tively, of irreversible ring closure were produced. Interest-
ingly, species 1 (X ) SO2) is much more amenable to ring
closure compared to radical 1 (X ) S) under similar
conditions, giving over 95% cyclized products. Nevertheless,
while the mode of cyclization of these radicals is of intrinisic
interest, the procedure does not represent a particularly
convenient entry into sulfur-substituted heterocyclic com-
pounds in view of the mixture of isomeric products so
obtained.
Table 1. Product Ratios from Bu3SnH Reduction of Radicals 1
and 4 (X ) S, SO2)
concn
(M)
T
precursor
radical
(°C) reduced 5-exo 6-endo
8a
4 (X ) S)
4 (X ) S)
4 (X ) S)
4 (X ) SO2) 0.011
4 (X ) SO2) 0.044
0.011
0.011
0.011
80
80
80
80
80
80
80
38.6
35.6
75.7
3.9
8.9
17.1
3.8
7.1
8.3
2.7
2.5
2.3
54.3
56.2
21.6
93.6
88.7
12.8
23.1
9a
9a,b
8a
8a
107
107
We now report a study on the regiochemistry associated
with the cyclization of 5-methyl-substituted radicals 4 (X )
S, SO2) (Scheme 2). By analogy with behavior of the
1 (X ) S)
1 (X ) SO2) 0.013
0.013
70.1
73.1
b
a From this work. Bu3SnH was used in 5-fold excess.
Scheme 2
Table 1. Inspection of the table reveals that the extent of
6-endo versus 5-exo ring closure is enormously enhanced
in the case of the sulfonyl radical 4 (X ) SO2) in which the
ratio of the 5-exo to 6-endo product is 1:37, with little of
the acyclic reduced species 5 (X ) SO2) detected. Under
experimental conditions designed to minimize even further
the extent of reduction,13 the reaction is found to give an
excellent yield (86%) of the substituted cyclohexane 7 (X
) SO2)14 uncontaminated by either isomer 5 (X ) SO2) or
isomer 6 (X ) SO2).
The data within Table 1 further demonstrate that the ratio
of 5-exo to 6-endo products remains constant for radicals 4
(X ) S, SO2) despite changes in concentration, thus
indicating that these reactions occur irreversibly under the
conditions employed in this study.
corresponding all-carbon 5-methyl-5-hexenyl system 4 (X
) CH2), it was anticipated that introduction of a methyl group
at C5 would be accompanied by a reduction in the rate of
5-exo ring closure of species 4 (X ) S, SO2), thus giving an
enhanced yield of the 6-endo product 7 (X ) S, SO2)
compared with isomer 6 (X ) S, SO2). Furthermore, it is
noteworthy that in the case of the R-sulfonyl-substituted
radical, the strong inductively withdrawing influence of the
SO2R group (sp value ) +0.7)8 would be expected to impart
significant electrophilic character onto the radical. Accord-
ingly, given that these reactions are predicted to proceed via
early transition states, regioselectivity is also expected to be
dependent upon the favored interaction of the frontier orbitals
of the reactants.
The selected precursors to the radicals 4 (X ) S, SO2)
under study were the selenides 8 (X ) S, SO2),9 which were
rigorously purified immeditely prior to use. Crich10 has
shown that the presence of PhSeH derived from reaction of
(PhSe)2 with Bu3SnH can introduce undesirable complica-
tions. As a precautionary measure, bistributyltin oxide was
added to the reaction mixture in view of its demonstrated
efficiency as a scavenger of PhSeH.11 Barton ester 9 was
also employed in order to determine whether adventitious
diphenyl diselenide might be influencing the extent of
cyclization.
It is also of interest to compare the behavior of radicals 4
(X ) S, SO2) with that of other R-substituted 5-methyl-5-
(8) Hansch, C.; Leo, A.; Taft, R. W. Chem. ReV. 1991, 91, 165.
(9) Synthesis of the required substrates and their potential products of
reaction will be described in the full paper. The identity of all compounds
was established by appropriate combustion data as well as GC and NMR
spectroscopic analysis. Spectral data and experimental procedures for new
compounds 5 (X ) SO2), 8 (X ) S, SO2), and 9 are provided in Suppporting
Information.
(10) Crich, D.; Yao, Q. J. Org. Chem. 1995, 60, 84.
(11) Maxwell, B. J.; Smith, B. J.; Tsanaktsidis, J. J. Chem. Soc., Perkin
Trans. 2 2000, 425.
(12) Stock solutions of 8 (X ) S, SO2) were prepared in benzene with
the addition of bistributyltin oxide11 and subjected to standard radical
cyclization conditions (catalytic amount of AIBN; 1.2 equiv of Bu3SnH
injected over 2 min; reaction temperature ) 80°C; reaction time ) 2 h;
cooling and then quenching with CCl4 (X ) S) or CH3I (X ) SO2)).
Reaction mixtures were analyzed by GC, and product ratios were then
determined by utilizing GC response factors for each of the expected
products; these response factors were determined using authentic samples
of 5-7, which had been prepared by standard procedures. In the case of
the Barton ester, precursor 9 was added to a refluxing solution of 1.2 equiv
of Bu3SnH.
(13) A 0.01M solution of (3-methyl-but-3-ene-1-sulfonylmethylselanyl)-
benzene 8 (X ) SO2) (0.135 g, 0.43 mmol) in benzene (40 mL) was
deoxygenated, heated to reflux (80 °C), and then treated with a solution of
Bu3SnH (0.16 g, 0.54 mmol) in deoxygenated benzene (3 mL) containing
a few crystals of AIBN over 15 min. The resulting solution was heated for
an additional 3 h and then cooled; the reaction was quenched with CH3I
and the solution concentrated in vacuo. GCMS analysis of the crude product
revealed the presence of 6 (X ) SO2) in minute amount (ca. 3%). Flash
chromatography on silica (hexane/ether) yielded pure sulfone 7 (X ) SO2)
as a white solid (0.055 g, 86%), mp 118-120 °C (lit.14a mp 119-120 °C),
whose 1H NMR14a and 13C NMR14b spectral data were identical to those
reported.
The results of the behavior of radicals 4 (X ) S, SO2)12
together with those for 1 (X ) S, SO2) are summarized in
(5) Vacher, B.; Samat, A.; Allouche, A.; Laknifli, A.; Baldy, A.; Chanon,
M. Tetrahedron 1988, 44, 2925.
(6) Ke, B.-W.; Lin, C.-H.; Tsai, Y.-M. Tetrahedron 1997, 53, 7805.
(7) Della, E. W.; Graney, S. D. Tetrahedron Lett. 2000, 41, 7987.
(14) (a) Kropp, P. J.; Breton, G. W.; Fields, J. D.; Tung, J. C.; Loomis,
B. R. J. Am. Chem. Soc. 2000, 122, 4280. (b) Barbarella, G.; Dembech, P.
Org. Magn. Res. 1984, 22, 402.
4066
Org. Lett., Vol. 4, No. 23, 2002