Organic Process Research & Development
ARTICLE
4-[(2-Amino-ethyl)-(5-chlorobenzoxazol-2-yl)amino]butan-
2-one-bis-MSA Salt (12). MVK adduct 10 (15.04 kg, 39.38 mol)
was charged to a vessel, and THF (150.4 L) was added. The re-
sulting slurry was cooled to 5 °C, and methane sulfonic acid
(7.57 kg, 78.8 mol) was added via a diaphragm pump maintaining
an internal temperature below 25 °C. The batch was heated
to 60 °C and allowed to age overnight. After aging overnight
HPLC showed the reaction to be 83% complete; thus, a further
charge of methane sulfonic acid (3.78 kg, 39.3 mol) was added via
a diaphragm pump, and heating was continued for a further 2 h. It
is important to note that charging 3 equiv of MSA directly to the
reaction mixture led to the product oiling out during the course of the
reaction. Initially charging 2 equiv of MSA allowed a seed bed to form
which controlled the crystallization during the remainder of the
reaction after the final equivalent of MSA had been added. After 2 h
HPLC confirmed complete consumption of the starting material.
The batch was cooled to 20 °C and then filtered, and the cake was
washed with THF (20 L). The resultant solid was dried in a
vacuum oven (T = 60 °C) for 36 h, to give 12 (17.4 kg, 36.7 mol,
94%) as a pale, tan solid. Mp 126.4 °C; 2% was lost to the liquors.
1H NMR (400 MHz, d6-DMSO): δ 7.86 (bs, 3H), 7.44 (d, 1H,
J = 8.5 Hz), 7.35 (d, 1H, J = 1.1 Hz), 7.06 (dd, 1H, J = 8.5 Hz, 1.1
Hz), 3.75 (t, 2H, J = 5.6 Hz), 3.69 (t, 2H, J = 6.9 Hz), 3.15 (m,
2H), 2.93 (t, 2H, J = 5.6 Hz), 2.36 (s, 6H), 2.14 (s, 3H). 13C
NMR (100.6 MHz, d6-DMSO): δ 207.8, 163.1, 147.7, 144.7,
128.6, 120.6, 115.7, 110.6, 46.7, 44.2, 41.3, 40.1 (identified by
dept), 37.4, 30.5. HRMS (ESI): m/z [Mþ þ H] calcd for
C13H16ClN3O2: 282.1009; found: 282.1012.
5-Chloro-2-(5-methyl-[1,4]diazepan-1-yl)-benzoxazole (rac-
11). Bis-MSA adduct 12 (17.19 kg, 36.27 mol), sodium acetate
(2.97 kg, 36.3 mol), and DCM (152 kg, 115 L) were charged to a
vessel. The resulting slurry was cooled to 15 °C ,and acetic acid
(26.8 kg, this was the amount required for dissolution of the salt)
was added, maintaining an internal temperature below 20 °C. The
batch was cooled to 15 °C, and Na(OAc)3BH (9.25 kg, 43.6 mol)
was charged via a glovebag over 30 min, maintaining an internal
temperature below 20 °C. The resulting solution was aged for 30
min at 20 °C; HPLC analysis showed the reaction to be complete.
The batch was then cooled to 10 °C, 2 N HCl (38.80 kg) was
added, andthe resultingsolution was agedfor30minat10°C. The
mixture was then adjusted to pH 9 using 5 N NaOH, maintaining
an internal temperature below 20 °C. Once at pH 9 the layers were
allowed to separate, and the organic lower layer was run off into a
clean drum. DCM (76.2kg, 57.5 L) was then added to the aqueous
phase; the layers were mixed and allowed to separate, and then the
lower layer was combinedwith the previous organics(this step was
repeated, but HPLC of the third DCM layer showed that this was
unnecessary as it only contained 12.5 g of rac-11 by assay). The
combined DCM fractions were then charged to a 160-L vessel and
were distilled to approximately 20 L (∼1 volume of DCM). THF
(44 kg) was charged to the vessel, and the resulting stream was
discharged into a clean drum and stored in the cold room prior to
the next step in the reaction sequence. The final solution was
found to contain 9.4 kg of rac-14 (98% yield, assay by weight).
This stream was sufficiently pure to be used directly in the
subsequent resolution.
(crystallization had already started), and the batch was aged at
20 °C for 4 h, being analyzed over time by HPLC. The batch
showed 72% ee and 35.8 mg/mL liquor loss after 3.5 h. The slurry
was filtered (filtration from the vessel took 3 h) and the vessel
washed with THF (25 kg), filtering and washing the solid (this
filtration took a further 2 h). The solid in the filter was dried with
a stream of nitrogen overnight, and then dried in a vacuum oven
(T = 40 °C) for 48 h. The DBT salt 16 (8.47 kg, 13.6 mol, 38.5%,
74% ee, 99 LCWP based on amine) was isolated as a white solid.
The liquors after filtration showed 53% ee.
Upgrade of the DBT salt 16. The DBT salt 16 (8.35 kg, 13.4
mol, 74% ee) was charged to a vessel. iPAc (137 kg) was added
over 10 min and the suspension stirred. MeOH (52 kg) was
added over 5 min (slight endotherm), and the batch was aged at
20 °C for 22 h, analyzed over time by HPLC. The batch showed
95% ee and 5.3 mg/mL liquor loss after 22 h. The slurry was filtered
and the vessel washed with iPAc (29 kg), filtering and washing
the solid (total filtration took 1.5 h). The solid was dried in a
vacuum oven (T = 40 °C) for 20 h. The DBT salt 16 (5.85 kg,
9.37 mol, 70%, 96% ee, 95 LCWP based on amine) was isolated
as a white solid. Mp 164.2 °C. The liquors after filtration were
14% ee. 1H NMR (400 MHz, d6-DMSO): δ 7.93 (d, 4H, J = 7.2
Hz), 7.62 (t, 2H, J = 7.4 Hz), 7.51-7.47 (m, 4H), (d, 1H, J = 8.0
Hz), 7.34 (d, 1H, J = 2.0 Hz), 7.03 (dd, 1H, J = 8.4, 2.0 Hz), 5.65
(s, 2H), 3.89-3.76 (m, 2H), 3.67-3.58 (m, 1H), 3.43-3.27 (m,
2H), 3.23-3.16 (m, 2H), 1.99-1.84 (m, 2H), 1.19-1.17 (d,
3H, J = 6.6 Hz). 13C NMR (100 MHz, d6-DMSO) δ 168.5, 165.0,
162.4, 147.3, 144.8, 133.4, 129.7, 129.3, 128.7, 128.1, 119.8,
115.3, 109.9, 73.1, 53.5, 44.5, 44.0, 43.8, 31.9, 19.0. HRMS (ESI):
m/z [Mþ þ H] calcd for C13H16ClN3O: 266.1060; found:
266.1058.
5-Methyl-2-[1,2,3]triazol-2-yl-benzoic Acid 5. The iodide 19
(6.04 kg, 23.0 mol), THF (45 L), and DMF (9.0 L) were charged
to a vessel. Copper iodide (218 g, 1.15 mol) and potassium
carbonate (7.94 kg, 57.4 mol) were added, and the mixture was
heated to an internal temperature of 40 °C. 1,2,3-Triazole (3.16
kg, 4.60 mol) was added as a solution in THF (6.0 L) over half an
hour (no exotherm), and heating continued to 65 °C (again no
exotherm observed), and the reaction was monitored by HPLC.
Once complete N,N-dimethylethylenediamine (244 mL, 2.30
mol) was added and mixture cooled to RT. Aqueous 3.6 M HCl
(36 L) was added (exotherm) and the mixture extracted twice
with ethyl acetate (2 ꢀ 30 L). The combined organics were
washed with LiCl solution (2 ꢀ 20 L). The acid solution assayed
for 3.79 kg of 5 (81%) and 4.64 kg of 5 and 20 combined (99%).
A solution of acids 5 and 20 (approximately 4.64 kg, 22.9 mol)
in THF and EtOAc (approximately 110 L) was concentrated to
low volume. THF (90 L) was added, and the solvent composition
was checked by 1H NMR to ensure most ethyl acetate had been
removed. Sodium tert-butoxide (2.42 kg, 25.2 mol) was added
slowly as a solid over 1-2 h (slight exotherm), allowing the
sodium salt to form, and was stirred overnight at RT. The liquors
showed a 45:55 ratio of 5:20, and the solid was collected by
filtration, washed with THF (2 ꢀ 20 L), and dried in a vacuum
oven (T = 40 °C) for 15 h to afford 4.22 kg of crude sodium salt
(72 wt %, corrected to 3.04 kg free acid equivalent, 65% from 19,
95.8 LCAP).
DBT Salt 16. Dibenzoyl-D-tartaric acid 15 (22.6 kg, 63.1 mol)
was charged to a vessel. THF (80 kg) was added over 20 min and
the suspension stirred (slight exotherm). The amine (9.37 kg,
crude from previous step in DCM [∼1 vol] and THF [44 kg,
50 L] (total weight 65.5 kg)) was added over 35 min (slight exo-
therm). The solution was seeded with enantiomerically pure 16
The crude sodium salt (4.22 kg, 14.9 mol) was charged to a
50-L vessel, and 3.6 N HCl (21.2 L) was added with cooling. The
slurry was then stirred at room temperature for 16 h and the off-
white solid isolated by filtration. The cake was washed with water
(11 L) and iPAc/heptane (2 ꢀ 5 L), and then was dried in a
374
dx.doi.org/10.1021/op1002853 |Org. Process Res. Dev. 2011, 15, 367–375