846 LIGHT AND LIFE 



consumed. Nor does correspondence of the chemiluminescence spec- 

 trum with the fluorescence spectrum serve to identify luminol as the 

 emitter of the hght, as was formerly thought. The correspondence 

 is spurious, since the fluorescence spectrum is observed in acid solu- 

 tion, the chemiluminescence spectrum in basic solution, where luminol 

 exists only as the mono anion, the fluorescence of which is essentially 

 nil. The 3-aminophthalate ion, however, does provide a good match 

 between chemiluminescent and fluorescent spectra at the same pH. In 

 addition to the above information bearing on the mechanism of the 

 reaction, there is evidence that free radicals are involved. Although 

 White proposes a mechanism for the luminol reaction in water, the 

 difficulties of studying the system are so great that he was led to try 

 various organic solvents in place of water. With certain weakly acid, 

 highly polar organic solvents, the most efficient of which proved to be 

 dimethyl sulfoxide, only luminol, oxygen, and a base are required, 

 the emission of light is superior to that in water, and the maximum 

 emission shifts from 430 m^u, to 480 ni/x (Fig. 4) . The products of 

 the reaction are light. No, and the 3-aminophthalate ion. The quan- 

 tum yield was O.I. A possible intermediate is the hyperoxide ion 

 shown below the main reaction in the figure. Since cyanide ions and 

 transition metals do not noticeably affect the reaction in dimethyl 

 sulfoxide, although they do in water, probably no free radicals are 

 produced in the former instance. However, the simple view that 

 the emitting molecule is excited during the ejection of the nitrogen, 

 and is therefore the phthalate ion itself, runs into difficulties. The 

 j^redicted agreement between the chemiluminescence of the luminol 



0%2^hv 



\ 11 / 



Vy*-° 



N© 



NH^ O 



Fig. 4. Tlic theiniluminescence of luminol in dimethyl sulfoxide. 



