Dicke-Narrowed Line Shapes in CO-Ar: Measurements, Calculations, and a Revised Interpretation.

New line shape calculations for CO buffered by Ar are compared to high-resolution measurements from a difference-frequency laser spectrometer, over a range of thermodynamic conditions relevant to the atmosphere. The calculations are based on solving the quantum kinetic (i.e. transport/relaxation) equation for the molecules within the impact approximation, and rely on the commonly used MOLSCAT and MOLCOL codes to determine the speed-dependent collisional relaxation rate. Velocity-changing effects are treated classically using a rigid sphere potential. The comparison initially reveals that the experimental profiles exhibit only 10% to 30% of the expected Dicke narrowing, which leads us to reevaluate our understanding of the narrowing process. A more subtle aspect of the disagreement between theory and experiment draws our attention to an assumption implicit in the calculation of the collisional relaxation rate: the assumption of a Maxwellian form for the velocity dependence of the off-diagonal elements of the density matrix (i.e. the optical coherences). This assumption allows for an analytical simplification of the problem, but eliminates velocity-changing effects (so that they must be added back in using a supplementary classical calculation, which is based here on a rigid sphere interaction). We find that the removal of the above-mentioned assumption should allow for accurate and fully quantum mechanical (but numerical) line shape calculations for systems like CO-Ar on existing computers. © 2006 American Institute of Physics [ABSTRACT FROM AUTHOR]