Highly stereoselective cationic cyclization assisted by a sulfenyl group, scope, limitation, and mechanism

Article abstract of DOI:10.1021/jo951128f

When 8-acetoxy-2-naethyl-9-(phenylthio)-2-nonene (la) was treated with an acid, followed by a base, alkylative cyclization proceeded to give a mixture of 1,2-disubstituted cyclohexanes: 2a, 3a, and 4a. The stereochemistry of the reaction was only slightly affected by the leaving group and the reaction conditions, such as the temperature, solvent, and acid. However, the bulkiness of the sulfenyl group had a great effect on the stereochemical course of the reaction. High trans selectivity was attained when 1c (a derivative of 1a with a bulkier sulfenyl group) was used as a substrate. On the other hand, the length and rigidity of the carbon chain of the substrate also had a major effect on the stereochemistry of the reaction; a high cis selectivity was observed when 10a (a one-carbon-fewer analog of Ia) or 15a (a derivative with one more double bond in the carbon chain than in 1a) was used as the substrate. The reaction proceeded via a 6,5- or 5,5-fused-ring intermediate. The sulfenyl-group-assisted reaction could be a useful method for the stereoselective cyclization of acetates of α-sulfenylated secondary alcohols.

Full text of DOI:10.1021/jo951128f

494
J . Org. Chem. 1996, 61, 494-502
High ly Ster eoselective Ca tion ic Cycliza tion Assisted by a Su lfen yl
Gr ou p . Scop e, Lim ita tion , a n d Mech a n ism
Changqing Liu, Kazuaki Kudo, Yukihiko Hashimoto, and Kazuhiko Saigo*
Department of Chemistry and Biotechnology, Graduate School of Engineering, The University of Tokyo,
Hongo, Bunkyo-ku, Tokyo 113, J apan
Received J une 21, 1995X
When 8-acetoxy-2-methyl-9-(phenylthio)-2-nonene (1a ) was treated with an acid, followed by a base,
alkylative cyclization proceeded to give a mixture of 1,2-disubstituted cyclohexanes: 2a , 3a , and
4a . The stereochemistry of the reaction was only slightly affected by the leaving group and the
reaction conditions, such as the temperature, solvent, and acid. However, the bulkiness of the
sulfenyl group had a great effect on the stereochemical course of the reaction. High trans selectivity
was attained when 1c (a derivative of 1a with a bulkier sulfenyl group) was used as a substrate.
On the other hand, the length and rigidity of the carbon chain of the substrate also had a major
effect on the stereochemistry of the reaction; a high cis selectivity was observed when 10a (a one-
carbon-fewer analog of 1a ) or 15a (a derivative with one more double bond in the carbon chain
than in 1a ) was used as the substrate. The reaction proceeded via a 6,5- or 5,5-fused-ring
intermediate. The sulfenyl-group-assisted reaction could be a useful method for the stereoselective
cyclization of acetates of R-sulfenylated secondary alcohols.
In tr od u ction
ity and that the reaction of the cation with silylated
carbon nucleophiles proceeded regioselectively due to the
neighboring participation of the sulfenyl group.8 How-
ever, such a reaction could not be applied to a saturated
secondary cation, even when the cation was stabilized
by the participation of a sulfenyl group; this failure would
arise from an instability of the cationic intermediate. We
thus considered that such an unstable intermediate
would successfully react with a nucleophilic moiety if the
reaction is designed as an intramolecular version, which
is entropically more favorable than an intermolecular
counterpart. On the basis of this consideration, we tried
to carry out the cyclization of some acetates of â-sulfe-
nylated secondary alcohols and found that the corre-
sponding cyclized products were obtained in good yields
with high stereoselectivity.9
Alkylative cationic cyclization is one of the attractive
methods for constructing carbocycles and has thus been
widely applied for terpene syntheses.1 Among them, the
simplest reaction is the biomimetic synthesis of limonene/
terpinolene by the cyclization of a nerol derivative. This
reaction was originally studied by using biointermediate
neryl pyrophosphate, itself, as a substrate.2 Since then,
the leaving group and/or activator was modified in order
to apply the strategy to synthetic organic chemistry.
Mukaiyama and co-workers showed that 2-halopyri-
dinium salts effectively activate nerol to give the corre-
sponding cyclized product in high yield.3 Nozaki et al.
reported on the cyclization of diethyl neryl phosphate by
using organoaluminum species4 and on the halogenative
cyclization of nerol by utilizing a TiCl4-amine complex.5
An enantioselective version of the cyclization was also
demonstrated by Yamamoto and co-workers by using a
chiral leaving group and a suitably designed aluminum
reagent.6 Thus, cationic cyclizations to cyclohexene
derivatives have been thoroughly investigated. However,
the formation of the corresponding cyclohexane deriva-
tives through cationic cyclization has been scarcely
reported, except for cases in which the initially formed
cations are tertiary or allylic;7 one of the drawbacks for
these simple cationic cyclizations is the dependence on
the substrate structure.
In this paper we report on the cationic cyclization
reaction in detail along with mechanistic aspects.
Resu lts a n d Discu ssion
When the acetate of â-sulfenylated secondary alcohol
1a was treated with a 1.1 molar equivalent of trimeth-
ylsilyl trifluoromethanesulfonate (TMSOTf) for 24 h, the
starting material was completely consumed. However,
no cyclized product was detected, but instead, a very
polar species was formed. The species decomposed quite
slowly upon standing at room temperature to give a much
less polar product, which was identified to be a mixture
of cyclized products: 2a , 3a , and 4a . On the basis of this
result, the reaction conditions were at first examined
concerning the conversion of the polar species into 2a ,
3a , and 4a . As a result, it was found that the decomposi-
tion was enhanced by treating the polar species with a
base.
We recently described that an allylic cation, to which
a sulfenyl group participated, showed appropriate stabil-
* Author to whom correspondence may be addressed: FAX, 03-5802-
3348; e-mail, saigo@chiral.t.u-tokyo.ac.jp
X Abstract published in Advance ACS Abstracts, December 15, 1995.
(1) Ho, T.-L. Carbocycle Construction in Terpene Synthesis; VCH
Publishers: New York, 1988; 277.
(2) Cramer, F.; Rittersdorf, W. Tetrahedron 1967, 23, 3015.
(3) Kobayashi, S.; Tsutsui, M.; Mukayama, T. Chem. Lett. 1976,
1137.
(4) Kitagawa, Y.; Hashimoto, S.; Iemura, S.; Yamamoto, H.; Nozaki,
H. J . Am. Chem. Soc. 1976, 98, 5030.
(5) Saito, T.; Itoh, A.; Ohshima, K.; Nozaki, H. Tetrahedron Lett.
1979, 3519.
The present reaction is the first example concerning
the formation of 1,2-dialkylcyclohexane by cationic cy-
clization; the reaction would proceed with the aid of a
neighboring group participation of the sulfenyl group.
(6) Sakane, S.; Fujiwara, J .; Maruoka, K.; Yamamoto, H. J . Am.
Chem. Soc. 1983, 105, 6154; Tetrahedron 1986, 42, 2193.
(7) Sutherland, J . K. Comprehensive Organic Synthesis; Pergamon
Press: Oxford, 1991; Vol. 3, p 341.
(8) Kudo, K.; Hashimoto, Y.; Houchigai, H.; Hasegawa, M.; Saigo,
K. Bull. Chem. Soc. J pn. 1993, 66, 848.
(9) A part of this work has been published previously: Kudo, K.;
Hashimoto, Y.; Saigo, K. Tetrahedron Lett. 1993, 34, 7063.
0022-3263/96/1961-0494$12.00/0 © 1996 American Chemical Society

Stereoselective Cationic Cyclization
J . Org. Chem., Vol. 61, No. 2, 1996 495
Ta ble 2. Cycliza tion of 1a -c
Ta ble 1. Cycliza tion of 1a u n d er Va r iou s Con d ition s
solvent yield/% 2a :3a :4a a
entry
substrate
R
yield/%
2:3:4a
entry Lewis acid
base
1
2
2
1a
1b
1c
Ph
85
84
78
31:63:6
20:76:4
4:96:0
1
2
3
4
5
6
7
8
9
10
11
12
13
TMSOTf
none
CH2Cl2
0
74
78
82
85
53
70
72
50
74
39
79
65
Mesb
Proton Spongeb
Et2NH
25:52:23
36:55:9
30:54:16
31:63:6
38:55:7
40:54:6
40:60:0
39:55:6
54:46:0
40:46:14
45:55:0
50:50:0
TIPPc
iBuNH2
Et3N
a
b
Determined by GC analysis Mes ) 2,4,6-trimethylphenyl
(mesityl). c TIPP ) 2,4,6-triisopropylphenyl.
hexane
toluene
CH3CN
Et2O
CHCl3
CH2Cl2
Ta ble 3. Cycliza tion of 5a ,c
TiCl4
SnCl4
Ph3CSbCl6
a
b
Determined by GC analysis. 1,8-Bis(dimethylamino)naph-
thalene.
This fact prompted us to thoroughly investigate various
reaction conditions, such as the base, temperature,
solvent, and Lewis acid as an activator. The results are
given in Table 1.
At first, the reaction was conducted at various tem-
peratures in CH2Cl2 using TMSOTf as a Lewis acid. The
reaction could not take place at temperatures below -23
°C, but proceeded quite slowly at -23 °C. Only at room
temperature could the reaction proceed at a reasonable
rate. Then, a subsequent study concerning the reaction
was carried out at room temperature. Although bases
could enhance the decomposition of the polar intermedi-
ate, in some cases the yield was low and the ratio of exo
isomer 4a increased (entries 2-4). In contrast, triethy-
lamine suppressed the formation of exo isomer 4a while
maintaining high yield (entry 5). The reaction was then
performed in various solvents with TMSOTf as an
activator, followed by treatment with Et3N (entries
5-10). The yield strongly depended on the solubility of
the polar intermediate in each solvent; the highest yield
was achieved when CH2Cl2 was used as a solvent. The
effect of a Lewis acid was also studied (entries 5, 11-
13). Among the Lewis acids examined, TMSOTf was the
most preferable from the viewpoint of the yield, although
the selectivity for diastereoisomers 2a and 3a was not
observed. On the basis of these results, we concluded
that the optimum reaction conditions for the present
cationic cyclization were those of entry 5 in Table 1;
subsequent reactions were carried out under such condi-
tions.
entry
substrate
R
yield/%
6:7:8:9a
1
2
5a
5c
Ph
86
91
40:60:0:0
0:91:0:9
TIPPb
a
b
Determined by GC analysis. TIPP ) Triisopropylphenyl.
the bulkiness of the sulfenyl group greatly affected the
stereoselectivity. The increase in the bulkiness of the
sulfenyl group resulted in a higher yield of the trans
isomer (Table 2); when 2,4,6-triisopropylbenzenesulfenyl
was used as a sulfenyl group, the cis/trans ratio reached
4:96 (entry 3).
This cationic cyclization could also be applied to
dimethylated substrate 5 (Table 3). In this case, the
same tendency was also observed concerning the effect
of the sulfenyl group on the stereoselectivity; the use of
the bulkier 2,4,6-triisopropylbenzenesulfenyl group re-
sulted in a higher trans selectivity. Noteworthy was that
the product mixture consisted of trans isomer 7c and
endo isomer 9c in a ratio of 91:9 and that neither the cis
nor exo isomer was detected (entry 2).
In the next stage, we examined the effect of the length
of the carbon chain on the stereoselectivity of the reac-
tion. When 10a , a one-carbon-fewer analog of 1a , was
treated with TMSOTf, the formation of a polar species
was also observed on TLC. Treatment of the reaction
mixture with Et3N gave a mixture of the corresponding
cyclized products in 70% yield, which comprised 83% of
cis isomer 11a , 17% of exo isomer 13a , and a trace
amount of endo isomer 14a ; surprisingly, the major
product was the cis isomer, and no trans isomer was
detected (Table 4, entry 1). We then optimized the
reaction conditions. The results are listed in Table 4.
Although the yield of a mixture of the cyclized products
and the ratio of the exo and endo isomers strongly
In the next stage, we examined the effects of the
leaving group and the sulfenyl group on the reaction.
When the leaving group was an acetoxyl group, the yield
was 85%, and the ratio of 2a , 3a , and 4a was 31:63:6
(Table 1, entry 5); when the leaving group was a hydroxyl
or trifluoroacetoxyl group, the yield and the isomer ratio
were 85% and 40:60:0 and 85% and 43:57:0, respectively.
These results indicate that the leaving group has little
effect on the stereoselectivity of the reaction. However,

496 J . Org. Chem., Vol. 61, No. 2, 1996
Liu et al.
Sch em e 1
Ta ble 4. Cycliza tion of 10a u n d er Va r iou s Con d ition s
citral as a 3:2 mixture of (Z)- and (E)-isomers, was treated
with 1.1 molar equivalents of TMSOTf in CH2Cl2 at room
temperature, the starting material was quickly con-
sumed, also giving a very polar species. The polar species
was decomposed by treatment with Et3N to give a
mixture of the cyclized products in 79% yield; cis isomer
16a was obtained with very high selectivity (the ratio of
cis/trans was 94:4) in spite of the stereochemical hetero-
geneity of 15a .
Treatment of 19a with TMSOTf in CH2Cl2, followed
by Et3N, gave a mixture of the cyclized products in 71%
yield; the cis isomer was also obtained with high selectiv-
ity (the ratio of cis/trans was 82:9), although there were
small amounts of two other isomers. Thus, the rigidity
of the carbon chain has a great effect on the stereochem-
istry of the reaction and strikingly increases the cis
selectivity.
entry Lewis acid
base
Et3N
Et3NH
iBuNH2
DBU
solvent yield/% 11a :12a :13a :14a a
1
2
3
4
5
6
7
8
TMSOTf
CH2Cl2
70
68
70
78
66
85
59
65
83:0:17:trace
77:0:45:3
51:0:20:4
58:0:37:5
40:0:57:3
52:0:42:6
13:0:85:2
34:0:62:4
iPr2NEt
EtONa
2,6-lutidine
Proton
Spongeb
Et3N
In order to clarify the mechanism of the reaction, we
tried to identify the polar intermediate by monitoring the
1
reaction of 1a by H NMR and 13C NMR.
1H NMR signals of 1a slowly faded, and new peaks
appeared upon the treatment of 1a with TMSOTf in CD2-
Cl2 at room temperature. The characteristic changes of
9
10
CH3CN
CH3NO2
CH2Cl2
53
64
47
77
70
87:0:9:4
76:0:18:6
90:0:10:0
32:66:2:0
6:85:9:0
11 Ph3CClO4
12 SnCl4
13 TiCl4
1
the H NMR signals are the following: (1) low-field shift
of the signals of the phenyl protons; (2) disappearance of
the signals of the two protons at around 5 ppm, which
can be assigned as the olefinic proton and the methyne
proton on the carbon adjacent to the acetoxyl group; (3)
disappearance of the signal of the methylene protons on
the carbon adjacent to the phenylthio group; (4) appear-
ance of new signals (at least six kinds) between 3.4 and
4.4 ppm, three of them are clear double doublets having
a large J value, which is of geminal couplings; (5) shift
of the signal of the methyl protons at 1.6 ppm to a higher
magnetic field, which splits into three distinct singlets
at 0.96, 1.17, and 1.20 ppm (the area ratio of the three
peaks was roughly 1:2:2); (6) split of the signal of the
other methyl protons at 1.7 ppm into three different
peaks at 1.63, 1.70, and 1.82 ppm (the area ratio was
roughly 2:2:1).
Moreover, the characteristic changes of 13C NMR
signals are the following: (1) disappearance of the signals
at 73 (the carbon adjacent to the acetoxyl group), 125 (the
olefinic carbon), and 127 ppm (the olefinic carbon); (2)
appearance of signals around 45, 80, and 120 ppm.
Taking these observations into account, the intermedi-
ate was considered to be sulfonium ions 26 (Scheme 2).
A precise investigation led us to conclude that intermedi-
ate 26 was a 2:3 mixture of cis diastereomeric sulfonium
a
b
Determined by GC analysis. 1,8-Bis(dimethylamino)naph-
thalene.
depended on the base and solvent used, the exclusive cis
selectivity was not affected by the solvent and base at
all (entries 1-10). It is, however, noteworthy that the
Lewis acid used had a great effect on the stereoselectivity.
When TMSOTf was used as a Lewis acid, cis isomer 11a
was exclusively obtained (entry 1), whereas when a metal
chloride was used as a Lewis acid, trans isomer 12a
became major; in the case of TiCl4, the ratio of cis/trans
reached 6:85 (entry 13). These results indicate that the
length of the carbon chain has a major effect on the
stereoselectivity of the reaction and that the cyclization
of a substrate, which gives a five-membered ring product,
can afford the cis or trans isomer selectively by using a
suitable Lewis acid for activation of the substrate.
Since the length of the carbon chain had a great effect
on the stereochemistry of the reaction, we considered that
the rigidity of the carbon chain would also have some
effect on the stereoselectivity of the reaction. On the
basis of this consideration, we carried out cyclizations of
substrates 15a and 19a (Scheme 1), which have one more
double bond in the carbon chain than does 1. When
substrate 15a , which could be easily synthesized from

Stereoselective Cationic Cyclization
J . Org. Chem., Vol. 61, No. 2, 1996 497
Sch em e 2
intermediates 26T and 26C, respectively, before an
interconversion between 25T and 25C takes place. Thus,
the trans product is produced as a major product under
kinetic control. Apparently, when the R1 group is bulkier,
the steric congestion in a cis transition state like 28C
would become more serious, resulting in the more pre-
dominant formation of 25T. On the other hand, the
leaving group does not affect the ratio of intermediates
26T and 26C at all, but rather affects the formation of
episulfonium intermediate 24. Hence, in these cycliza-
tions the leaving group ought to have no effect on the
stereochemistry, and a bulky sulfenyl group should be
favorable in the formation of the trans product via trans
intermediate 26T.
The fact that endo isomer 9c was obtained as a minor
product in the place of cis isomer 6c or exo isomer 8c in
the cyclization of 5c could be explained as follows. A
stable intermediate like 26C could not be formed due to
the severe interaction between the bulky 2,4,6-triisopro-
pylphenyl and the gem-dimethyl group as well as the 1,3-
diaxial interaction. Subsequently, cis cyclic exo cation
25C is converted to cyclic endo cation 27 via hydrogen
migration; a following deprotonation gives endo isomer
9c.
F igu r e 1. Possible transition states for the cyclization of 1a .
salts 26C and trans diastereomeric sulfonium salts 26T,
1
on the basis of a comparison of the H NMR spectrum
with those of 3,4,6,7,8,9-hexahydro-1,1,5-trimethylisoben-
zofuran isomers reported by Goldsmith and co-workers.10
Cis isomer 2a (via route a) and exo isomer 4a (via route
b) should be derived from cis intermediate 26C, whereas
trans isomer 3a should be produced from trans interme-
diate 26T, since the elimination reactions of these
intermediates are considered to occur in an anti manner.
The ratio of (2a + 4a )/3a (about 2:3) is in good agreement
with that of 26C/26T. On the basis of these results, a
possible mechanism for the cyclization would be as shown
in Scheme 2.
The facts that the leaving group has little effect on the
stereoselectivity and that a bulky sulfenyl group favors
the formation of the trans isomer could be explained on
the basis of a conformational argument (Figure 1).
Assuming a trans configuration for the episulfonium ion
part and a “chair” conformation for the carbon chain in
these cyclizations, transition state 28T, leading to trans
intermediate 25T, would be more favorable than transi-
tion state 28C, leading to cis intermediate 25C, because
of less steric congestion between the aryl group attached
to the sulfur and the gem-diethyl group at the developing
carbocationic center. Intermediates 25T and 25C would
be immediately converted into stable 6,5-fused-ring
1
The H NMR spectrum of the reaction mixture of 10a
and TMSOTf showed two singlets at 1.30 and 1.57 ppm,
which were assigned to the gem-dimethyl group of cis
diastereomeric intermediates 31. As described above, the
decomposition of 31 is responsible for the formation of
cis product 11a and exo product 13a . Subsequently, the
cyclization reaction of 10a is considered to proceed in a
mechanism similar to that of 1a (Scheme 3); intermedi-
ates 30C and 30T are irreversibly formed from 29.
Although cis cyclic cation 30C is considered to be less
favorable than trans cyclic cation 30T, due to a repulsion
of the two cis-substituted groups, it is well-known that
in a 5,5-fused system, a cis connection is much more
favorable than a trans connection. Therefore, cis cation
30C is attacked as soon as it is generated by the sulfur
atom to form stable diastereomeric 5,5-fused sulfonium
ion 31. On the other hand, 30T would be converted into
30C and then into 31 via cyclic endo cation 32, which is
formed by hydrogen migration of 30T, since the formation
of a trans 5,5-fused system can scarcely take place. As
a result, thermodynamically controlled cis product 11a
(10) Goldsmith, D. J .; Clark, B. C., J r.; J oines, R. C. Tetrahedron
Lett. 1967, 1211.

498 J . Org. Chem., Vol. 61, No. 2, 1996
Liu et al.
Sch em e 3
Sch em e 4
and exo isomer 13a were obtained almost exclusively.
Minor product 14a may be produced from cyclic endo
cation 32.11 When TiCl4 is used as a Lewis acid, trans
cation 30T would be captured as soon as it is generated
by a chloride ion to form neutral compound 33.12 Thus,
kinetically controlled trans product 12a was produced as
a main product. This proposed mechanism is strongly
supported by the fact that the treatment of a mixture,
obtained in entry 13 of Table 4 (cis isomer 11a :trans
isomer 12a :exo isomer 13a ) 6:85:9), with trifluoroacetic
acid in CH2Cl2 for 6 h, followed by quenching with
triethylamine, gave a mixture of cis isomer 11a and exo
isomer 13a in a ratio of 74:26, and no trans isomer 12a
was detected.
F igu r e 2. Possible transition states for the cyclization of 15a .
The reaction of 15a was monitored by 1H NMR to
explain the high cis selectivity. The signal of 15a quickly
faded upon the addition of TMSOTf, and new peaks
appeared. The 1H NMR spectrum of the reaction mixture
showed two sets of singlets at 1.18 and 1.63 ppm and at
1.14 and 1.84 ppm, respectively; the ratio of the peak
areas for the two sets of peaks was roughly 2:1. Although
we could not assign the peaks exactly, we deduced that
they belonged to the gem-dimethyl group of the two
diastereomeric sulfonium salts 35 shown in Scheme 4.
The decomposition of 35 would give rise to the cis and
exo isomers. The high selectivity, irrespective of the
stereochemical heterogeneity of the starting material
15a , indicates that in this reaction, an isomerization of
the double bond would take place in analogy with that
in the intermolecular reaction of (R-sulfenylmethyl)allyl
acetate.8 On the other hand, it is known that an
electrophilic attack toward a double bond occurs from the
direction where the π-orbital of the double bond is
maximally overlapped with the vacant lobe of the elec-
trophile.13 Two possible transition states for the cycliza-
tion of 15a are illustrated in Figure 2. From the figure
we can see that because of the rigidity of the carbon
chain, the π-electron of the double bond is much more
accessible to the electrophilic episulfonium cation in
transition state 36C (giving the cis isomer) than that in
transition state 36T (giving the trans isomer). Taking
these considerations into account, the mechanism for the
reaction of 15a would be as shown in Scheme 4.
Deter m in a tion of Rela tive Ster eoch em istr y of
P r od u cts
X-ray crystallographic analyses were performed for
sulfone 37 and sulfone 38, derived from 11a and 16a ,
respectively (Scheme 5). On the basis of these results,
(11) J ohnson, F.; Malhotra, S. K. J . Am. Chem. Soc. 1965, 87, 5492.
(12) A TLC analysis showed that the reaction intermediates con-
sisted of species less polar than the starting material.
(13) Nagase; S.; Ray; N. K.; Morokuma, K. J . Am. Chem. Soc. 1980,
102, 4536.

Stereoselective Cationic Cyclization
J . Org. Chem., Vol. 61, No. 2, 1996 499
Gen er al P r ocedu r e for th e P r epar ation of 1-(Ar ylth io)-
2,3-ep oxyp r op a n e: To a suspension of 1-chloro-2,3-epoxypro-
pane (0.3 mol) and sodium hydroxide (0.3 mol, pellet) in
dioxane (50 mL) was added arenethiol (0.1 mol) in dioxane
(50 ml) over a period of 20 min at room temperature with
stirring. Stirring was continued overnight. The precipitate
was removed by filtration. Evaporation of the solvent, followed
by distillation under reduced pressure, gave the title com-
pound.
Sch em e 5
1,2-Ep oxy-3-(p h en ylth io)p r op a n e: 95% yield; bp 110 °C/1
mmHg. 1H NMR was identical with that in the literature.14
1,2-Ep oxy-3-(m esitylth io)p r op a n e: 91% yield; bp 117-8
°C/1 mmHg; 1Η ΝΜR (270 MHz, CDCl3) 2.26 (s, 3H), 2.30 (dd,
1H, J ) 2.6, 4.5), 2.52 (s, 6H), 2.57 (dd, 1H, J ) 6.6, 13.5),
2.69 (ddd, 1H, J ) 0.7, 4.0, 4.2), 2.88 (ddd, 1H, J ) 0.7, 5.1,
13.5), 3.01-3.19 (m, 1H), 6.93 (s, 2H).
1,2-E p oxy-3-((2,4,6-t r iisop r op ylp h en yl)t h io)p r op a n e.
The purification was carried out by column chromatography
(eluent: hexane/AcOEt ) 20/1) instead of distillation: 61%
yield; pale-yellow oil; 1H NMR (270 MHz, CDCl3) 1.22 (d, 18H,
J ) 6.9), 2.27-2.50 (m, 1H), 2.50-2.97 (m, 4H), 2.97-3.14 (m,
1H), 3.97 (hept, 2H, J ) 6.9), 7.08 (s, 2H).
8-Acetoxy-2-m eth yl-9-(p h en ylth io)-2-n on en e (1a ). In a
similar manner to a method described in the literature,15
5-methyl-4-hexenal was obtained in 76% yield by the reaction
of 2-methyl-3-buten-2-ol with ethyl vinyl ether in the presence
of a catalytic amount of phosphoric acid at 150 °C under a
pressure of 5 atm for 2 h.
To the thus-obtained aldehyde (63.79 g, 0.57 mol) was added
sodium borohydride (8.93 g, 0.23 mol) in water (100 mL) in
such a way that the temperature of the system was maintained
at 40-50 °C with vigorous stirring. After the addition was
completed, stirring was continued until the temperature of the
system reached below 30 °C. The organic layer was then
separated, and the aqueous phase was extracted with ether
(3 × 20 mL). The organic and ethereal phases were combined,
washed with brine (20 mL), and dried over anhydrous sodium
sulfate. Evaporation of the solvent, followed by distillation
at reduced pressure (100-105 °C/44 mmHg), gave 47.78 g (0.42
mol, 74% yield) of 5-methyl-4-hexen-1-ol. The 1H NMR was
identical with that given in the literature.16
the relative configurations of 11a and 16a were confirmed
to be cis.
The relative configurations of the other products were
1
confirmed on the basis of a correlation of the H NMR
chemical shifts of selected peaks. The signals of the two
terminal vinyl protons appeared at 4.5-5.0 ppm as two
quasi-singlets. In the cases of the trans isomers, the
difference in the chemical shift between the quasi-
singlets was small, while it was large in the cases of the
cis isomers.
In conclusion, we have provided a useful method for
the highly stereoselective cyclization of secondary enyl
alcohols. By cyclization, trans-1,2-disubstituted cyclo-
hexane, cis-3,4-disubstituted cyclohexene, and cis- or
trans-1,2-disubstituted cyclopentane could be obtained
in fair-to-good yields with high stereoselectivity. Since
the products have easily functionalizable moieties in the
substituents, such as a sulfenyl group and a double bond,
this reaction would be potent in the preparation of
precursors for the synthesis of various kinds of cis- or
trans-1,2-disubstituted five- and six-membered carbocy-
cles.
The alcohol was transformed to 5-methyl-4-hexenyl bromide
by NBS-PPh3 in 47% yield according to the method described
in the literature.17 The 1H NMR was identical with that
reported in the literature.17
Exp er im en ta l Section
Gen er a l. The starting materials and reagents, purchased
from commercial suppliers, were used without further puri-
fication. All of the solvents were dried and distilled before
use and stored over sodium wire or molecular sieves. “Con-
centration under reduced pressure” refers to the use of a
rotatory evaporator at the water aspirator pressure. The
residual solvent was removed at a pressure of less than 5
mmHg using a vacuum pump. The reaction flasks were flame-
dried under a stream of argon. The syringes were oven-dried
and cooled to room temperature in a desiccator over phospho-
rous pentoxide. Analytical thin-layer chromatography (TLC)
was conducted on 1.0 × 4.0 cm precoated glass TLC plates
(0.2 mm layer thickness of Silica Gel 60 F-254). The spots
were visualized by an ultraviolet light, exposure to iodine
vapor, or heating after dipping in a 3-5% solution of phos-
phomolybdic acid in ethanol. Flash chromatography was
carried out using Silica Gel 60 (70-230-mesh ASTM). Pre-
parative TLC (PTLC) was carried out with Wakogel B-5F (0.75
mm layer thickness).
To a partially solidified solution of 5-methyl-4-hexenylmag-
nesium bromide in THF (10 mL), which was prepared from
the bromide (3.61 g, 20.4 mmol) and Mg (0.53 g, 22 mmol),
was added 1,2-epoxy-3-(phenylthio)propane (2.61 g, 15.7 mmol)
in THF (5 mL) at room temperature. Then, CuI (200 mg, 1.0
mmol) was added to the mixture. After an induction period
(ca. 1 min), an exothermic reaction occurred, and the solvent
was vigorously refluxed. The reaction then settled after a few
minutes. The reaction mixture was stirred for another 30 min,
poured into ice water, and partitioned between EtOAc and
water. The aqueous layer was extracted with EtOAc (3 × 30
mL). The combined organic layers were washed with brine
(50 mL), dried over anhydrous sodium sulfate, and concen-
trated under reduced pressure. Purification by column chro-
matography (eluent: hexane/EtOAc ) 10/1) gave 8-methyl-1-
(phenylthio)-7-nonen-2-ol (2.50 g, 60% yield): 1H NMR (270
MHz, CDCl3) 1.2-1.6 (m, 6H), 1.58 (s, 3H), 1.68 (s, 3H), 1.90-
2.03 (m, 2H), 2.40 (d, 1H, J ) 3.4), 2.84 (dd, 1H, J ) 8.8, 13.7),
3.15 (dd, 1H, J ) 3.3, 13.7), 3.60-3.72 (m, 1H), 5.04-5.15 (m,
1H), 7.18-7.42 (m, 5H); IR (neat) 3150-3600 (br), 3070, 2930,
2870, 1585, 1482, 1440, 1090, 1030, 740, 695. Anal. Calcd
for C16H24OS: C, 72.67; H, 9.14. Found: C, 72.65; H, 9.19.
Acetylation of the alcohol by Ac2O-pyridine in the presence
of a catalytic amount of 4-(dimethylamino)pyridine (DMAP),
The melting points were measured using a metal-block
apparatus and an open capillary tube and are uncorrected. GC
analyses were performed with a 25 m fused-silica capillary
column using cyanopropyl silicone as a stationary phase. NMR
spectra were measured on FT spectrometers operating at 270
or 400 MHz for 1H and 68 or 100 MHz for 13C or on a
1
1
continuous-wave instrument at 60 MHz for H. For H NMR,
the δ values are given in ppm with TMS as an internal
standard; the coupling constants are recorded in Hz. For 13C
NMR, the chemical shifts are reported in ppm relative to TMS
(14) Gu, X.-P.; Ikeda, I.; Okahara, M. Bull. Chem. Soc. J pn. 1987,
60, 667.
(15) Marbet, R.; Saucy, G. Helv. Chim. Acta 1967, 30, 218.
(16) Franke, L. R. R.-A.; Wolf, H. Tetrahedron 1984, 40, 3491.
(17) Somers, P. K.; Wandless, T. J .; Schreiber, S. L. J . Am. Chem.
Soc. 1991, 113, 8045.
or CDCl3 (δ 77.0). The unit for the values of IR spectra is cm-1
.
Low-resolution mass spectra were recorded at an ionization
potential of 70 eV, and the relative intensity is given in
parentheses after the corresponding m/ z value.

500 J . Org. Chem., Vol. 61, No. 2, 1996
Liu et al.
followed by purification by column chromatography (eluent:
hexane/EtOAc ) 15/1), gave 2.74 g (95% yield) of 1a : colorless
oil; 1H NMR (400 MHz, CD2Cl2) 1.2-1.4 (m, 4H), 1.61 (s, 3H),
1.69 (s, 3H), 1.6-1.75 (m, 2H), 1.95 (s, 3H), 1.90-2.00 (m, 2H),
3.07 (dd, 1H, J ) 5, 14), 3.13 (dd, 1H, J ) 6, 14), 4.90-5.00
(m, 1H), 5.00-5.10 (m, 1H), 7.20-7.40 (m, 5H); 13C NMR (CD2-
Cl2, 100 MHz) 17.7, 21.1, 25.2, 25.8, 28.2, 30.0, 33.5, 38.0, 73.1,
124.8, 126.6, 129.3, 129.9, 131.8, 136.7, 170.7; IR (neat) 3070,
2970, 2930, 2865, 1740, 1585, 1485, 1440, 1375, 1240, 1025,
740, 694. Anal. Calcd for C18H26O2S: C, 70.54; H, 8.55.
Found: C, 70.36; H, 8.60.
8-Acetoxy-9-(m esitylth io)-2-m eth yl-2-n on en e (1b). The
procedure was the same as that for 1a , except that 1,2-epoxy-
3-(mesitylthio)propane was used in place of 1,2-epoxy-3-
(phenylthio)propane: colorless oil; 62% yield for two steps from
5-methyl-4-hexenyl bromide; 1H NMR (270 MHz, CDCl3) 1.20-
1.37 (m, 4H), 1.58 (br s, 3H), 1.68 (br s, 3H), 1.64-1.75 (m,
2H), 1.94 (s, 3H), 1.87-2.01 (m, 2H), 2.25 (s, 3H), 2.49 (s, 6H),
2.80 (d, 1H, J ) 2.3), 2.82 (d, 1H, J ) 2.0), 4.84-4.93 (m, 1H),
5.04-5.11 (m, 1H), 6.91 (s, 2H); 13C NMR (68 MHz, CDCl3)
17.7, 20.92, 20.97, 21.9, 24.8, 25.7, 27.8, 29.6, 33.1, 38.8, 73.6,
124.4, 129.0, 130.1, 131.5, 138.0, 142.6, 170.6; IR (neat) 3030,
2970, 2930, 2870, 1740, 1600, 1460, 1375, 1240, 1022, 852.
Anal. Calcd for C21H32O2S: C, 72.37; H, 9.25. Found: C,
72.41; H, 9.18.
8-Acetoxy-2-m eth yl-9-((2,4,6-tr iisop r op ylp h en yl)th io)-
2-n on en e (1c). The procedure was the same as that for 1a ,
except that 1,2-epoxy-3-((2,4,6-triisopropylphenyl)thio)propane
was used in place of 1,2-epoxy-3-(phenylthio)propane: colorless
oil; 60% yield for two steps from 5-methyl-4-hexenyl bromide;
1H NMR (270 MHz, CD2Cl2) 1.24 (d, 18H, J ) 6.9), 1.15-1.38
(m, 4H), 1.59 (s, 3H), 1.68 (d, 3H, J ) 1.0), 1.63-1.73 (m, 2H),
1.90-2.30 (m, 2H), 2.74-2.88 (m, 2H), 2.86 (h, 1H, J ) 6.9),
3.90 (h, 2H, J ) 6.9), 4.98 (p, 1H, J ) 6.0), 5.04-5.15 (m, 1H),
6.99 (s, 1H); IR (neat) 3050, 2970, 2930, 2870, 1742, 1600,
1465, 1375, 1240, 1024, 940, 880. Anal. Calcd for C27H44O2S:
C, 74.95; H, 10.25. Found: C, 74.94; H, 10.18.
8-Acetoxy-2,6,6-tr im eth yl-9-(ph en ylth io)-2-n on en e (5a).
To a THF (8 mL) solution of tetramethylethylenediamine
(TMEDA) (0.81 g, 7 mmol) and thioanisole (0.87 g, 7 mmol)
was added n-BuLi (4.09 mL, 1.71 M in hexane) at 0 °C; the
resulting mixture was stirred at room temperature for 1 h.
3,3-Dimethyl-6-octenal (0.84 g, 5 mmol) in THF (4 mL), which
was prepared by the methylation of citral with Me2CuLi (73%
yield),18 was added to the mixture at -78 °C. After being
stirred at -78 °C for 15 min, the reaction mixture was allowed
to warm to room temperature. The reaction was then quenched
with a saturated NH4Cl aqueous solution (15 mL). The usual
workup gave 1.37 g of crude 1-(methylthio)-4,4,8-trimethyl-7-
nonen-2-ol. The alcohol was acetylated by Ac2O-pyridine in
the presence of a catalytic amount of DMAP to give 1.52 g of
5a after purification by column chromatography (eluent:
hexane/AcOEt ) 20/1): 91% yield; colorless oil; 1H NMR (270
MHz, CDCl3) 0.85 (s, 6H), 1.14-1.20 (m, 2H), 1.57 (s, 3H), 1.66
(s, 3H), 1.50-1.72 (m, 2H), 1.83-1.20 (m, 2H), 1.93 (s, 3H),
2.95 (dd, 1H, J ) 7.0, 13.5), 3.14 (dd, 1H, J ) 5.0, 13.5), 4.97-
5.08 (m, 1H), 7.15-7.52 (m, 5H); 13C NMR (68 MHz, CDCl3)
17.6, 21.2, 22.6, 25.7, 27.3, 32.6, 39.0, 42.4, 43.9, 70.4, 124.9,
126.2, 128.9, 129.4, 131.0, 135.8, 170.6; IR (neat) 3070, 2970,
2930, 2875, 1740, 1585, 1483, 1440, 1375, 1240, 1025, 740, 693.
Anal. Calcd for C20H30O2S: C, 71.80; H, 9.03. Found: C,
71.56; H, 8.84.
1598, 1465, 1370, 1240, 1022, 938, 880. Anal. Calcd for
C29H48O2S: C, 75.59; H, 10.05. Found: C, 75.40; H, 10.53.
7-Acetoxy-2-m eth yl-8-(p h en ylth io)-2-octen e (10a ). The
title compound was prepared in the same manner as that for
1a , except for the use of 4-methyl-3-pentenyl bromide19 in place
of 5-methyl-4-hexenyl bromide: 62% yield; colorless oil; 1H
NMR (400 MHz, CDCl3) 1.2-1.4 (m, 2H), 1.57 (s, 3H), 1.67 (s,
3H), 1.6-1.78 (m, 2H), 1.95 (s, 3H), 1.87-2.03 (m, 2H), 3.04
(dd, 1H, J ) 6, 14), 3.12 (dd, 1H, J ) 6, 14), 4.96-5.12 (m,
2H), 7.15-7.42 (m, 5H); 13C NMR (100 MHz, CD2Cl2) 17.7,
21.0, 256.4, 25.7, 27.6, 32.6, 37.7, 72.8, 124.0,126.3, 128.9,
129.6, 131.9, 136.0, 170.6; IR (neat) 3070, 2970, 2930, 2870,
1740, 1582, 1480, 1440, 1375, 124, 1025, 742, 695. Anal.
Calcd for C17H24O2S: C, 69.82; H, 8.27. Found: C, 69.99; H,
8.22.
8-Acet oxy-2,6-d im et h yl-9-(p h en ylt h io)-2,6-n on a d ien e
(15a ). The procedure was the same as that for 5a , except for
the use of citral in place of 3,3-dimethyl-6-octenal, giving 15a
as a 3:2 mixture of cis and trans isomers in 78% yield: colorless
oil; 1H NMR (400 MHz, CDCl3) 1.56 (s, 3H, minor), 1.59 (s,
3H, major), 1.649 (s, 3H, major), 1.652 (s, 3H, minor), 1.68 (s,
3H, major), 1.72 (d, 3H, J ) 1.2, major), 1.94-2.10 (m, 4H),
1.96 (s, 3H, minor), 1.97 (s, 3H, major), 3.01 (dd, 1H, J ) 6.1,
13.7, major), 3.02 (dd, 1H, J ) 5.5, 13.4, minor), 3.16 (dd, 1H,
J ) 7.0, 13.4, minor), 3.19 (dd, 1H, J ) 7.0, 13.7, major), 4.98-
5.12 (m, 1H), 5.13-5.20 (m, 1H), 5.60-5.71 (m, 1H), 7.15-
7.42 (m, 5H); IR (neat) 3070, 2975, 2930, 2860, 1740, 1582,
1480, 1440, 1372, 1240, 1020, 742, 695. Anal. Calcd for
C19H26O2S: C, 71.65; H, 8.22. Found: C, 71.37; H, 8.08.
1-[3-Acetoxy-4-(p h en ylth io)bu tyl]-2-(2-m eth yl-1-p r op e-
n yl)ben zen e (19a ). To a stirred solution of phthalide (2.68
g, 20 mmol) in CH2Cl2 (40 mL) at -90 °C was added over a
period of 5 min diisobutylaluminum hydride (0.93 M in hexane,
21.5 mL, 20 mmol). After being stirred for 2 h at -90 °C,
MeOH (3 mL) and a saturated potassium sodium tartrate
solution (16 mL) were successively added; the mixture was
allowed to warm up to room temperature. The usual workup,
followed by purification by column chromatography (eluent:
hexane/AcOEt ) 1/1), gave 2.20 g of 2-(hydroxymethyl)-
benzaldehyde in 81% yield as a 2:1 mixture of the ring-closed
hemiacetal and the free aldehyde. The 1H NMR was identical
with that described in the literature.20
After dimethyl sulfide (5 mL) was introduced into a flask
containing sodium hydride (55% dispersion in mineral oil,
0.436 g, 10 mmol), which was washed several times with
pentane, the mixture was heated at 75-80 °C for 45 min. To
the resulting solution of methylsulfenylmethyl carbanion,21
cooled with an ice-water bath, was added 4.32 g (10 mmol) of
isopropyltriphenylphosphonium iodide22 in 10 mL of warm
dimethyl sulfoxide (DMSO). After being stirred for 10 min at
room temperature, 0.68 g (5 mmol) of 2-(hydroxymethyl)-
benzaldehyde in DMSO (5 mL) was added to the mixture. The
mixture was first stirred at room temperature for 30 min and
then poured into a mixture of cold diluted hydrochloric acid
(20 mL) and AcOEt (20 mL). The usual workup, followed by
purification by column chromatography (eluent: hexane/
AcOEt ) 1/1), gave 0.65 g (80% yield) of 2-(2-methyl-1-pro-
penyl)benzenemethanol: 1H NMR (270 MHz, CDCl3) 1.70 (d,
3H, J ) 1.3), 1.66-1.78 (br, 1H), 1.92 (d, 3H, J ) 1.3), 4.63 (s,
2H), 6.32 (s, 1H), 7.10-7.50 (m, 4H).
According to a method described in the literature,23 2-(2-
methyl-1-propenyl)benzenemethanol was converted into 1-chlo-
romethyl-2-(2-methyl-1-propenyl)benzene in 75% yield by a
reaction with NCS/dimethyl sulfide, followed by purification
by column chromatography (eluent: hexane/AcOEt ) 20/1):
1H NMR (270 MHz, CDCl3) 1.69 (d, 3H, J ) 1.3), 1.93 (d, 3H,
J ) 1.3), 4.57 (s, 2H), 6.37 (s, 1H), 7.10-7.45 (m, 4H).
The procedure for the preparation of the title compound
(19a ) was the same as that for 1a , except for the use of
8-Acetoxy-2,6,6-tr im eth yl-9-((2,4,6-tr iisop r op ylp h en yl)-
th io)-2-n on en e (5c). The procedure was the same as that
for 5a , except that methyl(2,4,6-triisopropylphenyl)methyl
sulfide was used in place of thioanisole: 90% yield; colorless
oil; 1H NMR (270 MHz, CDCl3) 0.90 (s, 6H), 1.23 (d, 12H, J )
6.9), 1.24 (d, 6H, J ) 6.6), 1.18-1.30 (m, 2H), 1.56 (dd, 1H, J
) 2.6, 14.8), 1.61 (s, 3H), 1.68 (d, 3H, J ) 1.0), 1.76 (dd, 1H,
J ) 8.6, 14.8), 1.91 (s, 3H), 1.88-2.0 (m, 2H), 2.72 (dd, 1H, J
) 5.1, 12.2), 2.86 (hept, 1H, J ) 6.6), 2.88 (dd, 1H, J ) 6.2,
12.2), 3.91 (hept, 2H, J ) 6.9), 5.02-5.12 (m, 1H), 5.12-5.22
(m, 1H), 6.99 (s, 2H); IR (neat) 3050, 2970, 2930, 2875, 1740,
(19) Biernacki, W.; Gdula, A. Synthesis 1979, 37.
(20) Baillargeon, V. P.; Stille, J . K. J . Am. Chem. Soc. 1986, 108,
452.
(21) Corey, E. J .; Chaykovsky, M. J . Am. Chem. Soc. 1962, 84, 866.
(22) Wittig, G.; Wittenberg, D. J ustus Liebigs Ann. Chem. 1957, 606,
1. CA, 52, 1970i.
(18) Sakane, S.; Maruoka, K.; Yamamoto, H. Tetrahedron 1986, 42,
2203.
(23) Woodside, A. B.; Huang, Z.; Poulter, C. D. Org. Synth. 1987,
66, 211.

Stereoselective Cationic Cyclization
J . Org. Chem., Vol. 61, No. 2, 1996 501
1-chloromethyl-2-(2-methyl-1-propenyl)benzene in the place of
5-methyl-4-hexenyl bromide: 72% yield for two steps; colorless
oil; H NMR (270 MHz, CDCl3) 1.66 (d, 3H, J ) 1.3), 1.89 (d,
MHz, CD2Cl2) (only the characteristic peaks were recorded)
for 6a 0.90 (s, 3H), 0.97 (s, 3H), 1.71 (br s, 3H), 2.84 (dd, 1H,
J ) 8.9, 12.9), 2.94 (dd, 1H, J ) 6.1, 12.9), 4.72 (br s, 1H),
4.85-4.89 (m, 1H); for 7a 0.89 (s, 3H), 0.93 (s, 3H), 1.59 (br s,
3H), 2.44 (dd, 1H, J ) 9.0, 12.2), 3.10 (dd, 1H, J ) 2.3, 12.2),
4.73-4.78 (m, 2H); MS m/ z (rel intensity) for 6a 274 (M+, 11),
1512 (37), 123 (28), 109 (37), 95 (37), 69 (100), 55 (41); for 7a
274 (M+, 17), 151 (100), 123 (34), 109 (51), 95 (66), 69 (74), 55
(50); IR (neat) 3075, 2950, 2930, 2870, 1645, 1585, 1480, 1440,
890, 738, 696. Anal. Calcd for C18H26S: C, 78.76; H, 9.45.
Found: C, 78.94; H, 9.68.
1
3H, J ) 1.6), 1.96 (s, 3H), 1.86-2.0 (m, 2H), 2.50-2.67 (m,
2H), 3.03 (dd, 1H, J ) 6.3, 13.9), 3.15 (dd, 1H, J ) 5.9, 13.9),
4.93-5.07 (m, 1H), 6.23 (br s, 1H), 7.03-7.40 (m, 9H); 13C NMR
(68 MHz, CDCl3) 19.2, 21.0, 26.0, 29.1, 33.7, 37.4, 72.7, 123.5,
125.7, 126.3, 126.5, 128.7, 128.9, 129.7, 130.0, 135.7, 135.8,
137.5, 139.4, 170.6; IR (neat) 3070, 3020, 2970, 2930, 2870,
1740, 1585, 1485, 1442, 1378, 1240, 1030, 742, 695. Anal.
Calcd for C22H26O2S: C, 74.53; H, 7.39. Found: C, 74.51; H,
7.33.
Gen er a l P r oced u r e for Cycliza tion . To a solution of the
substrate (0.35 mmol) in CH2Cl2 (2 mL) was added a CH2Cl2
solution of TMSOTf (1.2 M, 0.32 mL, 0.38 mmol) at room
temperature; the resulting mixture was stirred for 24 h.
During this period the color of the solution turned black. Then,
triethylamine (0.5 mL, 3.6 mmol) was added to the mixture;
the resulting pale-brown solution was stirred for 8 h. The
mixture was partitioned between CH2Cl2 (5 mL) and a
saturated NH4Cl solution (5 mL). The aqueous layer was
extracted with CH2Cl2 (2 × 5 mL), and combined organic layers
were dried over anhydrous Na2SO4. After evaporation of the
solvent, the crude material was purified by means of a
preparative TLC (eluent: hexane) to give a mixture of the
cyclized compounds.
cis-1-((P h e n ylt h io)m e t h yl)-2-(2-p r op e n yl)cycloh e x-
a n e (2a ), tr a n s-1-((p h en ylth io)m eth yl)-2-(2-p r op en yl)-
cycloh exa n e (3a ), a n d 1-isop r op ylid en e-2-((p h en ylth io)-
m eth yl)cycloh exa n e (4a ): 1H NMR (400 MHz, CD2Cl2) (only
the characteristic peaks were recorded) for 2a 1.62 (s, 3H),
2.77 (dd, 1H, J ) 11, 13, PhSCHH), 2.89 (ddd, 1H, J ) 1, 3,
13, PhSCHH), 4.64 (s, 1H), 4.85 (s, 1H); for 3a 1.60 (s,3H),
1.86 (ddd, 1H, J ) 3, 12, 12, CH2C(CH3)CH), 2.51 (dd, 1H, J
) 10, 13, PhSCHH), 3.13 (dd, 1H, J ) 3, 13, PhSCHH), 4.72
(s, 1H), 4.75 (s, 1H); for 4a 1.90-2.00 (m), 2.5-2.6 (m), 3.0-
3.1 (m); MS m/ z (rel intensity) for 2a 246 (M+, 17), 164 (5),
149 (5), 137 (45), 123 (70),110 (21), 95 (40), 81 (91), 69 (100),
55 (54); for 3a 246 (M+, 12), 136 (17), 123 (100), 81 (52), 69
(31),67 (34), 55 (31); for 4a 246 (M+, 12), 136 (10), 123 (100),
81 (61), 67 (33), 55(21); IR (neat) 3075, 2930, 2860, 1640, 1584,
1480, 1440, 890, 735, 690. Anal. Calcd for C16H22S: C, 77.99;
H, 9.00. Found: C, 77.80; H, 8.82.
cis-1-((Mesit ylt h io)m et h yl)-2-(2-p r op en yl)cycloh ex-
a n e (2b), tr a n s-1-((m esitylth io)m eth yl)-2-(2-p r op en yl)-
cycloh exa n e (3b), a n d 1-isop r op ylid en e-2-((m esitylth io)-
m eth yl)cycloh exen e (4b): 1H NMR (400 MHz, CD2Cl2) (only
the characteristic peaks were recorded) for 2b 4.52 (br s, 1H),
4.74 (br s, 1H); for 3b 1.46 (br s, 3H), 2.24 (s, 3H), 2.48 (s,
6H), 2.45-2.55 (m, 1H), 2. 69 (dd, 1H, J ) 2.6, 12.2,
MesSCHH), 4.60 (br s, 1H), 4.62-4.67 (m, 1H), 6.90 (s, 2H);
for 4b 1.93-2.06 (m); MS m/ z (rel intensity) for 2b 152 (M+
- 136, 56), 137 (56), 119 (19), 95 (38), 81 (76), 69 (100); for 3b
288 (M+, 12), 273 (7), 165 (14), 152 (100), 137 (31), 123 (41),
119 (21), 95 (24), 81 (64), 69 (47); for 4b 152 (M+ - 136, 52),
123 (93), 81 (100), 67 (47); IR (neat) 3075, 2930, 2860, 1640,
1603, 1450, 13475, 890, 850. Anal. Calcd for C19H28S: C,
79.10; H, 9.78. Found: C, 78.95; H, 9.64.
tr a n s-1,1-Dim eth yl-4-(2-p r op en yl)-3-[((2,4,6-tr iisop r o-
p ylp h en yl)th io)m eth yl]cycloh exa n e (7c) a n d 4,4-d im e-
th yl-1-isop r op yl-2-[((2,4,6-tr iisop r op ylp h en yl)th io)m eth -
1
yl]cycloh exen e (9c): H NMR (400 MHz, CD2Cl2) (only the
characteristic peaks were recorded) for 7c 0.95 (s, 3H), 0.97
(s, 3H), 1.24 (d, 12H, J ) 6.9), 1.32 (d, 6H, J ) 6.9), 1.15-1.32
(m, 2H), 1.35-1.6 (m, 3H), 1.46 (br s, 3H), 1.62-1.8 (m, 2H),
1.85-1.95(m, 1H), 2.27 (dd, 1H, J ) 9.6, 11.8), 2.44 (dd, 1h, J
) 2.0, 9.6), 2.87 (hept, 1H, J ) 6.9), 3.90 (hept, 2H, J ) 6.90),
4.60-4.68 (m, 2H,), 6.99 (s, 2H); for 9c 3.24 (s, 2H), 7.00 (s,
2H); MS m/ z (rel intensity) for 7c 249 (M+ - 151, 27), 236
(90), 221 (30), 165 (23), 151 (47), 123 (13), 109 (72), 95 (40);
for 9c 249 (M+ - 151, 36), 236 (71), 151 (19), 121 (20), 109
(69), 95 (30); IR (neat) 3080, 2970, 2930, 2870, 1644, 1600,
1560, 1465, 1385, 890, 878. Anal. Calcd for C27H44S: C, 80.93;
H, 11.06. Found: C, 81.18; H, 11.24.
cis-1-((P h e n ylt h io)m e t h yl)-2-(2-p r op e n yl)cyclop e n -
ta n e (11a ), tr a n s-1-((p h en ylth io)m eth yl)-2-(2-p r op en yl)-
cyclop en t a n e (12a ), 1-isop r op ylid en e-2-((p h en ylt h io)-
m eth yl)cyclop en ta n e (13a ), a n d 1-isop r op yl-2-((p h en yl-
th io)m eth yl)cyclop en ten e (14a ): 1H NMR (400 MHz, CD2-
Cl2) for 11a 1.5-1.9 (m, 6H), 1.73 (s, 3H), 2.2-2.3 (m, 1H),
2.43 (t, 1H, J ) 12, PhSCHH), 2.4-2.5 (m, 1H), 2.86 (dd, 1H,
J ) 4, 12, PhSCHH), 4.74 (br s, 1H), 4.87 (br s, 1H), 7.1-7.3
(m, 5H); for 12a (only the characteristic peaks were recorded)
1.59 (s, 3H), 1.9-2.0 (m, 1H), 2.2 (t, 1H, J ) 8), 2.6-2.7 (m,
1H), 2.66 (dd, 1H, J ) 9, 12, PhSCHH), 3.16 (dd, 1H, J ) 3,
12, PhCHH), 4.59-4.71 (m, 2H); for 13a (only the character-
istic peaks were recorded) 2.63 (t, 1H, J ) 12, PhSCHH), 3.02
(dd, 1H, J ) 4, 12, PhSCHH); for 14a 0.87 (d, 6H, J ) 6.9),
3.62 (s, 2H); MS m/ z (rel intensity) for 11a 232 (M+, 21), 189
(7), 164 (19), 123 (100), 109 (49), 81 (88), 67 (97), 55 (60); for
12a 232 (M+, 11), 123 (58), 109 (100), 79 (39), 67 (43), 55 (30);
for 13a 232 (M+, 15), 123 (49), 109 (100), 81 (28), 67 (49), 45
(22); for 14a 123 (M+ - 109, 88), 81 (100); IR (neat) 3038, 2970,
2870, 1645, 1587, 1484, 1440, 1377, 1093, 1030, 895, 740, 695.
Anal. Calcd for C15H20S: C, 77.52; H, 8067. Found: C, 77.23;
H, 8048.
cis-1-Meth yl-3-((p h en ylth io)m eth yl)-4-(2-p r op en yl)cy-
cloh exen e (16a ), tr a n s-1-m eth yl-3-((p h en ylth io)m eth yl)-
4-(2-p r op en yl)cycloh exen e (17a ), a n d 4-isop r op ylid en e-
1-m eth yl-3-((p h en ylth io)m eth yl)cycloh exen e (18a ): 1H
NMR (400 MHz, CD2Cl2) (only the characteristic peaks were
recorded) for 16a 1.671 (s, 3H), 1.673 (s, 3H), 1.55-1.63 (m,
2H), 1.97-2.08 (m, 2H), 2.20-2.28 (m, 1H), 2.43-2.52 (m, 1H),
2.60 (dd, 1H, J ) 10.3, 12.0, PhCHH), 2.98 (dd, 1H, J ) 12.0,
3.6, PhSCHH), 4.71 (br s, 1H), 4.88 (br s, 1H), 5.68-5.73 (m,
1H), 7.10-7.35 (m, 5H); for 17a 4.78-4.83 (m, 2H), 5.38-5.43
(m, 1H); for 18a 5.54-5.9 (m, 1H); MS m/ z (rel intensity) for
16a 258 (M+, 27), 189 (11), 164 (11), 149 (33), 135 (57), 124
(38), 107 (76), 93 (100), 81 (71), 77 (59); for 17a 258 (M+, 22),
189 (8), 149 (13), 135 (36), 124 (46), 107 (64), 93 (100), 81 (38);
for 18a 258 (M+, 3), 135 (100), 107 (19), 97 (16); IR (neat) 3080,
2975, 2930, 1645, 1585, 1483, 1440, 1380, 895, 740, 693. Anal.
Calcd for C17H22S: C, 79.01; H, 8.58. Found: C, 79.22; H, 8.42.
cis-2-((P h en ylt h io)m et h yl)-1-(2-p r op en yl)-1,2,3,4-t et -
r ah ydr on aph th alen e (20a), tr a n s-2-((ph en ylth io)m eth yl)-
1-(2-pr open yl)-1,2,3,4-tetr ah ydr on aph th alen e (21a), 1-iso-
p r op ylid en e-2-((p h en ylt h io)m et h yl)-1,2,3-t r ih yd r on a p -
h th a len e (22a ), a n d 1-isop r op yl-2-((p h en ylth io)m eth yl)-
3,4-d ih yd r on a p h th a len e (23a ): 1H NMR (270 MHz, CDCl3)
(only the characteristic peaks were recorded) for 20a 1.45 (br
s, 3H), 1.45-1.6 (m, 2H), 1.85-2.0 (m, 1H), 2.26-2.38 (m, 1H),
2.71 (dd, 1H, J ) 9.0, 12.9, PhCHH), 2.73-2.90 (m, 2H), 3.29
(dd, 1H, J ) 3.6, 12.9, PhSCHH), 3.40 (d, 1H, J ) 9.5), 4.83-
cis-1-(2-P r op en yl)-2-[((2,4,6-tr iisop r op ylp h en yl)th io)-
m eth yl]cycloh exa n e (2c), tr a n s-1-(2-p r op en yl)-2-[((2,4,6-
tr iisopr opylph en yl)th io)m eth yl]cycloh exan e (3c): 1H NMR
(400 MHz, CD2Cl2) (only the characteristic peaks were re-
corded) for 2c 4.50, (br s, 1H), 4.71 (br s, 1H); for 3c 1.21 (d,
6H, J ) 6.9), 1.30 (d, 12H, J ) 6.9), 1.42 (br s, 3H), 1.18-
1.37(m, 4H), 1.40-1.64 (m, 2H), 1.67-1.84 (m, 3H), 2.21-2.34
(m, 1H), 2.28 (dd, 1H, J ) 9.6, 11.9), 2.59 (dd, 1H, J ) 2.3,
11.9), 2.86 (hept, 1H, J ) 6.9), 3.92 (hept, 2H, J ) 6.9), 4.61
(s, 1H), 4.62 (s, 1H), 6.98 (br s, 2H); MS m/ z (rel intensity) for
2c 372 (M+, 17), 329 (15), 249 (22), 236 (100), 221 (34), 137
(73), 123 (30, 95 (39); for 3c 372, (M+, 12), 329 (9), 249 (21),
236 (100), 221 (33), 137 (70), 123 (37), 95 (31); IR (neat) 3070,
2970, 2930, 2870, 1640, 1595, 1464, 1383, 1363, 890, 870.
Anal. Calcd for C25H40S: C, 80.58; H, 10.82. Found: C, 80.32;
H, 10.82.
cis-1,1-Dim et h yl-3-((p h en ylt h io)m et h yl)-4-(2-p r op en -
yl)cycloh exa n e (6a ), tr a n s-1,1-d im eth yl-3-((p h en ylth io)-
m eth yl)-4-(2-p r op en yl)cycloh exa n e (7a ): 1H NMR (400

502 J . Org. Chem., Vol. 61, No. 2, 1996
Liu et al.
4.90 (m, 1H), 5.02-5.07 (m, 1H); for 21a 1.61 (br s, 3H), 2.95
(dd, 1H, J ) 6.0, 13.5), 3.76 (d, 1H, J ) 4.9), 4.94-4.98 (m,
1H); for 22a 1.77 (s, 3H), 1.85 (s, 3H), 3.10 (dd, 1H, J ) 6.4,
13.1); for 23a 1.22 (d, 6H, J ) 7.3), 4.11 (s, 2H); MS m/ z (rel
intensity) for 20a 294 (M+, 34), 185 (31), 171 (100), 169 (82),
143 (65), 129 (80), 128 (63), 115 (36); for 21a 185 (M+ - 109,
10), 169 (8), 141 (8), 129 (12), 128 (7), 44 (40), 32 (100); for
22a 294 (M+, 18), 185 (11), 171 (31), 169 (15), 156 (10), 143
(18), 129 (20), 44 (29), 32 (100); for 23a 294 (M+, 15), 185 (40),
171 (20), 169 (23), 149 (34), 141 (29), 129 (55), 115 (29), 32
(100); IR (neat) 3075, 3020, 2925, 2860, 1645, 1480, 1440, 1375,
900, 740, 693. Anal. Calcd for C20H22S: C, 81.57; H, 7.53.
Found: C, 81.67; H, 7.58.
CHH), 2.94 (dd, 1H, J ) 1, 14, PhSO2CHH), 4.64 (s, 1H), 4.83
(s, 1H),7.5-7.7 (m, 3H), 7.85-7.98 (m, 2H); 13C NMR (68 MHz,
CDCl3) 22.1, 22.9, 217.1, 30.5, 35.6, 50.9, 56.6, 112.1, 128.0,
129.2, 133.5, 139.8, 144.5; IR (KBr pellet) 3070, 2970, 2880,
1648, 1450, 1310, 1155, 1090, 900, 755, 695. Anal. Calcd for
C15H20O2S: C, 68.14; H, 7.62. Found: C, 67.84; H, 7.69.
c-3-(Ben zen esu lfon ylm eth yl)-t-1-m eth yl-c-4-(2-p r op e-
n yl)-r -1,t-2-cycloh exa n ed iol (38):24 colorless crystals; mp
154-155 °C (from EtOH); 1H NMR (400 MHz, CDCl3) 1.30 (s,
3H), 1.10-1.95 (m, 6H), 2.45-2.60 (m, 2H), 2.95 (dd, 1H, J )
2.2, 14.3, PhSO2CHH), 4.06 (dd, 1H,
J ) 11.5, 14.3,
PhSO2CHH), 4.18 (br s, 1H), 4.64 (br s, 1H), 4.85 (br s, 1H),
7.50-7.70 (m, 3H), 7.85-7.98 (m, 2H); IR (KBr pellet) 3465,
3410, 3190, 3175, 2990, 2975, 2930, 2870, 1645, 1450, 1300,
1150, 1090, 1045, 907, 731. Anal. Calcd for C17H24O4S: C,
62.93; H, 7.45. Found: C, 62.81; H, 7.37.
cis-1-(Ben zen esu lfon ylm eth yl)-2-(2-pr open yl)cyclopen -
ta n e (37):24 colorless crystals; mp 49-50 °C (from EtOH); 1H
NMR (400 MHz, CDCl3) 1.4-1.9 (m, 6H), 1.57 (s, 3H), 2.4-
2.5 (m, 1H), 2.5-2.6 (m, 1H), 2.75 (dd, 1H, J ) 11, 14, PhSO2-
Ack n ow led gm en t. This work was financially sup-
ported by a Grant-in-Aid for Scientific Research (No.
06453132) from the Ministry of Education, Science,
Sports and Culture of J apan.
(24) The author has deposited atomic coordinates for the structures
of 37 and 38 with the Cambridge Crystallographic Data Centre. The
coordinates can be obtained, on request, from the Director, Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge, CB2 1EZ,
UK.
J O951128F