A highly enantioselective benzothiepine synthesis

Article abstract of DOI:10.1021/jo991786q

A highly enantioselective synthesis of benzothiepine 1a has been accomplished via an enantioen-riched sulfoxide intermediate obtained by asymmetric oxidation with a chiral oxaziridine in 89:11 er. The key step is a thermodynamically controlled asymmetric cyclization reaction that produces two new stereogenic centers. The (4R,5R) isomer 1a was obtained in 98:2 er.

Full text of DOI:10.1021/jo991786q

J . Org. Chem. 2000, 65, 2711-2715
2711
A High ly En a n tioselective Ben zoth iep in e Syn th esis
Ching-Cheng Wang,*,† J ames J . Li,‡ Horng-Chih Huang,† Len F. Lee,§ and David B. Reitz†
Medicinal Chemistry, Searle Research & Development, Monsanto Company, St. Louis, Missouri 63017
Received November 17, 1999
A highly enantioselective synthesis of benzothiepine 1a has been accomplished via an enantioen-
riched sulfoxide intermediate obtained by asymmetric oxidation with a chiral oxaziridine in 89:11
er. The key step is a thermodynamically controlled asymmetric cyclization reaction that produces
two new stereogenic centers. The (4R,5R) isomer 1a was obtained in 98:2 er.
Sch em e 1
In tr od u ction
Benzothiepines of which 1 (Scheme 1) is representative
are potent apical sodium co-dependent bile acid trans-
porter (ASBT) inhibitors useful for the treatment of
hypercholesterolemia.1 Separation of the enantiomers 1a
and 1b was accomplished initially by liquid chromatog-
raphy on a chiral stationary phase,2 and 1a was found
to be the significantly more potent isomer. Since larger
quantities of 1a were required for evaluation in animal
models, the development of an enantioselective synthesis
seemed very attractive. We now report an efficient
enantioselective synthesis of benzothiepines.
During our initial investigation of the synthesis of 1,
we found that cyclization of 2c (X ) NMe2) with 1.2 equiv
of potassium tert-butoxide in THF at -10 °C gave only
(()-1 with the hydroxyl and methoxyphenyl groups in a
cis relationship (Scheme 1). The selectivity can be
rationalized by invoking potassium ion complexation with
the sulfone and alkoxide oxygens, an effect which could
only exist in cis-(()-1 but not in a trans isomer. On the
other hand, either oxygen of the achiral sulfone would
be capable of directing the ring closure reaction by
forming the salt bridge; therefore, no enantioselectivity
would be anticipated. If the cyclization were possible with
a sulfoxide, a chiral ring closure could be envisioned with
an enantioenriched sulfoxide to preferentially provide one
enantiomer. To test this hypothesis, (()-2b was prepared
by oxidation of 2a with either 1 equiv of m-CPBA or 0.5
equiv of Oxone. Cyclization of (()-2b with potassium tert-
butoxide in THF at -10 °C yielded only the cis isomers
(()-3 (Scheme 1). This observation suggests that a chiral
sulfoxide (R)-2b should stereoselectively yield the desired
(1R,4R,5R) cyclized product.
ral products3 due to their high diastereoselectivity as
auxiliaries, yet so far there are no reported examples of
asymmetric cyclization utilizing internal chiral sulfox-
ides. Enantiomerically pure chiral sulfoxides can be
readily prepared by enzymatic4 oxidation using monooxy-
genases, chemical oxidation using Ti(IV)5 or Mn(III)6
complexes, or chiral oxaziridines7 to convert the prochiral
sulfides to the chiral sulfoxides.
Resu lts a n d Discu ssion
We chose to focus on the chemical oxidation of prochiral
sulfides by using Sharpless reagents, Davis’s oxaziri-
dines, and J acobsen’s catalyst. The aldehyde 4 was
selected as a model system to study the asymmetric
(3) (a) Hayes, P. Maignan, C. Tetrahedron: Asymmetry 1999, 10,
1041-1050. (b) Maezaki, N.; Sakamoto, A.; Tanaka, T.; Iwata, C.
Tetrahedron: Asymmetry 1998, 9, 179-182. (c) Nakamura, S.; Goto,
K.; Kondo, M.; Naito, S.; Tsuda, Y.; Shishindo, K. Bioorg. Med. Chem.
Lett. 1997, 7, 2033-2036. (d) Solladie, G.; Adamy, M.; Colobert, F. J .
Org. Chem. 1996, 61, 4369-4373. (e) Louis, C.; Hootele, C. Tetrahe-
dron: Asymmetry 1995, 6, 2149-2152.
(4) (a) Colonna, S.; Gaggero, N.; Carrea, G.; Pasta, P. Chem.
Commun. 1997, 5, 439-440. (b) Pasta, P.; Carrea, G.; Gaggero, N.;
Grogan, G.; Willetts, A. Biotechnol. Lett. 1996, 18, 1123-1128. (c) Kelly
D. R.; Knowles, C. J .; Mahdi, J . G.; Taylor, I. N.; Wright, M. A.
Tetrahedron: Asymmetry 1996, 7, 365-368.
(5) (a) Rosini, C.; Donnoli, M. I.; Superchi, S. J . Org. Chem. 1998,
63, 9392-9395. (b) Brunel, J . M.; Kagan, H. B. Bull. Soc. Chim. Fr.
1996, 133, 1109-1115. (c) Bulman Page, P. C.; Wilkes, R. D.;
Namwindwa, E. S.; Witty, M. J . Tetrahedron 1996, 52, 2125-2154.
(d) Kagan, H. B.; Zhao, S. H.; Samuel, O. Tetrahedron 1987, 43, 5135-
5144.
Chiral sulfoxides have been extensively used as key
intermediates for the enantioselective synthesis of natu-
† G. D. Searle.
‡ Current address: P.O. Box 5400, Bristol-Myers Squibb Company,
Princeton, NJ 08543-5400.
§ Monsanto Corporate Research.
(1) (a) Lee, L. F.; Banerjee, S. C.; Huang, H.; Li, J . J .; Miller, R. E.;
Reitz, D. B.; Tremont, S. J . U.S. Patent No. 5 994 391, 1999. (b) Reitz,
D. B.; Lee, L. F.; Li, J . J .; Huang, H.; Tremont, S. J .; Miller, R. E.;
Banerjee, S. C. World Patent 9733882. (c) Reitz, D. B.; Huang, H.;
Carpenter, A. J .; Tollefson, M. B.; Kolodziej, S. A.; Li, J . J .; Wang, C.;
Garland, D. J .; Mahoney, M. W.; Mischke, D. A.; Chen, F.; Reher, T.
R.; Reinhard, E. J .; Vernier, W. F.; Huang, W.; Beaudry, J . A.; Rapp,
S., R.; Smith, M. E.; Schuh, J . R.; Keller, B. T.; Glenn, K. C.; Tremont,
S. J .; Lee, L. F.; Banerjee, S. S.; Both, S. R.; Keller, K.; Miller, R. E.;
Wagner, G. M. Book of Abstracts, 216th National Meeting of the
American Chemical Society, Boston, August 23-27, 1998, MEDI-383.
(2) The separation was done on ChiralPak AD column with 10%
ethanol/hexane at 1.0 mL/min.
(6) J acobson, E. N.; Zhang, W.; Muci, A. R.; Ecker, J . R.; Deng, L.
J . Am. Chem. Soc. 1991, 113, 7063-7064.
(7) (a) Bohe, L.; Lusinchi, M.; Lusinchi, X. Tetrahedron 1999, 55,
155-166. (b) Davis, F. A.; Reddy, R. T.; Han, W.; Carroll, P. J . J . Am.
Chem. Soc. 1992, 114, 1428-1437. (c) Davis, F. A.; Reddy, R. T.;
Weissmiller, M. C. J . Am. Chem. Soc. 1989, 111, 5964-5965.
10.1021/jo991786q CCC: $19.00 © 2000 American Chemical Society
Published on Web 04/08/2000

2712 J . Org. Chem., Vol. 65, No. 9, 2000
Wang et al.
F igu r e 1. X-ray structure of racemic 3.
Sch em e 3a
Ta ble 1. Asym m etr ic Oxid a tion of Ar om a tic Su lfid e 5
method
temp, °C (R)-5/(S)-5a
A. CHP, Ti(Oi-Pr)4/(-)-DET/H2O (1:2:1)
B. TBHP, Ti(Oi-Pr)4/(-)-DET/H2O (1:2:1)
C. CHP, J acobsen’s Catalyst, PhI(OAc)2
D. (-)-(camphorsulfonyl)oxaziridine
-25
60/40
60/40
60/40
60/40
80/20
-25
-25 to rt
-25
E. (-)-(dichlorocamphorsulfonyl)oxaziridine -25
a
The enantiomeric ratios were determined by HPLC on (S,S)
Whelk-O 1 with ethanol-hexane (1:9). The absolute configuration
of the sulfoxides was derived from the configuration of 1a and 3.
oxidation of prochiral aromatic sulfide to the (R)- and (S)-
sulfoxide 5 (Table 1).
Oxidation of 4 at -25 °C with either cumene hydrop-
eroxide (CHP) or tert-butyl hydroperoxide (TBHP) in the
presence of Ti(Oi-Pr)4/(-)-DET/H2O (1:2:1) gave a 60:40
mixture of (R)- and (S)-sulfoxides (vida infra). Under
J acobsen’s conditions, CHP with iodobenzene diacetate
in acetonitrile also gave a 60:40 mixture of the sulfoxides
5. Oxidation with (-)-(camphorsulfonyl)oxaziridine in
CH2Cl2 was also not very selective (60:40 mixture);
however, the corresponding R,R-dichloro derivative under
the same conditions gave a much improved 80:20 mixture
of 5. This oxaziridine was selected for further investiga-
tion of an enantioselective synthesis of 1a .
ee
a
(i) SO3‚Py, TEA, DMSO, 25 °C (91%); (ii) KOt-Bu, THF, -15
°C (78%); (iii) m-CPBA, CH2Cl2, 0 °C (98%); (iv) (CH3)2NH, THF,
110 °C (91%); (v) recrystallization in hexane/ethyl acetate (80%).
1b in the same 89:11 mixture as observed initially for 7,
thus establishing that the ring-closure reaction is highly
enantioselective relative to the enantiomeric purity of the
sulfoxide. A single recrystallization provided 1a in 86%
yield with >98:2 er.
Sch em e 2
The absolute stereochemistries of the sulfoxides 5
(Table 1) and 7 (Scheme 2) can be derived as follows: The
X-ray crystal structure of the racemic 38 (Figure 1) shows
that the sulfoxide oxygen is cis to the hydroxyl and the
methoxyphenyl group, and the possible configuration of
the racemic pair 3a and 3b (Scheme 1) can only be
(1R,4R,5R) and (1S,4S,5S). Since the configuration of 1a
is (4R,5R) as determined by X-ray,9 the configuration of
C4 and C5 in its precursors 9 and 3a is also (R,R), leading
to the conclusion that the configuration of sulfoxide sulfur
atom in 3a is R. Moreover, since 3a is generated from
the asymmetric oxidation of the sulfide 6, (R)-7 and (R)-8
are established to be in the R configuration as well.
The success of our synthetic strategy relies on the
stereocontrolled cyclization of (R)-8 to 3a , in which the
er
Oxidation of 6 with (-)-(dichlorocamphorsulfonyl)-
oxaziridine and subsequent chromatographic purification
gave a 89:11 mixture of (R)-7 and (S)-7 (vida infra) in
95% yield (Scheme 2). Oxidation of the enantioenriched
(R)-7 with sulfur trioxide-pyridine gave the aldehydes
(R)-8 in 91% yield (Scheme 3). Subsequent cyclization
with KOt-Bu gave 3a in 78% yield which was oxidized
to 9 with m-CPBA in 98% yield. The nucleophilic dis-
placement of fluorine with dimethylamine gave 1a and
(8) Racemic 3 was prepared from the cyclization of racemic 8.
(9) The configuration of 1a was established from the X-ray structure
of corresponding phenol.

A Highly Enantioselective Benzothiepine Synthesis
J . Org. Chem., Vol. 65, No. 9, 2000 2713
Sch em e 4a
Sch em e 5
a
(i) Oxalyl chloride, TEA, DMSO, -78 °C to rt (40%); (ii)
NaBH4, CH3OH, -10 °C to rt.
stereochemistry of sulfoxide determines the formation of
the two stereogenic centers and yields only one out of
the four possible stereoisomers. The first step in elucidat-
ing the source of the selectivity is to determine whether
the products are kinetically or thermodynamically con-
trolled. Since the cyclization of sulfone 2c (Scheme 1) at
lower temperature (-30 °C) actually gives the trans
isomer as a kinetic-controlled product, the formation of
the cis isomer is thus suggested to be a thermodynamic-
controlled process.10 It is highly plausible that in the
cyclization of sulfoxide the preference for the formation
of cis isomer at -10 °C is also the result of a thermody-
namic-controlled process.
tions11 indicated that potassium ion forms a salt bridge
between the alkoxide and sulfoxide oxygens in 14 when
they are cis to each other, and that cis-14 is more stable
than trans-13 by more than 2 kcal/mol largely due to the
electrostatic term.
Con clu sion s
We have successfully developed an enantioselective
route to synthesize one of the four possible benzothiepine
enantiomers of benzothiepines in good ee and yield. This
stereoselective cyclization demonstrates the utility of a
chiral sulfoxide as a directing group in a cyclization
reaction. We have also provided evidence to support our
hypothesis that the stereoselective cyclization is directed
by the stereochemistry of the sulfoxide and that this is a
thermodynamically controlled process.
To gain additional insights into this highly enantiose-
lective cyclization process, we prepared and subjected the
presumed thermodynamic-controlled and kinetic-con-
trolled products (3 and 11, respectively) to the cyclization
conditions at -10 °C. If an equilibrium process is operat-
ing at -10 °C, then both materials should give a mixture
of same compositions as in the cyclization of 8. The trans
isomer 11 was prepared by Swern oxidation of the cis
alcohol 3 to give 10 in 40% yield Scheme 4). The ketone
10 was reduced by NaBH4 to give a 27:73 mixture of both
isomers favoring trans-11 in 80% yield.
Exp er im en ta l Section
1
Gen er a l. H NMR spectra were recorded in CDCl3 at 300
or 400 MHz and are reported as chemical shift (multiplicity,
coupling constants in hertz, integration) and chemical shifts
are reported as δ values in parts per million relative to CDCl3
(δ 7.26). 13C NMR spectra were recorded in CDCl3 at 75 MHz,
and chemical shifts are reported as δ values in parts per
million relative to CDCl3 (δ 77.00). Mass spectra were recorded
in fast atom bombardment (FAB) or electrospray ionization
(ESI) mode. Melting points are uncorrected. Analytical HPLC
was conducted on 4.6 mm × 250 mm Alltech Altima Silica 5 µ
column with hexane/ethyl acetate using UV detection. Chiral
HPLC was performed on 4.6 mm × 250 mm ChiralPak AD
from Daicel or 4.6 mm × 250 mm (R,R) Whelk-O 1 column
from Regis Technologies with ethanol/hexane using UV detec-
tion. Chromatographic purification was done on medium-
pressure liquid chromatography (MPLC) or Waters Prep500
using silica gel with hexane/ethyl acetate. X-ray crystal-
lography was done by the X-ray Diffraction Laboratory at the
Department of Chemistry in the University of MissourisSt.
Louis. All reagents were purchased from Sigma-Aldrich and
were used without further purification. All reactions involving
air- or moisture-sensitive reagents were conducted under
nitrogen.
In the isomerization experiments (Scheme 5), the cis
isomer 3, under the cyclization conditions, after 30 min
gave 77% of the cis isomer 3, 23% of decomposition
products, and no trans isomer 11. On the other hand,
trans-11 under the same conditions isomerized to give
75% of cis-3 and 25% of decomposition products. Cycliza-
tion of 8 under the same conditions also gave 77% of cis-
3. These experiments establish that the cis isomer is the
more thermodynamically stable isomer, and that 3, 8,
and 11 are in equilibrium.
We believe that the cyclization reaction occurs as
follows: The aldehyde 8 is deprotonated to give a
carbanionic intermediate 12, which can cyclize to either
the trans alkoxide 13 or the cis alkoxide 14; these
intermediates are in equilibrium with 12 and can be
protonated to products 11 and 3, respectively. Although
both alkoxides 13 and 14 could potentially be formed, the
rapidly established equilibrium would shift the product
formation to the thermodynamically more stable 14.
Monte Carlo conformational search and energy calcula-
(()-2-[[[2-(Br om om eth yl)-4-flu or op h en yl]su lfin yl]m e-
th yl]-2-bu tylh exa n a l (5). To a stirred solution of 7.3 g of
(11) Monte Carlo conformational search was performed on Macro-
model 6.5 with MM3 force field and chloroform as the solvent. Energy
calculation was performed on Sybyl 6.5 with Tripos force field and
Gasteiger-Huckel charge method.
(10) Unpublished results.

2714 J . Org. Chem., Vol. 65, No. 9, 2000
Wang et al.
2-[[[2-(br om om eth yl)-4-flu or op h en yl]th io]m eth yl]-2-bu -
tylh exa n a l (4) (20.2 mmol) in 50 mL of CH2Cl2 was added
5.2 g of m-CPBA (assumed 68% activity, 20.5 mmol) at 0 °C.
The mixture was stirred at 0 °C for 10 min and at room
temperature for 30 min. The mixture was quenched with
saturated Na2SO3 solution, and the CH2Cl2 layer was sepa-
rated, neutralized with saturated NaHCO3 solution, dried over
MgSO4, and concentrated in vacuo. Recrystallization from
hexane/ethyl acetate gave 6.8 g of racemic 5 (89%) as a
colorless solid: 1H NMR (CDCl3) δ 0.87-0.98 (m, 7H), 1.21-
1.39 (m, 8H), 1.76-1.85 (m, 3H), 1.96-2.20 (m, 1H), 3.08 (s,
2H), 4.57-4.70 (ABq, J ) 11.10 Hz, 2H), 7.16-7.28 (m, 2H),
8.01 (dd, J ) 5.40 Hz, 8.70 Hz, 1H), 9.56 (s, 1H). The structure
of 5 was also confirmed by X-ray crystallography. The solid
was dissolved in ethanol and chromatographed on (R,R)
Whelk-O 1 with ethanol-hexane (1:9) at 1 mL/min to give 2
peaks at 10.68 min (50%) and 12.08 min (50%).
Oxid a tion of 4. Meth od A. 5 (200 mg, 0.55 mmol) was
added into the solution of 1.5 mL of Ti(Oi-Pr)4 (5 mmol), 1.7
mL of (-)-diethyl tartrate (DET) (10 mmol), and 90 µL of H2O
(5 mmol), and the mixture was cooled to -25 °C. Cumene
hydroperoxide (CHP, 0.11 mL, 0.60 mmol) was added, and the
mixture was stirred for 16 h at -25 °C. TLC indicated the
conversion was nearly complete. The mixture was warmed to
room temperature, washed with brine, extracted with EtOAc,
dried over MgSO4, and concentrated in vacuo. The resulting
solid was dissolved in ethanol and chromatographed on (R,R)
Whelk O 1 to give 2 major peaks at 10.58 min (40%) and 11.89
min (60%).
Meth od B. tert-Butyl hydroperoxide (TBHP, 67 µL, 0.60
mmol) was added to the solution of 200 mg of 4 (0.55 mmol)
in the Ti(Oi-Pr)4/(-)-DET/H2O (1:2:1) solution (same as in
method A) at -25 °C and stirred for 16 h at -25 °C. The
workup procedure in method A was followed. The crude was
dissolved in ethanol and chromatographed on (R,R) Whelk-O
1 to give 2 peaks at 10.76 min (40%) and 12.17 min (60%).
Meth od C. J acobsen’s catalyst (18 mg, 0.03 mmol) was
added to a solution of 200 mg of 5 (0.55 mmol) in 3 mL of
acetonitrile and cooled to -25 °C. CHP (0.11 mL, 0.60 mmol)
was then added. The mixture was warmed to room tempera-
ture and stirred overnight. PhI(OAc)2 (193 mg, 0.60 mmol) was
added into the reaction mixture and stirred overnight. TLC
indicated disappearance of 5. The same workup procedure in
method A was followed, and the crude product was chromato-
graphed on (R,R) Whelk-O 1 to give 2 peaks at 10.69 min (40%)
and 12.08 min (60%).
Meth od D. (-)-(Camphorsulfonyl)oxaziridine (126 mg, 0.55
mmol) was added into a solution of 200 mg of 5 (0.55 mmol)
in 20 mL of CH2Cl2 at -25 °C. The mixture was stirred at
-25 °C overnight, and an aliquot was worked up and chro-
matographed on (R,R) Whelk-O 1 to give 2 peaks at 10.47 min
(40%) and 11.74 min (60%).
Meth od E. (-)-(Dichlorocamphorsulfonyl)oxaziridine (165
mg, 0.55 mmol) was added to a solution of 200 mg of 5 (0.55
mmol) in 20 mL of CH2Cl2 at -25 °C. The mixture was stirred
for 48 h at -25 °C, and an aliquot was worked up and
chromatographed on (R,R) Whelk-O 1 to give 2 peaks at 10.70
min (20%) and 11.74 min (80%).
2-Bu tyl-2-[[(R)-[4-flu or o-2-[(4-m eth oxyp h en yl)m eth yl]-
p h en yl]su lfin yl]m eth yl]-1-h exa n ol (7). To a stirred solu-
tion of 20.00 g (47.78 mmol) of 7 in 1 L of methylene chloride
was added 31.50 g of 96% (1R) (-)-(8,8-dichloro-10-camphor-
sulfonyl)oxaziridine (100.34 mmol) at 2 °C. After all the
oxaziridine dissolved, the mixture was placed into a -30 °C
freezer for 72 h. The solvent was evaporated, and the crude
solid was washed with 1 L of hexane. The white solid was
filtered off, and the hexane solution was concentrated in vacuo.
The crude oil was chromatographed on silica gel column
(hexane/ethyl acetate 17:3) to afford 19.00 g (95%) of the
product as a colorless oil: 1H NMR (CDCl3) δ 0.82-0.98 (m,
6H), 1.16-1.32 (m, 12H), 2.30 (d, J ) 13.80 Hz, 1H), 2.78 (d,
J ) 13.50 Hz, 1H), 3.45 (d, J ) 12.30 Hz, 1H), 3.69 (d, J )
12.30 Hz, 1H), (s, 3H), 3.97 (d, J ) 15.60 Hz, 1H), 4.07 (d, J )
15.90 Hz, 1H), 6.86 (m, 3H), 7.00 (d, J ) 8.10 Hz, 2H), 7.20
(m, 1H), 8.02 (m, 1H). 13C NMR (CDCl3) δ 14.17, 14.30, 23.50,
23.69, 25.20, 25.34, 32.78, 35.71, 36.89, 42.47, 55.47, 67.30,
68.06, 114.58, 115.59, 115.88, 117.74, 118.04, 127.27, 127.39,
129.88, 130.27, 138.42, 138.46, 140.60, 140,71, 158.82, 163.01,
166.35. Anal. Calcd for C25H36O3SF: C, 69.09; H, 8.12; S, 7.38.
Found: C, 68.90; H, 8.22; S, 7.38.
Enantiomeric excess was determined by chiral HPLC on a
(R,R) Whelk-O 1 column using 5% ethanol hexane as the
eluent. It showed to be 78% ee with the first eluting peak as
the major product.
(()-2-Bu tyl-2-[[[4-flu or o-2-[(4-m eth oxyp h en yl)m eth yl]-
p h en yl]su lfin yl]m eth yl]-1-h exa n ol (7) was prepared as
follows. To a stirred solution of 3.00 g of 6 (7.17 mmol) in a
mixture of 5 mL of CH3OH, 5 mL of THF, and 1 mL of H2O
was added 2.20 g of oxone (3.58 mmol). The mixture was
stirred at room-temperature overnight. The solvents were
evaporated, and the residue was dissolved in EtOAc, washed
with brine, dried over MgSO4, and concentrated in vacuo. The
crude was purified on silica gel (hexane/ethyl acetate 9:1) to
1
yield 2.74 g of (()-7 (88%). H and 13C NMR are identical to
that of 7 from the above. HRMS (ES+) calcd of C25H36O3SF
(M + H) 435.2369 found 435.2325.
2-Bu tyl-2-[[(R)-[4-flu or o-2-[(4-m eth oxyp h en yl)m eth yl]-
p h en yl]su lfin yl]m eth yl]-1-h exa n a l (8). 7 (78% ee, 19.00 g,
43.72 mmol) and 20.96 g of sulfur trioxide-pyridine (131.16
mmol) were added to a solution of 13.27 g of triethylamine
(131.16 mmol) in 200 mL of DMSO at room temperature. The
mixture was stirred at room temperature for 48 h. H2O (500
mL) was added to the mixture and stirred vigorously. The
mixture was then extracted with 500 mL of EtOAc twice. The
EtOAc layer was separated, dried over MgSO4, and concen-
trated in vacuo. The crude oil was filtered through 500 mL of
silica gel using hexane/ethyl acetate (17:3) to give 17.30 g (91%)
of the product as a light orange oil: 1H NMR (CDCl3) δ 0.85-
0.95 (m, 6H), 1.11-1.17 (m, 4H), 1.21-1.39 (m, 4H), 1.59-
1.76 (m, 4H), 1.89-1.99 (m, 1H), 2.57 (d, J ) 14.10 Hz, 1H),
2.91 (d, J ) 13.80 Hz, 1H), 3.97 (d, J ) 15.90 Hz, 1H), 4,12 (d,
J ) 15.90 Hz, 1H), 6.86 (m, 3H), 7.03 (d, J ) 8.4 Hz, 2H), 7.19
(td, J ) 8.4 Hz, 2.4 Hz, 1H), 8.02 (dd, J ) 8.7 Hz, 5.7 Hz, 1H),
9.49 (s, 1H). 13C NMR (CDCl3) δ 14.03, 14.12, 23.33, 23.36,
25.78, 26.03, 32.48, 33.33, 36.84, 52.26, 55.50, 60.84, 114.51,
115.51, 115.80, 117.49, 117.78, 127.20, 127.32, 130.03, 130.54,
139.44, 139.47, 140.93, 141.04, 158.78, 163.00, 166.34, 204.27.
(()-2-Bu tyl-2-[[4-flu or o-2-[(4-m eth oxyp h en yl)m eth yl]-
p h en yl]su lfin yl]m eth yl]-1-h exa n a l (8) was also prepared
from the oxidation of (()-7 by sulfur trioxide-pyridine and
triethylamine in DMSO. Yield: 81%. 1H and 13C NMR are
identical to those of 8 from the above.
(1R,4R,5R)-3,3-Dibu tyl-7-flu or o-2,3,4,5-tetr a h yd r o-5-(4-
m eth oxyp h en yl)-1-ben zoth iep in -4-ol 1-Oxid e (3). To a
stirred solution of 17.30 g of 8 (39.99 mmol) in 300 mL of THF
at -15 °C was added 47.99 mL of 1.0 M KOt-Bu in THF (1.2
equiv) under N2. The solution was stirred at -15 °C for 4 h.
The solution was then quenched with 100 mL of H2O and
neutralized with 4 mL of concentrated HCl solution at 0 °C.
The THF layer was separated, dried over MgSO4, and con-
centrated in vacuo. The product was purified by silica gel
chromatography (hexane/ethyl acetate 17:3) to give 13.44 g
(78%) of the product as a white solid: mp 166-167 °C. 1H NMR
(CDCl3) δ 0.87-0.97 (m, 6H), 1.16-1.32 (m, 4H), 1.34-1.48
(m, 4H), 1.50-1.69 (m, 4H), 1.86-1.96 (m, 1H), 2.88 (d, J )
13 Hz, 1H), 3.00 (d, J ) 13 Hz, 1H), 3.85 (s, 3H), 4.00 (s, 1H),
(s, 1H), 6.52 (dd, J ) 9.9 Hz, 2.4 Hz, 1H), 6.94 (d, J ) 9 Hz,
2H), 7.13 (td, J ) 8.4 Hz, 2.4 Hz, 1H), (d, J ) 8.7 Hz, 2H),
7.82 (dd, J ) 8.7 Hz, 5.7 Hz, 1H). 13C NMR (CDCl3) δ 14.21,
23.40, 23.56, 24.52, 25.21, 32.00, 35.53, 45.09, 46.76, 55.55,
59.68, 74.68, 114.62, 114.96, 115.25, 118.35, 118.67, 124.83,
124.94, 129.95, 133.79, 138.46, 138.56, 140.36, 159.12, 162.54,
165.85. HRMS (ES+) calcd for C25H34O3SF (M+H) 433.2213
found 433.2246. Anal. Calcd for C25H33O3SF: C, 69.41; H, 7.69.
Found: C, 69.29; H, 7.66.
r el-(1R,4R,5R)-3,3-Dibu tyl-7-flu or o-2,3,4,5-tetr a h yd r o-
5-(4-m eth oxyp h en yl)-1-ben zoth iep in -4-ol 1-oxid e (()-3
was also prepared from the cyclization of (()-8 by KOt-Bu in
1
THF at -10 °C. Yield: 77%. H and 13C NMR are identical to
those of 4 from the above.

A Highly Enantioselective Benzothiepine Synthesis
J . Org. Chem., Vol. 65, No. 9, 2000 2715
(4R,5R)-3,3-Dib u t yl-7-flu or o-2,3,4,5-t et r a h yd r o-5-(4-
m eth oxyp h en yl)-1-ben zoth iep in -4-ol 1,1-Dioxid e (9). To
a stirred solution of 13.44 g (31.07 mmol) of 3 in 150 mL of
CH2Cl2 was added 9.46 g of m-CPBA (assumed 68% activity,
37.28 mmol) at 0 °C. After stirring at 0 °C for 2 h, the mixture
was allowed to warm to room temperature and stirred for 4
h. Saturated aqueous Na2SO3 (50 mL) was added to the
mixture and stirred for 30 min. The solution was then
neutralized with 50 mL of saturated NaHCO3 solution. The
methylene chloride layer was separated, dried over MgSO4,
and concentrated in vacuo to give 13.00 g (97%) of the product
(t, J ) 7.2 Hz, 3H), 1.07-1.15 (m, 2H), 1.21-1.32 (m, 5H),
1.47-1.62 (m, 3H), 1.90-1.98 (m, 1H), 2,46-2.53 (m, 1H), 3.05
(d, J ) Hz, 1H), 3.35 (d, J ) 12.8 Hz, 1H), 3.82 (s, 3H), 5.73
(s, 1H), 6.88 (dd, J ) 2.4 Hz, 9.2 Hz, 1H), 6.92 (d, J ) 12.0 Hz,
2H), 7.11 (td, J ) 2.5 Hz, 8.3 Hz, 1H), 7.55 (d, J ) 8.8 Hz,
2H), 7.78 (dd, J ) 5.4 Hz, 8.6 Hz, 1H). 13C NMR (CDCl3) δ
13.84, 14.13, 23.25, 23.39, 25.59, 26.66, 33.82, 34.28, 52.48,
55.51, 56.61, 66.72, 114.35, 114.86, 115.19, 115.38, 115.68,
125.46, 125.58, 126.44, 131.14, 138.66, 138.76, 139.06, 159.60,
163.19, 166.52, 206.72. HRMS (ES+) calcd for C25H32O3SF (M
+ H) 431.2056 found 431.2031.
r el-(1R,4S,5R)-3,3-Dibu tyl-7-flu or o-2,3,4,5-tetr a h yd r o-
5-(4-m eth oxyp h en yl)-1-ben zoth iep in -4-ol 1-Oxid e (11). To
a stirred solution of 0.51 g of 10 (1.18 mmol) in 10 mL of CH3-
OH was added 45 mg of NaBH4 (1.18 mmol) at -20 °C in a
dry ice-acetone bath. The mixture was stirred at -20 °C for
30 min and then slowly warmed to room temperature and
stirred for another 1 h. The mixture was quenched with H2O
at 0 °C, and CH3OH was evaporated. The mixture was
extracted with EtOAc, and the extracts were combined, washed
with brine, dried over MgSO4, and concentrated in vacuo to
give 0.5 g of the mixture of 3% 10, 27% of 3, and 70% of 11 (by
1H NMR). The mixture was chromatographed on silica gel
(hexane/ethyl acetate 17:3) to yield 0.12 g of 3 and 0.32 g of
11: 1H NMR (CDCl3) δ 0.83 (t, J ) 7.2 Hz, 3H), 0.99 (t, J )
7.2 Hz, 3H), 1.16-1.24 (m, 4H), 1.28-1.46 (m, 4H), 1.51-1.75
(m, 4H), 1.98-2.06 (m, 1H), 3.02-0.09 (m, 2H), 3.83 (s, 3H),
3.90 (d, J ) 9.2 Hz, 1H), 4.50 (d, J ) 9.6 Hz, 1H), 6.75 (dd, J
) 2.2 Hz, 10.6 Hz, 1H), 7.00 (d, J ) 8.8 Hz, 2H), 7.08 (td, J )
2.4 Hz, 8.2 Hz, 1H), 7.24 (d, J ) 3.2 Hz, 2H), 7.79 (dd, J ) 5.6
Hz, 8.8 Hz, 1H). A satisfactory analytical result was not
obtained due to instability of 11.
1
as a white solid: mp 159-160 °C. H NMR (CDCl3) δ 0.89-
0.95 (m, 6H), 1.09-1.42 (m, 12H), 2.21 (m, 1H), 3.08 (d, J )
15.6 Hz, 1H), 3.19 (d, J ) 15.6 Hz, 1H), 3.87 (s, 3H), 4.18 (s,
1H), (s, 1H), 6.54 (dd, J ) 10.2 Hz, 2.4 Hz, 1H), 6.96-7.07 (m,
3H), 7.40 (d, J ) 8.1 Hz, 2H), 8.11 (dd, J ) Hz, 5.85 Hz, 1H).13C
NMR (CDCl3) δ 14.18, 14.24, 23.39, 24.61, 25.15, 29.96, 35.39,
44.37, 46.23, 55.55, 56.91, 113.90, 114.19, 114.78, 119.24,
119.56, 129.94, 130.10, 134.25, 135.99, 136.05, 143.56, 143.66,
159.17. HRMS (ES+) calcd for C25H34O4SF (M + H) 449.2162
found 449.2184.
(4R,5R)-3,3-Dibu tyl-7-(d im eth yla m in o)-2,3,4,5-tetr a h y-
d r o-5-(4-m eth oxyp h en yl)-1-ben zoth iep in -4-ol 1,1-Dioxid e
(1a ). To a solution of 13.00 g (28.98 mmol) of 9 in 73 mL of
2.0 M of dimethylamine in THF (146 mmol) in a Parr reactor
was added ca. 20 mL of neat dimethylamine. The mixture was
sealed and stirred at 110 °C overnight and cooled to ambient
temperature. The excess dimethylamine and THF were evapo-
rated. The crude oil was dissolved in 200 mL of EtOAc and
washed with 100 mL of water, dried over MgSO4, and
concentrated in vacuo. Purification on a silica gel column using
hexane/ethyl acetate (4:1) gave 12.43 g (91%) of the product
1
Isom er iza tion of 3 a n d 11. To a stirred solution of 100
mg of 3 (0.23 mmol) in 3 mL of THF was added 0.25 mL of 1.0
M KOt-Bu in THF (0.25 mmol) at -10 °C. The mixture was
stirred at -10 °C for 30 min, and an aliquot was quickly
transferred into a vial cooled at -10 °C in a dry ice-acetone
bath. The aliquot was quenched with H2O and extracted with
ether. The ether layer was subject to HPLC equipped with a
Alltech Silica column (ethanol/hexane 2:98 at 1.0 mL/min) and
gave 3 at 16.98 min (76.8%). The whole mixture was quenched
with H2O, treated with saturated NH4Cl solution, extracted
with EtOAc, dried over MgSO4, and concentrated in vacuo. 1H
NMR (CDCl3) showed the major component to be 3, some
decomposition products, a small amount of 8, and no 11. To a
stirred solution of 100 mg of 11 (0.23 mmol) in 3 mL of THF
was added 0.25 mL of 1.0 M KOt-Bu in THF (0.25 mmol) at
-10 °C. The mixture was stirred at -10 °C for 30 min, and
an aliquot was quickly transferred into a vial cooled at -10
°C in a dry ice-acetone bath. The aliquot was quenched with
H2O, extracted with ether, and analyzed by HPLC and gave 3
at 16.20 min (74.8%). The mixture was quenched with H2O,
treated with saturated NH4Cl solution, extracted with EtOAc,
dried over MgSO4, and concentrated in vacuo. 1H NMR (CDCl3)
showed the major component to be 3, some decomposition
products, a small amount of 8, and no 11.
as a white solid: mp 134-135 °C. H NMR (CDCl3) δ 0.87-
0.93 (m, 6H), 1.10 (m, 12H), 2.17-2.25 (m, 1H), 2.81 (s, 6H),
2.99 (d, J ) 15.3 Hz, 1H), 3.15 (d, J ) 15.3 Hz, 1H), 3.84 (s,
3H), 4.11 (d, J ) 7.5 Hz, 1H), 5.49 (s, 1H), 5.99 (d, J ) 2.4 Hz,
1H), 6.51 (dd, J ) 8.7 Hz, 2.4 Hz, 1H), (d, J ) 8.7 Hz, 2H),
7.42 (d, J ) 8.4 Hz, 2H), 7.90 (d, J ) 8.7 Hz, 1H). 13C NMR
(CDCl3) δ 14.24, 14.30, 23.45, 23.47, 24.69, 25.19, 30.04, 35.61,
39.97, 44.36, 46.52, 55.53, 57.69, 76.67, 108.84, 114.28, 114.81,
126.23, 129.45, 130.28, 135.28, 139.94, 153.34, 158.75. HRMS
(ES+) calcd for C27H40NO4S (M + H) 474.2678 found 474.2684.
Anal. Calcd for C27H39NO4S: C, 68.46; H, 8.30; N, 2.96.
Found: 68.54; H, 8.29; N, 2.98. The product was determined
to have 89:11 er by chiral HPLC on a ChiralPak AD column
using 5% ethanol/hexane as the eluent. Recrystallization of
this solid from hexane/ethyl acetate gave 1.7 g of the racemic
product. The remaining solution was concentrated to give 10.6
g of the enantioenriched product as a white solid. Enantiomeric
ratio of this solid was determined by chiral HPLC to be 98:2
with the first eluting peak as the major product.
r el-(1R,5R)-3,3-Dibu tyl-7-flu or o-2,3-d ih yd r o-5-(4-m eth -
oxyp h en yl)-1-ben zoth iep in -4-(5H)-on e 1-Oxid e (10). To a
stirred solution of 0.46 mL of DMSO (6.48 mmol) in 1.0 mL of
CH2Cl2 was added 2.4 mL of 2.0 M oxalyl chloride in CH2Cl2
solution dropwise at -78 °C in a dry ice-acetone bath. The
mixture was stirred at -78 °C for 30 min. To the mixture was
added 1.4 g of (()-8 (3.24 mmol) in 10 mL of CH2Cl2 dropwise,
and the resulting mixture was stirred at -78 °C for 2 h. To
the stirred mixture was added 1.8 mL of triethylamine (12.96
mmol), and the mixture was stirred at -78 °C for 30 min. The
dry ice-acetone bath was removed, and the mixture was
stirred for 1 h. The mixture was quenched with H2O at 0 °C
and extracted with 100 mL of CH2Cl2. The extracts were
combined, washed with brine, dried over MgSO4, and concen-
trated in vacuo. The residue was chromatographed on silica
gel (hexane/ethyl acetate 95: 5) to afford 0.56 g of 10 (40%)
as a colorless oil: 1H NMR δ 0.77 (t, J ) 7.0 Hz, 3H), 0.86
Ack n ow led gm en t. We thank Professor Peter Beak,
Professor Victor Snieckus, and Dr. Andrew J . Carpenter
for helpful discussions and Dr. Sastry A. Kunda for
providing the X-ray crystals of the phenol of 1a .
Su p p or tin g In for m a tion Ava ila ble: One full-scale NMR
spectrum of each new compound; X-ray crystallographic data
ORTEP diagrams. This material is available free of charge via
the Internet at http://pubs.acs.org.
J O991786Q