Knowledge Bank

An emission system first observed over a hundred years ago by Schuster in various DC discharges in flowing $NH3$ has been restudied at the NRC in Ottawa, where mixed isotopic spectra using both D and $\mathrm{<mrow\; data-mjx-texclass="ORD">15</mrow>}N$ allowed the carrier of the spectrum to be positively $identified1$ as $NH4$. The results of ab initio $calculations2$ and a preliminary rotational $analysis1$ led to a provisional assignment of one hand as a $(3d)2F2-(3s)2A1$ Rydberg transition, even though such a transition would be parity forbidden in the united atom limit. The present work attempts to place the original rotational analysis on a more quantitative basis. A computer program at l'Universite de Dijon, France, originally $developed3$ for vibrational transitions in tatrahedral molecules, was used to carry out a least squares fit of the rotational structure of the observed electronic transition in $NH4$. Group theoretical arguments alone permit the Dijon rotation-vibration Hamiltonian to be used as a phenomenological rotation-electronic Hamiltonian. However, while the precise meaning of the various scalar and tensor operators in the Dijon $formalism3$ is well understood for the rotation-vibration problem in $CH4$, the significance of these parameters for the rotation-electronic problem in $NH4$ has yet to be determined. Various theoretical points and the status of the least squares fit will be discussed.