Approximations based on the adiabatic treatment of rotation for resonances

In the adiabatic treatment of overall rotational motion, the rotational energy is obtained by diagonalization of the inertia tensor at each nuclear configuration, and subsequent insertion of the rotation constants into the standard formalism for the energy for a symmetric or asymmetric top. We have tested this approximation previously for bound states and resonances in HCO, and found it to be quite accurate. This adiabatic approximation is justified here by deriving an approximation very similar to it (but less accurate) for a triatomic molecule. We then consider further approximations to the adiabatic rotation approximation. In one we assume that rotation constants for each resonance are independent of the angular momentum state J. This approximation requires a minimum of two calculations of resonance positions and widths for nonzero J in addition to the one for J=0. The second approximation we consider is standard first-order perturbarion theory. The adiabatic rotational energy is the perturbation relative to the J=0 Hamiltonian, and the complex L2 eigenfunctions of this Hamiltonian are the zero-order states. These two approximations are tested for HCO bound states and resonances, where those obtained from the full adiabatic rotation approximation are assumed to be the benchmark calculations.