The study of collisions between nuclear particles has developed to a remarkable extent with the discovery of the neutron and the introduction of artificial methods for effecting nuclear disintegration. It has been found in the last few years that the interpretation of the observed results is by no means as simple as was first expected. This situation is most apparent when the explanation of the variation of probability of capture of slow neutrons by different nuclei is considered. This probability varies in a very irregular manner from element to element and pronounced selective effects occur in certain cases. Attempts to explain (Elsasser and Perrin 1935; Bethe 1935) these resonance phenomena in terms of the usual approximations of quantum collision theory were soon found to be inadequate, All such attempts were based on the assumption that the Chance of a nuclear collision being elastic is high compared with that of its resulting in capture or excitation. A high probability of capture (with emission of radiation) or excitation could then only appear together with a high probability of elastic collision and this is frequently contradicted by the experimental results. The sharpness of the observed resonance phenomena was also difficult to understand on this basis. It was first pointed out by Bohr (1936) that the initial assumptions concerning the probability of elastic collisions, virtually involving the treatment of the elastic scattering as a one-body problem in the first approximation, cannot be valid for nuclei in which the particles, even is existing separately in the nuclei at all, are so closely packed. On making a close collision with a nucleus a particle, such as an
α
-particle, neutron or proton, comes into close and strong interaction with a number of nuclear particles and its incident energy becomes distributed among them. It is only when a particular particle receives sufficient energy to leave the quasi-stable complex formed that a disintegration particle is emitted. (This may of course be the original incident particle, in which case the collision would be an elastic or excitation one.) Otherwise the surplus energy is emitted as radiation. The resonance phenomena arise from the energy levels of the quasi-stable complex. If the incident energy is such that the total energy is equal or nearly equal to that of one of these energy levels, the range of interaction and hence the collision cross-section is quite large. This point of view must be adopted not only when dealing with neutron collisions but in all cases in which the impinging particle does not possess an energy greatly in excess of the minimum necessary for the process to occur. Disintegrations produced by charged particles, in which resonance effects have been observed for some time (Feather 1937, P. 154), must therefore be capable of description in this way.
Formation of HCN+ in collisions of N+ and N2+ with a self-assembled propanethiol surface on gold