Ion pair formation in alkali-halogen collisions at high velocities, ch. 3

In a molecular beam apparatus, ion pair formation in alkali atom-halogen molecule collisions has been studied. Total cross-sections for negative ion production have been measured in collisions of Na, K and Cs with C12, Br2, I2, IC1 and IBr at relative velocities from threshold to 2 x 105 m/s. In all total cross-sections, a maximum at about 8 x 103 m/s is observed. In some systems a second maximum at 9 x 104 m/s is observed, the value of which is approximately half the first one. The fractions of atomic ions of the total halogen ion yield are also determined. They strongly depend on the specific halogen molecule and are less sensitive to the velocity and to the alkali atom. In case of heteronuclear halogens IX the I- fractions show a rapid decrease at 7 x 103 m/s, falling off to zero at higher velocities. Data are given in graphs. The results are interpreted in terms of a two state model. The transition probability is calculated using the Landau-Zener formula, attention being paid to the orientational dependence of the coupling matrix element H12 and to the increase of the crossing radius Rsub(c) due to stretching of the XY- bond during the collision. In the calculations performed for the higher velocities, only the groundstate of XY- is considered. The maximum of the total cross-section at low velocity is attributed to bondstretching while the maximum at high velocity is considered to be a Landau-Zener maximum. If no second maximum is observed, it is shown that it is concealed due to bondstretching. From the values of the fraction we have deduced the vertical electron affinity of XY (Cl2: 1.02 eV, Br2: 1.47 eV, I2: 1.72 eV, ICll: 1.48 eV and IBr: 1.62 eV). For the Na-XY systems, accurate values of the coupling matrix element H12 are reported (Na-Br2: 0.42 eV, Na-I2: 0.40 eV, results are presented and compared with theoretical predictions on the relation between H12 and Rsub(c). It is shown that the symmetric expression of H12 in terms of the ionization potential and the electron affinity as given by the relation of Olson et al is not valid