486 PHOTOCHEMISTRY OF PIGMENTS IN VITRO CHAP. 18 



dissociation, at least when dealing with the excitation by yellow or red 

 light, but have to keep in mind the possibilities of tautomerization, 

 reversible oxidation (or reduction), and dismutation. 



Kautsky, Hirsch, and Flesch (1935), when they first suggested the existence of 

 long-Hved excited states in dyestuff solutions, thought of metastable electronic states 

 (similar to those of free atoms). But because of the density and mutual overlapping 

 of energy levels of complex molecules, states of this type are unlikely to be long-lived 

 in organic molecules. Kautsky (1937) thought electronic excitation may become long- 

 lived, in concentrated dyestuff solutions, in consequence of a continuous exchange of 

 excitation energy between colhding molecules. However, a photon cannot avoid being 

 re-emitted in this way any more than a man can increase his life expectancy by changing 

 his address often. The hypothesis of metastable (triplet) electronic states of organic 

 molecules has been revived by Lewis and Kasha (1944); however, it seems that if the 

 rule which prohibits singlet-triplet transitions does not preclude the formation of the 

 triplet state from the excited singlet state within < 10"^ sec, it is unlikely to delay its 

 transformation into the singlet ground state for as long as several seconds or even 

 minutes. 



2. Reversible Photochemical Reactions 



When light absorption leads to a chemical change, we may assume 

 that the reaction product does not absorb light in exactly the same 

 spectral region as the original species. The color of intensely illuminated 

 dyestuff solutions which undergo reversible photochemical transforma- 

 tions, must therefore be different from their color in dim light. In 

 extreme cases (high yield of decolorization, slow back reaction) the 

 result may be a complete (but nevertheless reversible) loss of color in 

 light (as observed in illuminated thionine solutions in presence of ferrous 

 ions; cf. pages 77 and 152). Since the maximum hght absorption, 

 realizable in photochemical experiments with intensely colored pigments, 

 is of the order of ten absorption acts per molecule per second {cf. Vol. II, 

 Chapter 25), decolorization can be observed visually only if the back 

 reaction requires at least a tenth of a second. If the back reaction 

 occurs in 0.01 sec, decolorization can still be detected by photometry. 



Air-saturated chlorophyll solutions (in methanol) do not show re- 

 versible bleaching. But if oxygen is driven out by pure nitrogen, a 

 reversible bleaching becomes detectable by photometric methods (Porret 

 and Rabinowitch 1937; Livingston 1941). Porret and Rabinowitch, 

 using a 2000-watt carbon arc, observed, in a 2 X 10"^ M chlorophyll solu- 

 tion, a reversible bleaching of about 1% (measured in red light). Thus: 



(18.6) nh.a7T ^ 0.01 



where rih, stands for the frequency of light absorption, a for the reduction 

 in absorbing power in the red caused by bleaching, 7 for the quantum 

 yield, and t for the mean life of the bleached state. The frequency, 

 Wh,, was approximately 1 {i. e., each molecule absorbed once in a second). 

 The speed with which the color returned in the dark indicated that t 



