Toward the laboratory identification of the not-so-simple NS2 neutral and anion isomers

Russell Thackston
Ryan Fortenberry, Georgia Southern University
Joseph Francisco, Purdue University
Timothy Lee, NASA Ames Research Center

Abstract

The NS2 radical is a simple arrangement of atoms with a complex electronic structure. This molecule was first reported by Hassanzadeh and Andrew’s group [J. Am. Chem. Soc. 114, 83 (1992)] through Ar matrix isolation experiments. In the quarter century since this seminal work was published, almost nothing has been reported about nitrogen disulfide even though NS2 is isovalent with the common NO2. The present study aims to shed new insight into possible challenges with the characterization of this radical. No less than three potential energy surfaces all intersect in the C2vC2v region of the SNS radical isomer. A type-C Renner-Teller molecule is present for the linear 2ΠuΠu state where the potential energy surface is fully contained within the 2.05 kcal/mol lower energy X̃ 2A1X̃2A1 state. A C2vC2v, 12B112B1 state is present in this same region, but a double excitation is required to access this state from the X̃ 2A1X̃2A1 state of SNS. Additionally, a 12A′12A′ NSS isomer is also present but with notable differences in the geometry from the global minimum. Consequently, the rovibronic spectrum of these NS2isomers is quite complicated. While the present theory and previous Ar matrix experiments agree well on isotopic shifts, they differ notably for the absolute fundamental vibrational frequency transitions. These differences are likely a combination of matrix shifts and issues associated with the neglect of non-adiabatic coupling in the computations. In either case, it is clear that high-resolution gas phase experimental observations will be complicated to sort. The present computations should aid in their analysis.