The entry was reevaluated with respect to the first entry from Jan. 2017 as briefly described below. The rotational transition frequencies were taken from
(1) T. Sakaizumi, H. Mure, O. Ohashi, and I. Yamaguchi, 1990, J. Mol. Spectrosc. 140, 62;
and from
(2) L. Kolesniková, J. L. Alonso, C. Bermúdez, E. R. Alonso, B. Tercero, J. Cernicharo, and J.-C. Guillemin, 2016, Astron. Astrophys. 591, Art. No. A75.
The blended line option, available also in erham, was used in the present treatment. Two lines were added as components of blends. Only two lines were omitted whose residuals exceeded 0.4 MHz, some lines with residuals around 0.3 MHz were retained in the present calculation. Data from (1) were used in the present evaluation, as earlier, but in contrast to (2). The parameter ε₂ and its distortion corrections were omitted. Three fewer parameters than in (2) were sufficient to achieve a fit of superior quality than that in (2). The experimental lines were not merged, but the can be inspected in the archive directory. The molecules displays commonly splitting caused by methyl internal rotation; state number 0 signals the A symmetry component, 1 the E symmetry component. Calculations with uncertainties exceeding 0.3 MHz should be viewed with substantial caution.
The dipole moment components were determined in (1). However, b-type transitions can not be predicted reliably at present and have been omitted.