An efficient, direct bis-ortho-chlorination of 4-(difluoromethoxy)aniline and its application to the synthesis of BMS-665053, a potent and selective pyrazinone-containing corticotropin-releasing factor-1 receptor antagonist

Article abstract of DOI:10.1021/op2003198

An efficient scale-up synthesis of (S)-5-chloro-1-(1-cyclopropylethyl)-3- (2,6-dichloro-4-(difluoromethoxy)phenylamino)-pyrazin-2(1H)-one, 1 (BMS-665053), is described. This new process features a one-step direct bis-ortho- chlorination of 4-(difluorometh

Full text of DOI:10.1021/op2003198

Article
pubs.acs.org/OPRD
An Efficient, Direct Bis-ortho-chlorination of 4-(Difluoromethoxy)-
aniline and Its Application to the Synthesis of BMS-665053, a Potent
and Selective Pyrazinone-Containing Corticotropin-Releasing
Factor-1 Receptor Antagonist
Jianqing Li,* Daniel Smith, Subramaniam Krishnananthan, Richard A. Hartz, Bireshwar Dasgupta,
Vijay Ahuja, William D. Schmitz, Joanne J. Bronson, Arvind Mathur, Joel C. Barrish, and Bang-Chi Chen
Medicinal Chemistry, Bristol-Myers Squibb Research and Development, Princeton, New Jersey, United States
ABSTRACT: An efficient scale-up synthesis of (S)-5-chloro-1-(1-cyclopropylethyl)-3-(2,6-dichloro-4-(difluoromethoxy)-
phenylamino)-pyrazin-2(1H)-one, 1 (BMS-665053), is described. This new process features a one-step direct bis-ortho-
chlorination of 4-(difluoromethoxy)aniline with HCl and H2O2, and a palladium-catalyzed coupling of 2,6-dichloro-4-
(difluoromethoxy)aniline 2 and (S)-3,5-dichloro-1-(1-cyclopropylethyl)pyrazin-2(1H)-one 3. The process was applied to the
preparation of batches of 1 for preclinical toxicology studies.
INTRODUCTION
RESULTS AND DISCUSSION
Original Synthesis of 2,6-Dichloro-4-(difluoro-
■
■
Novel pharmacological approaches leading to the development
of improved treatments for stress-related diseases, such as
anxiety and depression, are of significant interest to the
pharmaceutical industry. Despite the availability of currently
marketed anxiolytic and antidepressant drugs, issues with
currently available therapies, such as a delayed onset of action,
undesirable side effects, and a lack of efficacy in subgroups of
patients, stimulate continued interest in novel treatments. One
such potential novel method of treatment for stress-related
disorders includes corticotropin-releasing factor-1 (CRF1)
receptor antagonists.1−6
methoxy)aniline, 2. The preparation of (S)-5-chloro-1-
(1-cyclopropylethyl)-3-(2,6-dichloro-4-(difluoromethoxy)-
phenylamino)-pyrazin-2(1H)-one, 1, is a convergent synthesis
developed by our medicinal chemists. It requires two advanced
intermediates, 2,6-dichloro-4-(difluoromethoxy)aniline, 2,
and (S)-3,5-dichloro-1-(1-cyclopropylethyl)pyrazin-2(1H)-one,
3 (Scheme 1).7
Scheme 1. Convergent synthesis of 1
During our discovery program,7 (S)-5-chloro-1-(1-cyclo-
propylethyl)-3-(2,6-dichloro-4-(difluoromethoxy)-
phenylamino)-pyrazin-2(1H)-one, 1 (BMS-665053), Figure 1,
In the original synthesis, aniline intermediate 2 was prepared
from commercially available 3,5-dichlorophenol 4 (Scheme 2).7
Nitration of 2 with sodium nitrite in H2SO4 and water at reflux
afforded a mixture of 2- and 4-nitro regiomeric products 5 and
6 in a 2:1 ratio, from which the desired 4-nitro isomer 6 was
isolated by selective crystallization in 25% yield. Reaction of 6
with methyl difluorochloroacetate in DMF in the presence of
K2CO3 gave 1,3-dichloro-5-(difluoromethoxy)-2-nitrobenzene,
7, in 72% yield. Hydrogenation of 7 with palladium over
charcoal in methanol provided 2 in 47% yield after chromato-
graphic separation.
Figure 1. (S)-5-Chloro-1-(1-cyclopropylethyl)-3-(2,6-dichloro-4-
(difluoromethoxy)phenylamino)-pyrazin-2(1H)-one, 1.
was identified as a novel pyrazinone-containing CRF1 receptor
antagonist.7 Compound 1 is a high-affinity CRF1 receptor
antagonist with an IC50 of 1.0 nM and is a potent inhibitor
of CRF-stimulated cyclic adenosine monophosphate (cAMP)
production in human Y-79 retinoblastoma cells (IC50 = 4.9 nM).
It was selected as our lead candidate for further evaluation.
Herein, we describe a new and efficient scale-up synthesis of 1
to support preclinical toxicology studies.
Received: November 7, 2011
Published: December 16, 2011
© 2011 American Chemical Society
156
dx.doi.org/10.1021/op2003198 | Org. ProcessRes. Dev. 2012, 16, 156−159

Organic Process Research & Development
Article
the most frequently used conditions for o-chlorination of anilines.
Surprisingly, no product was detected by LC/MS. The starting
material, 8, mostly decomposed to a complex mixture (Table 1,
entry 1). The use of 6 N aqueous HCl resulted in a trace
amount of product (Table 1, entry 2). Further increasing the
HCl concentration to 12 N and adding the HCl solution at
0 °C (to minimize the escape of HCl gas) afforded the isolated
product 2 in 11% yield (Table 1, entry 3). Addition of the
concentrated HCl and 30% H2O2 at rt followed by refluxing for
16 h gave 2 in 22% isolated yield (Table 1, entry 4). From the
above experiments, we observed the following important
phenomena: (1) a higher concentration of HCl gave a better
yield; (2) the reaction was slightly exothermic and occurred
above room temperature; (3) both the addition of HCl and
H2O2 were exothermic, and prolonged aging above 50 °C
resulted in excessive oxidation of the aniline 8. Disappointed by
these results, we decided to use a different source of HCl for
the reaction. Thus, the commercially available 2 N HCl in ether
instead of aqueous HCl was used in the reaction. To our
delight, the reaction was much cleaner, and the desired product
dichloroaniline 2 was isolated in 43% yield (Table 1, entry 5).
Next, we decided to explore any potential solvent effects on this
reaction. When DMF was used as the solvent instead of
methanol, the reaction was very clean and was completed in
less than 2 h. The product 2,6-dichloro-4-(difluoromethoxy)-
aniline, 2, was isolated in 93% yield after extractive workup and
crystallization from hexane (Scheme 3 and Table 1, entry 6).
Scheme 2. Original synthesis of 2,6-dichloro-
4-(difluoromethoxy)aniline intermediate, 2
The original synthesis of intermediate 2 was amenable for
small-scale preparation. However, there were several issues
pertaining to scale-up. First, the nitration at an elevated temper-
ature was a safety concern, and the yield was low due to poor
regioselectivity. Second, while the difluoromethylation of 6 gave
satisfactory yield of the desired product 7, the possibility exists
for a potential competitive side reaction (polymerization
through nucleophilic aromatic substitution, i.e., phenolate
displacing chlorines which are activated by the o-nitro
group)8 under the reaction conditions especially during scale-
up. Third, des-halogenation occurred in the hydrogenation
step, and tedious column purification was required resulting in
the low isolated yield. Thus, a more efficient alternative scale-up
synthesis of 2,6-dichloro-4-(difluoromethoxy)aniline 2 was
desired.
Scheme 3. New improved synthesis of 2,6-dichloro-
4-(difluoromethoxy)aniline, 2
Alternative Synthesis of 2,6-Dichloro-4-(difluoro-
methoxy)aniline, 2. Our focus in the development of an
alternative synthesis of 2 was on the direct bis-ortho-
chlorination of 4-(difluoromethoxy)aniline, 8, which is
commercially available (Table 1). The direct ortho chlorination
Synthesis of (S)-3,5-Dichloro-1-(1-cyclopropylethyl)-
pyrazin-2(1H)-one, 3, and Final Coupling of 2 and 3.
With the one-step synthesis of intermediate 2 in place, we
turned our focus to the synthesis of intermediate 3 and the final
coupling of 2 with 3 (Scheme 4). It was reported that
substituted 3,5-dichloro-2(1H)-pyrazinones were prepared by
treatment of hydrochlorides of 2-sec-aminoalkanenitriles with
oxalyl chloride in o-dichlorobenzene at 80−100 °C.14 The 2-sec-
aminoalkanenitriles were obtained by Strecker synthesis.
However, this method has some scale up issues, such as the
use of potassium cyanide, the high boiling point and hazardous
solvent o-dichlorobenzene and column chromatography for
purification. We decided to attempt the nucleophilic sub-
stitution of chloroacetonitrile 10 with commercially available
(S)-1-cyclopropylethylamine hydrochloride, 9, to give inter-
mediate 11 and subsequently to optimize the reaction
conditions of 11 to 3 (Scheme 4).
Table 1. HCl source and solvent screen for the direct
bis-ortho-chlorination of 4-(difluoromethoxy)aniline, 8
time
(h)
entry
HCl source
solvent
temperature
yield (%)
1
2
3
4
5
6
1 N in water
6 N in water
12 N in water
12 N in water
2 N in Et2O
2 N in Et2O
MeOH
MeOH
MeOH
MeOH
MeOH
DMF
rt to reflux
rt to reflux
0 °C to reflux
rt to reflux
rt to 38 °C
rt to 38 °C
16
16
24
16
8
no product
trace
11
22
43
2
93
of aniline derivatives is a well-known reaction. A number of
chlorination reagents or systems such as Cl2,9 HCl/H2O2,10
ZnCl2/NaBiO3,11 NaOCl/FeCl3,12 and NCS13 have been
utilized for this transformation. Despite the fact that there
was no specific method reported for the direct chlorination of
p-alkoxyanilines, we were attracted by the potential efficiency of
this new transformation (one step vs three in the original
synthesis). We were particularly interested in using the
combination of HCl/H2O2 for the direct bis-ortho-chlorination
of 4-(difluoromethoxy)aniline, 8, because both HCl and H2O2
are readily available and environmentally benign.
Reaction of 9 with 10 in the presence of potassium iodide
and potassium carbonate in acetonitrile at 50 °C for 16 h
afforded the free base of 11 as an oil after aqueous workup which
1
contained ∼5% dialkylated product based on H NMR. The
crude oily free base was treated with 2 N HCl in ether to give
pure (S)-2-(1-cyclopropylethylamino)acetonitrile hydrochlor-
ide, 11, as a white crystalline solid in 91% yield. The reaction of
11 with oxalyl chloride was carried out in toluene at 55 °C for
Thus, 8 was treated with 4 equiv of 1 N aqueous HCl at
room temperature followed by addition of 30% H2O2 in
methanol, and the mixture was heated at reflux overnight,
157
dx.doi.org/10.1021/op2003198 | Org. ProcessRes. Dev. 2012, 16, 156−159

Organic Process Research & Development
Article
In conclusion, an efficient and direct bis-ortho-chlorination
of 4-(difluoromethoxy)aniline, 8, with HCl and H2O2 was
developed. Subsequent palladium-catalyzed coupling of 2,6-
dichloro-4-(difluoromethoxy)aniline, 2, with (S)-3,5-dichloro-
1-(1-cyclopropylethyl)pyrazin-2(1H)-one, 3, was found to be
superior to other base-mediated reaction conditions to give 1.
This process was successfully applied to the larger-scale
preparation of CRF1 receptor antagonist, BMS-665053.
Scheme 4. Improved synthesis of 11 and (S)-5-chloro-1-
(1-cyclopropylethyl)-3-[2,6-dichloro-4-(difluoromethoxy)-
phenylamino]-pyrazin-2(1H)-one, 1 (BMS-665053)
EXPERIMENTAL SECTION
■
All reagents were obtained from Aldrich Chemical Co. and used
without further purification unless otherwise stated. All
reactions were performed under a nitrogen atmosphere. All
glassware was dried and purged with nitrogen or argon before
use. All reactions were monitored by Shimadzu LC/MS system
using the following method: Phenomenex C18 column 10 μm
4.6 mm × 50 mm. Solvent: A = 10% methanol/90% water with
0.1% TFA; B = 90% methanol/10% water with 0.1% TFA.
Gradient: 0−100% B over 4 min. Flow: 4 mL/min, wavelength:
220 nm. HPLC analyses were performed using a Shimadzu system
1
(model SPD 10AV). All H NMR and 13C NMR spectra were
recorded on a Bruker 300 or 400 MHz spectrometer using
DMSO-d6 or CDCl3 as the solvents.
18 h. After extractive workup, the crude product was purified by
crystallization from 70% ethanol to give the product 3 in 65%
yield with 98% HPLC purity.
2,6-Dichloro-4-(difluoromethoxy)aniline, 2. 4-
(Difluoromethoxy)aniline, 8 (63.7 g, 0.4 mol), was dissolved
in dry DMF (800 mL). To the solution was added a solution of
2 N HCl in diethyl ether (800 mL, 1.6 mol) over a period of
0.5 h (rt to ∼32 °C). After cooling to rt, a solution of 30%
H2O2 (95.2 mL, 0.84 mol) was added dropwise over a period of
1 h (temperature rose to 38 °C). After stirring at ∼38 °C for 2 h,
the reaction mixture was cooled to rt and diluted with water
(2 L). The mixture was then extracted with heptane (2 × 1 L).
The combined organic extracts were washed with brine (2 L)
and concentrated in vacuo (∼30 mmHg) at 50 °C to give the
crude product (92.0 g) which was crystallized from heptane
(200 mL) at −20 °C to afford 85.0 g of 2,6-dichloro-4-
(difluoromethoxy)aniline, 2, as an off-white solid (98% HPLC
area purity, 93% yield). 1H NMR (400 MHz, CDCl3) δ 7.04 (s,
2H), 6.36 (t, J = 72.0 Hz, 1H), 4.40 (s br, 2H). 13C NMR (100
MHz, CDCl3) δ 141.0, 138.5, 120.8, 119.4, 115.8 (t, J = 260.4
Hz). LRMS (ESI) m/e 225.9 [(M − H)−, calcd for
C7H4NOCl2F2 226.0]. (Authors urge readers to follow MSDS
sheet regarding safe handling of 30% peroxide as well as related
waste stream.)
With substantial amounts of key intermediates 2 and 3 in
hand, screening was performed to determine the optimal
coupling conditions of 2 and 3 (Table 2). In the absence of a
base, heating of 2 and 3 at 120 °C for 1.5 h did not provide any
product (Table 2, entry 1). Using excess amount of K2CO3 as
the base, the reaction was carried out in toluene at reflux for
16 h, however, only a trace amount of the product was detected
by LC/MS (Table 2, entry 2). Treatment of compound 2 with
NaHMDS in THF and subsequent reaction with 3 for 16 h
only yielded 50% of the desired API (Table 2, entry 3). Using
Pd(OAc)2 and BINAP as the catalyst and Cs2CO3 as the base,
the reaction in toluene at reflux for 16 h afforded the product 1
in 28% yield. By switching the base to K2CO3, the reaction
went to completion (Table 2, entry 5). After workup,
trituration with heptane and crystallization from EtOH, the
final product (S)-5-chloro-1-(1-cyclopropylethyl)-3-[2,6-di-
chloro-4-(difluoromethoxy)phenylamino]-pyrazin-2(1H)-one,
1 (BMS-665053), was isolated in 76% yield with 99.5% HPLC
purity.
Table 2. Reagent and condition screening for coupling 2 and 3
entry
reagent and condition
isolated yield % (conversion %)
1
2
3
4
5
NMP, 120 °C, 1.5 h
K2CO3 (6.7 equiv), toluene, reflux, 16 h
NaHMDS (2 equiv), THF, rt, 16 h
no product
trace amount
50 (57)
3 mol % Pd(OAc)2, 3 mol % BINAP, Cs2CO3 (6.7 equiv), toluene, reflux, 16 h
3 mol % Pd(OAc)2, 3 mol % BINAP, K2CO3 (6.7 equiv), toluene, reflux, 16 h
28 (40)
76 (100)
158
dx.doi.org/10.1021/op2003198 | Org. ProcessRes. Dev. 2012, 16, 156−159

Organic Process Research & Development
Article
(S)-2-(1-Cyclopropylethylamino)acetonitrile Hydro-
chloride, 11. To a suspension of (S)-1-cyclopropylethylamine
hydrochloride, 9 (41.4 g, 340.2 mmol), in anhydrous
acetonitrile (900 mL) at room temperature was added
potassium carbonate (141.3 g, 1024 mmol), potassium iodide
(62.4 g, 376 mmol), and chloroacetonitrile, 10 (21.6 mL, 341.3
mmol). The reaction mixture was heated at 50 °C for 16 h with
vigorous stirring. The resulting mixture was then cooled to
room temperature and was filtered through a pad of Celite with
acetonitrile rinsing (100 mL). The filtrate was concentrated in
vacuo (∼30 mmHg) at 50 °C, and the residue was partitioned
between CH2Cl2 (1 L) and water (1 L). The organic phase was
separated, washed with water (1 L) and brine (1 L), dried over
Na2SO4, filtered, and concentrated in vacuo (∼30 mmHg) at
40 °C to give a brown oil. The oil was dispersed in diethyl ether
(900 mL), and the solution was treated with 2 N HCl in ether
(300 mL). The resulting suspension was cooled to 10 °C and
then filtered, and the filter cake was rinsed with cold ether
(80 mL). After drying by house vacuum for 2 h, the product
(S)-2-(1-cyclopropylethylamino)acetonitrile hydrochloride, 11
product, 1 (BMS-665053), as a slightly pink solid (99.5% HPLC
area purity, 76% yield). Mp 164−165 °C; [α]25 +20.2 (c 0.353,
D
1
CHCl3); H NMR (400 MHz, CDCl3) δ 7.69 (s, 1H), 7.21
(s, 2H), 6.90 (s, 1H), 6.52 (t, J = 72.6 Hz, 1H), 4.31−4.25
(m, 1H), 1.44 (d, J = 6.7 Hz, 3H), 1.12−1.04 (m, 1H), 0.79−0.73
(m, 1H), 0.61−0.54 (m, 1H), 0.51−0.45 (m, 1H), 0.41−0.35
(m, 1H). HRMS (ESI) m/e 424.0181 [(M + H)+, calcd for
C16H15N3O2Cl3F2 424.0198].
AUTHOR INFORMATION
Corresponding Author
*To whom correspondence should be addressed. Telephone:
609-252-3409. Fax: 609-252-7410. E-mail: jianqing.li@bms.com.
■
REFERENCES
■
(1) Dzierba, C. D.; Hartz, R. A.; Bronson, J. J. Recent advances in
corticotropin-releasing factor receptor antagonists. In Annu. Rep. Med.
Chem.; Macor, J. E., Ed.; Academic Press: San Diego, 2008; Vol. 43, pp
1− 23.
(2) Tellew, J. E.; Luo, Z. Small molecule antagonists of the
corticotropin-releasing factor (CRF) receptor: recent medicinal
chemistry developments. Curr. Top. Med. Chem. 2008, 8, 506−520.
(3) Grigoriadis, D. E. The corticotropin-releasing factor receptor: a
novel target for the treatment of depression and anxiety-related
disorders. Expert Opin. Ther. Targets 2005, 9, 651−684.
(4) Gilligan, P. J.; Li, Y.-W. Corticotropin-releasing factor
antagonists: recent advances and exciting prospects for the treatment
of human diseases. Curr. Opin. Drug Discovery Dev. 2004, 7, 487−497.
(5) Grigoriadis, D. E.; Haddach, M.; Ling, N.; Saunders, J. The CRF
receptor: structure, function and potential for therapeutic intervention.
Curr. Med. Chem. 2001, 1, 63−97.
(6) Gilligan, P. J.; Robertson, D. W.; Zaczek, R. Corticotropin
releasing factor (CRF) receptor modulators: progress and oppor-
tunities for new therapeutic agents. J. Med. Chem. 2000, 43, 1641−
1660.
(7) Hartz, R. A.; Ahuja, V. T.; Rafalski, M.; Schmitz, W. D.; Brenner,
A. B.; Denhart, D. J.; Ditta, J. L.; Deskus, J. A.; Yue, E. W.; Arvanitis,
A. G.; Lelas, S.; Li, Y.-W.; Molski, T. F.; Wong, H.; Grace, J. E.; Lentz,
K. A.; Li, J.; Lodge, N. J.; Zaczek, R.; Combs, A. P.; Olson, R. E.;
Mattson, R. J.; Bronson, J. J.; Macor, J. E. J. Med. Chem. 2009, 52,
4161−4172.
(8) Tao, Z-F; Sowin, T. J.; Lin, N.-H. Synlett 2007, 18, 2855−2858.
(9) Chang, Y.-S.; Deinzer, M. L. Synth. Commun. 1990, 20, 2501−
2506.
(49.6 g, 91% yield), was collected as an off-white solid.
1
Mp 115−116.1 °C; [α]25 −25.0 (c 0.779, CHCl3); H NMR
D
(400 MHz, DMSO-d6) δ 9.01 (s br, 2H), 4.14 (s, 2H), 2.41−
2.36 (m, 1H), 1.25 (d, J = 6.8 Hz, 3H), 0.91−0.82 (m, 1H),
0.62−0.55 (m, 1H), 0.55−0.41 (m, 2H), 0.24−0.18 (m, 1H).
LRMS (ESI) m/e 125.0 [(M + H)+, calcd for C7H13N2 125.1].
Anal. Calcd for C7H13ClN2: C, 52.34; H, 8.16; N, 17.44.
Found: C, 52.03; H, 8.24; N, 17.16.
(S)-3,5-Dichloro-1-(1-cyclopropylethyl)pyrazin-2(1H)-
one, 3. (S)-2-(1-Cyclopropylethyl-amino)acetonitrile hydro-
chloride (11) (49.5 g, 309 mmol) was suspended in dry toluene
(1.5 L). The mixture was cooled to 5 °C with an ice bath.
Oxalyl chloride (133.8 mL, 1542 mmol) was added. The
reaction mixture was then heated to 55 °C for 18 h. After
cooling to rt, the reaction mixture was concentrated in vacuo
(∼30 mmHg) at 60 °C to 200 mL. The mixture was then
added slowly to a cold (5 °C) saturated solution of KH2PO4
(1.0 L). The mixture was diluted with CH2Cl2 (1 L) and then
stirred for 20 min. The organic phase was separated and
concentrated in vacuo (∼30 mmHg) at 40 °C. The yellow
residue was crystallized from 70% aqueous ethanol (60 to 0 °C)
to afford 46.5 g of the product (S)-3,5-dichloro-1-(1-
cyclopropylethyl)pyrazin-2(1H)-one (3) as a white solid
(10) Vyas, P. V.; Bhatt, A. K.; Ramachandraiah, G.; Bedekar, A. V.
Tetrahedron Lett. 2003, 44, 4085−4088.
(11) Muathen, H. A. Helv. Chim. Acta 2003, 86, 164−168.
(12) Lu, S.-R. Hecheng Huaxue 2001, 9, 86−88.
(13) Moseley, J. D.; Brown, D.; Firkin, C. R.; Jenkin, S. L.; Patel, B.;
Snape, E. W. Org. Process Res. Dev. 2008, 12, 1044−1059.
(14) Vekemans, J.; Pollers-Wieers, C.; Hoornaert, G. J. Heterocycl.
Chem. 1983, 20, 919−923.
(65% yield, 98% HPLC area purity). [α]25 +27.6 (c 0.476,
D
1
CHCl3); H NMR (300 MHz, CDCl3) δ 7.47 (s, 1H), 4.26−
4.20 (m, 1H), 1.46 (d, J = 6.6 Hz, 3H), 1.13−1.06 (m, 1H),
0.85−0.77 (m, 1H), 0.64−0.59 (m, 1H), 0.58−0.50 (m, 1H),
0.39−0.32 (m, 1H). LRMS (APCI) m/e 233.0 [(M+H)+, calcd
for C9H11N2OCl2 233.0].
(S)-5-Chloro-1-(1-cyclopropylethyl)-3-[2,6-dichloro-4-
(difluoromethoxy)phenylamino]-pyrazin-2(1H)-one, 1
(BMS-665053). A mixture of Pd(OAc)2 (0.72 g, 3.21 mmol)
and BINAP (2.0 g, 3.21 mmol) in toluene (1 L) was stirred at rt
for 10 min. 2,6-Dichloro-4-(difluoromethoxy)aniline, 2 (26.9 g,
118 mmol), (S)-3,5-dichloro-1-(1-cyclopropylethyl)pyrazin-
2(1H)-one, 3 (25.0 g, 107 mmol and K2CO3 [100.0 g, 725
mmol]), were added, respectively. The mixture was heated to
reflux for 16 h. After cooling to rt, the mixture was filtered
through a Celite pad and rinsed with EtOAc (500 mL). The
filtrate was washed with brine (2 × 1 L) and then concentrated
in vacuo (∼30 mmHg) at 50 °C to give the crude product,
which was triturated with heptane and then crystallized from
EtOH (150 mL, 70 °C to −20 °C) to afford 34.5 g of the
159
dx.doi.org/10.1021/op2003198 | Org. ProcessRes. Dev. 2012, 16, 156−159