silylbromide produced the silylated product 10 in good yield
(63%). Other unidentifiable products were produced when
TMSCl or TMSI was used.14 The above result suggests that
the reactivity of the benzyl carbanion derived from 8 is higher
at the carbon atom than at the sulfinyl oxygen position.
Table 2. Alkylation Reactions of Compound 8
Most interestingly, when the closely related benzyl sul-
foxide 9 (R1 ) Me) was treated under an identical set of
conditions, trimethylsilyloxy-thioether 11 was the major
product isolated (entry 5, Table 1) and is the result of a
vinylogous Pummerer rearrangement.15 The exclusive forma-
tion of 11 suggests that the relative reactivity of the two
nucleophilic centers (carbon and oxygen) in the benzyl
carbanion derived from 9 is inverted with respect to the
situation encountered with the carbanion derived from 8. In
the case of sulfoxide 9, the sulfinyl oxygen atom corresponds
to the most reactive site of the sulfinyl benzyl carbanion.
Notably, the enantiomeric purity of the resulting product 11
has the remarkably high value of 98% ee. This value was
determined by cleavage of the Si-O bond using Bu4NF
followed by an examination of the NMR spectrum of the
Mosher ester derived from the corresponding alcohol.16 In
an attempt to improve the overall yield of the reaction, we
evaluated the role of different parameters such as concentra-
tion, solvent, and the nature of the silylated electrophile
(Table 1). TMSCl was found to be the most efficient
electrophile (entry 5), whereas TMSI and TMSOTF gave
much lower yields of product (entries 7 and 8). The optimal
conditions consisted of using redistilled TMSCl at a con-
centration of 0.1 M and at -78 °C (entry 9). Lower or higher
concentrations resulted in a diminished yield.
entry
alkyl halide
methyl iodide
ethyl bromide
ethyl iodide
benzyl bromide
2-phenylethyl bromide PhCH2CH2 (14)
2-phenylethyl iodide
2-(3,4-dimethoxy-
phenyl)ethyliodide
product (R)
CH3 (9)
CH3CH2 (12)
CH3CH2 (12)
PhCH2 (13)
yield (%)
1
2
3
4
5
6
7
98
85
92
93
59
81
85
PhCH2CH2 (14)
(3,4-dimethoxy-
phenyl)CH2CH2
(15)
starting material
starting material
CH2CHdCH2 (16)
CH2OMe (17)
8
9
2-bromopropane
1-bromoethylbenzene
allyl bromide
10
11
12
13
14
82
94
15
55
51
ICH2OMe
BrCH2CH2OTIPS
ICH2CH2OTIPS
ICH2CH2NPhta
CH2CH2OTIPS (18)
CH2CH2OTIPS (18)
CH2CH2NPht (19)
a Pht ) phthalimide.
bromides. It was not possible to carry out an alkylation using
secondary halides due to a competitive elimination.17 Inter-
estingly, when 2-phenethyl iodide or bromide was used, the
substitution product 14 (entry 5/6) was obtained and only
trace quantities of styrene were detected in the crude reaction
mixture.18 The last five entries in Table 2 demonstrate the
flexibility of the alkylation when more complicated halides
are used as the electrophiles.
With these sulfinyl compounds in hand, we proceeded to
investigate their vinylogous Pummerer behavior. Using the
conditions described in Table 1, the reactions were found to
occur rapidly and furnished a wide variety of benzylic
alcohols (Table 3) in yields ranging between 50 and 70%
for the three-step sequence. Complete stereoselectivity (ee
> 98%) was observed in all of the cases examined. The
presence of a heteroatom on the side chain was found to
influence the reaction course. Thus, 2-p-tolylsulfinyl styrene
(27) was the only product obtained starting from 17 (entry
8), thereby indicating that elimination of the methoxy group
is preferred over the rearrangement. In contrast, the homolo-
gous compound 18 afforded diol 26 in 56% yield under the
reaction conditions. Finally, the more heavily substituted
amide 19 failed to undergo the reaction.19
To further evaluate the scope and generality of the
asymmetric vinylogous Pummerer reaction, it became neces-
sary to prepare several more highly substituted sulfinyl
derivatives (i.e., 12-19). The method previously used to
synthesize 812a and 912b was based on the sulfinylation of
the Grignard derivatives derived from 2-bromotoluene and
2-bromoethylbenzene. This procedure is severely limited
using the more highly substituted bromides due to the
unavailability of the starting materials. Instead, we opted to
carry out a base-induced alkylation of ortho-sulfinyl toluene
8 that proceeded in high yield (Table 2). Thus, the reaction
of 8 with LDA in THF at -78 °C furnished the expected
anion, which readily reacted with various halides. The
reaction proceeded best when alkyl iodides were employed,
although the yield was respectable with the corresponding
(13) Garc´ıa Ruano, J. L.; Alema´n, J.; Aranda, M. T.; Ferna´ndez-Iba´n˜ez,
M. A.; Rodr´ıguez-Ferna´ndez, M. M.; Maestro, M. C. Tetrahedron 2004,
60, 10067.
(14) R3N and R3SiX have also been used in some cases to produce the
normal Pummerer reaction; see: Tokitoh, N.; Igarashi, Y.; Ando, W.
Tetrahedron Lett. 1987, 28, 5903.
(15) Under similar conditions, ethylphenylsulfoxide underwent C-sily-
lation; see: Miller, R. D.; Hassig, R. Tetrahedron Lett. 1984, 25, 5351. An
alternative vinylogous silicon-Pummerer rearrangement could have also been
invoked (see: Brook, A. G.; Anderson, D. G. Can. J. Chem. 1968, 46,
2115. Shainyan, B. A.; Kipichenko, S. V.; Freeman, F. J. Am. Chem. Soc.
2004, 124, 11456) to explain the stereochemical results encountered.
However, we could not obtain any evidence indicating the formation of a
C-silyl derivative, even when working in the presence of HMPA, which
would enhance the nucleophilicity at the carbon atom.
(17) This behavior is also found with the alkylation of other sulfinyl
carbanions; see: Drabowicz, J.; Kielbasinski, P.; Mikolajczky, M. In The
Chemistry of Sulphones and Sulphoxides; Patai, S., Rappoport, Z., Stirling,
C., Eds.; Wiley: Chichester, 1988; pp 305-317.
(18) There are some examples in the literature where substitution rather
than elimination was observed with phenylethyl-substituted systems; see:
Solladie´-Cavallo, A.; Martin-Cabrejas, L. M.; Caravatti, G.; Lang, M.
Tetrahedron: Asymmetry 2001, 12, 967.
(16) Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Am. Chem.
Soc. 1991, 113, 4092.
(19) Stabilization of the carbanionic intermediate by the nitrogen atom
may be responsible for the lack of reactivity of 19.
Org. Lett., Vol. 7, No. 1, 2005
21