15min:

A. L. KALEDIN, W. G. LAWRENCE, M. C. HEAVEN, Q. CUI, J. E. STEVENS AND K. MOROKUMA, Cherry L. Emerson Center for Scientific Computation and Department of Chemistry, Emory University, Atlanta, GA 30322, U.S.A.

We report a rigorous ab initio study of the ground and low-lying excited-state potential energy surfaces (PES) of the Cl₃ radical at CASSCF, CASPT2, and internally contracted (IC) MRSDCI levels of theory with the Dunning's avdz and avtz basis sets. ( There is a previous study which examined only restricted areas of the PES. ) The ground state PES has two true minima, both of which are van der Waals complexes between Cl and Cl₂. The linear asymmetric minimum is of ² symmetry, and the bent asymmetric minimum is ²A'. At the ICMRSDCI/ avdz level of theory with counterpoise correction for energy, the Jacobi coordinates for the linear minimum are: r=3.90 au, R=8.47 au, D_e(Cl₂(X)-Cl)= 250 cm^-1, and those of the bent minimum are: r=3.90 au, R=6.85 au, \gamma=66.47^o, D_e(Cl₂(X)-Cl)=230 cm^-1. Addition of spin-orbit interaction as a perturbation predicts that the global minimum is linear ²_3/2 stabilized by 280 cm^-1 whereas bent ²E_1/2 is stabilized by only 136 cm^-1.

Excited valence states exhibit only one strongly bound minimum: ²_g with the bond length of 4.67 au at the ICMRSDCI/ avdz level of theory. This state is located 1.1 eV above the van der Waals minima and exhibits a strong (1 au) dipole transition moment to the below-lying ²_u state. It is bound by approximately 5,000 cm^-1 with respect to Cl₂(³_u)-Cl asymptote. This minimum is due to the avoided crossing between two valence bond structures: Cl-Cl Cl and Cl Cl-Cl.

Attempts are made in connection to previous experimental and ab initio studies to explain the on-going hypothesis of a long-lived ground state intermediate formed by the reaction: Cl₂ + Cl \Leftrightarrow Cl₃. According to our ground state calculations this postulate is strongly undermined. In addition, we attempt to explain a 470 nm long-lived laser-induced fluorescence observed after isolating Cl and Cl₂ in Ar matrix. Our calculations do not show any evidence that the trichlorine radical is responsible for this fluorescence. However, the possibilities of Cl₃ playing some role in this process are examined.