p. C. WILHELMSEN, R. LUMRY AND H. EYRING 77 



most of the kinetic studies have been done on this compound. How- 

 ever, there is evidence that the mechanism is the same in each case, 

 and most information concerning the reaction of one of the DPD's 

 can be extended to the remainder. This similarity has been found 

 true in the investigations of Zellner and Dougherty (1937) where it 

 was found that the rate of oxidation of many of the DPD's is the same 

 when measured by the rate of evolution of nitrogen. The apparent 

 difference in rates, when the intensities of the luminescence are 

 compared, is probably due largely to the diflFerent efficiencies of the 

 emitters produced in each case. An additional possibility is that there 

 are two or more reaction routes and that different fractions of some 

 of the DPD's proceed by the luminous and nonluminous reactions, 

 in which case some forms of information about one of the DPD's 

 could not be simply extended to the others in the series. Further 

 investigations are needed in order to deterr.iine the extent to which 

 this second possibility must be considered. 



The Role of Oxygen 



Many different substances and procedures have been employed in 

 connection with the DPD's to produce chemiluminescence. Fre- 

 quently, hydrogen peroxide is used either alone or with various other 

 agents such as hemin, potassium ferricyanide, or sodium hypochlorite 

 ( Albrecht, 1928 ) . Chemiluminescence can also be produced by adding 

 solutions such as ferricyanide or hypochlorite to an alkaline solution 

 of the DPD. Some investigators have employed electrolysis (Harvey, 

 1929; Spruit, 1950; Bremer, 1953) and supersonics (Prudhomme, 1949; 

 Flosdorf et ah, 1936). However, in the final analysis, either oxygen or 

 a compound that readily decomposes to form oxygen, e.g., hydrogen 

 peroxide or sodium hypochlorite, appears to be indispensable. It might 

 seem that the oxidation of the DPD's by ferricyanide does not appear 

 to require oxygen. However, as shall shortly be seen, the chemilu- 

 minescent reaction is largely dependent on dissolved oxygen. 



Inasmuch as oxygen is so intimately involved in the chemilumines- 

 cence of the DPD's, it will be interesting to study the reaction 

 between oxygen and luminol. Some authors (Harvey, 1929; Drew, 

 1938) have reported that no light was observed when oxygen was 

 bubbled through a luminol solution. Their failure to observe the light 



