2.2. Attachment of weakly bound Rydberg electrons - the dissociative charge transfer
It is generally assumed that the nature of electron-molecule interaction does not change when the free electrons are replaced by the weakly bound electrons in Rydberg states. Indeed, for states with high enough main quantum numbers n, the distance between the electron and the ionic core of the Rydberg atom is very large (102-103 Bohr radii). Therefore it is usually assumed that the core has little influence on the transfer process of the Rydberg electron . The collision of the Rydberg atom with the molecule is then considered as an attachment of a quasi-free electron to the molecule, the electron kinetic energy being equal to its (very low) classical kinetic energy on the Rydberg orbit [3, 6]. The role played in this process by the remaining ionic core of the Rydberg atom is, however, often neglected or underestimated and not well understood.
Here we are interested in the relevance of the two different mechanisms through which the dissociative charge (i.e. electron) transfer (DCT) from Rydberg atom to the molecule can take place:
The quasi-free electron attachment model (2a ) assumes the formation of an excited Na2-* ion through capture of the Rydberg electron by Na2, followed by the decay of the negative ion through one of the channels (1a) or (1b ) (see also Fig. 1a). The dynamics of this process is considered to be virtually undisturbed by the distant Na+ core ion of the Rydberg atom apart from the fact that product Na- ion will use part of its kinetic energy to overcome the weak Coulomb attraction with the former. The mechanism (2b), in contrast, fully accounts for the three-body nature of the DCT process. It assumes that Na2 + Na** collision leads to formation of an intermediate highly excited triatomic Na3** collision complex which further decays to yield charged fragments (see Fig. 1b). The existence of such complex was directly evidenced in our recent observations of formation of Na3+ ions in collisions of excited Na atoms and molecules, whereby the Na3+ ions are a product of autoionisation of Na3** [7 ,8].
Simple asymptotic energy considerations predict two different threshold binding energies of the Rydberg electron Eth at which the formation of Na- ions in the processes (2a) and (2b) is still possible. In the quasi-free electron model the kinetic energy released in the dissociation process of Na2-* is determined by the sum of initial vibrational excitation E(v") of the neutral Na2 before the attachment of the electron and the lowering of the electronic energy of Na- with respect to neutral Na (called also the electron affinity - EEA), less the dissociation energy De of Na2. This energy is equally distributed to Na- and Na fragments, whereas Na- alone has to overcome the Coulomb attraction with the core ion, equal to the binding energy of the captured Rydberg electron. Thus, in the quasi-free electron model the threshold binding energy for a given initial vibrational excitation E(v") of Na2 is given by
Eth = 1/2 ( E(v??) + EEA - De) (3a)
In the Na3 ** collision complex model, however, the triatomic molecule decomposes as a whole via a non-adiabatic coupling of the covalent Na3** potential surface with the surface connected asymptotically to Na- + Na + Na+. The threshold binding energy is thus determined by the corresponding asymptotic dissociation energy at a given initial vibrational excitation of Na2:
Ethcompl = ( E(v??) + EEA - De ) (3b)
The experimentally measured Na- ion yield in Na2(v") + Na**(nl) collisions as a function of binding energy of the electron in the nl state is shown in Fig. 2 for three different vibrational levels v" = 13, 14, and 22. The formation of Na- ions beyond the threshold energy given by (3a) clearly shows that the negative ion yield cannot be explained by the quasi-free electron model alone, and the three body effects are non-negligible.
Information on the kinetic energy distribution of the product Na ions would be very helpful in order to identify whether the mechanism (2a) or (2b) dominates the DCT, as well as to get more insight into dynamics of the process. From the above discussion it is clear that the mechanisms (2a) and (2b) will yield Na ions with different kinetic energies. Such energy analysis can be done using the ion imaging technique [9 ]. Additional requirements for implementation of this technique for studies of the Na2(v") + Na**(nl ) system are: (a) the energy resolution should be better than 10 meV (half of the vibrational level spacing of Na2) and (b) the zone where the reaction takes place should be kept free from electric fields to avoid field ionisation of the highly excited Na**(nl) atoms. The former condition allows the use of Franck-Condon pumping (FCP) scheme for vibrational excitation of the molecules (since individual vibrational levels can be distinguished), which is easier to implement than the STIRAP technique. The latter condition ensures that no free electrons due to field ionisation interact with the molecules. The design and test of such ion imaging setup is discussed in the following section.