The molecule has a non-classical structure with one hydrogen bonded to both carbon atoms. Therefore, the name vinylium is inappropriate as it refers to the classical structure which has not been detected spectroscopically thus far.
The transition frequencies have been summarized in
(1) M. Cordonnier and L. H. Coudert, 1996, J. Mol. Spectrosc., 178, 59.
Most of them were taken from
(2) M. Bogey, M. Cordonnier, C. Demuynck, and J. L. Destombes, 1992, Astrophys. J., 399, L103.
Since the reported uncertainties in (1) and (2) were considerably too conservative their values have been reduced by factors of 2 to 3.
Additional infrared ground state combination differences were also used in the fit. These data were reported by
(3) C. M. Gabrys, D. Uy, M.-F. Jagod, T. Oka, and T. Amano, 1995, J. Phys. Chem. 99, 15611.
Predictions with uncertainties larger than 500 kHz should be viewed with caution. Moreover, transitions involving K_a = 2 are very uncertain because the data set does not permit D_K to be determined reliably. The origin of the K_a = 2 – 1 branch are uncertain to at least 100 MHz. Transitions with even higher K_a have been omitted from the entry but the levels were considered in the calculation of the partition function.
Note: The molecular ion exists in ortho and para forms with an intensity ratio of 3 : 1. The para levels are described by K_a + K_c being even. The 1₀₁ level is the lowest ortho level and is 2.1885 cm^–1 above ground.
Note further: The ortho levels are split by the internal rotation of the three equivalent H atoms. In most cases, the splitting is expected to be less than about 350 kHz. However, the J = 1 – 1 transition has a splitting of almost 500 kHz, see (1).
The dipole moment was calculated ab initio by
(4) T. J. Lee and H. F. Schaefer III, 1986, J. Chem. Phys. 85, 3437.