ON THE USE OF DISTORTED FRANCK-CONDON FACTORS IN THE INTERPRETATION OF NEUTRALIZATION-REIONIZATION EXPERIMENTS

Neutralization by collision with an appropriate target gas is a vertical process. However, it takes place between two deformed potential energy surfaces, i.e., that of a polyatomic ion perturbed by an atomic collision gas (Hg or Xe) and that of the neutral polarized by the receding Hg+ or Xe+ ion. Hence, it is argued that the Franck-Condon factors which determine the yield of production of the neutral do not involve the overlap integral between the vibrational wavefunctions of the free species (ion and neutral). Instead, the quantity to be considered is |<0|φ0p(τ)≫|2, where |0≫ is the vibrational wavefunction of the unperturbed neutral, τ is the time during which the newly created polyatomic neutral is effectively polarized by the departing Hg+ or Xe+ ion, and |φ0p≫ the vibrational wavefunction of the polarized neutral. A sequence of two consecutive relaxation processes induced by the neutralizing collision can, at least in favourable cases, bring about an increase in the yield of production of a marginally stable radical. These ideas have been applied to the controversial stability of the chloronium ylide. MNDO calculations reveal that the neutralization step is well represented by the Demkov model. This implies that the yield of neutralization is controlled not only by the energy difference Δ between the ionization energy of the neutralizing gas (Hg, Xe, Zn, Na) and the recombination energy of the polyatomic ion, as usually assumed, but also by an additional coupling element H12 whose value depends on the nature of the neutralizing gas. © 1990.