Further progress in fitting 13000 torsion-wagging-rotational MW and IR VT = 0,1 transitions in CH3NH2 using the hybrid (tunnelling + internal rotation) program

Hougen, Jon T.

Permalink

https://hdl.handle.net/2142/100597 Copy

Description

Title

Further progress in fitting 13000 torsion-wagging-rotational MW and IR VT = 0,1 transitions in CH3NH2 using the hybrid (tunnelling + internal rotation) program

Author(s)

Hougen, Jon T.

Contributor(s)

Kleiner, Isabelle

Issue Date

2018-06-19

Keyword(s)

Large amplitude motions, internal rotation

Abstract

A few years ago, the authors wrote a hybrid program to fit rotational levels in molecules with one CH3 internal-rotation large-amplitude motion, one NH2 inversion large-amplitude motion, and symmetry described by the G12 PI group. This program was applied with success to the MW spectrum of 2-methylmalonaldehyde, but the rather small data set for this molecule did not provide a stringent test of the model. More challenging is the application of the hybrid program to CH3NH2, since this molecule has a much larger data set, containing both MW and IR transitions, as well as having a more extensive vt, J, and K quantum number coverage. In our ISMS talk this year we will first give an overview of our best least-squares fit to date: The data set contains slightly more than 2500 MW and 11000 IR transitions with J ≤ 32 and K ≤ 14, which are fit to a weighted standard deviation of 1.64 using 71 parameters. Next, we present an assessment of this fit’s strong points (e.g., significantly less parameters, ability to predict spectra in higher torsional states) and weak points (e.g., somewhat larger standard deviation, greater parameter correlation) when compared to the best all-tunneling-model fit in the literature. Based on this assessment, we believe that our fit, as well as the predictive abilities of the program, are sufficiently good that we can now begin considering collaborations with measurement and assignment campaigns of vt = 1 MW data and vt = 2, 3 IR data already underway in other laboratories. Finally, we will present a slightly modified ordering scheme for the operators in this hybrid program, and describe the need for devising a contact transformation treatment to specify determinable parameters in the hybrid Hamiltonian, in order to reduce parameter-correlation problems during the trial-and-error fitting process. A knowledge of determinable parameters would be particularly useful here, since there is little previous experience to guide the choice of “a good set” of higher-order constants to float when carrying out large fits having over 10000 lines, Jmax = 40, Kmax = 15, vt = 0 and 1 torsional states, A1, A2, B1, B2, E1, E2 symmetry species, and nearly 100 parameters.

Publisher

International Symposium on Molecular Spectroscopy

Type of Resource

text

Language

eng

Permalink

http://hdl.handle.net/2142/100597

DOI

10.15278/isms.2018.TK07

Copyright and License Information

Copyright 2018 Jon T. Hougen

Owning Collections