Abstract
The authors have determined relative associative ionisation (AI) rate constants kAI for a series of Na(3s)+Na*(nl) collisions where 17AI values are scaled to earlier absolute determinations. They compare the behaviour of kAI as a function of n and l with a rigorous application of a model suggested earlier by Janev and Mihajlov (1980). Disagreements between measured and calculated rate constants demonstrate that, even for moderately high n, the spectator Rydberg electron viewpoint for associative ionisation is not a complete picture.