102 PRIMARY PROCESSES 



cule, and for, say, a 10-Mev alpha particle the total number of ejected 

 atoms (including ejections by the ejected atoms themselves) will be of 

 the order of 100. For high-energy particles the number of ejected atoms 

 is invariably very much smaller than the total number of excited and 

 ionized atoms (as in the example cited, where the ratio is of the order 

 of 10~^), and since the latter entities are usually chemically effective, nu- 

 clear collisions are usually unimportant and are properly neglected in 

 considering the primary processes. Whether cases exist in which elec- 

 tronic energy transfer is so ineffectual in causing chemical or biological 

 change that the nuclear collisions are of consequence is not known. It 

 is certainly not inconceivable that in some instances energy transfer to 

 nuclei might play a decisive role in effecting changes in very large, stable 

 molecules, for the ejection of an atom, particularly if this in turn ejects 

 other atoms from the same molecule, will severely and probably perma- 

 nently damage the molecular structure. It is evidently much less likely, 

 in general, that a molecule will recover from loss of an atom than from 

 loss of an electron. 



Fast electrons also may lose energy in nuclear collisions, but even if 

 they have very great energy (their mass then significantly exceeding the 

 small rest mass) the number of collisions sufficiently violent to eject 

 atoms from molecules is extremely small — much smaller, relative to the 

 number of ionizations produced, than for energetic heavy particles. 

 The nuclear-collision mechanism is therefore certainly negligible for 

 electron or gamma-ray irradiation. 



For heavy ions, such as fission fragments, the energy loss in nuclear 

 collisions is relatively greater than for bare nuclei like protons or alpha 

 particles, because, compared to the ionic charge, the nuclear charge of 

 the particle is higher and the internuclear Coulomb interaction there- 

 fore of more consequence. This suggests that for these radiations the 

 process might be more likely to achieve importance in determining chemi- 

 cal and biological transformations. The same conclusion applies for 

 slowly moving recoil ions — for example, some of the recoils from fast 

 neutron collisions — and the effect would seem to merit study, for in- 

 stance in evaluating dosages for these recoil ions. 



In order to provide quantitative illustration of some of the matters 

 discussed above. Figs. 1-3 (pp. 111-112) present information on the 

 nuclear collisions of protons in water, which was chosen for convenience 

 in computation and also because it is a representative medium (in this 

 respect) for radiobiology. The figures, and the calculations upon which 

 they are based, are explained in the appendix below. Similar calcula- 

 tions could be made readily for any medium the atomic composition of 

 which is known. 



