Abstract

Methanol data for ν t = 0 (below the barrier) and ν t = 1 (straddling the barrier) have been treated, using a program based on the formalism of Herbst et al. ( J. Mol. Spectrosc. 108, 42-57, 1984). Altogether, 909 microwave lines (mostly assigned a 50 kHz measurement uncertainty) and 197 tunable far-infrared lines (mostly assigned 100 or 200 kHz uncertainties) in the range K ≤ 12, J ≤ 20 and 5549 Fourier-transform far-infrared lines (assigned an uncertainty of 0.0002 cm −1 = 6 MHz) in the range K ≤ 14, J ≤ 20 have been globally fit to give a unitless weighted standard deviation of 1.032, using 56 adjustable and 8 fixed parameters. Some difficulties remain in fitting high K and J b-type tunable far-infrared lines. Although these difficulties may arise from deficiencies in the model, it also seems possible that some measurement problems still remain in the tunable far-infrared data set. Apart from the high K and J difficulties with this latter data set, a very satisfactory fit has been achieved, leading us to the conclusion that up through the first excited torsional level, a traditional one-dimensional internal rotation model is capable of accurately describing methanol energy levels, without consideration of other large-amplitude-motion effects, such as those arising, for example, when the OH bending vibration carries the molecule to a C 3ν configuration and the internal rotation barrier vanishes.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call