- and right-handed circular polarization being present
in the laser light. Each potential corresponds to a linear combination of
magnetic sublevels that results from a diagonalization of the Hamiltonian
including these Raman couplings.
Examination of the potentials shows where the Lambda/8 spatial frequency comes
from. At every place where the "un-diagonalized," or diabatic, energy
levels would have a crossing, the
adiabatic potentials have an avoided
crossing that acts as a local potential minimum and concentrates the atoms.
While this picture is essentially correct, a full analysis involving a
quantum Monte Carlo calculation shows that dynamical effects also play an
important role. The motion of the atoms in the potential is fast enough that
non-adiabatic transitions can and do occur in the avoided crossings, leading to
a somewhat more complicated picture. Nevertheless, when all dynamics are taken
into account, good qualitative agreement is found between calculations and
experiment.
![]()
Jabez J. McClelland - NIST
Robert J. Celotta - NIST
![]()
Rajeev Gupta
![]()
Peter Marte - Harvard University
Online: September 1996
Last Updated: February 2008
Website Comments:egpwebmaster@nist.gov