The data are based on a combined fit by
(1) H. S. P. Müller, 2023, unpublished.
With respect to the second entry of Jan. 2017, hyperfine-resolved transition frequencies in the ground vibrational state were added to the data set. These were reported by
(2) O. Asvany, C. R. Markus, A. Roucou, S. Schlemmer, S. Thorwirth, and C. Lauzin, 2021, J. Mol. Spectrosc. 378, Art. No. 111447.
With respect to the first entry of June 2006, we have adjusted some of the uncertainties, provide calculations for v = 1, and included fairly recent submillimeter data from
(3) J. Cernicharo, S. Bailleux, E. Alekseev, A. Fuente, E. Roueff, M. Gerin, B. Tercero, S. P. Treviño-Morales, N. Marcelino, R. Bachiller, and B. Lefloch, 2014, Astrophys. J. 795, Art. No. 40.
We have attributed 20 kHz as uncertainties to these data in the present calculations. This may still be somewhat conservative. Infrared transition frequencies from
(4) W. C. Ho, I. Ozier, D. T. Cramb, and M. C. L. Gerry, 1991, J. Mol. Spectrosc. 149, 559;
and from
(5) M. López-Puertas, J.-M. Flaud, J. Peralta-Calvillo, B. Funke, and S. Gil-López, 2006, J. Mol. Spectrosc. 237, 218
were also used in the fit.
The ground state calculations should be reliable up to at least 2.0 THz; the v = 1 data are essentially unaffected by the new data and should be viewed with some caution throughout.
Quantum-chemical calculations of the dipole moments were taken from
(6) R. Polák and R. Fiser, 2004, Chem. Phys. 303, 73.
The ¹⁴N hyperfine splitting may be resolvable at lower frequencies or at lower quantum numbers. Therefore, a separate calculation with hyperfine structure is available up to 480 GHz and with J" up to 3. The spin-multiplicity of the ¹⁴N nucleus was considered in the calculation of the partition function. Contributions from the first excited vibrational state to the partition function were considered also, but are essentially negligible at 300 K.