5569
Table 2. The oxidation ofalkenyl sulfides (2) and the chemical shifts of vinyl protons
Run
2
4
Yield
%
(Chemical shift of vinyl proton (8))a)
(Chemical shift of vinyl proton (15))a)
2c
2d
( E;
5.33 Z; 5.36 )
4c
4d
( E; 7.00, Z; 5.83 )
95
90
Z; 5.60 )b,c)
Z; 5.35 )b,d)
Z, 5.94 )b,e)
Z; 5.77 )b,f)
Z; 6.23 )
( E; 5.60
( E; 5.35
( E; 6.00
( E; 5.77
( E; 5.93
(E; 5.13
( E, > 6.3, Z, 5.90 )b,c)
( E; > 6.3, Z; 5.83 )b,d)
( E, > 6.6, Z; 5.94 )b,e)
( E; > 6.3, Z; 5.90 )b,f)
( E; > 6.7, Z; 5.98 )
( E; 6.80, Z; 5.54 )
2e
4e
75
2f
2g
2h
4f
4g
4h
93
87
91
Z; 5.13 )
(E; 5.71,Z; 6.01 )
( E; > 6.5, Z; 5.74 )
a) Measured in CC14. b) The signals were assigned by the comparison with the NMR spectrum of the authentic
compound prepared from the corresponding aromatic ketone.6) c) 2,4-Xylyl. d) 2,6-Xylyl. e) p-Tolyi, f) o-
Tolyl.
oxidation (Table 2). It is rational to assume that the former is the E-isomer and the latter is Z-isomer. This
assumption is also supported by the prediction of the chemical shifts using Z-factors.7)
The stereocbemical results listed in Table I indicate that the number of substituents on benzene ring is a
dominant factor to determine the main stereoisomer. Furthermore, the mixtures with the same proportions of
stereoisomers were formed from both Z and E- 1a. This fact suggests that the initial products of the present
reaction were isomerized to the more thermally stable ones under the reaction conditions. In fact, it was
confirmed that the alkenyl sulfide (2f (E : Z ffi 60 : 40 )) was thermally isomerized to its equilibrium composition
(22 : 78) by heating in benzene for 16 h. Furthermore, the distributionofstereoisomers of authentic alkenyl
sulfides (2) prepared by the reaction of aromatic ketones with thiopbenol in the presence of phosphorus
pentoxide, which was expected to be a thermodynamically controlled process, was found to be similar to that
observed in the reaction of 1 with arenes ( 2d (2,4-xylyl), E : Z ffi 66 : 34; 2d (2,6-xylyl), E : Z ffi 79 : 21; 2e
(o-tolyl), E : Z ffi 59 : 41; 2e (p-tolyl), E : Z = 20 : 80; 2t", E : Z = 19 : 81 ).
As noted above, 1-(phenylthio)vinylstannane (Z-la) react with SnCI4 to form some intermediate. In order to
determine the structure of intermediate, Z-I a was treated with SnCI4 in anisole at r.t. After evaporation of the
volatile materials under reduced pressure, the NMR spectrum of the residue was measured in CDCI3 and the sole
triplet of vinyl proton was observed at ~ = 7.03 which corresponded to neither the starting material (Z- I a) (15 =
6.43) nor (Z)- 1-chloro-1-(phenylthio)-l-pentene (5) (5 = 6.27). Further, (Z)- 1-(phenylthio)- l-(triphenyl-
stannyl)-1-pentene was isolated in 82% yield by the treatment of the intermediate with excess phenylmagnesium
bromide in anisole. These results suggest that the intermediate of the present reaction is the trichlorovinyl-
stannane (6). It was confirmed that 6 was thermally stable and only a trace amount of the vinyl chloride (5) was
detected when it was heated in refluxing m-xylene (see Table 1, run 6).
Some of the alkenyl sulfides were hydrolyzed to the corresponding aromatic ketones to determine the ratio of
position isomers. Since the distribution of position isomers was similar to that of common Friedel-Crafis type
reactions (Table 3), it is reasonable to assume that the present reaction proceeds through the initial formation of
the arenium ion (7) by the attack of arene on the trichlomvinylstannane (6) or the vinyl cation species (8) formed
from 6 as illustrated in Eq. 2.