Rotational transition frequencies were reported by
(1) F. C. Wyse, W. Gordy, and E. F. Pearson, 1970, J. Chem. Phys. 52, 3887;
and by
(2) J. Hoeft, F. J. Lovas, E. Tiemann, and T. Törring, 1970, Z. Naturforsch. 25a, 1029.
Also used in the model were rovibrational transition frequencies from
(3) A. G. Maki and F. J. Lovas, 1982, J. Mol. Spectrosc. 95, 80;
and from
(4) H. G. Hedderich and P. F. Bernath, 1992, J. Mol. Spectrosc. 153, 73.
The highest-J ground state transition of (1) was omitted because of a large deviation between measured frequency and that calculted from the final spectroscopic parameters.
The calculated transitions should be accurate enough for all observational purposes. Some caution may be advised for those at high J or high v. Al hyperfine splitting may be resolvable at very low J, but these transitions are likely too weak to be observable.
The first 10 vibrational states were included in the calculation of partition function values. These may be slightly too low at the highest temperatures, but still higher than classically calculated values.
The dipole moment from a quantum-chemical calculation was reported by
(5) M. Yousefi and P. F. Bernath, 2018, Astrophys. J. Suppl. Ser. 273, Art. No. 8.
An experimental ground state value is compatible within the uncertainties.