Free-radical cyclization of enantiomerically enriched 2-p-tolylthio derivatives of 2-allylcyclohexanones with Mn(III): Asymmetric synthesis of bridged bicyclic ketones and thiochroman-3-ones

Article abstract of DOI:10.1016/S0957-4166(99)00467-X

Mn(III)-based oxidative intramolecular cyclization of enantiomerically enriched 2-allyl-2-(p-tolylsulfonyl)-cyclohexanone 4 and 2-allyl-2-(p- tolylsulfenyl)cyclohexanone 9 are reported. The observed chemoselectivity (reaction on the allylic double bond yielding bridged bicyclic ketones vs. reaction on the aromatic ring of the p-tolyl group affording thiochroman-3- ones) depends on the sulfur function (sulfone or thioether, respectively), which determines the electronic density of the p-tolyl ring and the conformational preferences of the starting compounds. The nature of the substituent at C-2 is related to the endo/exo selectivity of the cyclization as well as the regioselectivity in the formation of the enones. (C) 1999 Published by Elsevier Science Ltd.

Full text of DOI:10.1016/S0957-4166(99)00467-X

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 10 (1999) 4427–4436
Free-radical cyclization of enantiomerically enriched 2-p-tolyl-
thio derivatives of 2-allylcyclohexanones with Mn(III):
asymmetric synthesis of bridged bicyclic ketones and thio-
chroman-3-ones†
J. L. García Ruano ∗ and A. Rumbero
Departamento de Química Orgánica, Universidad Autónoma, Cantoblanco, 28049 Madrid, Spain
Received 19 October 1999; accepted 25 October 1999
Abstract
Mn(III)-based oxidative intramolecular cyclization of enantiomerically enriched 2-allyl-2-(p-tolylsulfonyl)-
cyclohexanone 4 and 2-allyl-2-(p-tolylsulfenyl)cyclohexanone 9 are reported. The observed chemoselectivity
(reaction on the allylic double bond yielding bridged bicyclic ketones vs. reaction on the aromatic ring of the
p-tolyl group affording thiochroman-3-ones) depends on the sulfur function (sulfone or thioether, respectively),
which determines the electronic density of the p-tolyl ring and the conformational preferences of the starting
compounds. The nature of the substituent at C-2 is related to the endo/exo selectivity of the cyclization as well as
the regioselectivity in the formation of the enones. © 1999 Published by Elsevier Science Ltd. All rights reserved.
1. Introduction
Bicyclo[3.3.1]nonan-9-one derivatives are interesting intermediates leading to natural and biologically
valuable products,1 novel bridged cyclic amino acids as pharmaceutical agents,2 silaspiro compounds for
liquid crystal composition,3 etc. The problem of the synthesis of these compounds has recently been
solved in an elegant way by Snider4 making use of a Mn(III)-based oxidative free-radical cyclization5 of
1-allyl-2-oxocycloalkanecarboxylate 1 yielding a readily separated mixture of three olefins (Scheme 1).
The ester group controls the regioselectivity of the α-allylation of the cycloalkanone yielding the starting
compound 1. Concerning the cyclization step, the role of the ester group, which avoids the problems
derived from the regioselectivity of the initially formed radical A, is evident, but its plausible influence
∗
†
Corresponding author. E-mail: joseluis.garcia.ruano@uam.es
In memoriam, Professor J. de Pascual Teresa.
0957-4166/99/$ - see front matter © 1999 Published by Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(99)00467-X
tetasy 3103

4428
J. L. García Ruano, A. Rumbero / Tetrahedron: Asymmetry 10 (1999) 4427–4436
in the composition of the final olefin mixtures, achieved from its evolution with Cu(OAc)2·H2O, has not,
so far, been clarified.
Scheme 1.
Some conclusions could be drawn from this problem by reproducing Snider’s reaction sequence
on substrates where the ester has been replaced by other functional groups able to stabilize enolates,
thus also controlling the regioselectivity of the α-allylation. Additionally, in those cases where such
groups were able to induce asymmetric α-alkylation, non-racemic bicyclo[3.3.1]nonan-9-one derivatives
could be obtained. On the basis of the highly stereocontrolled methylation of enantiomerically pure β-
ketosulfoxides,6 we envisioned that the use of the sulfinyl group instead of the ester could be suitable
for preparing enantiomerically enriched bicyclo[3.3.1]nonan-9-one derivatives. In this paper we report
our studies concerning the reactions of 2-thio derivatives (sulfoxides, sulfones, and thioethers) of 2-
allylcyclohexanones with Mn(OAc)3/Cu(OAc)2 under conditions similar to those reported by Snider
(Scheme 2).
2. Results and discussion
The synthesis of compound 3, used as the starting material in our experiments, was carried out from
(R)-2-(p-tolylsulfinyl)-1-cyclohexanone 2, previously reported by our research group,7 by phase-transfer
catalyzed alkylation.8 The reaction afforded an 80:20 mixture of 39 and the undesired O-allylation
product. The ready pyrolytic elimination of the SOTol group from compound 3 hindered its isolation and
complete characterization. When the crude product was treated under the Snider cyclization conditions,
only decomposition products were obtained. These results indicated that the sulfinyl derivatives could
not be used as the starting products of cyclizations due to their instability.
The in situ oxidation of the crude reaction mixture containing 3 (m-CPBA/CH2Cl2, 0°C) afforded the
sulfone 4 (72%, isolated yield) which was isolated and characterized. Oxidative cyclization of 4 with 2
equiv. of Mn(OAc)3·2H2O and 1 equiv. of Cu(OAc)2·H2O in acetic acid at 90°C (conditions described
by Snider) led to the bridged bicyclohexanone 5 (≥95% yield). The reductive elimination of the sulfone
5 with Na2HPO4-buffered sodium amalgam10a takes place with simultaneous reduction of the carbonyl
group10b affording a 3:2 mixture of two isomers 6, which was subsequently oxidized with PCC (CH2Cl2,
rt) to give bicyclo[3.3.1]non-2-en-9-one 7. Several syntheses of the bridged bicycloalkenone 7 have been
reported4c,11 due to its interest as an intermediate in the preparation of spiropyrans as photochromic
substances.12 To our knowledge, the method that we describe herein is the first asymmetric synthesis of
this compound.
1
The structure of the compounds 4–6 and the complete assignment of H and 13C signals were
1
unambiguously established by using H and 13C NMR spectra as well as the 2D NMR techniques:

J. L. García Ruano, A. Rumbero / Tetrahedron: Asymmetry 10 (1999) 4427–4436
4429
Scheme 2. (a) PhCH2N+(Et)3Cl−, allyl bromide, NaOH (50%), CH2Cl2, 0°C; (b) m-CPBA, CH2Cl2, 0°C; (c) 2
Mn(OAc)3·2H2O, 1 Cu(OAc)2·H2O, AcOH, 90°C; (d) Na–Hg (6%), Na2HPO4, MeOH, rt; (e) PCC, silica gel, CH2Cl2, rt;
(f) P2I4, CH2Cl2, rt; (g) (CF3CO)2O, NaI, –40°C, acetone; (h) NaH, allyl bromide, –40°C, DMF
1H–1H COSY, HMQC, HMBC, and NOESY. The structure of 7 was established by comparison with
previously reported spectral data.4c Their enantiomeric excesses (see Scheme 2) were determined by
chiral HPLC.13 Racemic compounds 4–7, required for chromatographic analysis, were synthesized as
depicted in Scheme 2. The oxidation of sulfoxide 2 with m-CPBA in CH2Cl2 at 0°C yielded sulfone 8
(ca. 98%), which was transformed into allylated compound (±)-4 under the same reaction conditions
indicated for 3. Compound (±)-4 was further converted into (±)-5 and (±)-7. Although the ee of
compound 5 cannot be determined,13 it must be similar to those of compounds 4 and 7.
The most significant fact related to cyclization is the exclusive formation of the endo regioisomer 5, in
contrast with the mixture of compounds (two endo- and one exo-isomers) formed in the case of compound
1 (Scheme 1). Moreover, the reactivity of 4 (requiring 8 h to give 5 in 95% yield) is clearly higher than
that of 1 (18 h, 75% yield). These facts suggest a significant role of the substituent at C-2 in the control of
the composition of the reaction mixture. To verify the possible influence of the size of the substituent at
C-2 in the course of these reactions, a study of thioether 9 was performed. Treatment of the crude reaction
mixture containing 3 with diphosphorus tetraiodide14 yielded sulfenyl compound 9 (70% isolated yield).
The enantiomeric excess (86%) was determined by chiral HPLC,13 and the configuration of the major
enantiomer was unequivocally established as (S) by comparison of its specific rotation with that of an
authentic sample of (R)-(+)-9.15 This assignment allows us to establish the (S) configuration at C-2 for
sulfoxide 3 and sulfone 4. Unexpectedly, the reaction of (S)-9 with 2Mn(OAc)3·2H2O/Cu(OAc)2·H2O
in acetic acid at 90°C afforded compound 10 (80% isolated yield and 86% ee) instead of the expected
bicyclo[3.3.1]nonan-9-one (Scheme 2). The structures of compounds 9 and 10 were unambiguously
1
1
determined from their H and 13C NMR spectra, as well as by 2D NMR techniques: H–1H COSY,
HMQC, HMBC, and NOESY. Their enantiomeric excesses were determined by chiral HPLC.13 Racemic

4430
J. L. García Ruano, A. Rumbero / Tetrahedron: Asymmetry 10 (1999) 4427–4436
(±)-9 and (±)-10 were obtained as depicted in Scheme 2. Reduction of sulfoxide 2 with trifluoroacetic
anhydride and sodium iodide16 and further reaction of the resulting thioether 11 with sodium hydride
and allyl bromide17 yielded (±)-9, which was subsequently transformed into (±)-10 under the condition
indicated for S-(−)-9.
Taking into account previous observations in α-sulfinyl carbanion chemistry18 we propose a plausi-
ble mechanism for these Mn(III)-mediated cyclizations, represented in Scheme 3. The stereoselective
allylation of the starting sulfinyl cyclohexanone 2 must be a consequence of the steric differentiation of
the diastereotopic faces of the enolate, due to the spatial arrangement adopted by its oxygens in order to
minimize their electrostatic repulsion. This explanation was proposed to justify the highly stereoselective
methylation of acyclic β-ketosulfoxide6 and is in agreement with the above-mentioned predictions of (S)
configuration at C-2 for the sulfoxide 3. The sulfonyl group is clearly bulkier than the allyl one19 and,
therefore, will adopt an equatorial arrangement. This was unequivocally established by interpretation
of the HMBC experiment.20 The oxidation of 4 with Mn(III) yields intermediate 4a, which reacts with
the double bond mainly affording the endo radical 4b21 instead of the more congested five-membered
exo radical. The increase in the size of the substituent could explain the observed improvement in the
endo/exo selectivity, which explains why it was higher for sulfone 4 than for ester 1. According to the
results obtained by Kochi,22 the formation of an alkylcopper(III) intermediate, such as 4c, could be the
following step in the reaction sequence, previous to the β-hydride elimination leading to (+)-(1S,5S)-5-
(p-tolylsulfonyl)bicyclo[3.3.1]non-2-en-9-one 5. The regioselectivity of this last step could be related to
the allylic strain of the substituent at C-2, 5 always being favored with respect to 50. As the SO2pTol
group is bulkier than CO2Et the evolution of sulfone 4 must be more selective (only 5 was detected) than
that of ester 1 (66:7 mixture of regioisomers was formed, Scheme 1).
Scheme 3.
The formation of (−)-(2S,4R)-2-allyl-6-methyl-2,4-propanethiochroman-3-one 10 could be explained

J. L. García Ruano, A. Rumbero / Tetrahedron: Asymmetry 10 (1999) 4427–4436
4431
by assuming that the conformation favored for compound 9 is that displaying the p-tolylsulfenyl group in
an axial position. The smaller size of the sulfenyl group with respect to the alkyl one19 would justify this
conformational preference, which could be confirmed by the 1H–1H NOESY spectrum of compound 9.23
The axial arrangement of the p-tolylsulfenyl group prevents the attack of the α-carbonyl radical to the
allylic double bond. Taking into account that the electrophilic character of the α-keto radical intermediate
preferentially reacts with nucleophilic double bonds,24 the intramolecular reaction of the radical center
at 9a (Scheme 3) with the electron enriched ring of its p-tolylsulfenyl group is not unexpected. A similar
evolution of 4a in the right conformation exhibiting the SO2pTol group in an axial arrangement would
be unlikely, due to lower electronic density of the p-tolylsulfonyl ring.25
As a final conclusion, we have added to the knowledge on the chemoselectivity of the Mn(III)-based
oxidative free-radical cyclization of unsaturated ketones. The different evolution of compounds 4 and
9 could be explained by assuming the conformation of the substrates and the different nucleophilic
character of the double bonds acting as radical acceptors. The size of the substituent at C-2 could
be related to the endo/exo selectivity of the cyclization, as well as the regioselective formation of the
resulting olefins. Therefore, the different oxidation states (S, SO2) of the sulfur atom have a definitive
influence in the chemoselectivity of these intramolecular cyclizations, and the starting sulfinyl group
is responsible for the enantiomeric excess of the final products. In summary, asymmetric synthesis of
bicyclo[3.3.1]non-2-en-9-one and thiochroman-3-one derivatives has been developed.
3. Experimental
3.1. Materials and general procedure
Melting points were obtained in open capillary tubes and are uncorrected. 1H NMR (300 MHz) and 13
C
NMR (75 MHz) spectra were recorded using TMS as an internal reference. All reactions were monitored
by TLC which was performed on precoated sheets of silica gel 60, and flash column chromatography was
carried out with silica gel (230–400 mesh). Eluting solvents are indicated in the text. β-Ketosulfoxide 2
was prepared according to the described method.7
3.2. (−)-(2S)-2-Allyl-2-(p-toluenesulfonyl)cyclohexanone 4
A solution of sulfoxide 2 (100 mg, 0.43 mmol), benzyltriethylammonium chloride (110 mg, 0.48
mmol), allyl bromide (59 mg, 0.48 mmol), and 50% NaOH (38 µL) in 7 mL of CH2Cl2 was stirred at
0°C for 1 h and 10% HCl was added. The layers were separated and the organic layer was washed with
cold water. To this solution of sulfoxide 3 in CH2Cl2 at 0°C was added dropwise m-chloroperoxybenzoic
acid (115 mg) in 2.5 mL of CH2Cl2. The mixture was stirred at 0°C for 2 h and then a saturated Na2SO3
solution was added. The organic layer was washed (NaHCO3 solution), dried (anhydrous MgSO4), and
evaporated in vacuo. Flash chromatography on silica gel by eluting with hexane:ethyl acetate (4:1) gave
90.5 mg (72%) of 4 as a colorless oil. [α]D20=−163.3 (c 0.5, CHCl3), ee 80%. 1H NMR (CDCl3) δ ppm:
1.64 (1H, qt, J=13.1, 3.8 Hz, H-5ax), 1.78 (1H, m, H-4eq), 2.04 (1H, ddd, J=15.3, 12.8, 4.3 Hz, H-3ax),
2.10 (1H, m, H-5eq), 2.13 (1H, m, H-7a), 2.35 (1H, tt, J=12.6, 3.7 Hz, H-4ax), 2.44 (3H, s, CH3), 2.53
(1H, m, H-6eq), 2.62 (1H, ddt, J=13.5, 5.3, 1.6 Hz, H-7b), 2.70 (1H, ddd, J=15.3, 6.2, 3.8 Hz, H-3eq),
3.05 (1H, ddd, J=15.8, 13.0, 6.3 Hz, H-6ax), 5.01 (1H, br d, J=16.9 Hz, H-9t), 5.08 (1H, br d, J=10.1
Hz, H-9c), 5.43 (1H, m, H-8), 7.32 (2H, d, J=8.1 Hz, H-30), 7.59 (2H, d, J=8.1 Hz, H-20); 13C NMR
(CDCl3) δ ppm: 21.42 (C-4), 21.64 (CH3), 25.21 (C-5), 29.73 (C-3), 38.09 (C-7), 41.50 (C-6), 76.57

4432
J. L. García Ruano, A. Rumbero / Tetrahedron: Asymmetry 10 (1999) 4427–4436
(C-2), 120.26 (C-9), 129.37 (C-30), 130.18 (C-20), 131.66 (C-8), 132.16 (C-40), 145.27 (C-10), 205.01
(C-1). Anal. calcd for C16H20SO3: C, 65.73; H, 6.85. Found: C, 65.45; H, 6.61.
3.3. (+)-(1S,5S)-5-(p-Toluenesulfonyl)bicyclo[3.3.1]non-2-en-9-one 5
Mn(OAc)3·2H2O (720 mg, 2.68 mmol), Cu(OAc)2·H2O (268 mg, 1.34 mmol), and β-ketosulfone 4
(391 mg, 1.34 mmol) were stirred in 20 mL of degassed glacial acetic acid at 90° C for 8 h under argon.
The mixture was diluted with water and a 10% solution of NaHSO3 was added. The resulting solution
was extracted with CH2Cl2. The organic layer was washed (saturated NaHCO3, brine), dried (Na2SO4),
and evaporated in vacuo to give 350 mg (90%) of 5 as white solid, mp 169–171°C. [α]D20=+13.2 (c 1,
CHCl3). 1H NMR (CDCl3) δ ppm: 1.72 (1H, m, H-7a), 1.80 (2H, m, H-8), 2.04 (1H, m, H-7b), 2.18 (1H,
m, H-6a), 2.36 (1H, m, H-6b), 2.45 (3H, s, CH3), 2.75 (1H, ddd, J=18.4, 3.6, 1.5 Hz, H-4a), 2.93 (1H, dt,
J=5.9, 2.9 Hz, H-1), 3.30 (1H, br d, J=18.4 Hz, H-4b), 5.58 (1H, m, H-2), 5.59 (1H, dt, J=9.6, 3.5 Hz,
H-3), 7.98 (1H, d, J=8.3 Hz, H-20), 7.99 (1H, d, J=8.3 Hz, H-30);13C NMR (CDCl3) δ ppm: 17.46 (C-7),
21.61 (CH3), 32.35 (C-8), 36.72 (C-6 and C-4), 49.06 (C-1), 72.99 (C-5), 126.46 (C-2), 127.83 (C-3),
129.15 (C-30), 131.50 (C-20), 133.11 (C-40), 144.88 (C-10), 205.59 (C-9). Anal. calcd for C16H18SO3: C,
66.18; H, 6.24. Found: C, 66.13; H, 6.34.
3.4. Bicyclo[3.3.1]non-2-en-9-ols 6
To a solution of sulfone 5 (60 mg, 0.207 mmol) in MeOH (5 mL) were added Na2HPO4 (916 mg, 5.76
mmol) and powdered 6% Na–Hg (660 mg). The solution was vigorously stirred at room temperature for
48 h. Then the resulting Hg was decanted, and the reaction mixture was diluted with CH2Cl2 (50 mL)
and washed with 1% aqueous NaOH and brine. The organic layer was dried (MgSO4) and evaporated in
vacuo to yield 20 mg (70%) of 6 as a 2:3 mixture of two isomers. Major isomer: 1H NMR (CDCl3) δ
ppm: 1.24 (1H, m, H-8a), 1.30 (1H, m, H-6a), 1.35 (1H, m, H-7a), 1.62 (1H, m, H-7b), 1.87 (1H, m, H-
8b), 1.92 (1H, m, H-6b), 2.00 (1H, m, H-5), 2.08 (1H, m, H-4a), 2.33 (1H, m, H-1), 2.51 (1H, m, H-4b),
3.94 (1H, t, J=3.2 Hz, H-9), 5.55 (1H, m, H-2), 5.81 (1H, dt, J=10.1, 3.6 Hz, H-3); 13C NMR (CDCl3) δ
ppm: 17.07 (C-7), 22.20 (C-8), 27.03 (C-6), 33.30 (C-5), 33.61 (C-4), 36.20 (C-1), 69.84 (C-9), 128.01
(C-2), 129.07 (C-3). Minor isomer: 1H NMR (CDCl3) δ ppm: 1.30 (1H, m, H-7a), 1.47 (1H, m, H-7b),
1.49 (1H, m, H-6a), 1.52 (2H, m, H-8), 1.62 (1H, m, H-6b), 1.83 (1H, m, H-4a), 2.10 (1H, m, H-5),
2.41 (1H, m, H-4b), 2.46 (1H, m, H-1), 3.68 (1H, dt, J=3.5, 1.2 Hz, H-9), 5.53 (1H, m, H-2), 5.97 (1H,
dt, J=9.8, 3.4 Hz, H-3); 13C NMR (CDCl3) δ ppm: 16.23 (C-7), 28.14 (C-4), 28.98 (C-8), 32.91 (C-6),
34.16 (C-5), 37.68 (C-1), 71.58 (C-9), 125.97 (C-2), 130.30 (C-3).
3.5. (−)-(1S, 5R)-Bicyclo[3.3.1]non-2-en-9-one 7
A solution of 6 (20 mg, 0.145 mmol) in 5 mL of dry CH2Cl2 was added to a stirring mixture of silica
gel (55 mg) and pyridinium chlorochromate (55 mg, 0.25 mmol) containing 5 mL of dry CH2Cl2 in one
portion at room temperature. The mixture of reaction was stirred for 2 h and finally diluted with dry
diethyl ether (20 mL). The supernatant solution was decanted and evaporated at room temperature and
atmospheric pressure. The brown residue was purified by silica gel column chromatography (CH2Cl2) to
afford 10 mg of 7 (51%). [α]D20=−13.3 (c 0.6, CHCl3), ee 80%. 1H NMR (CDCl3) δ ppm: 1.53 (2H,
m, H-7), 1.88 (2H, m, H-6), 1.91 (2H, m, H-8), 2.46 (1H, ddd, J=18.5, 3.8, 1.4 Hz, H-4a), 2.58 (1H, m,
H-5), 2.75 (1H, br dd, J=18.5, 7.3 Hz, H-4b), 2.85 (1H, m, H-1), 5.58 (1H, dddd, J=9.6, 6.0, 2.4, 1.7 Hz,
H-2), 5.94 (1H, dt, J=9.6, 3.4 Hz, H-3); 13C NMR (CDCl3) δ ppm: 16.93 (C-7), 33.21 (C-6), 36.89 (C-4

J. L. García Ruano, A. Rumbero / Tetrahedron: Asymmetry 10 (1999) 4427–4436
4433
and C-8), 45.52 (C-5), 47.72 (C-1), 127.17 (C-2), 129.99 (C-3), 216.62 (C-9). The data are identical to
those reported previously.4c
3.6. (±)-2-Allyl-2-(p-toluenesulfonyl)cyclohexanone 4
(a) To a solution of sulfoxide 2 (74 mg, 0.31 mmol) in 5 mL CH2Cl2 at 0°C was added dropwise
m-chloroperoxybenzoic acid (100 mg) in 3 mL of CH2Cl2. The mixture was stirred at 0° C for 2 h. After
standard workup, ketosulfone 8 (75 mg, 96%) was obtained as a colorless oil. 1H NMR (CDCl3) δ ppm:
1.7–2.6 (m, 6H), 2.43 (s, 3H), 2.83 (m, 2H), 3.81 (t, J=5.0 Hz, 1H), 7.35 (d, J=8.3 Hz, 2H), 7.79 (d,
J=8.3 Hz, 2H). (b) By a procedure identical to that described for preparation of (−)-4, the reaction of 8
(75 mg, 0.30 mmol), benzyltriethylammonium chloride (80 mg, 0.35 mmol), allyl bromide (43 mg, 0.35
mmol), and 50% NaOH (28 µL) in 5 mL of CH2Cl2 was stirred at 0° C for 1 h. After standard workup,
(±)-4 (78 mg, 89%) was obtained.
3.7. (±)-5-(p-Toluenesulfonyl)bicyclo[3.3.1]non-2-en-9-one 5
By a procedure identical to that described for preparation of (+)-5, the reaction of (±)-4 (22 mg, 0.075
mmol), Mn(OAc)3.2H2O (40 mg, 0.15 mmol), and Cu(OAc)2.H2O (15 mg, 0.075 mmol) in 3 mL of
degassed glacial acetic acid afforded 20 mg of (±)-5 (92%).
3.8. (−)-(2S)-2-Allyl-2-(p-toluenesulfenyl)cyclohexanone 9
A solution of sulfoxide 2 (200 mg, 0.86 mmol), benzyltriethylammonium chloride (220 mg, 0.96
mmol), allyl bromide (118 mg, 0.96 mmol), and 50% NaOH (76 µL) in 15 mL of CH2Cl2 was stirred at
0° C for 1 h. The organic layer was washed with 10% HCl, cold water, and dried (MgSO4). This solution
was added in one portion to a stirred suspension of diphosphorus tetraiodide (245 mg, 0.43 mmol) in
1 mL of dry CH2Cl2 at room temperature under argon atmosphere. After 30 min the reaction mixture
was quenched by adding water. The organic layer was separated, washed with an aqueous solution of
sodium thiosulfate, dried over MgSO4, and evaporated in vacuo. Flash chromatography on silica gel by
eluting with hexane:ethyl acetate (10:1) gave (−)-9 (152 mg, 68%) as a colorless oil. [α]D20=–290 (c
0.4, CHCl3), ee 86%. 1H NMR (CDCl3) δ ppm: 1.64 (1H, m, H-5a), 1.71 (1H, m, H-4a), 1.95 (1H, m,
H-3a), 2.09 (1H, m, H-5b), 2.10 (2H, m, H-3b and H-4b), 2.31 (1H, m, H-6eq), 2.32 (2H, m, H-7), 2.33
(3H, s, CH3), 3.38 (1H, ddd, J=14.6, 14.6, 6.2 Hz, H-6ax), 5.05 (1H, br d, J=16.8 Hz, H-9t), 5.10 (1H,
br d, J=10.1 Hz, H-9c), 5.85 (1H, m, H-8), 7.10 (2H, d, J=8.1 Hz, H-30), 7.22 (2H, d, J=8.1 Hz, H-20);
13C NMR (CDCl3) δ ppm: 21.25 (CH3), 21.42 (C-4), 26.83 (C-5), 36.48 (C-3), 37.65 (C-6), 39.60 (C-7),
60.07 (C-2), 118.28 (C-9), 126.53 (C-10), 129.71 (C-30), 133.97 (C-8), 136.29 (C-20), 139.72 (C-40),
207.01 (C-1). HMRS found m/z: 260.1234 (M+); C16H20SO requires: 260.1234.
3.9. (−)-(2S,4R)-2-Allyl-6-methyl-2,4-propanethiochroman-3-one 10
By a procedure identical to that described for preparation of (+)-5, the reaction of (−)-9 (50 mg, 0.19
mmol), Mn(OAc)3·2H2O (102 mg, 0.38 mmol), and Cu(OAc)2·H2O (38 mg, 0.19 mmol) in 5 mL of
degassed glacial acetic acid at 90°C under argon for 36 h afforded 39 mg of (−)-10 (80%) as a colorless
oil. [α]D20=–148.8 (c 0.5, CHCl3), ee 86%. 1H NMR (CDCl3) δ ppm: 1.63 (1H, m, H-10a), 2.01 (1H,
m, H-9a), 2.09 (2H, m, H-11), 2.10 (1H, m, H-10b), 2.37 (1H, m, H-9b), 2.52 (3H, s, CH3), 2.58 (2H, dd,
J=7.5, 3.8 Hz, H-12), 3.56 (1H, br t, J=2.7 Hz, H-4), 5.15 (2H, m, H-14), 5.83 (1H, m, H-13), 6.70 (1H,

4434
J. L. García Ruano, A. Rumbero / Tetrahedron: Asymmetry 10 (1999) 4427–4436
br s, H-5), 6.91 (1H, d, J=8.2 Hz, H-7), 6.95 (1H, d, J=8.2 Hz, H-8); 13C NMR (CDCl3) δ ppm: 18.07
(C-10), 20.71 (CH3), 38.11 (C-11), 40.71 (C-12), 42.91 (C-9), 52.51 (C-4), 55.25 (C-2), 119.38 (C-14),
123.23 (C-8), 128.17 (C-7), 130.14 (C-5), 130.55 (C-4a), 132.21 (C-13), 134.47 (C-6), 137.51 (C-8a),
208.41 (C-3). HMRS found m/z: 258.1080 (M+); C16H18SO requires: 258.1078.
3.10. (±)-2-Allyl-2-(p-toluenesulfenyl)cyclohexanone 9
(a) A solution of trifluoroacetic anhydride (311 µL, 2.2 mmol) in acetone (1 mL) was added dropwise
into a stirred suspension of sulfoxide 2 (103 mg, 0.43 mmol) and sodium iodide (195 mg, 1.31 mmol) in
the same solvent (3 mL) at −40°C under argon atmosphere. The reaction mixture was stirred for 30 min
at −40°C, and an excess of saturated aqueous Na2SO3 and Na2CO3 was added. Acetone was removed in
vacuo, the aqueous layer was extracted with CH2Cl2, and the combined organic extracts were dried over
anhydrous Na2SO4 and evaporated in vacuo to provide compound 11 (88 mg, 93%). 1H NMR (CDCl3) δ
ppm: 1.6–2.4 (m, 6H), 2.31 (s, 3H), 2.92 (m, 2H), 3.78 (t, J=5.0 Hz, 1H), 710 (d, J=8.3 Hz, 2H), 7.31 (d,
J=8.3 Hz, 2H); 13C NMR (CDCl3) δ ppm: 21.04 (CH3), 22.40 (CH2), 27.25 (CH2), 33.73 (CH2), 38.87
(CH2), 56.91 (CH), 129.75 (2×CH), 132.68 (2×CH), 137.76 (C), 207.65 (C). (b) To a suspension of NaH
(10 mg, 60% oil dispersion, 0.24 mmol) in 1 mL of dry DMF was added a solution of 11 (52 mg, 0.24
mmol) in 2 mL of dry DMF. The mixture was stirred at −40°C for 1 h under argon. Then allyl bromide
(25 µL, 0.25 mmol) was added and the reaction mixture was stirred for 2 h at the same temperature. To
this reaction mixture was added water and 10% HCl, and the solution was extracted with CH2Cl2. The
organic layer was washed (brine), dried (MgSO4), and evaporated in vacuo. Flash chromatography on
silica gel by eluting with hexane:ethyl acetate (10:1) yielded (±)-9 (76%).
3.11. (±)-2-Allyl-6-Methyl-2,4-propanethiochroman-3-one 10
By a procedure identical to that described for preparation of (+)-5, the reaction of (±)-9 (33 mg, 0.125
mmol), Mn(OAc)3·2H2O (67 mg, 0.25 mmol), and Cu(OAc)2·H2O (25 mg, 0.125 mmol) was stirred in
4 mL of degassified glacial acetic acid at 90°C under argon for 36 h. After standard workup, compound
(±)-10 (25 mg, 78%) was obtained.
Acknowledgements
Financial support from the DGICYT, Spain (Project PB96-0035), is gratefully acknowledged.
References
1. (a) Fukuyama, Y.; Minami, H.; Kuwayama, A. Phytochemistry 1998, 49, 853. (b) Abe, F.; Yamauchi, T. Chem. Pharm. Bull.
1979, 27, 1604. (c) Rao, M. M.; Meshulam, H.; Zelnik, R.; Lavie, D. Phytochemistry 1975, 14, 1071. (d) Lavie, D.; Levy,
E. C.; Rosito, C.; Zelnik, R. Tetrahedron 1970, 26, 219.
2. Horwell, D. C.; Bryans, J. S.; Kneen, C. O.; Morrell, A. I.; Ratcliffe, G. S. (Warner-Lambert Company). PCT Int. Appl. WO
Patent 97 33859, 1997; CA 127, 278461, 1997.
3. Shimizu, T.; Kano, T.; Ogiwara, T.; Kaneko, T.; Nakajima, M.; Kurihara, H. (Shinetsu Chem. Ind. Co, Japan). Jpn. Kokai
Tokkyo Koho JP Patent 0820588, 1996; CA 124, 328568, 1996.
4. (a) Snider, B. B.; McCarthy Cole, B. J. Org. Chem. 1995, 60, 5376. (b) Snider, B. B.; Kiselgof, E. Y. Tetrahedron 1996,
52, 6073. (c) McCarthy Cole, B.; Han, L.; Snider, B. B. J. Org. Chem. 1996, 61, 7832. (d) O’Neil, S. V.; Quickley, Ch. A.;
Snider, B. B. J. Org. Chem. 1997, 62, 1970.

J. L. García Ruano, A. Rumbero / Tetrahedron: Asymmetry 10 (1999) 4427–4436
4435
5. (a) De Klein, W. J. In Organic Synthesis by Oxidation with Metal Compounds; Mijs, W. J.; de Jonge, C. R. H. I., Eds.;
Plenum Press: New York, 1986, p. 216. (b) Melikyan, G. G. Synthesis 1993, 833. (c) Iqbal, J.; Bhatia, B.; Nayyar, N. K.
Chem. Rev. 1994, 94, 519. (d) Snider, B. B. Chem. Rev. 1996, 96, 339.
6. Kosugi, H.; Kanno, O.; Uda, H. Tetrahedron: Asymmetry 1994, 5, 1139.
7. Carreño, M. C.; García Ruano, J. L.; Pedregal, C.; Rubio, A. J. Chem. Soc., Perkin Trans. 1 1989, 1335.
8. Pohmakotr, M.; Chancharunee, S. Tetrahedron Lett. 1984, 25, 4141.
9. Allylation in conditions reported in Ref. 6 (NaH, allyl bromide, DMF) are also satisfactory. Nevertheless, the instability of
3 in the conditions required to eliminate the solvent (DMF), previous to the cyclization step, made it advisable to use phase
transfer conditions in dichloromethane. Under these conditions, compound 3 was obtained as a mixture of epimers at C-2,
with des ranging between 80 and 86%.
10. (a) Trost, B. M.; Arndt, H. C.; Strege, P. E.; Verhoeven, T. R. Tetrahedron Lett. 1976, 39, 3477. (b) The use of other reagents
to cleave the C–S bond without affecting the carbonyl group, such as Al–Hg/THF–H2O and Zn/NH4Cl, was unsuccessful.
11. (a) Foote, C. S.; Woodward, R. B. Tetrahedron 1964, 20, 687. (b) Brevis, S.; Hughes, P. P. Chem. Commun. 1966, 6. (c)
Martin, J.; Parker, W.; Shroot, B.; Stewart, T. J. Chem. Soc. (C) 1967, 101. (d) Erman, W. F.; Kretschmar, H. C. J. Org.
Chem. 1969, 33, 1545. (e) Doyle, M.; Parker, W. J. Chem. Soc., Perkin Trans. 1, 1977, 858. (f) Heumann, A.; Kraus, W.
Tetrahedron 1978, 34, 405. (g) Taeschler, C.; Sorensen, T. S. J. Org. Chem. 1998, 63, 5704.
12. Momota, J; Imura, T.; Kobayakawa, T. Eur. Pat. Appl. EP Patent 678517, 1995; CA 124, 145901, 1996 and Jpn. Kokai
Tokkyo Koho JP Patent 07258245, 1995; CA 124, 175832, 1996.
13. HPLC analysis was performed with: (a) Chiral column Chiralpak AS (25 cm×0.46 cm) eluting with a mixture of n-
hexane:isopropanol (90:10), flow rate=1.0 mL/min, detection 220 nm, retention times: (−)-4, t=9.1 min; (+)-4, t=11.2
min; (−)-7, t=8.6 min; (+)-7, t=12.0 min. (b) Chiral column Chiralcel OD (25 cm×0.46 cm) eluting with a mixture of n-
hexane/isopropanol (99:1), flow rate=1.0 mL/min, detection 254 nm, retention times: (−)-9, t=4.7 min; (+)-9, t=5.2 min; (−)-
10, t=5.4 min; (+)-10, t=7.7 min. (c) The ee of compound 5 could not be determined by chiral HPLC due to its insolubility
in the eluting mixture.
14. (a) Krief, A.; Denis, J. N. Tetrahedron Lett. 1979, 3995. (b) Posner, G. H.; Asirvatham, E. J. Org. Chem. 1985, 50, 2589.
15. (a) Hiroi, K.; Koyama, T.; Anzai, K.; Chem. Lett. 1990, 235. (b) Hiroi, K.; Abe, J.; Suya, K.; Sato, S.; Koyama, T. J. Org.
Chem. 1994, 59, 203.
16. Bravo, P.; Pregnolato, M.; Resnati, G. J. Org. Chem. 1992, 57, 2726.
17. Snider, B. B.; Yu-Fong Wau, B.; Buckman, B. O.; Foxman, B. M. J. Org. Chem. 1991, 56, 328.
18. (a) Solladié, G. Synthesis 1981, 185. (b) Solladié, G. Pure Appl. Chem. 1988, 60, 1699. (c) House, S.; Jenkins, P. R.; Fawcett,
J.; Russell, D. R. J. Chem. Soc., Chem. Commun. 1987, 1844. (d) Mioskowski, C.; Solladié, G. Tetrahedron 1980, 36, 227.
19. Crabbe, P.; Hirsch, J.A.; Mislow, K.; Raban, M.; Schlögl, K. Topics in Stereochemistry, vol. 1; Allinger, N. L.; Eliel, E. L.,
Eds.; Interscience Publishers: New York, 1967.
1
3
20. H–13C correlation experiment (HMBC) was optimized for observing JCH≈8 Hz. The HMBC spectrum shows a large
difference in the three-bond H–13C coupling constants between the C-7 (δ 38.09) and the protons H-3ax (δ 2.04, ddd,
1
J=15.3, 12.8 and 4.3 Hz) and H-3eq (δ 2.70, ddd, J=15.3, 6.2 and 3.8 Hz). The three-bond correlation from H-3ax to C-7
(3J≈8 Hz) typical of a disposition anti (Pretsch, E.; Clerc, T.; Seibl.; Simon W. Tablas para la elucidación estructural de
compuestos orgánicos por métodos espectroscópicos; Alhambra, Madrid, 1989) and, additionally, the fact that no correlation
was observed between H-3eq and C-7, allowed the unambiguous assignment of conformation I for compound 4.
21. The formation of the more congested five-membered exo radicals would be difficult because of the size of substituent at
C-2.

4436
J. L. García Ruano, A. Rumbero / Tetrahedron: Asymmetry 10 (1999) 4427–4436
22. (a) Kochi, J. K.; Bemis, A.; Jenkins, C. L. J. Am. Chem. Soc. 1968, 90, 4616. (b) Kochi, J. K.; Bacha, J. D. J. Org. Chem.
1968, 33, 2746. (c) Jenkins, C. L.; Kochi, J. K. J. Am. Chem. Soc. 1972, 94, 843. (d) Kochi, J. K. In Free Radicals; Kochi,
J. K., Ed.; Wiley: New York, 1973; chapter 11. (e) Kochi, J. K. Science 1967, 155, 415.
23. The 1H–1H NOESY spectrum of compound 9 is in full agreement with the axial arrangement of the sulfenyl group, showing
connectivities between the proton H-20 and the protons H-4ax and H-6ax and also between the proton H-8 and the protons
H-3ax and H-3eq.
24. Fossey, J.; Lefort, D.; Sorba, J. In Free Radicals in Organic Chemistry; John Wiley & Sons: Chichester, 1995.
25. The different evolution of compounds 4 and 9 could be explained by assuming only the different nucleophilic character
of the double bonds acting as radical acceptors (pTolS>allylic double bond>pTolSO2) regardless of the conformational
behavior of the substrates. Nevertheless, the fact that reaction time required for the complete evolution of compound 9 was
larger (36 h) than that for 4 (8 h) suggests to us that this is not the case and other factors must be involved.