1122 J . Org. Chem., Vol. 65, No. 4, 2000
Darbeau and White
An alternative proposal4c-e is that the nature of the
transition states leading to alkylarenes varies with the
electrophilicity of the active species and with the nucleo-
philicity of the aromatic substrate.4c-e By this scheme,
in reactions with relatively weak electrophiles or with
weakly nucleophilic aromatics, the transition state of
highest energy (σ-type) lies late on the reaction coordi-
nate. Conversely, reactions with strongly electrophilic
reagents and/or strongly nucleophilic aromatics proceed
via an early (π-type) transition state of highest energy.4c-e
The π-type transition states may or may not proceed to
the σ-complex via a discrete π-complex (intermediate).4e,k
Benzylation of benzene-toluene mixtures has been
studied particularly extensively1b,4,5 because placing sub-
stituents on the aromatic nucleus of the benzyl moiety
allows elucidation of the electronic effects operating in
the reaction.1b,4,5 The relative amounts of alkylated
toluene and benzene formed (kT/kB) as well as the isomer
distribution are typically measured. For the 4-R-benzyl
cations (R ) Me, H, Cl, NO2), kT/kB ranges from ∼2.5 to
20 and % meta isomer ranges from ∼2.8 to 6.2 have been
reported.4,5
There are several general problems associated with the
standard F-C approach. These include uncertainty of the
identity of the active electrophile, variable product
distributions,4,5 acid-catalyzed isomerizations4e and dispro-
portionations,4e over-alkylation,4b extreme sensitivity to
traces of water,5c rate of mixing,4d insolubility of the
catalyst in the neat aromatic,4,5 secondary alkylations in
reactive cosolvents,1b etc. The identity of the actual
alkylating agent has not been established in most cases
or even specified; carbocations are often assumed to be
the active species, but ion pairs5a,c and polarized alkyl
halides4a,d,g,5c have also been invoked. Displacement reac-
tions on the latter would then lead to the observed
alkylarene (eq 1).4a,d,g,5a
zylating agent and with the catalyst. If the reactions were
under kinetic control, then the active benzylating agent
in the standard F-C approach is not a discrete carboca-
tion because if the same species were generated in each
case, the same product distributions would be expected.
The active electrophile is probably a benzyl halide-Lewis
7
acid complex. Thus, the marked (230%) rise in kT/kB as
the halide is changed from F to Br with AlCl3-CH3NO2
as catalyst is accounted for by assuming that the kT/kB
is inversely proportional to the extent of polarization in
the complex. On the other hand, the isomer distribution,
in particular the % meta value from the same experi-
ments, remains essentially constant as the halide ion is
varied.7 This latter observation is consistent with the
formation of carbocations, but this conclusion cannot be
made in light of the changing kT/kB values. Thus the
identity of the active electrophile in these benzylations
is uncertain.
Standard F-C alkylations of benzene-toluene often
utilize cosolvents such as acetonitrile and nitromethane
to reduce complications from phase separation,5a,b an
extreme sensitivity to traces of water,4,5 isomerizations,4a
and disproportionations.4a Such reactive solvents inter-
cept deaminatively generated benzyl cations1b,c,8a,9 to form
less reactive, longer-lived onium ions, which cause (sec-
ondary) benzylations1b,c,9 that are more selective than the
(primary) benzylations via the first-formed benzyl cation.1b
1b
Thus higher values of kT/kB and decreased values of %
meta are observed.1b A similar situation is likely to exist
in the standard F-C alkylations in the presence of
reactive cosolvents. Thus much, if not all, of the benzy-
lations observed in those systems may stem from second-
ary electrophiles. Consequently, data obtained from these
systems may not be valid for use in defining the chem-
istry of the F-C alkylation because the observed product
distribution may not reflect (only) the first-formed elec-
trophile.
Dea m in a tive Ap p r oa ch to F r ied el-Cr a fts Ben -
zyla tion . Deamination involves the loss of N2 (or N2O)
from an organic molecule or fragment. In these reactions,
alkyl diazonium ions dediazoniate to form a highly
reactive, essentially free cation (where the latter is
stabilized by resonance, etc. as in the present case).1b,8,9
The alkyl cation is believed to be the active electrophile
because (1) predominant retention of configuration is
observed in the ester from deamination of phenylethyl-
N-nitrosonaphthamide8b (and its 2-butyl analog8b) and
18O-studies of “intramolecular inversion”8d require the
finite existence of a carbocation; (2) deamination of
bridgehead amines leads to solvent-derived product
(SDP),8c,10a which is impossible if the diazonium ion were
the alkylating agent; and (3) the decomposition of N-
methyl-N-nitrosotoluenesulfonamide (Diazald) in toluene
does not yield the xylenes (although the sterically acces-
sible methyl diazonium ion is believed to be the
intermediate).10b
Many standard F-C benzylations exhibit third-order
kinetics:6 first-order each in aromatic substrate, catalyst,
and alkylating agent.6 In these cases, a mechanism
involving rate-determining carbocation formation fol-
lowed by attack by the carbocation on the aromatic
nucleus is excluded because the substrate would not
appear in the rate expression. However, formation of an
alkyl electrophile-Lewis acid complex followed by rate-
determining reaction between the electrophile and the
aromatic substrate is consistent with the observed kinet-
ics. Presumably, as the degree of charge separation in
the complex rises, the rate equation becomes zero-order
in the aromatic and second-order overall. However, the
latter kinetics is also consistent with slow carbocation
formation and subsequent fast reaction with the aromatic
substrate.
(7) Data taken from Table 11, ref 4e
(8) (a) White, E. H.; De Pinto, J . T.; Polito, A. J .; Bauer, I.; Roswell,
D. F. J . Am. Chem. Soc. 1988, 110, 3708. (b) White, E. H.; Field, K.
W.; Hendrickson, W. H.; Dzadzic, P.; Roswell, D. F.; Paik, S.; Muller,
P. W. J . Am. Chem. Soc. 1992, 114, 8023. (c) White, E. H.; McGirk, R.
H.; Aufdermarsh, C. A.; Tiwari, H. P.; Todd, M. J . J . Am. Chem. Soc.
1973, 95, 8107. (i) White, E. H.; Aufdermarsh, C. A., J r. J . Am. Chem.
Soc. 1961, 83, 1174.
The data7 from standard F-C runs in pure and diluted
aromatic solvents show that kT/kB varies with the ben-
(9) Darbeau, R. W.; White, E. H.; Nunez, N. P.; Coit, B.; Daigle, M.
A. J . Org. Chem. 2000, 1115.
(6) (a) Choi, S. U.; Brown, H. C. J . Am. Chem. Soc. 1963, 85, 2596.
(b) DeHaan, F. P. et al. J . Org. Chem. 1986, 51, 1587.
(10) (a) Kirmse, W. J .; Moench, D. Chem. Ber. 1991, 124, 4, 237. (b)
Brosch, D.; Kirmse, W. J . J . Org. Chem. 1991, 56, 907.