The molecule is nearly T-shaped with the NaN bond slightly shorter than the NaC bond. Therefore, the molecule should be considered more as sodium isocyanide rather than as sodium cyanide.
The first entry from Apr. 2007 has been revised based on an analysis from
(1) H. S. P. Müller, D. T. Halfen, and L. M. Ziurys, 2012, J. Mol. Spectrosc. 272, 23.
The experimental measurements were taken from
(2) J. J. van Vaals, W. L. Meerts, and A. Dymanus, 1984, Chem. Phys. 86, 147;
and from
(3) D. T. Halfen and L. M. Ziurys, 2011, Astrophys. J. 730, Art. No. 107.
As described in (1), the reported uncertainties in (2) were assumed to be 3 σ values. Uncertainties for the lower K_a transition frequencies from (3) were probably judged somewhat conservatively. On the other hand, three transition frequencies from (3) were omitted fromthe final fit. The predicted transition frequencies should be fairly reliable as long as their uncertainties do not exceed 1 MHz.
While in the laboratory b-type transitions have been observed, their intensities are likely very small because the b dipole moment component is about 1.5 orders of magnitude smaller than the a component; hence, transitions will be about three orders of magnitude weaker.
Hyperfine structure splitting caused by the ¹⁴N nucleus or the ²³Na nucleus are not of relevance for astronomical observations.
The ab initio dipole moment was taken from
(4) H. S. P. Müller, 2007, unpublished.
Note: in the original detection paper, the dipole moment from an ab initio paper was taken as being in units of Debye, whereas it was given in atomic units; in Debye, the dipole moment is larger by a factor of 2.542 !