16 



PLATZMAN: Is emission in that region observed under bombardment? 



KASHA: No, there is little in that region. 



PLATZMAN: Are you saying that you think the luminescence dies off to- 

 wards shorter wavelengths and then starts again in the ultraviolet? 



KASHA: No* I would say that the secondary electrons may cause the emis 

 sion of higher frequency light first, which then can be absorbed competitively 

 by the solvent and the solutes. The solutes re-emit this absorbed high-fre- 

 quency radiation. 



PLATZMAN: How do the electrons cause this high-frequency emission? 



KASHA: Possibly by recombination. 



LINSCHITZ: One point I might make is that the absorption of such radiation 

 may be quite inefficient because the excited states that you can reach by parti- 

 cle bombardment or fragment recombination need not correspond to those that 

 might be reached by optical excitation, and that might also give regions of trans- 

 parency in the solvent. 



PLATZMAN: Why do you believe that the recombination transition should be 

 radiative rather than non- radiative ? 



KASHA: There are radiative recombinations of molecules. 



PLATZMAN: Yes, but always less probable than non-radiative ones. We 

 are discussing neutral recombination, not positive ions and electrons. Are you 

 not thinking of two fragments? 



KASHA: Oh, yes, molecular fragments. 



PLATZMAN: They would be even less likely to radiate. 



ONSAGER: There is a chance that they would, nevertheless. 



BURTON: What is the order of magnitude of the chance of radiation on re- 

 combination of the pair of radicals? 



MAGEE: It could be up to 100 per cent, I would think. If there is enough 

 binding energy in a particular excited state formed on association so that it will 

 not dissociate thermally, then it has nothing to do but radiate. 



LINSCHITZ: There are certain chemiluminescent reactions in which radical 

 recombination seems to be responsible for the electronic excitation. Radiation 

 must of course compete with thermal degradation for this energy. Relatively 

 few chemiluminescent efficiencies have been measured, but even in the best 

 cases the light yields are not greater than a few per cent, in condensed systems. 

 Generally, the efficiency seems to be very much lower, particularly if complex, 

 non-fluorescent molecules are involved. 



BOAG: A year or two ago we looked for recombination radiation from water 

 in the following rough way. A 1500-kev electron beam was allowed to pass into 

 water contained in a quartz dish. TheCerenkov light emission was considerable 

 and enabled a good spectrum to be photographed (on a Hilger medium quartz 

 spectrograph with a narrow slit) in about 40 seconds. The electron energy was 

 then reduced to below 250 kev and the same total electron current was passed 



