DEPENDENCE OF ROTATIONAL RELAXATION RATES ON THE MOLECULAR SPEED, M LEVEL, AND M DEGENERACY

Author Institution: Department of Chemistry, Harvard UniversitySeveral time-resolved experiments have been performed on the $J = 0-1$ and $J = 1-2$ transitions of OCS and on the $J = 0-1$ transitions of $CH_{3}CI$. The analytical evaluation of decay curves due to wall-collisions and the Doppler effect developed in this work allows an analysis of non-exponential collision-induced relaxation. The speed dependence of $T_{2}$ relaxation rates and of some $T_{1}$ rates has been determined to first order by assuming a relaxation rate changing linearly with the speed, including in the analysis the resultant changes in wall-collision and Doppler effects due to changes in the average speed of the monitored population. Non-zero speed dependences imply non-Lorentzian pressure-broadened lineshapes (observed in $OCS^{1}$, OCS $T_{2}$ and $T_{1}$ decays have a significant speed dependence, reflecting contributions from non- (dipole-dipole) processes. $CH_{3}C1$ $T_{2}$ decays for the $J = 0-1 F = 3/2-3/2$ transition show a zero speed dependence. In OCS, the results of two 2-level $T_{1}$ experiments and of one 3-level double resonance experiment imply that the $T_{1}$ rates for the three individual levels ($M = 0, J = 0, 1,$ and 2) are identical. Therefore, the $T_{1}$ decays for these three experiments should be exponential in the absence of a speed dependence. The $T_{1}$ speed dependence is significant but smaller than the $T_{2}$ speed dependence for these transitions. We have also found in the OCS $J = 1-2$ that $T_{1}(M = 0)$ is shorter than $T_{2}(M = \pm 1)$, which implies the occurrence of significant $\bigtriangleup J = 0, \Delta M = \pm~ 2$ transitions. Also, for all OCS $T_{1}$ times, $T_{1}$ (Stark voltage on) is longer than $T_{1}$ (Stark voltage off). That is, reducing the number of level degeneracies reduces the collision $efficiency.^{1}$R. A. Creswell, S. R. Brown, and R. H. Schwendeman, J. Chem. Phys. 64, 1820 (1976)