214 



R. L. Hinman and P. Frost 



out, however, because an authentic specimen of oxindole-3-acetic 

 acid showed the typical spectrum of an oxindole [X,nas (O.LV HCl) 249 

 rtifi, shoulder 280; log Emax 4.00, 3.35], and underwent no change when 

 treated with hydrogen peroxide at pH 1. 



The available information leads us to believe that the interme- 

 diate formed first in the reaction of lAA and hydrogen peroxide in 

 acidic media is: 



OH 



CH2CO2H 



OH 



CH2CO2H 



V^N^ or its salt: \A®N^ X 



e 



H 



Neither of these is the final product since the spectra of known com- 

 pounds of this type (16) do not correspond to those observed in this 

 reaction. Moreover, such intermediates do not explain, of them- 

 selves, the requirement of a two-carbon carboxyl-bearing side chain. 

 They do, however, afford a variety of potential pathways, such as 

 cyclization, dehydration, decarboxylation, etc., in which the carboxyl 

 group may participate."' These pathways are now under investigation. 



A number of attempts have been made to scale up the reaction 

 of lAA and hydrogen peroxide at pH 1 and in acetic acid to permit 

 isolation of the products. At concentrations of 0.\M, however, the 

 spectra of the products showed only broad general absorption in the 

 240 to 260 m/i. region. The products themselves were intractable tars. 



The spectra of a variety of other compounds which might have 

 been among the products have been compared to those obtained in 

 the reactions described herein. o-Aminoacetophenones, Bz-hydroxy-o- 

 aminoacetophenones, anthranilic acid, hydroxyanthranilic acids, 2- 

 and 4-quinolones, dihydro-2- and -4-quinolones, indolenines. and di- 

 oxindoles were considered, but their spectra differed in significant 

 ways from those reported herein. 



COMPARISON OF THE PEROXIDASE SYSTEM 

 AND THE H.Oo/H SYSTEM 



As shown in Figure 3, the similarity of the spectra obser\ed near 

 the ends of the two reactions is striking. There are, however, certain 

 differences to be noted. Following the disappearance of the indole 



' III considering tlic possibility tliat reaction occurs by oxidation of the ben- 

 zene ring, it is difficult to sec liow tlie carboxyl group could participate selectively 

 ill such a reaction. The most important consideration in ruling out this pathway, 

 however, is the known preference of most reagents for the hctcro ring (4). 



