Dimer formation during reactions of benzylic halides with lithium naphthalene and mechanisms of dimer formation from reactions of benzylic halides with benzylic carbanions (original) (raw)

1976, Journal of The American Chemical Society

. (3) Similar results have been obtained by anodic oxidation in acetonitrile containing fluoroboric acid as supporting electrolyte [E. Kotani, N. Takeuchi, and S. Tobinaga, J. Chem. Soc., Chem. Commun., 380 (1973)]. (41 M. A. Schwartz. E. F. Rose, and B. Vishnuvaiiala, J. Am. Chem. SOC., 95, 612 (1973). B 30. 89 11976) (5) 0. Hammerich, V. D. Parker, and A. Ronlan, Acta Chem. Scand., Ser. I -. -I -, _~. ~. (6) €,., (phenol ) ,> €,*Jphenol ether) for phenols with not more than one &alkyl substituent and €,.Jphenol) < E,.Jphenol ether) for 2,6dialkylphenols. (7) The diarylalkanes containing terf-butyl groups (6 and 8) looses one tertbutyl group in the presence of TFMS. Cyclic voltammograms run at various times alter the preparation of a 1 mM solution of 6 in CH2C12-TFMS (1 % ) at -50 OC showed that the rate constant for this dealkylation is about 4 X s-' under these conditions. In a preparative oxidation of 8 in CHpCIz-TFMS at -50 OC, compound 35 was obtained in 60 % yield with 100 % conversion (two stereoisomeric spirbdienones should be obtained from the diarylpropane with one tert-butyl group. However the same product (35) is formed by dienone-phenol rearrangement of these spirodienones). From this we conclude that 35 is formed by cyclization of the mono-terf-butyl compound. (8) U. Svanholm. A. Ronlan, and V. D. Parker, J. Am. Chem. SOC., 96, 5108 (1974). (9) A. Ronlan, 0. Hammerich, and V. D. Parker, J. Am. Chem. SOC., 95, 7132 (1973). (10) This appears a reasonable assumption since our measurements in CH2Cl2-FSO3H (ref 5) have shown that the oxidation potential for the oxidation of a cation radical of a phenol or a phenol ether is at least 300-400 mV more anodic than the oxidation potential for the oxidation of the phenol or phenol ether to the cation radical. Preparative experiments were carried out at the latter potential. (1 1) In our preliminary communication (ref 2) we argued on the basis of the difference in oxidation potential between a phenol and its methyl ether that the initial one-electron transfer from a phenol to an anode with for-mation of a cation radical is generally followed by rapid deprotonation to a phenoxy radical which can dimerize (path e in Scheme I), react as an electrophile (path d), or become further oxidized to a phenoxonium ion (path f). Our finding ref 5) that the reversible potential (in CHpClp-FS03H) for the redox reaction -e phenol phenol cation radical +e is about the same as for the reversible redox reaction (also in CHpCIp-FS03H), -e phenol ether +phenol ether cation radical +e of the corresponding methyl ether supports this hypothesis. (12) A. Ronlan. J. Coleman, 0. Hammerich, and V. D. Parker, J. Am. Chem. SOC., 96, 845 (1974). (13) This very rough estimate is obtained by applying the Tafel equation assuming that the phenol and its methyl ether have the same transfer coefficient (0.5) and the same exchange current. (14) U. Palmquist and A. Ronlan, to be submitted for publication. (15) V. D. Parker, U. Palmquist. and A. Ronlan. Acta Chem. Scand., Ser. 8. 28, 1241 (1974); V. D. Parker and A. Ronlan, J. Am. Chem. SOC., 97, 4714 (1975). (16) We assume that phydroxy-and pmethoxyphenyl groups are equally efficient nucleophiles. (17) E. Kotani, F. Miyazaki, and S. Tobinaga, J. Chem. SOC., Chem. Com mun., 300 (1974). (18) Our previous investigations of the anodic coupling and cyclization reactions of phenol ethers (ref 15) have shown that phenol ether cation radicals preferentially couple or cyclize through positions para to methoxy groups and that coupling or cyclization does not occur if these positions are substituted. Abstract: Reactions of benzylic fluorides and chlorides with lithium naphthalene in THF have been found to give dimers in 61 -79%