G . S.HAMMOND, C. E. BOOZER, C. E. HAMILTON AND J . N . SEN
3238
[A CONTRIBUTION
FROM THE
CHEMICAL LABORATORY OF IOlVA
STATE
Vol. 77
COLLEGE]
Air Oxidation of Hydrocarbons. 111. Mechanism of Inhibitor Action in Benzene and Chlorobenzene Solutions
s. HAMMOND, CHARLES E. BOOZER, CHESTER E. HAMILTOX ,\ND
B Y GEORGE
JYOTIRINDRA
?J. S E N
RECEIVEDNOVEMBER G , 19.54 Most potent antioxidants possess one of the two structurcs, ArOH or ArNHR. The 1:tbilc hydrogetis have beell viewccl as a critical part of the structures and characterization of inhibition products shows that they arc indeed removed durillg the course of the reactions. However, several lines of evidcncc indicate that hydrogcii abstraction is not a discrcte first step in the inhibition process. The key observations are: (1),there is 110 kinetic isotope effect noted if S-D-S-methyl;Lililine and N-D-diphenylamine are compared with the corresponding undeuterdtcd aminc; ( 2 j specifically designed compou~ids such as hydrazobenzene, d o not prove t o be as potent as inhibitors as their close relatives despite the fact that the former apparently should have a good aptitude for the hydrogen abstraction reaction; ( 3 ) the effects of substitueiits 011 the inhibitory efficiency of aromatic amines and phenols can be correlated with the Hamniett equation in a mariner which suggests t h a t electron removal from the inhibitor is of paramount importance; (4)despite the absence of labile hydrogens dimethylaniline shows distinct, but weak, inhibitory action and N,S'-tetramethyl-~-plieii~letiedi~mii1c is a strong inhibitor; a d ( 5 ) it has been found that the latter compound is converted to a complex which may be hydrolyzcd to form a Wurster dye. These data are held to be consistent with the view that the first stcp of the inhibition process is the reversible formation of a loosc molecular complex betwecii the inhibitor and R O r . . This complex then is destroyed by reaction with a second ROz. radical.
The empirical generalization that most of the valuable commercial antioxidants are either phenols or aromatic amines which retain at least one hydrogen attached to nitrogen has led t o the view that the first and rate-controlling step of the inhibition reaction is hydrogen abstraction from the hetero atom.'
inechanisni it is possible to account for the experimental observations in a satisfactory manner. I n this last mechanism it is assumed that ROY reacts reversibly with the inhibitor t o give a complex which is then destroyed in an irreversible reaction with a second alkylperoxy radical.
Results Deuterated Amines.-It is reasonable to expect (2) that reaction 1 should show a kinetic isotope effect if While these reactions are quite unlike the first the labile hydrogen is replaced by d e ~ t e r i u m . ~ , ~ steps involved in the inhibition of vinyl polymeriza- A normal isotope effect has bee11 observed in the tion2-j it is evident that there must be mechanistic rate of oxidation of a heavily deuterated olefin9 differences in the two reactions, since compounds which can be attributed to a decreasc in the rate of which are potent polymerization inhibitors are in- the chain-carrying step of the reaction, equatiorl 3, effective as antioxidants. Furthermore, the hy- when hydrogen is substituted for deuterium. I ' drogen abstraction reactions would in general bc It also has been found" that the rate of the dcgraexothermic with R02. as the attacking radical,G dative chain transfer reaction of growing polywhereas the corresponding steps in vinyl poly- allylacetate radicals, equation 4, is clecrcased by a merization would lead to the formation of C-H factor of 2.7 by deuteration of the monomer in the bonds and in some eases actually would be endo- allylic position. thermic, which is in itself sufficient t o eliminate H(D) I such reactions from consideration as possible inRoy, + C C !=C< --+ K O y H ( U ) + >C-C=C hibitory processes. However, these facts merely serve to demonstrate that reactions 1 and 2 should (8) be considered as possible termination steps. No TZ. $- CH?:-=CHCHOAc --+ evidence of a compelling nature ever has been preI sented to further sustain this view. As the result I-{(U ) of the present investigation we have reached thc RH(I1) + ClI:!---CIICHOAc (-1) conclusion that the simple hydrogel1 abstraction Figure 1 deinotistrates that N-deuterated dimechanism is inadequate to account for our experiinental observations. A logical alternative mecha- phenylamine is not significantly different from the nism involving prior addition of alkylperoxy undeuterated amine in its antioxidant action. radicals to the aromatic nucleus also has been found Similar results have been reported previously'" to give a poor account of the experimental observa- in the case of N-deuterated N-methylaniline. Special Inhibitors.-Several compounds were tions. By a modest modification of the second studied because they possessed structures which (1) For key references see: (a) C. E. Boozer a n d G . S. Hammond, lead to rather definite predictions as to their beTHIS J O U R N A L . 76, 3861 (1954); (b) C . E. Boozer, G. S. H a m m o n d ,
+ R01.--+ ArO. + R02H ArNHR + ROz. +ArGR + ROzH BrOH
(1)
C. E. Hamilton a n d J. N. Sen, ibid., 77, 3233 (1055). ( 2 ) P. D . B a r t l e t t , G. S. Hammond and H. K w a r t , Disc. F a r u d o y Soc.. 2, (1947). (3) P. D. Bartlett and H . K w a r t , THISJ O U R N A L . 72, 1051 (1950). , (4) S. G. Cohen, J . Polymer Sci., 2, 511 (1947); THISJ O U R N A L67, 17 (1945); 69, 1057 (1950). ( 5 ) P. J. Flory, "Principles of Polymer Chemistry." Cornell Uni1953. p , 162. versity Press, I t h a c a . N. Y., (6) Because of t h e resonance energy of t h e radicals Are. and ArXCHB.
(7) 11. C. Urey a n d G. K. Teal, Rev. Mmf. P h y s . , 6 , 34 (1935). ( 8 ) F. H. Westheimcr a n d N. Xicolaides, THIS J O U R N A I . , 71, 2,; (1949): 0. Reitz, 2 . pkysik. Ckem., 176, 3(i3 (193G); C . L. Wilson, J . Ckem. SOL.,1850 (1936). (9) R. A . M a x and F. E . Deatherage, J . >4m.Oil Ckenrisls' .SO
