57 



poses according to two mechanisms, part of it giving what is called the "forward 

 reaction", with products H2 and H 2 2 and the other giving the "radical reac- 

 tion", with products H and OH (10). Later Dr. Allen will tell you something 

 about the chemical evidence of this. I really hesitate to discuss it in his pres- 

 ence, since he was the originator of this set of reactions. In equation form we 



H 2 -W^_ ^ iH 2 + iH 2 2 F 



have 



H 2 -V*— > H+ OH R 



where the symbol — S* — > means "under the influence of high-energy radiation 

 gives". 



These reactions can be separated by the introduction of scavengers into the 

 medium, which are very effective in reacting with all of the radicals which are 

 formed. You will find, even if there is a high concentration of scavenger, you 

 get some of the F reaction. It has also been demonstrated, with isotopes (11), 

 that the H 2 and H 2 2 comes from the water. It does not come from any kind of 

 solute in the water, even when you are scavenging the radicals, but comes di- 

 rectly from the water. 



It seemed attractive to us to put together a model in which you get the same 

 number of radicals per unit energy regardless of the type of radiation - gam- 

 ma, tritium -beta, or alpha - and a radical distribution depending on the 

 ionization density of the radiation. For gamma radiation, it is known experimen- 

 tally that 20 per cent of the radicals must go into the F reaction, and around 80 

 per cent into the R reaction; while in the case of alpha, of course, something 

 like 90 per cent must go into the F reaction and only 10 per cent into the R re- 

 action. 



It looked as if there were a possibility to explain these facts on a geometrical 

 basis, i.e. with the use of radical diffusion only, and we started out to do just 

 that. I will indicate how the theory goes and apply it to the case of gamma rays. 



You see how this has to work. For the case of the gamma rays, a typical 

 primary event will form several radical pairs. Let's say they are formed in 

 some kind of a region which I will indicate as spherical. We will say this sphere 

 is drawn about the center of an event. 



FANO: I should like to take issue with your saying a typical primary event 

 has several radical pairs. The typical primary event has but one radical pair. 

 Most frequently a primary event has but one radical pair. It is the usual mis- 

 understanding that a typical ion cluster in air has three ion pairs. Three is not 

 typical. 



ALLEN: I think the point is that the fraction of radicals which are formed in 

 these isolated pairs is not very large. 



MAGEE: Well, let's go on with this. I want to get the model out before you 

 first, and then you can tear it to bits. You understand that if I were following 

 the Lea description I was telling you about, I would have said the thing to do is to 

 take the OH radicals as in a Gaussian distribution, say with one b_ parameter, 

 and the H atoms as in another Gaussian distribution with another_b_ parameter. 



BOAG: Both the H and the OH radicals are assumed to be in Gaussian density 

 distributions having maxima on the axis and differing only in half width. However, 



