J. Org. Chem. 1989,54, 3770-3771
3770
donor is apparently due to semiquinone formation (Scheme IV). As predicted from Scheme IV, addition of a base such as DABCO causes the reaction product to change completely, and only the substituted quinone 2 is formed in a slow reaction because of the poor driving force for the transfer of an electron from 2'- (E, 0) to t-BuHgC1. Only 2 is observed in Me2S0, albeit in low yield. Substitutive (oxidative) free radical alkylations of quinones by tert-alkyl iodides has also been described in the systems
RI/H202/Fe(II)/Me2S0, RI/t-BuOOH/Fe(III), RI/ H202/Me2CO/H+,and RI/CH3C02-/S208~/H20?It has also been demonstrated that the reaction of trialkylboranes with 1,4-quinones to form upon hydrolysis the 2-alkylhydroquinones proceeds by a free-radical chain mechanism.'"
-
(9) Minisci, F.; Vismara, E.; Fontana, F. NATO ASZ Series C 1989, 257, 29. (10) Kalbalka, G. W. J. Organomet. Chem. 1971,33, C25.
Observation of A1C13-CatalyzedTrialkylsilylation of Benzene and Toluene with Chlorotrialkylsilanes in the Presence of Hunig Bases' George A. Olah,* Thorsten Bach, and G. K. Surya Prakash Donald P. and Katherine B. Loker Hydrocarbon Research Institute and Department of Chemistry, University of Southern California, Los Angeles, California 90089-1661 Received June 8, 1989
Summary: A1C13-catalyzedtrialkylsilylation of benzene and toluene has been shown to occur using chlorotrialkylsilanes in the presence of Hunig bases. Sir: Electrophilic silylation of aromatic compounds such as benzene and toluene under typical Friedel-Crafts alkylation conditions using halosilanes and a Lewis acid catalyst has never been observed.2 This is probably due to the great ease of protodesilylation of any arylsilane formed under the reaction conditions (i.e., the protodesilylation rate k-l is faster than that of silylation kl,i.e., k-' > k,) as has been confirmed by the work of Eaborn3 and Szelea4 On the other hand, some examples of direct electrophilic silylations of nonbenzenoid, highly activated aromatic systems such as ferrocene6and pyrrole6have been reported. H I
Si% I
One way we considered for achieving silylation of benzene or toluene is to slow down the rate of protodesilylation by trapping the proton eliminated in the silylation reaction by a suitable hindered base. Following this strategy we now report observation of the direct silylation of benzene and toluene with a series of chlorotrialkylsilanes under Friedel-Crafts conditions in the presence of Hunig bases (hindered tertiary amines) as proton acceptors. When excess benzene or toluene was heated with chlorotrialkylsilane and aluminum trichloride in the presence of diisopropylethylamine at 150 "C in a sealed heavy walled glass tube for 24 h, the corresponding aryltrialkylsilanes were obtained in low, but detectable (1) Aromatic Substitution. 56. For part 55, see: Olah, G. A.; Ernst, T. D. J. Org. Chem. 1989,54, 1203. (2) Gilman, H.; Dunn, G. E. Chem. Reu. 1953,52,77. Russell, G. A. J. Am. Chem. SOC.1959,81,4831. Effenberger, F.; Hilbich, D. Synthesis 1979. - . -, 841. - - -. (3) Bott, R. W.; Eaborn, C.; Greasley, P. M. J. Chem. SOC. 1964,4804. (4) Szele, I. Helu. Chim. Acta 1981, 64, 2733 and references therein. 1967,89, 5054. (5) Sollott, G. P.; Peterson, W. R. J . Am. Chem. SOC. (6) Simchen, G.; Frick, U. Synthesis 1984,929. Also see: Olah, G. A.; Rach, T.; Prakash, G. K. S., manuscript in preparation.
0022-3263/89/1954-3770$01.50/0
Table I. Trialkvlsilvlation of Benzene and Toluene substr R'3SiCl, R' = yield: % o:m:p R = H
R=H R = H RzCH3
CHS
CHzCHs CH(CHsh
CH3
0.3 1.2 1.6 1.0
0.6:34.6:64.8
'Based on silylating agent; GC yields, calibrated by internal standards.
amounts. The mole ratio of R3SiCkA1C1,:base was generally 1:42. The reaction mixture was worked up by quenching with aqueous NaHC03 followed by extraction with hexane. The product arylsilanes were identified by GC/MS7 and the yields were determined by GC8 using internal standards. The authentic compounds used for comparison were prepared according to literature metho d ~ .In~ the case of trimethylsilylation the reaction was performed on a preparative scale with the isolation of the aryltrimethylsilanesand characterization by NMR,Et,and mass ~pectrometry.'~~J" The yield of trimethylsilylation of benzene was 0.390,whereas that of toluene was 1%. The related triethylsilylation of benzene gave a 1.2% yield and triisopropylsilylation 1.6%. Although these yields, based on the amount of starting trialkylchlorosilanes, are too low to be considered of preparative value, it should be pointed out that the trialkylchlorosilanes under the reaction conditions undergo substantial disproportionation in the presence of AlCl, as the major reaction, explaining in part the low yields. Regardless, the first observation of electrophilic trialkylsilylation of benzene and toluene is significant both in regard of extending our knowledge of (7) The GC-MS analyses were carried out on a Finnigan Model INCOS-50 spectrometer interfaced with a Varian Associates Model 3400 gas
chromatograph. (8) GC analysis were done on a Varian Associates Model 3700 gas chromatograph equipped with a DB-1 glass capillary column. (9) Clark, H. A.; Gordon, A. F.; Young, C. W.; Hunter, M. J. J. Am. Chem. SOC.1951, 73,3798. Gilman, H.; Ingham, R. K.; Smith, A. G. J. Org. Chem. 1953,18, 1743. (10)(a) NMR spectra were obtained on a Varian Associates Model VXR-200 NMR spectrometer equipped with a 5-mm'H/'@F-broad band probe. IR spectra were obtained on a Perkin-Elmer 1550 spectrometer. (b) We were unable to verify these resulta by subjecting the authentic isomers to the reaction conditions. Under these conditions the major reaction was protolytic desilylation. However, acid-catalyzed isomerizations of 1,2-bis(trimethylsilyl)benzeneshave been reported, see ref 15.
0 1989 American Chemical Society
J. Org. Chem., Vol. 54, No. 16, 1989 3771
Communications aromatic substitution and in probing the nature of the electrophilic silylating agent involved.
r
r
-I
R = CH,, C2HS,K 3 H 7
In the absence of added hindered tertiary amines as proton traps no trialkylsilylation was observed. Reactions where the weaker and bulkier 2,6-di-tert-butyl-4-methylpyridine was used gave very low (10.1%) yields of the silylated arenes. Use of 1,8-bis(dimethylamino)naphthalene Proton Sponge or hexamethyldisilazane as bases did not give any silylated arene product. Clearly the nature of base appears to be important with a combination of lowest possible nucleophilicity (facilitated by steric bulk) and highest possible kinetic basicity (proton affinity). The ratio of competitive trimethylsilylation of toluene and benzene gave a kTJkEratio -3.3, indicating that the reaction has low substrate selectivity due to the strongly electrophilic character of the silylating agent. At the same time observed isomer distribution of the silylation of toluene was 0.6% ortho, 34.6% meta, and 64.8% para, indicating both the bulkiness of the silylating agent and substantial isomerizatiodob (Table I). Nucleophilic displacement by the aromatic compound on silicon (from CH3 to CzH5to i-C3H7)in an SN1-Sior SN2-Si type mechanism'l is less likely since increased steric bulk of the alkyl groups on silicon has actually a beneficial effect on the reaction. This effect, however, also can be due to the decreasing tendency for disproportionation. It is more probable that initial ionization to a short-lived trivalent silicenium ion occurs in an SN1-Simanner followed by fast silylation of the aromatics to a silylated arenium ion. However, our data based on steric arguments alone cannot rule out the involvement of a five-coordinate siliconium ion type intermediate.12 Studies of the facile prot~desilylation~~~ of silyl-substituted aromatics have postulated protonated silylarenium ions as intermediates in the course of the reaction, although they were never observed as long-lived ions. Deprotonation of the silylated arenium ions in the present study is facilitated by added hindered tertiary amines thus allowing observation of arylsilanes (in competition with desilylation), albeit in low yield. Other reasons for the inefficiency of the reaction are following (i) extensive disproportionation of the silylating agent under the reaction conditions to di- and trichlorosilanes and (ii) competing alkylation of the arenes by the alkyl groups of the hindered amine. Attempts to suppress these side reactions by lowering the reaction temperature or using other silylating (11) For a comprehensive treatise, see: Sommer, L. H. Stereochemistry, Mechanism and Silicon; McGraw-Hill: New York, 1965. (12) Corriu, R. 3. P.; Henner, M. J . Organomet. Chem. 1974, 74, 1. Corey, J. Y.;West, R. J . Am. Chem. SOC.1963, 85, 4034.
agents13were so far unsuccessful, The possible mechanism involving the intermediacy of a trivalent silicenium ion is supported by a recent study of the gas-phase silylation of benzene and toluene by Fornarini14who observed a k T / k B ratio of 3.8,which compares rather well with the present solution data. The high amount of meta-silylated product observed in the solution trimethylsilylationof toluene (~35%)is indicative of either intramolecular 1,2-trimethylsilylshift15in the intermediate arenium ion leading to the thermodynamically stable meta product or intermolecular isomerization. Such rearrangements are common in Friedel-Crafts alkylations particularly of tert-alkyl systems. Methyl migration in the (trimethylsily1)toluenium ions appears to be less likely since protonation prefers to occur at the carbon to which silicon is attached. It is interesting to point out that whereas the present study of Friedel-Crafts trialkylsilylation of benzene and toluene seems to suggest s N 1 type silylation, similar silylation of ferrocene, a much more nucleophilic substrate, takes place through s N 2 like displacement of trialkylchlorosilane-aluminum chloride complex by the aromatic.16 Benzene, toluene, and other alkylbenzenesi6 may not be sufficiently nucleophilic to displace the moderately polarized silyhting agent in an sN2 fashion. Their silylations, therefore, necessitate probable ionization to the reactive, but short-lived trialkylsilicenium ions. This takes place, however, only to a limited degree and under rather harsh conditions where side reactions render yields of silylated products very low. Acknowledgment. Support of our work by the National Science Foundation is gratefully acknowledged. T.B. wishes to thank the Konrad-Adenauer Foundation, Bonn, West Germany, for a scholarship and the Dr. SophieBernthsen Foundation, Heidelberg, West Germany, for an award. (13) Other silylating reagents such as (trimethylsily1)trifluoromethanesulfonate or hexamethyldisilazane did not give the silylated arenes. The reaction was not successful below 150 O C . Lewis acids such as AlBr8 also are applicable but yields do not improve. (14) Fornarini, S.J. Org. Chem. 1988,53, 1314. (15) Seyferth, D.; White, D. L. J . Am. Chem. SOC.1972, 94, 3132. (16) To be published. See also Bach, T. Diplomarbeit, University of Heidelberg, 1988.
