Search

The shapes of Dicke-narrowed spectral lines in the fundamental P -branch of CO in Ar are studied by comparing high-resolution measurements and theoretical calculations. The measured spectra were recorded at temperatures between 214 and 324 K, and at pressures between 0.025 and 1 atm. The calculations are based on solving a transport/relaxation equation for the appropriate off-diagonal element of the density matrix; they use a realistic intermolecular potential to determine the speed-dependent collisional broadening, and a rigid sphere potential to determine the Dicke narrowing. It is found that the calculations can reproduce the measured spectra within the experimental noise under all conditions, but that the magnitude of the Dicke narrowing in the measured spectra is 70–90% less than predicted from the mass diffusion constant. A revised view of the collision operator resolves the discrepancy in principle, and leads to a better understanding of the line shape problem in general.

The collisional shifts and widths of several P -branch spectral lines in the fundamental band of CO–Ar have been measured at temperatures between 214 and 324 K and pressures between 0.025 and 1 atm. The widths have been determined using a line shape model based on the solution of the transport/relaxation equation for the appropriate off-diagonal element of the density matrix. The model uses a realistic molecular potential energy surface to calculate the speed dependence of the collisional broadening, and a rigid sphere potential to calculate the translational motion. It is found that both the shifting and broadening coefficients follow a power law dependence on the temperature. Additionally, it is demonstrated that studies have tended to overestimate the accuracy of collisional widths when the line shape model used to obtain the widths involves multiple fitted line shape parameters or fails to fit the measured spectra within the experimental noise.

The 03 1 0 ← 01 1 0 parallel Q branch of N 2 O has been studied at 297 K and over the pressure range 1–130 torr. Absorption spectra were recorded using a high resolution (1.5 MHz or 5 × 10 −5 cm −1 ) and high signal-to-noise (>3500:1) mid-infrared spectrometer based on difference-frequency infrared generation in AgGaS 2 . In the low-pressure range (1–11 torr) we obtained accurate values for the line strengths, the broadening coefficients, the weak mixing coefficients, and the overall shifting of the branch. The medium pressure results, ranging from 23 to 130 torr, were analyzed by treating the band as a whole, using a relaxation matrix formalism, based on an energy gap scaling law. We find, effectively, that only 36% of the rotationally inelastic collisions are associated with Q branch mixing, the rest presumably being associated with Q – P and Q – R mixing in the same vibrational band. The pressure shifting coefficient of the 03 1 0 ← 01 1 0 Q branch as a whole was also determined and found to be 5.8 × 10 −3 cm −1 /atm towards lower frequencies.

The publisher regrets that some corrections requested by the author at the proof stage were not made. On page 57, the first full sentence under Eq. (10) should read: ‘‘Unlike Eq. (8), Eq. (9) leads to a line shape with the correct behaviour in the limit as the velocity-changing collision rate goes to zero.’’ On page 58, lines 9 and 10 under the Section 3 heading should read: ‘‘. . . (see, for example, [4, 38, 42, 43, 49, 51, 52]).. . .’’