84 



molecule, which may also be observed in rigid solvents. Since this phosphor- 

 escence is known to arise from triplet states of the molecule, we thus have di- 

 rect evidence of the formation of such triplets by electron-radical recombination 

 tion. We will return to this matter later. The main point that I want to make 

 at this time is that surprisingly low excitation energies may result in photo- 

 ionization or photo-oxidation of these organic molecules in solution. For many 

 aromatic molecules with reasonably stable semiquinones (i.e. , those which are 

 stable in rigid solvents at liquid nitrogen temperatures), radicals or radical- 

 ions are formed by excitation in the middle ultraviolet. A knowledge of these 

 threshold energies is of course necessary for evaluating possible mechanisms 

 of radiation action. Photochemical studies in rigid solvents, of the type just 

 described, combined with data on relative oxidation-reduction potentials, should 

 help establish the ionization energies of a variety of organic molecules in solu- 

 tion. 



KASHA: May I make a comment? Glasses or mixtures of organic liquids 

 which when supercooled at a liquid nitrogen temperature form a homogeneous 

 glass have been used for quite a long time in spectroscopic work. Lewis and 

 Lipkin were the first to develop glasses which did not crack at low temperature, 

 and they in general used ternary systems, the more common of which was ether- 

 isopentane-ethanol. They also substituted ethylamine for the alcohol. 



LINSCHITZ: In these mechanisms discussed so far, the transfer of energy 

 takes place by the formation of high-energy fragments (either by unimolecular 

 or bimolecular processes) followed by further reaction of these fragments. I 

 should like to move along now to processes involving several molecules - in 

 particular, resonative energy transfer. This kind of "broadcasting" process 

 has been known for a long time, going back to the work of Franck, Cario and 

 others on sensitized fluorescence. Kallmann and London calculated long ago 

 that, in the case of completely sharp resonance, for instance between an ex- 

 cited and normal sodium atom, cross sections for energy transfer could be 

 obtained of the order of 1000 times the ordinary gas -kinetic cross section. 

 More recently attention has been drawn to this process by the work of Kallmann 

 and his group on scintillation counters (12). It is found that certain organic 

 hydrocarbon crystals emit light under ionizing radiation. If the crystal is 

 melted, the luminescence is immediately quenched. But the addition now to the 

 liquid solvent of small amounts of fluorescent substances, in particular those 

 having absorption bands lying to the red of the original solvent absorption, again 

 restores the luminescence in high yield. In this case, practically all of the ra- 

 diation is absorbed by the solvent and the problem is to find how much is trans- 

 ferred to the solute. 



Kallmann has measured energy yields for light production in alpha or 

 gamma irradiated anthracene crystals. His value is certainly correct to -i 50%, 

 which is all we need for our order-of-magnitude argument, and he finds that 

 the energy efficiency for the conversion of gamma energy into light is about 

 10%. 



That was for anthracene crystals, for which the fluorescence yield (ultra- 

 violet light) is nearly 100%. 



For solutions of anthracene in xylene, the optimum concentration for gam- 

 ma-light conversion is about 1 molecule of anthracene per 1000 of xylene. At 

 this concentration, the (ultraviolet excited) fluorescence yield is about 20%. If 

 all the energy were transferred from xylene to anthracene, we would therefore 

 expect to obtain an energy efficiency of 20% that of the crystal, or 2% over-all. 

 Actually, the observed yield is 1%. So these figures mean that you are getting 

 roughly 50% of the energy transferred out of the xylene into the anthracene. 



