A scalable synthesis of CE-157119 HCl Salt, an SRI/5-HT2A antagonist

Article abstract of DOI:10.1021/op3002273

A scalable synthesis of CE-157119 HCl salt (1), an SRI/5-HT_2A antagonist, was developed via the regioselective S_NAr etherification between a phenol and an N-methylamide. This early development route shortened the original 5-step synt

Full text of DOI:10.1021/op3002273

Article
pubs.acs.org/OPRD
A Scalable Synthesis of CE-157119 HCl Salt, an SRI/5-HT2A Antagonist
Yong Tao,* Daniel W. Widlicka, Paul D. Hill, Michel Couturier, and Gregory R. Young
Chemical Research and Development, Groton Laboratories, Pfizer Worldwide Research and Development, Eastern Point Road,
Groton, Connecticut 06340, United States
ABSTRACT: A scalable synthesis of CE-157119 HCl salt (1), an SRI/5-HT2A antagonist, was developed via the regioselective
SNAr etherification between a phenol and an N-methylamide. This early development route shortened the original 5-step
synthesis to three steps, eliminated all chromatography and increased the overall yield from 15% to 34%. The process was
implemented for API manufacture from 100-g scale to multikilogram scale.
INTRODUCTION
RESULTS AND DISCUSSION
■
■
In our continuous effort to find new antidepressants that
combine potent 5-HT2A receptor blocking activity with the
established efficacy and safety of a serotonin reuptake inhibitor
(SRI), CE-157119 HCl salt (1, Figure 1) was identified as a
Improving the regioselectivity of the SNAr etherification was
critical for development of a scalable synthesis. It was
hypothesized that the coordination of the carbonyl oxygen
with the potassium cation of the phenolate facilitates the
regioselectivity at C-2 position over C-4 as depicted in Figure 2.
We anticipated that replacing the aldehyde function at C-1
position with a weaker electron-withdrawing group, such as
methyl amide 10 or ethyl ester 3, would enrich the electronic
density at carbonyl oxygen to benefit such interaction and
therefore to improve the regioselectivity. Methyl amide 10 was
selected for development4 as the route would provide a
straightforward three-step synthesis as shown in Scheme 2.
Amide 10 was easily accessed from either commercially
available acid 25 or ester 3 (Scheme 3). Activation of 2 with
CDI in THF followed by reaction with aqueous methylamine
afforded 10 cleanly. A similar result was obtained through ester
aminolysis of 3 with aqueous methylamine in THF at ambient
temperature. The isolation of 10 from either approach was
straightforward. Concentration of the reaction mixture followed
by addition of water precipitated 10 as a white crystalline
powder in ∼90% yield. The approach employing acid 2 was
chosen for scale-up as the starting material itself is an easy-to-
handle solid. In finding that the bulk lot of 2 contained 5%
water, azeotropic distillation was performed at the outset to
remove water before CDI activation. Two batches of 10 (6.33
and 6.40 kg) with >98% purity were obtained in 85% and 86%
yields respectively.
Phenol 7 was prepared by either monomethylation of 2-
methylresorcinol (6) or monodemethylation of 2,6-dimethox-
ytoluene (11) as shown in Scheme 4. Neither method was
sufficiently chemoselective, as methylation provided 11 and
demethylation provided 6 as byproducts. Instead, conventional
purification methods were developed to isolate 7. By modifying
the reported monodemethylation method,3 7 was first prepared
on 100-g scale for the dose-range finding study. Thus,
treatment of 6 with dimethyl sulfate (1.1 equiv.) and aq
NaOH (2.4 equiv) at reflux afforded a 3:1:1 mixture of 7:6:11.
The overmethylated byproduct 11 was removed from the basic
aqueous reaction mixture by extraction into MTBE. The
Figure 1. Structure of 1.
promising lead compound.1 On the basis of its good in vitro/in
vivo neuropharmacological activities and preclinical pharmaco-
kinetics, 1 was predicted to offer greater efficacy for a broad
range of patients, as well increased safety and tolerability over
the current selective serotonin reuptake inhibitors (SSRI), such
as sertraline (Zoloft).2 Thus, we prepared 1 on 100-g scale for a
dose-range finding toxicity study, followed by multikilogram
manufacture to support regulatory toxicity and early clinical
studies.
Inhibitor 1 was originally synthesized in 15% overall yield by
medicinal chemists using the route shown in Scheme 1.
Aldehyde 5 was prepared from acid 2 in three steps via
esterification, reduction, and oxidation. Phenol 7 was obtained
through monomethylation of 2-methylresorcinol (6).3 Cou-
pling of 5 and 7 via nucleophilic aromatic substitution (SNAr)
formed a 4:1 mixture of desired ether 8 and 4-substituted side
product 9, which required chromatographic separation.
Reductive amination of 8 with methylamine gave free amine
1a, which was then converted to the target HCl salt (1). While
this route was suitable for analogue synthesis with various
amino side chains, poor regioselectivity in the SNAr reaction
was the major obstacle for our scale-up efforts. In addition, the
three-step preparation of aldehyde 5 was not practical with use
of the toxic oxidant PCC; intermediates 5, 7, 8, and 1a were all
purified by chromatography; and the final HCl salt formation
offered poor recovery, as a significant amount of 1 remained in
the mother liquor.
Received: August 15, 2012
Published: October 14, 2012
© 2012 American Chemical Society
1805
dx.doi.org/10.1021/op3002273 | Org. Process Res. Dev. 2012, 16, 1805−1810

Organic Process Research & Development
Article
Scheme 1. Original synthesis of 1
toxic and a volatile side product.6 The reaction was quenched
with water, and the DCM layer was isolated. Extraction of the
DCM phase with aq NaOH separated 7 and 6 from the side
products, 11 and methyl bromide.7 Acidification of the aqueous
layer to pH 2−3 with aq citric acid followed by DCM extraction
further upgraded 7 to ∼95% purity while leaving most of 6 in
the acidic aqueous layer. Crystallization from hexanes/DCM
isolated 7 (>98% purity) as a white powder. Two batches of 7
(3.1 and 3.3 kg) were manufactured in 57% and 61% yields,
respectively.
With both 7 and 10 in hand, we set out to screen the base
and solvent effects on regioselectivity for the key SNAr reaction
(Scheme 5). Unlike aldehyde 5, methyl amide 10 generated
almost exclusively the desired C-2 substituted product (12)
under all conditions tested (Table 1). With equimolar ratio of 7
and 10, the reaction reached completion in 18 h at 120 °C in
the presence of K2CO3 in DMF or NMP but required several
days for high conversion with t-BuOK at reflux in 2-methyl
THF, THF or toluene. The K2CO3/DMF conditions (entry 1)
were selected for scale-up, and two slightly different isolation
procedures were developed. On 100-g scale, we diluted the
mixture with 4 volumes of water and 1 volume of isopropyl
ether (both relative to DMF) to precipitate 12 as a 97% pure
tan granular solid in 63% yield. For multikilogram-scale
manufacture, adding 1 volume of water (relative to DMF)
resulted in precipitation of crude 12 in 78% yield. This isolated
crude contained 94.3% of 12 with two ∼1% unidentified
Figure 2. Proposed coordination state between the carbonyl oxygen
and K+.
aqueous layer was then acidified with HCl and extracted with
DCM. The organic extracts were concentrated and passed
through a pad of silica gel with a wash of 1:1 (v/v) DCM/
hexanes. This purification provided 7 as a 95% pure waxy solid,
which was adequate for the subsequent etherification, in 53%
yield.
Alternatively, the monodemethylation of 11 was developed
for further scale-up as this process allowed for more control
over the reaction end point and afforded pure 7 as a crystalline
solid. Dimethyl ether 11 was treated with BBr3 in DCM at −10
°C for 24 h, at which point >70% of 11 was consumed, and the
typical reaction profile was 6:1:2 to 6:1:1 of 7:6:11. An aqueous
ethanolamine scrubber was applied to trap methyl bromide, a
Scheme 2. Proposed process route
1806
dx.doi.org/10.1021/op3002273 | Org. Process Res. Dev. 2012, 16, 1805−1810

Organic Process Research & Development
Article
Scheme 3. Two approaches to amide 10
through recrystallization, salt formation, or chromatography.
Satisfactory results were obtained by reduction with two in situ
generated reducing reagents: sodium acetoxyborohydride from
NaBH4 and acetic acid,8 or diborane from BF3·OEt2 and
NaBH4.9 Both methods offered clean conversion to 1a without
formation of byproduct 13 in similar yields (∼80%) on lab
scales. The first method was used for the 100-g-scale
preparation with replacement of 1,4-dioxane with 2-methyl
THF. Thus, reaction of 12 with NaBH4 (5 equiv) and acetic
acid (5 equiv.) in 2-methyl THF at reflux for 20 h generated a
mixture of 1a and the corresponding boron−amine complex,
which was neutralized with aq NaOH and extracted into 2-
methyl THF. Treatment of the organic extracts with 2 N aq
HCl at reflux broke the complex and precipitated crude 1
(∼96% purity) in 83% yield.
Scheme 4. Two approaches to phenol 7
Scheme 5. Regioselective SNAr etherification to 12
Table 1. Base and solvent screen for the regioselective SNAr
The second method was developed for kilogram-scale
manufacture as it required shorter reaction time and used less
reducing agent. Slow addition of a solution of 12 and BF3·OEt2
(2.5 equiv) in THF to a 55 °C suspension of NaBH4 (1.9
equiv) in THF, followed by 3 h at reflux, cleanly converted 12
to 1a as a boron−amine complex. After an aqueous quench, the
complex was extracted into 2-methyl THF from the acidic
media. Breaking the complex with aq HCl at reflux offered 8.36
kg of crude 1 (96% purity) in 69% yield in the kilo lab.
Conducting a HCl salt break and reformation further
enhanced the quality of 1 to meet the API delivery criteria.10
Crude 1 was neutralized with aq NaOH and extracted into 2-
methyl THF. Addition of conc HCl (1.1 equiv) and azeotropic
distillation to remove water led to the recrystallization of 1 as
the desired polymorph (Form A).11 CE-157119 HCl salt (1,
7.36 kg) with 99.8% purity was obtained as a white powder in
88% yield.
etherification
base
entry (equiv)
solvent,
temp
reaction
time
conversion
isolated yield
(%) (HPLC)
of 12 (%)
1
2
3
4
5
6
K2CO3
(2.2)
DMF,
120 °C
NMP,
120 °C
DME,
85 °C
THF,
65 °C
2-MeTHF,
78 °C
12 h
18 h
24 h
6 d
96
88
40
82
92
93
63
NA
NA
NA
62
K2CO3
(2.2)
K2CO3
(2.2)
tBuOK
(1.1)
tBuOK
(1.1)
5 d
tBuOK
(1.1)
toluene,
110 °C
4 d
NA
impurities, which could be carried over to the API. A brief
solvent screen unfortunately indicated that isopropyl ether was
the solvent providing the best purging efficiency with good
recovery. In order to minimize the potential safety hazard
caused by peroxides, freshly made isopropyl ether (with BHT
as stabilizer) was purchased and used quickly upon receipt. This
additional reslurry operation in isopropyl ether integrated the
purity of 12 over 98% with 87% recovery. Two batches of 12
(5.5 and 5.8 kg) were therefore obtained in 63% and 67%
yields.
Our initial attempts to reduce amide 12 to free base 1a were
conducted with LiAlH4. Unfortunately, this reduction gen-
erated a ∼4:1 mixture of desired 1a and dechlorinated side
product 13 (Scheme 6), which were difficult to separate
CONCLUSIONS
■
An improved synthesis of CE-157119 HCl salt (1) was
achieved through the regioselective SNAr etherification between
phenol 7 and methyl amide 10, followed by amide reduction
with in situ-generated diborane (Scheme 7). This early
development route shortened the original five-step synthesis
to three steps, eliminated all chromatography, and increased the
overall yield from 15% to 34%. The process was implemented
for API manufacture from 100-g to multikilogram scale.
Scheme 6. Amide reduction to free base 1a
1807
dx.doi.org/10.1021/op3002273 | Org. Process Res. Dev. 2012, 16, 1805−1810

Organic Process Research & Development
Article
Scheme 7. Improved three-step synthesis of 1
citric acid (30 kg, 60 mol) and extracted with DCM (30 L).
The DCM layer was dried with MgSO4 (3 kg) and
concentrated under vacuum (30 °C, 0.1 bar) to ∼7 L of the
residual volume. Hexanes (60 L) were added, and the mixture
was seeded. The resulting slurry was stirred at 20 °C for 3 h and
at 4 °C for 17 h and then filtered. The cake was washed with
hexanes (10 L) and dried in a vacuum oven (20 °C) for 16 h to
obtain 7 (3.27 kg, 23.7 mol) with 98.7% purity as a white
EXPERIMENTAL SECTION
■
Proton NMR spectra were recorded at 400 MHz, and carbon
NMR spectra were recorded at 100 MHz. HPLC analyses were
carried out using a Zorbax SB-CN column (4.6 mm × 150 mm,
5 μm) at 30 °C with 0.2% aq H3PO4 (A) and MeCN (B) as
mobile phase in linear gradient from 95% of A to 5% of A over
15 min at 2 mL/min flow rate; HPLC purity was detected at
210 nm wavelength and reported by area %. All chemicals were
of commercial quality and used as received.
1
powder in 61% yield: mp 43.9−44.5 °C. H NMR (400 MHz,
CDCl3): δ 7.06 (t, J = 8.2 Hz, 1H), 6.52 (d, J = 8.2 Hz, 1H),
6.48 (d, J = 8.2 Hz, 1H), 4.65 (brs, 1H, exchanged with D2O),
3.85 (s, 3H), 2.16 (s, 3H). 13C NMR (100 MHz, CDCl3): δ
159.69, 154.43, 126.45, 112.15, 108.08, 103.12, 55.73, 7.98.
GC−MS: 138 (M+). Elemental Analysis: Calcd for C, 69.54; H,
7.30; Found C, 69.25; H, 7.65.
N-Methyl 2,5-difluoro-4-chloro-benzamide (10), from
2 on Kilogram Scale. 2,5-Difluoro-4-chloro-benzoic acid 2
(7.0 kg, 36.4 mol) was dissolved in THF (175 L), and 105 L of
THF was atmospherically distilled off.12 The residual solution
was cooled to 20 °C, and CDI (6.48 kg, 40 mol) was added.
The mixture was stirred at 20 °C for 2 h.13 Methylamine (11.3
kg, 145 mol, 40 wt % in water) was added, and the mixture was
stirred at 20 °C for 2 h. Forty-two liters of solvent were distilled
off under vacuum (<30 °C, 0.1 bar), and water (140 L) was
added. The resulting slurry was cooled to 2 °C, stirred for 2 h,
and filtered. The cake was washed with water (14 L) and dried
in a vacuum oven (35 °C) for 24 h to obtain 10 (6.33 kg, 30.8
mol) with 99.4% purity as a white powder in 85% yield: mp
127−128 °C. 1H NMR (400 MHz, CDCl3): δ 7.92 (dd, J = 9.2,
6.8 Hz, 1H), 7.24 (dd, J = 10.6, 5.6 Hz, 1H), 6.73 (brs, 1H),
3.05 (d, J = 4.8 Hz, 3H). 13C NMR (100 MHz, CDCl3): δ
161.84, 155.61 (dd, JFC = 246, 3 Hz), 154.85 (dd, JFC = 247, 2
Hz), 125.21 (dd, JFC = 21, 2 Hz), 121.05 (dd, JFC = 15, 6 Hz),
118.98 (dd, JFC = 26, 3 Hz), 118.31 (d, JFC = 31 Hz), 26.97.
GC−MS: 205.0 (M+). Elemental Analysis: Calcd for C, 46.74;
H, 2.94; N, 6.81; Found C, 46.94; H, 2.98; N, 6.72.
N-Methyl 2,5-difluoro-4-chloro-benzamide (10), from
3 on lab scale. Ethyl 2,5-difluoro-4-chloro-benzoate 3 (154 g,
0.70 mol) was dissolved in THF (250 mL). Water (250 mL)
was added, followed by methylamine (303 mL, 3.5 mol, 40 wt
% in water). The mixture was stirred at 20 °C for 2 h. Water
(500 mL) was added, and most of THF was distilled off under
vacuum (<30 °C, 0.05 bar). The resulting slurry was stirred at
20 °C for 1 h and filtered. The cake was washed with water
(500 mL) and dried in a vacuum oven (40 °C) for 16 h to
obtain 10 (132.6 g, 0.64 mol) as a white powder in 92% yield.
2-Methyl-3-methoxyphenol (7), through Monodeme-
thylation of 11 on Kilogram Scale. The solution of 2,6-
dimethoxytoluene 11 (6.0 kg, 39 mol) in DCM (60 L) was
cooled to −6 °C. A scrubber containing ethanolamine (4.34 kg,
71.1 mol) and water (17.4 L) was connected to the reaction
system. 1 M BBr3 in DCM (21.2 L, 21.2 mol) was added over 2
h, while the reaction temperature was maintained between −8
to −4 °C. The mixture was agitated at −6 °C for 24 h until 70%
of 11 was consumed. Water (60 L) was slowly added, while the
reaction temperature was maintained below 10 °C. The DCM
layer was separated and then extracted with 1 N aq NaOH (42
L). The aqueous solution was adjusted to pH 2−3 with 2 N aq
2-Methyl-3-methoxyphenol (7), through Monomethy-
lation of 6 on Lab Scale. A mixture of 2-methylresorcinol 6
(550 g, 4.43 mol) and NaOH (212 g, 5.30 mol) in water (2.65
L) was heated to 90 °C. Dimethyl sulfate (462 mL, 4.88 mol)
was added over a period of 45 min, followed by 10 wt % aq
NaOH (2.2 kg, 5.5 mol) over a period of 30 min, while the
reaction temperature was maintained between 90 to 100 °C.
The mixture was stirred at 100 °C for additional 30 min and
then cooled to 20 °C. The mixture was extracted with MTBE
(2 L). Conc HCl (800 mL) was added slowly to the aqueous
layer to adjust pH <2. The aqueous mixture was extracted with
DCM twice (2 L each) and the combined extracts were washed
with water (1 L) and dried over MgSO4. The DCM layer was
concentrated to ∼1 L of the residual volume, and the resulting
oil was passed through a pad of silica gel (1 kg, 16 cm D × 14
cm H) with a wash of 1:1 (v/v) hexanes/DCM (6 L). The
filtrates were concentrated to dryness under vacuum (<40 °C,
0.1 bar) to provide 7 (324 g, 2.35 mol) with 95.0% purity as an
off-white, waxy solid in 53% yield.
4-Chloro-5-fluoro-2-(3-methoxy-2-methyl-phenoxy)-
N-methyl-benzamide (12), Using DMF As Solvent on
Kilogram Scale. 2-Methyl-3-methoxyphenol 7 (3.70 kg, 26.7
mol), N-methyl 2,5-difluoro-4-chloro-benzamide 10 (5.54 kg,
26.9 mol), and K2CO3 powder (6.23 kg, 44.6 mol) were added
to DMF (46.5 L). The mixture was heated to 130 °C and
stirred for 12 h. The mixture was then cooled to 20 °C, and
water (46.5 L) was added slowly over 1 h. The resulting slurry
was stirred at 20 °C for 2 h and at 2 °C for 2 h and then
filtered. The cake was washed with water (15.5 L) and dried in
a vacuum oven (40 °C) for 32 h to obtain crude 12 (6.75 kg) as
a beige powder. This crude was added to isopropyl ether (26.8
L, CAUTION: isopropyl ether may contain explosive peroxides.
Use appropriate safety precautions). The slurry was stirred at
50 °C for 30 min, slowly cooled to 0 °C over 1 h, further stirred
at 0 °C for 2 h, and then filtered. The cake was washed with
isopropyl ether (6.7 L) and dried in a vacuum oven (30 °C) for
24 h to obtain 12 (5.45 kg, 16.8 mol) with 98.8% purity as an
1
off-white powder in 63% yield: mp 126.7−128.1 °C. H NMR
1808
dx.doi.org/10.1021/op3002273 | Org. Process Res. Dev. 2012, 16, 1805−1810

Organic Process Research & Development
Article
(400 MHz, CDCl3): δ 8.10 (d, J = 9.8 Hz, 1H), 7.81 (brs, 1H),
7.12−7.36 (m, 1H), 6.83 (d, J = 8.3 Hz, 1H), 6.49−6.74 (m,
2H), 3.92 (s, 3H), 3,04 (d, J = 4.8 Hz, 3H), 2.10 (s, 3H). 13C
benzamide 12 (11.3 kg, 34.6 mol) was dissolved in THF (33.9
L). BF3·OEt2 (12.3 kg, 86.4 mol) was added. The solution was
slowly transferred to a heated (55 °C) suspension of NaBH4
(2.45 kg, 64.8 mol) in THF (28.2 L) over 3 h while the
temperature of the receiving reactor was maintained between
50 to 60 °C. The mixture was stirred at 60 °C for 3 h and then
cooled to 5 °C. A 10:1 (w/w) mixture of water and THF (20.6
kg) was slowly added, while the temperature of the receiving
reactor was maintained between 5 to 20 °C. After the mixture
stirred at 20 °C for 2 h, water (45.2 L) was slowly added over
30 min, and the mixture was extracted with 2-methyl THF
(45.2 L). The organic layer was washed with water (56.5 L) and
brine (65 kg). MeOH (11.3 L) and conc HCl (6.81 kg, 69.1
mol) were then added, and the mixture was slowly heated to 60
°C over 2 h and stirred for 1 h. The mixture was concentrated
under vacuum (35 °C, 0.1 bar) to ∼25 L of the residual
volume, and 2-methyl THF (113 L) was added. The mixture
was concentrated atmospherically at 80 °C to ∼65 L of the
residual volume. The resulting slurry was cooled to 10 °C,
stirred for 2 h, and filtered. The cake was washed with 2-methyl
THF (11 L) and blown with nitrogen for 12 h to give crude 1
(8.36 kg, 24.1 mol) with 96% purity as a white powder in 69%
yield.
Purification of [4-Chloro-5-fluoro-2-(3-methoxy-2-
methyl-phenoxy)-benzyl]methylamine Hydrochloride
(1). A solution of 0.67 N aq NaOH (40 L, 26.3 mol) was
added to a slurry of crude [4-chloro-5-fluoro-2-(3-methoxy-2-
methyl-phenoxy)-benzyl]methylamine hydrochloride 1 (8.36
kg, 24.1 mol) in 2-methyl THF (80 L). The mixture was stirred
for 30 min until all solids dissolved and the organic layer was
separated. The organic layer was washed with water twice (40 L
each) and filtered through a 0.5 μm filter to a speck-free reactor
with a rinse of 2-methyl THF (5 L). Conc HCl (2.77 kg, 28.1
mol) was added, and the mixture was concentrated under
vacuum (35 °C, 0.1 bar) to ∼18 L of the residual volume.
Additional 2-methyl THF (50 L) was added, and the mixture
was concentrated atmospherically at 80 °C to ∼42 L of the
residual volume. The slurry was slowly cooled to 5 °C over 1 h,
stirred for 2 h, and filtered. The cake was washed with 2-methyl
THF (8 L) and dried in a vacuum dryer (40 °C) for 20 h to
obtain 1 (7.36 kg, 21.3 mol) with 99.8% purity as a white
powder (Form A) in 88% yield: mp 186.5−188.8 °C. 1H NMR
(400 MHz, CD3OD): δ 7.44 (d, J = 9.1 Hz, 1H), 7.23 (t, J = 8.3
Hz, 1H), 6.88 (d, J = 8.3 Hz, 1H), 6.49−6.73 (m, 2H), 4.32 (s,
2H), 3.86 (s, 3H), 2.77 (s, 3H), 2.03 (s, 3H). 13C NMR (100
MHz, CD3OD): δ 160.93, 154.78, 154.67 (d, JFC = 3 Hz),
154.62 (d, JFC = 244 Hz), 128.87, 124.14 (d, JFC = 19 Hz),
121.84 (d, JFC = 7 Hz), 120.70 (d, JFC = 24 Hz), 120.08, 118.34,
113.94, 108.96, 56.53, 47.90, 33.87, 9.22. LC−MS: 310.4 (ES+).
Elemental Analysis: Calcd for C, 55.51; H, 5.24; N, 4.05; Cl,
20.48; F, 5.49; Found C, 55.49; H, 5.13; N, 3.99; Cl, 20.33; F,
5.49.
NMR (100 MHz, CDCl3): δ 163.63, 159.37, 153.56 (d, JFC
=
244 Hz), 152.73, 152.07 (d, JFC = 3 Hz), 127.54, 124.73 (d, JFC
= 20 Hz), 122.25 (d, JFC = 6 Hz), 119.38 (d, JFC = 23 Hz),
117.72, 113.06, 107.80, 55.86, 26.87, 8.99. LC−MS: 324.4
(ES+). Elemental Analysis: Calcd for C, 59.36; H, 4.67; N, 4.33;
Found C, 59.62; H, 4.53; N, 4.23.
4-Chloro-5-fluoro-2-(3-methoxy-2-methyl-phenoxy)-
N-methyl-benzamide (12), Using DMF As Solvent on
Lab Scale. 2-Methyl-3-methoxyphenol 7 (12.42 g, 89.9
mmol), N-methyl 2,5-difluoro-4-chloro-benzamide 10 (18.48
g, 89.9 mmol), and K2CO3 powder (27.33 g, 197.7 mmol, <325
mesh) were added to DMF (170 mL). The slurry was stirred at
120 °C for 12 h and then cooled to 20 °C. Water (680 mL) was
added slowly, followed by isopropyl ether (150 mL). The
resulting slurry was stirred at 20 °C for 2 h and filtered. The
cake was washed with water (100 mL) and isopropyl ether (50
mL), and dried in a vacuum oven (40 °C) for 16 h to obtain 12
(18.18 g, 56.1 mmol) as a light-tan granule in 63% yield.
4-Chloro-5-fluoro-2-(3-methoxy-2-methyl-phenoxy)-
N-methyl-benzamide (12), Using 2-Methyl THF As
Solvent. 2-Methyl-3-methoxyphenol 7 (242 g, 1.75 mol), N-
methyl 2,5-difluoro-4-chloro-benzamide 10 (343 g, 1.67 mol)
and t-BuOK (196 g, 1.75 mol) were added to 2-methyl THF
(3.4 L). The suspension was refluxed for 5 d. The mixture was
cooled to 20 °C. Water (1.5 L), brine (1.5 L), and 1 N aq HCl
(900 mL) were added. The organic layer was separated, and the
aqueous layer was extracted twice with 2-methyl THF (1 L
each). The combined organic extracts were washed with brine
(1.5 L), dried with MgSO4, and concentrated to dryness.
Isopropyl ether (1.8 L) was added, and the mixture was stirred
at reflux for 1 h and cooled to 5 °C slowly over 1 h. The slurry
was further stirred at 5 °C for 2 h and filtered. The cake was
washed with isopropyl ether (500 mL) and dried in a vacuum
oven (40 °C) for 16 h to provide 12 (333.2 g, 1.03 mol) with
98% purity as an off-white powder in 62% yield.
[4-Chloro-5-fluoro-2-(3-methoxy-2-methyl-phenoxy)-
benzyl]methylamine hydrochloride (1), through Reduc-
tion with Sodium Acetoxyborohydride Generated in
Situ on Lab Scale. NaBH4 (58.2 g, 1.54 mol) was added to a
solution of 4-chloro-5-fluoro-2-(3-methoxy-2-methyl-phenoxy)-
N-methyl-benzamide 12 (100 g, 0.31 mol) in 2-methyl THF
(700 mL). The suspension was cooled to 2 °C. A solution of
acetic acid (88 mL, 1.5 mol) in 2-methyl THF (300 mL) was
added over a period of 1 h. The mixture was heated to reflux
and stirred for 22 h. The mixture was then cooled to 10 °C and
slowly poured into stirred ice/water (1.4 L); then 50 wt % aq
NaOH (120 g, 1.5 mol) was slowly added. The organic layer
was separated, and the aqueous layer was extracted with 2-
methyl THF (1 L). The combined organic extracts were
washed with brine (1 L) and concentrated to ∼300 mL of the
residual volume. 2 N aq HCl (1 L) was added, and the mixture
was refluxed for 2 h. The resulting slurry was cooled to 20 °C
and filtered. The cake was washed with water (200 mL) and 2-
methyl THF twice (200 mL each) and dried in a vacuum oven
(40 °C) for 24 h to obtain 1 (88.6 g, 0.256 mol) with 95.8%
purity as a white powder in 83% yield.
AUTHOR INFORMATION
■
Corresponding Author
*yong.tao@pfizer.com
Notes
The authors declare no competing financial interest.
[4-Chloro-5-fluoro-2-(3-methoxy-2-methyl-phenoxy)-
benzyl]methylamine Hydrochloride (1), through Reduc-
tion with Diborane Generated in Situ on Kilogram Scale.
4-Chloro-5-fluoro-2-(3-methoxy-2-methyl-phenoxy)-N-methyl-
ACKNOWLEDGMENTS
■
The authors thank the Groton Kilo Lab staffs for their
successful execution of the process; Mr. Jack King and Dr.
1809
dx.doi.org/10.1021/op3002273 | Org. Process Res. Dev. 2012, 16, 1805−1810

Organic Process Research & Development
Article
Chris Foti for their efficient analytical supports; and Dr. Bryan
Li, Dr. Joshua Dunetz and Dr. Jeffrey Sperry for their valuable
inputs to the manuscript.
REFERENCES
■
(1) Claffey, M. M.; Fliri, A. F. J.; Gallaschun, R. J.; O’Donnell, C. J.
PCT Int. Appl. 2006, WO 2006027684. Chem. Abstr. 144:292415.
(2) Selected reviews: (a) McRae, A. L.; Brady, K. T. Expert Opinion
on Pharmacotherapy 2001, 2, 883. (b) Owen, R. T. Drugs of Today
1992, 28, 439. (c) Sheehan, D. V.; Kamijima, K. Int. Clin.
Psychopharmacology 2009, 24, 43. (d) Dorgan, D. P.; Caillard, V. J.
Clin. Psychiatry 1988, 49 (Suppl.), 46. (e) Hirschfeld, R. M. Depression
and Anxiety 2000, 11, 139. Synthesis: (f) Welch, W. M. Adv. Med.
Chem. 1995, 3, 113. (g) Quallich, G. J. Chirality 2005, 17 (Suppl.),
S120. (h) Quallich, G. J. Eur. Pat. Appl. 2000, EP 1059287. Chem.
Abstr. 134:41980.
(3) Kirkemo, C. L.; White, J. D. J. Org. Chem. 1985, 50, 1316.
(4) Tao, Y.; Widlicka, D. PCT Int. Appl. 2007, WO 2007010350.
Chem. Abstr. 146:184243.
(5) Preparation of acid 2: (a) Qian, X.; Chen, W.; Cao, S.; Tang, X.
Faming Zhuanli Shenqing Gongkai Shuomingshu 2009, CN 101381301
A 20090311. (b) Zhao, H. Huaxue Shijie 2011, 52, 30.
(6) Hettenbach, K.; Am Ende, D. J.; Leeman, K.; Dias, E.;
Kasthurikrishnan, N.; Brenek, S. J.; Ahlijanian, P. Org. Process Res.
Dev. 2002, 6, 407.
(7) Adding ethanolamine solution in the scrubber to the organic layer
converted the volatile methyl bromide to mono- or dimethylethanol-
amine.
(8) (a) Umino, N.; Iwakuma, T.; Itoh, N. Tetrahedron Lett. 1976,
763. (b) Furniss, B.; Hanaford, A,; Smith, P.; Tatchell, A. Vogel’s
Textbook of Practical Organic Chemistry, 5th Edition, p774.
(9) (a) Kanth, J. V. B.; Brown, H. C. Inorg. Chem. 2000, 39 (8), 1795.
(b) Sengupta, S.; Sahu, D. P.; Chatterjee, Sunil K. Indian J. Chem., Sect.
B: Org. Chem. Incl. Med. Chem. 1994, 33B (3), 285.
(10) At the stage of early development API should have >98% HPLC
purity with no individual unidentified impurity over 0.2%.
(11) Form A is the only polymorph of CE-157119 HCl salt (1)
obtained up to date.
(12) After azeotropic distillation, the solution contained <0.2% of
water.
(13) The completion of acid activation was monitored by HPLC
through derivatization with benzylamine. The reaction reached to
>98% conversion.
1810
dx.doi.org/10.1021/op3002273 | Org. Process Res. Dev. 2012, 16, 1805−1810