Synthesis of a selective estrogen receptor β-modulator via asymmetric phase-transfer catalysis

Article abstract of DOI:10.1016/j.tet.2007.03.074

An efficient asymmetric synthesis of selective estrogen receptor β-modulator (S)-4-bromo-9a-butyl-8-chloro-6-fluoro-7-hydroxy-1,2,9,9a-tetrahydro-fluoren-3-one was developed. The route features a chemoselective aromatic chlorination reaction, an asymmetri

Full text of DOI:10.1016/j.tet.2007.03.074

Tetrahedron 63 (2007) 4459–4463
Synthesis of a selective estrogen receptor b-modulator via
asymmetric phase-transfer catalysis
*
Mark A. Huffman, Jonathan D. Rosen, Roger N. Farr and Joseph E. Lynch
Department of Process Research, Merck & Co., Inc., PO Box 2000, Rahway, NJ 07065-0900, USA
Received 24 January 2007; revised 5 March 2007; accepted 12 March 2007
Available online 16 March 2007
Abstract—An efficient asymmetric synthesis of selective estrogen receptor b-modulator (S)-4-bromo-9a-butyl-8-chloro-6-fluoro-7-hydroxy-
1,2,9,9a-tetrahydro-fluoren-3-one was developed. The route features a chemoselective aromatic chlorination reaction, an asymmetric phase-
transfer-catalyzed alkylation of an indanone with efficient ee upgrade by racemate crystallization, and a robust bromination reaction using
imidazole as an in situ bromine trap to avoid overreaction. The synthesis proceeds in 34% yield over 8 steps from 2-fluoroanisole, and provides
material with >99.5% ee.
Ó 2007 Elsevier Ltd. All rights reserved.
O
Br
1. Introduction
O
F
F
Hormone replacement therapy (HRT) has been a widely
used and effective treatment for the symptoms of meno-
pause.1 However the Women’s Health Initiative2 identified
adverse effects associated with chronic HRT, which high-
light the need for improved therapies. The discovery of two
estrogen receptor subtypes (a and b)3 with different tissue
distributions4 raises the possibility that subtype-specific es-
trogen receptor modulators might avoid some of the adverse
effects of non-selective HRT. A program to identify b-selec-
tive ligands5 led to the identification of tetrahydrofluor-
enones including 1 as potent and subtype-selective estrogen
receptor b-modulators.6 This paper presents the develop-
ment of an asymmetric synthesis of 1 that allowed for the
preparation of larger quantities to support advanced preclin-
ical studies.
MeO
HO
Cl
Cl
1
2
O
F
F
MeO
MeO
Cl(H)
Cl(H)
3
4
Scheme 1.
The required methoxy indanone 2 could be accessed by a
Friedel–Crafts acylation/alkylation sequence, and a late-
stage bromination and deprotection would complete the
synthesis.
Racemic syntheses of 1 and related tetrahydrofluorenones
have been achieved through Robinson annulation of 2-
substituted-5-methoxy indanones.5a,6,7 In one case, targeting
a candidate for brain edema treatment, extension of the
catalytic asymmetric phase-transfer alkylation of indanones
to alkylation with 1,3-dichloro-2-butene led to an efficient
asymmetric Robinson annulation.8 Although the substrate
scope of the asymmetric annulation had not been defined,
this reaction might form the basis of an attractive approach
to 1 (Scheme 1).
2. Results and discussion
The route used during drug discovery6 began with Friedel–
Crafts acylation of 1-chloro-6-fluoroanisole (4a) with
hexanoyl chloride (Scheme 2). This reaction was sluggish
due to the electron deficiency of 4a; cleavage of the O-methyl
group followed by O-acylation of the resulting phenol com-
peted with the desired reaction. Reaction of the product
3a with formaldehyde in methanol gave a mixture of
methoxymethyl and methylene substituted products 5a and
6a. Acid-catalyzed intramolecular alkylation was only
moderately regioselective, giving indanones 2 and 7a in a
ratio of 85:15.
Keywords: Phase-transfer catalysis; Asymmetric catalysis; Bromination;
Chlorination.
* Corresponding author. Tel.: +1 732 594 3452; fax: +1 732 594 5170;
e-mail: mark_huffman@merck.com
0040–4020/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2007.03.074

4460
M. A. Huffman et al. / Tetrahedron 63 (2007) 4459–4463
O
O
F
MeO
F
AlCl3
Cl
CH2O
K2CO3, MeOH
+
CH2Cl2
MeO
X
X
4a, X=Cl
4b, X=H
3a, X=Cl
3b, X=H, 92%
O
O
O
O
F
X
F
F
H2SO4
toluene
+
+
MeO
X
MeO
MeO
OMe
MeO
F
X
X
5a, X=Cl
5b, X=H
6a, X=Cl
6b, X=H
2, X=Cl
2:7a = 85:15
8:7b >99:1
7a, X=Cl
7b, X=H
NaOCl
77%
8, X=H, 88%
Scheme 2.
To address these issues, we began our synthesis with 2-
fluoroanisole (4b). Friedel–Crafts acylation was facile and
highly para-selective, and 3b was crystallized in 92%
yield. After reaction with formaldehyde to give the mixture
of 5b and 6b, intramolecular aromatic substitution
occurred away from the fluorine substituent, giving inda-
none 8 with complete regioselectivity and 88% yield after
crystallization.
bromide as catalyst (Scheme 4). The desired (S) enantiomer
was obtained as expected,8,9 but in only 65% ee compared
with 92% ee previously obtained for alkylation of 6,7-di-
chloro-5-methoxy-2-propyl-1-indanone. The reaction was
optimized by reducing the catalyst charge to 5%, reducing
the large excess of dichlorobutene to a charge of 1.25 equiv,
and running the alkylation at 10 ꢀC for 30 h, which gave
a 95% yield and 76% ee. A further improvement to 80%
ee was possible by running the reaction under very dilute
conditions (100 mL toluene/g indanone), but this was
impractical for larger scale use.
At this stage, selective chlorination of 8 was required. In ad-
dition to the aromatic 4-position of indanone, the 2-position
a to the ketone, the benzylic 3-position and the aromatic 6-
position were all potential sites for reactivity with a chlorine
source. In practice, only reaction at a to the carbonyl com-
peted with the desired 4 chlorination (Scheme 3).
Cl
O
F
Cl
Cl
NaOH/tol
PTC
+
2
MeO
Cl
O
O
F
F
Cl
Cl
11
+
+
8
2
O
MeO
MeO
Cl
F
9
10
1) H2SO4, tol, 0 °C
2) H2SO4, H2O
tol, 60 °C
MeO
Scheme 3.
Cl
12
Sulfuryl chloride, NCS, and dichlorodimethyl hydantoin all
reacted selectively at the 2-position to give undesired 9
under anhydrous conditions. In the presence of aqueous
acid, the two N-chloro reagents also reacted at the 4-position
giving some desired 2 and dichloro derivative 10. tert-Butyl
hypochlorite in acetic acid gave good selectivity toward 2.
Similar selectivity was seen with aqueous sodium hypochlo-
rite in acetic acid and with gaseous chlorine in aqueous
acetic acid. Both NaOCl and Cl2 favored 9 and 10 under
non-acidic conditions.
Scheme 4.
A few other cinchona-based phase-transfer catalysts were
evaluated without improvement in selectivity. It is note-
worthy that N-(9-anthracenylmethyl)-dihydrocinchoninium
bromide and O-(9)-allyl-N-(9-anthracenylmethyl)-cinchoni-
dinium bromide catalysts, which carry out highly enantio-
selective alkylations of tert-butylglycinate benzophenone
imine,10 were poorly selective in the alkylation of indanone
2, providing a maximum ee of 26%.
The most practical procedure involved addition of aqueous
NaOCl to a solution of 8 in acetic acid. The product was
crystallized from the reaction mixture and could be isolated
directly in 77% yield.
Cyclization of alkylation product 11 to the tetrahydrofluor-
enone 12 was accomplished with sulfuric acid and water in
toluene.8 The enantiomeric purity of 12 could be increased to
97% ee by crystallization of the insoluble racemate from
heptane. In practice, however, the upgrade was deferred to
a later stage where it was found to be even more efficient
(vide infra).
The key asymmetric alkylation of 2 with 1,3-dichloro-2-
butene was initially tried under the literature conditions,
using
10%
N-(p-trifluoromethylbenzyl)cinchoninium

M. A. Huffman et al. / Tetrahedron 63 (2007) 4459–4463
4461
Conversion of 12 to the final product 1 required installation
3. Conclusion
of the bromine atom and cleavage of the phenolic methyl
group (Scheme 5). In the original synthesis,6 bromination
was performed first to give racemic 13, which was resolved
by chromatography. Deprotection with BBr3 then gave 1.
The drawback for large-scale asymmetric synthesis was
that 13, like 12, was only crystalline in racemic form. In con-
trast, deprotection of 12 with AlCl3 gave free phenol 14,
which could be crystallized. We felt that isolation at this
stage would provide valuable purification prior to the final
step. Racemic 14 is extremely insoluble, and the enantio-
meric purity was upgraded to 99% ee by filtration of the
quenched reaction mixture prior to crystallization of the
(S) isomer from acetonitrile. Compound 14 was obtained
in 61% yield over the three steps from indanone 2 with
23% having been removed as racemate.
Selective estrogen receptor b-modulator 1 was synthesized
in eight steps and 34% overall yield, starting from 2-fluoro-
anisole. The synthesis features a chemoselective aromatic
chlorination reaction, an asymmetric phase-transfer-cata-
lyzed alkylation with efficient ee upgrade by racemate crys-
tallization, and a robust bromination reaction, which uses
imidazole as an in situ bromine trap to avoid overreaction.
The route was used to prepare kilogram quantities of 1 to
allow for advanced preclinical studies.
4. Experimental
4.1. General methods
All reactions were conducted under an atmosphere of dry
nitrogen. All reagents and solvents are available from
commercial sources and were used as received unless other-
wise indicated. 1,3-Dichloro-2-butene was prepared accord-
ing to a method described in a patent.12 Dichloromethane
O
Br
F
MeO
and toluene were dried over 4 A molecular sieves. 1H
˚
Cl
13
NMR and 13C NMR spectra were recorded at 400 MHz
and 100 MHz, respectively. Chemical shifts are referenced
to solvent signals (proton d 7.27 for CHCl3, carbon d 77.0
for CDCl3).
12
1
AlCl3
O
Br2
imidazole
F
4.1.1. 1-(3-Fluoro-4-methoxyphenyl)hexan-1-one (3b).
To a solution of 2-fluoroanisole (158 g, 1250 mmol) in
dry dichloromethane (1.25 L) was added AlCl3 (196 g,
1470 mmol). The solution was then cooled on a water bath
while hexanoyl chloride (194 g, 1440 mmol) was added
dropwise. The resulting solution was stirred for 30 min at
room temperature, then cooled to ꢁ10 ꢀC. Water (1.25 L)
was added slowly to quench. Caution: quenching is highly
exothermic. The organic layer was separated and washed
with aqueous 5 N NaOH (625 mL) followed by water
(625 mL). The final organic layer was evaporated to low vol-
ume and the residue was dissolved in hot heptane (850 mL).
Crystallization occurred during cooling and the solid was
collected at 0 ꢀC, rinsing with cold heptane. Drying under
vacuum gave 3b as white crystals (258 g, 92%). Mp 61–
62 ꢀC; 1H NMR (CDCl3) d 7.74 (m, 1H), 7.69 (dd,
J¼12.0, 2.1 Hz, 1H), 6.99 (t, J¼8.4 Hz, 1H), 3.95 (s, 3H),
2.87 (t, J¼7.4 Hz, 2H), 1.72 (m, 2H), 1.35 (m, 4H), 0.91
(m, 3H); 13C NMR (CDCl3) d 198.3, 152.1 (d, J¼248 Hz),
151.7 (d, J¼11 Hz), 130.5 (d, J¼5 Hz), 125.3 (d, J¼3 Hz),
115.8 (d, J¼19 Hz), 112.4 (d, J¼1 Hz), 56.3, 38.3, 31.6,
24.2, 22.6, 14.0. Anal. Calcd for C13H17FO2: C, 69.62; H,
7.64. Found: C, 69.79; H, 7.68.
HO
Cl
14
Scheme 5.
Bromination to give 1 could be carried out by addition of
precisely 1 equiv of bromine to a mixture of 14 and sodium
acetate in ethanol/acetic acid. Addition of water then crystal-
lized the product. Stoichiometry was critical because any ex-
cess bromine catalyzed formation of a dimeric by-product.11
On the other hand, if less than 0.995 equiv of bromine was
added, the remaining 14 was not adequately removed in
the crystallization. Each of these scenarios required a differ-
ent recrystallization procedure to achieve target purity.
The fact that the dimeric impurity forms only after all of 14
is consumed indicates that this side reaction is significantly
slower than the consumption of bromine in the formation of
1. This raises the possibility that a bromine trapping reagent
of intermediate reactivity might be used to prevent the side
reaction without interfering with the desired bromination.
About 20 potential bromine traps were evaluated, including
olefinic, aromatic, and heteroaromatic compounds. From
these experiments, imidazole emerged as having the appro-
priate rate of reactivity. Imidazole has the additional benefit
of the proper basicity to replace sodium acetate in the role of
base for the bromination reaction, which liberates HBr. Un-
der the optimized conditions, 1.75 equiv of imidazole are
combined with 14 in ethanol/acetic acid and bromine is
added slowly. One equivalent of bromine is sufficient for
complete conversion, but excess can be used and is con-
verted to mono- and polybromo imidazoles. Addition of wa-
ter crystallizes 1 in 90% yield, or 88% after a decolorizing
carbon treatment.
4.1.2. 1-(3-Fluoro-4-methoxyphenyl)-2-methoxymethyl-
hexan-1-one (5b) and 1-(3-fluoro-4-methoxyphenyl)-2-
methylenehexan-1-one (6b). To a solution of 3b (253 g,
1130 mmol) in methanol (1.13 L) were added K2CO3
(156 g, 1130 mmol) and aqueous 37% formaldehyde
(110 g, 1355 mmol). The suspension was stirred at 50 ꢀC for
18 h, and then cooled to room temperature. Toluene (1.13 L)
and water (1.13 L) were added and the layers were sepa-
rated. The organic layer was washed again with water
(1.13 L), and then concentrated to a volume of 500 mL.
The solution was used in the next step without further

4462
M. A. Huffman et al. / Tetrahedron 63 (2007) 4459–4463
purification. NMR data were acquired on the mixture. Com-
pound 5b: 1H NMR (CDCl3) d 7.78 (ddd, J¼8.3, 2.0, 0.8 Hz,
1H), 7.74 (dd, J¼12.0, 2.0 Hz, 1H), 7.01 (t, J¼8.3 Hz, 1H),
3.96 (s, 3H), 3.71–3.61 (m, 2H), 3.49 (q, J¼3.8 Hz, 1H),
3.29 (s, 3H), 1.71 (m, 1H), 1.51 (m, 1H), 1.27 (m, 4H), 0.85
(t, J¼7.2 Hz, 3H). Compound 6b: 1H NMR (CDCl3)
d 7.62 (ddd, J¼8.4, 2.1, 1.0 Hz, 1H), 7.59 (dd, J¼12.0,
2.1 Hz, 1H), 6.99 (t, J¼8.4 Hz, 1H), 5.75 (q, J¼1.2 Hz,
1H), 5.50 (s, 1H), 3.97 (s, 3H), 2.46 (t, J¼7.4 Hz, 2H),
1.38 (m, 2H), 1.27 (m, 2H), 0.93 (t, J¼7.2 Hz, 3H).
1H), 1.91 (m, 1H), 1.52 (m, 1H), 1.42–1.25 (m, 3H), 0.92
(t, J¼7.2 Hz, 3H); 13C NMR (CDCl3) d 198.7, 155.1 (d,
J¼12 Hz), 152.9 (d, J¼251 Hz), 148.1, 126.1 (d, J¼6 Hz),
111.3 (d, J¼19 Hz), 109.2, 71.0, 56.5, 42.9, 38.4, 26.8,
22.7, 13.9. Anal. Calcd for C14H16ClFO2: C, 62.11; H,
5.96. Found: C, 62.09; H, 6.00.
4.1.6. 2-Butyl-2,4-dichloro-6-fluoro-5-methoxyindan-1-
one (10). To a solution of 8 (59 mg, 0.25 mmol) in acetoni-
trile (0.5 mL) was added NCS (91 mg, 0.68 mmol) and conc
HCl (42 mL). After stirring for 48 h, the mixture was parti-
tioned between ethyl acetate and water. The organic layer
was washed with water and aqueous 5% NaHCO3, then di-
luted with an equal volume of heptane and filtered through
a pad of silica gel. Evaporation gave an oil (68 mg, 89%).
1H NMR (CDCl3) d 7.50 (d, J¼9.6 Hz, 1H), 4.13 (d,
J¼2.8 Hz, 3H), 3.44 (m, 2H), 2.14 (m, 1H), 1.91 (m, 1H),
1.56 (m, 1H), 1.43–1.25 (m, 3H), 0.93 (t, J¼7.2 Hz, 3H);
13C NMR (CDCl3) d 198.2, 155.8 (d, J¼253 Hz), 150.8 (d,
J¼14 Hz), 145.1 (d, J¼2 Hz), 128.7 (d, J¼7 Hz), 125.4 (d,
J¼4 Hz), 111.4 (d, J¼21 Hz), 70.0, 61.7 (d, J¼7 Hz),
41.9, 38.1, 26.8, 22.7, 13.9. HRMS calcd for C14H15Cl2FO2:
305.0511 (M+H). Found: 305.0518 (M+H).
4.1.3. 2-Butyl-6-fluoro-5-methoxyindan-1-one (8). The
solution of 5b and 6b in toluene (estimated 1130 mmol
combined with 500 mL) was diluted with toluene (2 L).
Concentrated sulfuric acid (250 mL) was added and the
two-phase mixture was stirred at 50 ꢀC for 3 h. The mixture
was then cooled to 5 ꢀC and slowly quenched with water
(1.25 L), maintaining the temperature below 20 ꢀC. Layers
were separated and the organic layer was washed again
with water (1.25 L). The final toluene solution was concen-
trated to low volume and the residue was dissolved in hot
heptane (1 L). The product crystallized upon cooling and
was isolated by filtration at 0 ꢀC, rinsing with cold hep-
tane/toluene 9:1. Drying under vacuum gave 8 as off-white
1
crystals (231 g, 88%). Mp 82–83 ꢀC; H NMR (CDCl3)
4.1.7. (S)-2-Butyl-4-chloro-2-(3-chlorobut-2-enyl)-6-fluoro-
5-methoxyindan-1-one (11). To a solution of 2 (92.5 g,
342 mmol) in toluene (1.4 L) was added N-[4-(trifluoro-
methyl)-benzyl]cinchoninium bromide (9.1 g, 17 mmol).
Aqueous 50% NaOH (465 mL) was added and the mix-
ture was cooled to 10 ꢀC. 1,3-Dichloro-2-butene (534 g,
427 mmol, mixture of Z and E isomers) was then added
and the mixture was vigorously stirred for 30 h. The reaction
was then quenched by slow addition of water (1.4 L), which
raised the temperature to 20 ꢀC. The aqueous layer was
removed. The organic layer was washed with water
(450 mL) followed by 1 M citric acid (450 mL). The final
toluene layer was filtered to remove a small amount of pre-
cipitate, then concentrated to a volume of 500 mL. The crude
solution was used directly in the next step. 1H NMR (CDCl3)
(signals reported for major Z isomer) d 7.39 (d, J¼9.6 Hz,
1H), 5.29 (m, 1H), 4.11 (d, J¼2.4 Hz, 3H), 2.92 (d,
J¼1.5 Hz, 2H), 2.50 (qdd, J¼14.9, 6.8, 1.4 Hz, 2H), 2.35
(br d, J¼8.0 Hz, 1H), 2.05 (d, J¼1.2 Hz, 3H), 1.64 (m,
3H), 1.30–0.98 (m, 4H), 0.85 (t, J¼7.2 Hz, 3H).
d 7.38 (d, J¼9.8 Hz, 1H), 6.94 (d, J¼7.2 Hz, 1H), 3.96 (s,
3H), 3.24 (dd, J¼17.0, 7.7 Hz, 1H), 2.74 (br d, J¼17.0 Hz,
1H), 2.64 (m, 1H), 1.92 (m, 1H), 1.48–1.29 (m, 5H), 0.90
(t, J¼7.1 Hz, 3H); 13C NMR (CDCl3) d 207.0 (d, J¼3 Hz),
153.9 (d, J¼20 Hz), 152.6 (d, J¼252 Hz), 151.3 (d,
J¼5 Hz), 129.6 (d, J¼6 Hz), 110.1 (d, J¼18 Hz), 109.3 (d,
J¼1 Hz), 56.4, 47.9, 32.8, 31.3, 29.5, 22.8, 14.0. Anal.
Calcd for C14H17FO2: C, 71.16; H, 7.25. Found: C, 71.28;
H, 7.35.
4.1.4. 2-Butyl-4-chloro-6-fluoro-5-methoxyindan-1-one
(2). A solution of 8 (140 g, 593 mmol) in glacial acetic
acid (1.77 L) was cooled to 16 ꢀC in a water bath. Aqueous
7.57% NaOCl (1.14 kg, 1160 mmol) was added dropwise
over 4 h. Water (800 mL) was then added over 30 min to
complete the crystallization. The suspension was filtered,
rinsing with 1:1 AcOH/water, water, and finally with meth-
anol/water 2:1. Drying under vacuum gave 2 as an off-white
solid (124 g corrected for 99% purity, 77%). An analytically
pure sample was obtained by recrystallization from metha-
1
nol. Mp 56–57 ꢀC; H NMR (CDCl3) d 7.39 (d, J¼9.5 Hz,
4.1.8.
(S)-9a-Butyl-8-chloro-6-fluoro-7-methoxy-
1H), 4.09 (d, J¼2.4 Hz, 3H), 3.27 (qd, J¼9.8, 1.4 Hz, 1H),
2.76–2.66 (m, 2H), 1.94 (m, 1H), 1.52–1.33 (m, 5H),
0.93 (t, J¼7.1 Hz, 3H); 13C NMR (CDCl3) d 206.4,
155.7 (d, J¼252 Hz), 149.8 (d, J¼14 Hz), 148.4, 132.2
(d, J¼7 Hz), 125.6 (d, J¼4 Hz), 109.8 (d, J¼20 Hz), 61.6
(d, J¼7 Hz), 47.5, 31.8, 31.2, 29.4, 22.7, 14.0. Anal. Calcd
for C14H16ClFO2: C, 62.11; H, 5.96. Found: C, 62.22; H,
5.98.
1,2,9,9a-tetrahydrofluoren-3-one (12). The solution of 11
(325 mmol) in toluene (500 mL) was cooled to 0 ꢀC with
an ice bath. Concentrated H2SO4 (350 mL) was added over
1 h. Caution: addition is exothermic and HCl is liberated.
The mixture was stirred for 1 h more at 35 ꢀC. Water
(11 mL) was then added and the mixture was heated to
65 ꢀC for 2 h. The mixture was then cooled to 20 ꢀC and
slowly added into cold water (750 mL). The layers were sep-
arated and the organic layer was washed with aqueous 5%
NaHCO3 (500 mL). Assays showed 105 g (95% from 2)
and 76% ee. The toluene solution was carried into the next
step. A small sample of racemic 12 was collected by evapo-
ration of the toluene solution and crystallization from hep-
tane. 1H NMR (CDCl3) d 7.19 (d, J¼10.1 Hz, 1H), 6.10 (s,
1H), 4.02 (d, J¼1.8 Hz, 3H), 3.08 (d, J¼16.7 Hz, 1H),
2.66 (d, J¼16.7 Hz, 1H), 2.61–2.43 (m, 2H), 2.31 (ddd,
J¼13.2, 5.1, 1.9 Hz, 1H), 1.98 (td, J¼13.2, 5.6 Hz, 1H),
4.1.5. 2-Butyl-2-chloro-6-fluoro-5-methoxyindan-1-one
(9). To a solution of 8 (118 mg, 0.50 mmol) in acetic acid
(0.5 mL) was added sulfuryl chloride (74 mg, 0.55 mmol).
After 90 min, the reaction was quenched with water and ex-
tracted with ethyl acetate. The organic layer was washed
twice with water and evaporated to a yellow oil (118 mg,
1
87%). H NMR (CDCl3) d 7.50 (d, J¼9.5 Hz, 1H), 6.93
(d, J¼7.0 Hz, 1H), 3.99 (s, 3H), 3.46 (s, 2H), 2.10 (m,

M. A. Huffman et al. / Tetrahedron 63 (2007) 4459–4463
4463
1.62 (m, 1H), 1.46 (m, 1H), 1.31–1.17 (m, 4H), 0.86 (t,
4.2. Measurement of enantiomeric excess
J¼7.0 Hz, 3H); 13C NMR (CDCl3) d 198.6, 171.5 (d,
J¼3 Hz), 155.8 (d, J¼249 Hz), 147.0 (d, J¼14 Hz), 141.8
(d, J¼2 Hz), 133.5 (d, J¼9 Hz), 125.4 (d, J¼4 Hz), 117.5,
109.1 (d, J¼21 Hz), 61.7 (d, J¼6 Hz), 46.5, 42.4, 37.5,
33.8, 31.9, 27.5, 23.2, 14.0. Anal. Calcd for C18H20ClFO2:
C, 66.97; H, 6.24. Found: C, 67.00; H, 6.25.
Enantiomer ratios were measured by SFC using isocratic
methods with methanol in CO2 at 200 bar and 35 ꢀC with
a flow rate of 1.5 mL/min and column dimensions of
250ꢂ4.6 mm. For 12: Chiralpak AD column, 15% metha-
nol, monitoring 320 nm. For 14: Chiralpak AS column,
27% methanol, monitoring 330 nm. For 1: Chiralcel OJ
column, 25% methanol, monitoring 340 nm.
4.1.9. (S)-9a-Butyl-8-chloro-6-fluoro-7-hydroxy-1,2,9,9a-
tetrahydrofluoren-3-one (14). A toluene solution of 12
(30.0 g, 92.9 mmol, 76% ee) was azeotropically dried and
concentrated to a volume of 170 mL. The solution was
cooled to 5 ꢀC and AlCl3 (18.6 g, 139 mmol) was added.
Once the exotherm subsided, the mixture was warmed to
40 ꢀC for 2.5 h. The reaction was then quenched at 40 ꢀC
by slow addition of 1-propanol (29 mL). The mixture was
stirred for 1 h more at 45 ꢀC, and then cooled to 20 ꢀC.
A pre-mixed combination of aqueous 10% citric acid
(310 mL) and aqueous 2 N HCl (35 mL) was added, very
slowly at first with cooling to keep the temperature between
20 and 50 ꢀC. The mixture was then stirred for 1 h more at
20 ꢀC. The precipitated racemate was removed by filtration,
rinsing with 5% 1-propanol in toluene. The filtrate was
allowed to settle and the aqueous layer was removed. The
organic layer was washed again with the aqueous citric acid/
HCl mixture (225 mL) followed by 0.1 N HCl (225 mL),
adding 1-propanol (14 mL) to the organic phase before
each wash. The final toluene solution was evaporated with
simultaneous addition of acetonitrile until toluene was re-
moved and the final volume was 132 mL. The mixture was
heated to 55 ꢀC, and then cooled slowly to induce crystalli-
zation. The solid was collected at ꢁ20 ꢀC, rinsing with cold
acetonitrile. Drying under vacuum gave 14 as a tan solid
(18.23 g, 64%, 99% ee). Mp 121–122 ꢀC; 1H NMR
(CDCl3) d 7.49 (br s, 1H), 7.21 (d, J¼9.3 Hz, 1H), 6.12 (s,
1H), 3.08 (d, J¼16.7 Hz, 1H), 2.68 (d, J¼16.7 Hz, 1H),
2.64–2.47 (m, 2H), 2.31 (ddd, J¼13.2, 5.1, 1.9 Hz, 1H),
2.00 (td, J¼13.2, 5.7 Hz, 1H), 1.63 (m, 1H), 1.47 (m, 1H),
1.31–1.16 (m, 4H), 0.86 (t, J¼7.0 Hz, 3H); 13C NMR
(CDCl3) d 199.7, 173.2, 151.5 (d, J¼245 Hz), 144.4 (d,
J¼16 Hz), 142.1 (d, J¼2 Hz), 129.9 (d, J¼8 Hz), 119.0 (d,
J¼4 Hz), 116.3, 108.4 (d, J¼20 Hz), 46.6, 42.3, 37.7,
33.7, 31.9, 27.6, 23.2, 14.0. Anal. Calcd for C17H18ClFO2:
C, 66.13; H, 5.88. Found: C, 65.90; H, 5.81.
References and notes
1. Warren, M. P.; Halpert, S. Best Pract. Res. Clin. Endrocrinol.
Metabol. 2004, 18, 317.
2. (a) Women’s Health Initiative. JAMA 2002, 288, 321; (b)
Women’s Health Initiative. JAMA 2004, 291, 1701.
3. (a) Mosselman, S.; Polman, J.; Dijkema, R. FEBS Lett. 1996,
392, 49; (b) Kuiper, G. G. J. M.; Enmark, E.; Pelto-Huikko,
M.; Nilsson, S.; Gustafsson, J.-A. Proc. Natl. Acad. Sci.
U.S.A. 1996, 93, 5925.
4. (a) Couse, J. J.; Lindzey, J.; Grandien, K.; Gustafsson, J.-A.;
Korach, K. S. Endocrinology 1997, 138, 4613; (b) Kuiper,
G. G. J. M.; Carlsson, B.; Grandien, K.; Enmark, E.;
Haggblad, J.; Nilsson, S.; Gustafsson, J.-A. Endocrinology
1997, 138, 863.
5. (a) Wilkening, R. R.; Ratcliffe, R. W.; Tynebor, E. C.;
Wildonger, K. J.; Fried, A. K.; Hammond, M. L.; Mosley,
R. T.; Fitzgerald, P. M. D.; Sharma, N.; McKeever, B. M.;
Nilsson, S.; Carlquist, M.; Thorsell, A.; Locco, L.; Katz, R.;
Frisch, K.; Birzin, E. T.; Wilkinson, H. A.; Mitra, S.; Cai, S.;
Hayes, E. C.; Shaeffer, J. M.; Rohrer, S. P. Bioorg. Med.
Chem. Lett. 2006, 16, 3489; (b) Norman, B. H.; Dodge, J. A.;
Richardson, T. I.; Borromeo, P. S.; Lugar, C. W.; Jones,
S. A.; Chen, K.; Wang, Y.; Durst, G. L.; Barr, R. J.;
Montrose-Rafizadeh, C.; Osborne, H. E.; Amos, R. M.; Guo,
S.; Boodhoo, A.; Krishnan, V. J. Med. Chem. 2006, 49, 6155;
(c) Manas, E. S.; Unwalla, R. J.; Xu, Z. B.; Malamas, M. S.;
Miller, C. P.; Harris, H. A.; Hsiao, C.; Akopian, T.; Hum,
W.-T.; Malakian, K.; Wolfrom, S.; Bapat, A.; Bhat, R. A.;
Stahl, M. L.; Somers, W. S.; Alvarez, J. C. J. Am. Chem. Soc.
2004, 126, 15106.
6. Parker, D. L. Jr.; Ratcliffe, R. W.; Wilkening, R. R.; Wildonger,
K. J. Patent WO0182923, 2001.
7. Cragoe, E. J., Jr.; Woltersdorf, O. W., Jr.; Gould, N. P.;
Pietruszkiewicz, A. M.; Ziegler, C.; Sakurai, Y.; Stokker,
G. E.; Anderson, P. S.; Bourke, R. S.; Kimelberg, H. K.;
Nelson, L. R.; Barron, K. D.; Rose, J. R.; Szarowski, D.;
Popp, A. J.; Waldman, J. B. J. Med. Chem. 1986, 29, 825.
8. Bhattacharya, A.; Dolling, U.-H.; Grabowski, E. J. J.; Karady,
S.; Ryan, K. M.; Weinstock, L. M. Angew. Chem., Int. Ed. Engl.
1986, 25, 476.
9. Enantiomeric ratio was measured after cyclization to form 12.
10. (a) Corey, E. J.; Xu, F.; Noe, M. C. J. Am. Chem. Soc. 1997,
119, 12414; (b) Lygo, B.; Wainwright, P. G. Tetrahedron
Lett. 1997, 38, 8595.
11. The by-product mass corresponds to a dimer of 1 formed with
loss of HBr; HRMS calcd for C34H33BrCl2F2O4 (M+H):
693.0986. Found: 693.1008 (M+H).
12. Hagedorn, F.; Wedemeyer, K.; Mayer-Mader, R. German
Patent 2318115, 1973.
4.1.10.
(S)-4-Bromo-9a-butyl-8-chloro-6-fluoro-7-
hydroxy-1,2,9,9a-tetrahydrofluoren-3-one (1). To a solu-
tion of 14 (10.0 g, 32.4 mmol) in acetic acid (25 mL) and
ethanol (50 mL) was added activated charcoal Darco
KB-B (2.0 g). The mixture was stirred 1 h, filtered, and
rinsed with 1:2 acetic acid/ethanol (35 mL). Imidazole
(3.86 g, 56.7 mmol) was added and the solution was cooled
to 0 ꢀC. Bromine (1.59 mL, 4.97 g, 31.1 mmol) was added
dropwise, maintaining the temperature below 10 ꢀC. The
solution was then warmed to 21 ꢀC and water (100 mL)
was added slowly to induce crystallization. The solid was
collected by filtration and rinsed with 1:1 acetic acid/water.
Drying under vacuum at 40 ꢀC gave 1 as an off-white
solid (11.06 g, 88%, >99.5% ee). Anal. Calcd for
C17H17BrClFO2: C, 52.67; H, 4.42. Found: C, 52.73; H,
4.45.