The molecule displays internal rotation of the two non-equivalent methyl groups. The experimental data were reproduced quite well with the erham program. The present analysis is based on the data from
(1) L. Bonah, O. Zingsheim, H. S. P. Müller, J.-C. Guillemin, F. Lewen, and S. Schlemmer, 2022, J. Mol. Spectrosc. 388, Art. No. 111674.
The data are predominantly from that study. Lower MW data were taken from
(2) F. J. Lovas and P. Groner, 2006, J. Mol. Spectrosc. 236, 173.
The analysis is the one extending to 1090 GHz. The calculations should be sufficiently accurate for all observational purposes. Transitions with calculated uncertainties exceeding 0.1 MHz should be viewed with some caution. Please note that the merged transitions refer to measured transition frequencies, but the uncertainties are the calculated ones.
Please note also that erham uses two σ values to label each symmetry component. This is incompatible with the VAMDC-CDMS requirements and does not agree with the convention of the classical CDMS to use one state number to label one symmetry species (or one vibrational state etc.). The following relations exist:
0 / 0 0 / A / AA,
1 / 0 1 / E1 / EA,
2 / 1 0 / E2 / AE,
3 / 1 1 / E3 / EE1,
4 / 1 2 / E4 / EE2.
Please note also that the partition function refers to the ground vibrational state. Vibrational correction factors for a posteriori corrections were derived in (1) in the harmonic approximation.
The dipole moment was assumed to be the same as for the main isotopic species which was determined by
(3) R. Peter and H. Dreizler, 1965, Z. Naturforsch. A 20, 301.