Process Research for (+)-Ambrisentan, an Endothelin-A Receptor Antagonist

Article abstract of DOI:10.1021/acs.oprd.8b00184

An efficient and robust synthetic route to (+)-ambrisentan ((+)-AMB) was designed by recycling the unwanted isomer from the resolution mother liquors. The racemization of AMB in the absence of either acid or base in the given solvents was reported. The recovery process was developed to produce racemates with purities over 99.5%. The mechanism of the formation of the process-related impurities of (+)-AMB is also discussed in detail. (+)-AMB was obtained in 47% overall yield with >99.5% purity and 99.8% e.e. by chiral resolution with only one recycling of the mother liquors on a 100-g scale without column purification.

Full text of DOI:10.1021/acs.oprd.8b00184

Subscriber access provided by UNIV OF DURHAM
Full Paper
Process Research for (+)-Ambrisentan, an Endothelin-A Receptor Antagonist
weidong feng, songming zhuo, jun yu, chuanmeng zhao, and Fu-Li Zhang
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.8b00184 • Publication Date (Web): 06 Aug 2018
Downloaded from http://pubs.acs.org on August 6, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 34
Organic Process Research & Development
1
2
3
4
5
6
Process Research for (+)-Ambrisentan, an Endothelin-A
Receptor Antagonist
7
8
9
Wei-Dong Feng†, Song-Ming Zhuo‡, Jun Yu§, Chuan-Meng Zhao§, Fu-Li Zhang†§*
†Collaborative Innovation Center of Yangze River Delta Region Green
Pharmaceuticals, Zhejiang University of Technology, 18 Chaowang Road, Hangzhou
310014, China
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
‡Department of Pharmaceutial Engineering, China Pharmaceutical University, 24
Tongjiaxiang, Nanjing 210009, China
§Shanghai Institute of Pharmaceutical Industry, China State Institute of Pharmacetical
Industry, 285 Gebaini Road, Pudong, Shanghai 201203, China
*Eꢀmail: zhangfuli1@sinopharm.com
1
ACS Paragon Plus Environment

Organic Process Research & Development
Page 2 of 34
1
2
3
4
TOC graphic
5
6
7
8
Green, Efficient and Robust Process for (+)-Ambrisentan
NH2
9
NH3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
COOH
COO
O
COOH
N
N
N
N
MeO
MeO
MeO
O
O
90%
40%
N
N
purity>99.5%
(+)-Ambrisentan
ICH grade quality
purity>99.5%
ee>99.8%
90%
Recycle
Mother Liquors
2
ACS Paragon Plus Environment

Page 3 of 34
Organic Process Research & Development
1
2
3
4
ABSTRACT
5
6
7
8
An efficient and robust synthetic route to (+)ꢀambrisentan ((+)ꢀAMB)
was designed by recycling the unwanted isomer from the resolution
mother liquors. The racemization of AMB in the absence of either acid or
base in the given solvents was reported. The recovery process was
developed to produce racemates with purities over 99.5%. The
mechanism of the formation of the processꢀrelated impurities of (+)ꢀAMB
is also discussed in detail. (+)ꢀAMB was obtained in 47% overall yield
with >99.5% purity and 99.8% e.e. by chiral resolution with only one
recycling of the mother liquors on a 100ꢀg scale without column
purification.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
KEYWORDS Process, Recovery, Ambrisentan, Pulmonary arterial
hypertension
INTRODUCTION
Pulmonary arterial hypertension (PAH) is a devastating chronic disease
characterized by vasoconstriction, remodeling of the pulmonary arteries,
progressive increases in pulmonary vascular resistance, and thrombosis.1
The median life expectancy from the time of diagnosis in patients with
PAH without targeted treatments is 2.8 years.2 There are three targets that
are used by drugs such as endothelin receptor antagonists,
phosphodiesterase typeꢀ5 inhibitors, and prostacyclin derivatives to
control the disease.3
3
ACS Paragon Plus Environment

Organic Process Research & Development
Page 4 of 34
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(+)ꢀAMB, which has the brand name Letairis®, is a selective
endothelinꢀA (ETA) receptor antagonist and was approved by the FDA in
June 2007 for the treatment of PAH. It has far fewer side effects than
bosentan, a nonselective antagonist for type A (ETA) and type B (ETB)
receptors.4
Scheme 1. General synthetic route to (+)-AMB
The general route5 for the synthesis of (+)ꢀAMB, as shown in Scheme
1, involves the formation of epoxide 3 via the Darzens reaction with
benzophenone and methyl chloroacetate in the presence of NaOMe in
THF and subsequent acidꢀcatalyzed ringꢀopening with MeOH to produce
ester 4. The hydrolysis of ester 4 to carboxylic acid 5 followed by chiral
resolution with various amines yields optically active carboxylic acid 6,
which undergoes nucleophilic substitution with pyrimidine 7 to give final
4
ACS Paragon Plus Environment

Page 5 of 34
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
product 1. The selection of the optical resolution agent has been studied a
number of times in the past few years. The use of methyl Lꢀprolinate and
(S)ꢀ1ꢀ(4ꢀnitrophenyl)ethanꢀ1ꢀamine as optical resolution agents on a
laboratory scale has been reported, and a yield of 35% based on the
racemates and an optical purity of 99.8% were achieved.6 However,
when the described reaction step is scaled up (several kg to 100 kg),
additional workꢀup procedures become necessary to ensure a high optical
purity. The diastereomeric salt of (S)ꢀ2ꢀhydroxypropionic acid 6 and
(S)ꢀ1ꢀ(4ꢀnitrophenyl)ethanꢀ1ꢀamine is difficult to crystallize in the
scaleꢀup process, and it cannot been smoothly removed by filtration. As a
result, some of the mother liquor containing the undesired enantiomer
remains in the crystals. Only when the crystals are subsequently stirred in
a tank together with fresh solvent and are filtered to give a fresh solid,
which is then rewashed several times with solvent, is the required optical
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
purity
obtained.
The
chiral
separation
of
5
using
(S)ꢀ1ꢀ(4ꢀchlorophenyl)ethanꢀ1ꢀamine as the optically active base
successfully avoids the above problem;7 however, the amine is a chemical
byꢀproduct and has not been widely commercialized. Furthermore, the
maximum yield can only be 50% from 5 to 6 via chiral resolution, and
discarding the isomer of 6 as waste is not environmentally friendly.
Methodologies for the preparation (+)ꢀAMB via asymmetric
techniques, such as asymmetric epoxidation,8 asymmetric Darzens
5
ACS Paragon Plus Environment

Organic Process Research & Development
Page 6 of 34
1
2
3
4
5
6
7
8
9
reaction,9 and asymmetric reduction,10 have been reported. However,
these asymmetric methods have many drawbacks, such as (a) the starting
material or the chiral catalyst not being available commercially, (b) the
synthetic route being too long, and (c) the enantioselectivity not being
sufficiently high. Moreover, even though it only accounts for 1% of the
final product, (ꢀ)ꢀAMB is not easily purged by recrystallization because
the solubility of (±)ꢀAMB is very poor in alcoholic solvents. Thus, none
of the above approaches are suitable for manufacturing scaleꢀup.
Developing an economical, efficient, practical and robust process route
for commercialization is highly desirable.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RESULT AND DISCUSSION
We herein report an improved synthetic route for the preparation of
(+)ꢀAMB (1), as shown in Scheme 2.
Scheme 2. Improved synthetic route to (+)-AMBa
6
ACS Paragon Plus Environment

Page 7 of 34
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
aReagents and conditions: (a) NaOMe, ClCH2COOMe, THF, ꢀ10~ꢀ5 °C, 3 h, 95%; (b)
conc. HCl, MeOH, 25~30 °C, 3 h, 90%; (c) K2CO3, DMF, 85 °C, 8 h, 94%; (d) KOH,
H2O, dioxane, 80 °C, 4 h, 96%; (e) Sꢀαꢀphenylethylamine, recrystallized with IPA and
MTBE, 40%; (f) (1) HCl, H2O, IPA; (2) crystallized with IPA and H2O, 90%; (g) HCl,
H2O, IPA; (h) MeCN, reflux, 3 h, 90%.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Preparation of methyl 3,3-diphenyloxirane-2-carboxylate (3). The
Darzens condensation was carried out with excess ClCH2COOMe and
NaOMe at 0~5 °C.5b Under the above conditions, approximately 14.9%
benzophenone remained by HPLC because the competing Williamson
reaction preferentially occurred between ClCH2COOMe and NaOMe at
0~5 °C, as shown in Table 1 (entry 3). The temperature had to be
appropriately controlled to avoid the Williamson reaction as much as
possible. The temperature screening experiments indicated that the
reaction proceeded much faster at ꢀ10~ꢀ5 °C, as shown in Table 1 (entry
2). However, when the reaction was run at lower temperatures even with
an extended reaction time, the conversion rate did not increase
significantly, as shown in Table 1 (entry 1). Since the reaction requires
excess ClCH2COOMe and NaOMe, the stoichiometry of the reagents was
also investigated, and the results are summarized in Table 2. The best
result was obtained with 1.7 eq of both ClCH2COOMe and NaOMe
(Table 2, entry 3). If the amounts of ClCH2COOMe and NaOMe were
7
ACS Paragon Plus Environment

Organic Process Research & Development
Page 8 of 34
1
2
3
4
5
6
7
8
9
further increased, the reaction did not reach completion, and 3.2% of the
starting material remained, as shown in Table 2 (entry 4).
Table 1. Screening of the reaction temperature for the Darzens
reaction
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
HPLC (%)b
entrya
T (°C) T (h)
ꢀ20~ꢀ15 6
2
3
1
2
3
7.2
3.2
14.9
92.8
96.8
85.1
ꢀ10~ꢀ5
0~5
3
3
aReaction conditions: ClCH2COOMe (2.0 eq), NaOMe (2.0 eq). bOnly peak areas (%)
of 2 and 3 were integrated on HPLC.
Table 2. Screening of reaction stoichiometry for the Darzens reaction
HPLC (%)b
entrya
eq
t (h)
6
2
3
1
2
3
1.3
1.5
1.7
25.2
15.9
3.5
74.8
84.1
96.5
3
3
8
ACS Paragon Plus Environment

Page 9 of 34
Organic Process Research & Development
1
2
3
4
5
4
2
3
3.2
96.8
aReaction temperature: ꢀ10 °C~ꢀ5 °C. bOnly peak areas (%) of 2 and 3 were integrated
6
7
on HPLC.
8
9
The workꢀup procedure involved dilution of the reaction mixture with
MTBE followed by cold water washes to neutralize the pH to minimize
the hydrolysis of epoxide 3, and this was followed by concentration of the
mixture and a solvent swap with methanol. The minor quantities of 2 and
the methoxymethylacetate impurity were removed in the next step. The
hydrolytic impurity of 3 was detected by LCꢀMS to be less than 1.0%,
and it can be removed by washing with water. The organic phase was
concentrated and exchanged with MeOH under reduced pressure. The
weight content of 3 in MeOH was determined by HPLC analysis using
the external standard method to obtain an accurate yield.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Preparation of methyl 2-hydroxy-3-methoxy-3,3-diphenylpropanoate
(4). The ring opening of epoxides using nucleophiles under acid catalysis
is well known. Intermediate 4 was obtained from epoxide 3 in the
presence of an acid catalyst and MeOH due to the formation of a stable
diphenylmethylium species, and MeOH not only served as the solvent but
also as the nucleophilic reagent. The ringꢀopening of epoxide 3 was
catalyzed with either a Lewis or Brønsted acid, as described in the
literature.8,11 Catalysts such as BF3, conc. HCl, and H2SO4 were screened
in our study, and the results shown in Table 3 indicated that the reaction
9
ACS Paragon Plus Environment

Organic Process Research & Development
Page 10 of 34
1
2
3
4
5
6
7
8
9
proceeded with high conversion in all cases. Since the quenching of BF3
required the addition of water, which produced highly toxic HF smog, it
was unsuitable for scaleꢀup at a plant. Reactions of pꢀTsOH or MeSO2OH
with methanol would generate pꢀTsOMe or MeSO2OMe, which are
genotoxic to humans.12 Similarly, the potential esterification of MeOH
with H2SO4 produces Me2SO4, a highly toxic carcinogen.13 HCl was
chosen as the safe acid catalyst for largeꢀscale commercial processes
because unlike H2SO4 and HNO3, it is not an oxidant.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 3. Preparation of 4 with various acid catalysts
entry
acid
T/t (°C/h) Yield (%)a
1
2
3
4
5
6
BF3
50/4
35/3
86
88
91
90
86
89
TsOH
MeSO2OH 30/2
conc. HCl
HNO3
30/3
25/3
25/2
H2SO4
aIsolated yield of 4.
The hydrolytic impurity of 4 was detected on LCꢀMS and HPLC and
accounted for approximately 2.0% of the material. The ringꢀopening
impurities resulting from Clꢀ and H2O as nucleophilic reagents were also
detected by LCꢀMS to be less than 1.0%. These impurities were
10
ACS Paragon Plus Environment

Page 11 of 34
Organic Process Research & Development
1
2
3
4
5
6
successfully removed along with those carried over from the previous
step by recrystallization in MeOH.
7
8
9
Preparation
of
methyl
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2-((4,6-dimethylpyrimidin-2-yl)oxy)-3-methoxy-3,3-diphenylpropano
ate (8). The coupling reaction of 4 and 7 in the presence of a base at
appropriate temperatures afforded ether 8. Initially, we followed the
reaction conditions for the preparation of 1 as shown in Scheme 1, and a
strong base, such as NaH or NaNH2, was used in the literature.14 The
hydrolytic byproducts of 4 and 8 were detected at approximately 2.3%
and 1.2% on HPLC, respectively, because of the presence of a small
amount of water from the solvent and atmosphere in the reaction mixture.
The byproducts from the elimination of MeOH and the CꢀO cleavage of
the pyrimidyl ether moiety of 8 were detected by LCꢀMS. Furthermore,
the reaction mixtures of NaH or NaNH2 and DMF have potential risks of
explosion in pilot plants,15 and therefore, various alternative bases and
solvents were screened for this reaction. If an inorganic base is used as
the deprotonation reagent, aprotic solvents are preferable over protic
solvents. We have attempted this coupling reaction using solvents such as
MeCN, THF, and DMF, as summarized in Table 4. When THF or toluene
was used, the formation of a highly viscous mass hampered the reaction
and caused incomplete conversion of the starting material (entries 2 and
3). The coupling reaction performed well using MeCN as the solvent but
11
ACS Paragon Plus Environment

Organic Process Research & Development
Page 12 of 34
1
2
3
4
5
6
7
8
9
required a large quantity of MeCN. The stirring efficiency of the reaction
mixture was improved, and the yield sharply increased when DMF was
used as the solvent (entry 5). Experimental results suggested that the
reaction conversion was considerably higher with solvents of high
polarity; therefore, DMF was selected for this reaction. Not surprisingly,
no desired product was found in the absence of base even if the solvent
was DMF (entry 9). Different bases, such as NaHCO3, Na2CO3 and
Cs2CO3, were investigated during optimization studies. Generally, higher
conversions were observed with stronger bases, as shown in Table 4
(entries 5, 6, 7 and 8). As expected, potassium and cesium were both
better counterions than sodium, as they provided higher conversions in
the substitution reaction. With the additional consideration of the high
cost and hygroscopicity of Cs2CO3, K2CO3 was selected as the preferred
base. When the temperature was lowered, a longer reaction time was
required (entry 10). To summarize the above results, the optimized
reaction conditions for the coupling of 4 and 7 were K2CO3 as the base in
DMF at 90 °C for 8 h.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 4. Coupling of 4 and 7 under various conditions
12
ACS Paragon Plus Environment

Page 13 of 34
Organic Process Research & Development
1
2
3
4
entry solvent
base (1.0 eq) T/t (°C/h)
HPLCa
1
2
MeCN
THF
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
NaHCO3
Na2CO3
Cs2CO3
none
80/16
65/16
90/16
90/16
90/8
95.5
79.7
73.8
94.7
97.3
82.4
88.9
96.5
ND
5
6
7
8
3
toluene
dioxane
DMF
DMF
DMF
DMF
DMF
DMF
9
4
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
5
6
90/16
90/16
90/8
7
8
9
90/16
10
K2CO3
70/16
95.3
aHPLC purity of the reaction mixture.
The previously mentioned methanol elimination and CꢀO cleavage
byproducts were not detected by LCꢀMS under the present reaction
conditions. The hydrolytic impurities of 4 and 8 were both less than 1.0%
by HPLC analysis, and the hydrolytic byproduct of 8 was racemic 9,
which was the product of the next step. The hydroxyꢀsubstituted impurity
of 7 was also detected by LCꢀMS (<1.0% on HPLC). Intermediate 8 was
a hydrophobic organic compound so it could be precipitated from the
mixture of DMF and H2O. However, despite the water being added to the
reaction mixture very slowly, a sticky bulk solid appeared, which would
cause difficulties during scaling up in the pilot plant. In view of this, it
was not practical to directly add water to the reaction mixture. The
addition of a third solvent that was not only miscible with DMF and H2O
but also partially dissolved intermediate 8 was considered. Ultimately,
13
ACS Paragon Plus Environment

Organic Process Research & Development
Page 14 of 34
1
2
3
4
5
6
7
8
9
MeOH was selected through solvent screening. In the end, the most
practical workꢀup procedure was dilution of the reaction mixture with
MeOH followed by slow addition of water to generate the precipitate, and
desired intermediate 8 was easily filtered from the mixture. No remaining
desired product was detected in the filtration mother liquors by HPLC.
The crude product was further purified by recrystallization in methanol
and water to give intermediate 8 with over 99.5% purity on HPLC.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Preparation of racemic (±)-AMB (9). The steric hindrance from the
ester moiety of 8 was larger than that of 4 and disfavored the hydrolysis.
Therefore, harsh ester hydrolysis conditions were applied in the step. A
large amount of 8 was unreacted even when refluxing in a mixture of
aqueous KOH and THF for 16 hours, as shown in Table 5 (entry 1),
because the solubility of 8 was poor in aqueous THF. The hydrolysis of 8
was very rapid in NMP (entry 3), but byproduct 13 from MeOH
elimination was detected by HPLC (approximately 31%). The data in the
table showed that the content of impurity 13 increased with increasing
solvent polarity. Impurity 13 was likely formed via the elimination of
MeOH from 8 followed by ester hydrolysis in aqueous potassium
hydroxide because the carbonyl αꢀH of 8 was more susceptible to
deprotonation than its counterpart in acid 9. Similarly, impurity 13 was
also detected (approximately 14% on HPLC) when using MeOH as the
reaction solvent (entry 2). When dioxane was eventually chosen as the
14
ACS Paragon Plus Environment

Page 15 of 34
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
reaction solvent, the level of impurity 13 decreased substantially (only
approximately 7% on HPLC). In all cases, hydroxylꢀcontaining impurity
5 was detected (<5%) and was presumably formed by the nucleophilic
displacement of the pyrimidinoloxy moiety by a hydroxide anion,
simultaneously generating impurity 12. The formation mechanism of
impurity 2 along with 12 was proposed as in Scheme 3. The
deprotonation of the methine group adjacent to the carboxylic ester in the
basic environment could lead to the elimination of the methoxy group to
afford α,βꢀunsaturated ester 8a, which upon Michael addition with
hydroxide would lead to 8b. This intermediate finally fragmented into
impurities 2 and 12 under the strong basic conditions as indicated.
Interestingly, experiments showed that desired product 9 was not
degraded in the strong basic conditions since the elimination of the
methoxy group in 9 was unlikely due to the reduced acidity of the
carbonyl αꢀH of 9.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 5. Hydrolysis of 8 using various solvents
OH
O
COOH
OH
COOH
COOMe
COOH
KOH
H2O
N
N
N
MeO
MeO
MeO
N
N
+
O
O
O
N
N
N
8
9
12
5
13
2
HPLCb
entrya solvent
T/t(°C/h)
65/16
8
9
12
5
13
2
1
2
THF
82.56
0.74
16.08
82.53
0.21
0.13
0.41
1.47
0.17
14.41
0.60
0.70
MeOH
65/16
15
ACS Paragon Plus Environment

Organic Process Research & Development
Page 16 of 34
1
2
3
4
5
6
7
3
4
NMP
80/1
0.27
0.28
61.16
91.61
2.36
0.47
4.43
0.70
30.84
6.89
0.95
0.04
dioxane 80/5
aKOH (2 equiv), 15% NaOH aqueous solution. bHPLC purity of the reaction mixture.
8
9
Scheme 3. Proposed formation mechanism of impurities 2 and 12
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
After the reaction was completed, the reaction mixture was diluted
with water and extracted with DCM to remove nonpolar impurities. If
dilute hydrochloric acid was directly added to the separated water layer, a
substantial amount of viscous matter precipitated, which hampered
stirring during scaleꢀup. The improved workꢀup procedure was as follows:
the reaction mixture was first diluted with IPA, and the acid was then
slowly added to precipitate the product without the formation of the
viscous matter.
Preparation of (+)-AMB (1). Commercially available and inexpensive
Sꢀαꢀphenylethylamine was first chosen as the chiral resolution agent. To
our delight, the optical purity of the diastereomeric salt 10 was up to 98.5%
in MeOH, but the best yield was only 20%. When it was replaced by
EtOH and IPA, the yield increased to 24% and 30%, respectively. The
16
ACS Paragon Plus Environment

Page 17 of 34
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
yield was further improved by adding an antisolvent such as nꢀheptane,
IPE or MTBE. The experiments indicated that the chiral purity of
diastereomeric salt 10 was highest when the chiral resolution was
performed in a mixture of IPA and MTBE. The other isomer was
undetectable by chiral HPLC following a simple recrystallization of the
crude salt in the same solvent, and the yield of the chiral resolution was
up to 40%.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Generally, diastereomeric salt 10 was dissociated with dilute
hydrochloric acid, and the free (+)ꢀAMB was extracted with solvents
such as EA, toluene or DCM, but its solubility was very poor in most of
these solvents. Although it was highly soluble in DCM, partial
racemization of (+)ꢀAMB was observed on chiral HPLC. In addition, a
large amount of solvent would need to be distilled, which is not
convenient for scaleꢀup production. The improved procedure was as
follows: diastereomeric salt 10 was dissolved in IPA at 50 °C, and then
dilute hydrochloric acid was slowly added to the resulting clear solution
to precipitate desired free acid 1 without using extraction or distillation
operations.
Purification of crude 1 to obtain highly pure (+)ꢀAMB was a
challenging task. Our primary goal was to reduce the production cost by
minimizing the loss of the product during the purification process.
Therefore, we tested a variety of recrystallization conditions using
17
ACS Paragon Plus Environment

Organic Process Research & Development
Page 18 of 34
1
2
3
4
5
6
7
8
9
solvents such as MeOH, IPA, and H2O and mixtures of these solvents.
Finally, IPA and H2O in a 1:1 ratio provided good yield (~90%) of highly
pure (+)ꢀAMB of ICHꢀgrade quality. The polymorphic stability of
(+)ꢀAMB was examined under different temperatures and humidity levels
by powder Xꢀray diffraction (PXRD) studies and thermogravimetric
analysis (TGA). PXRD analysis demonstrated that the product had the
same crystal form as that reported for (+)ꢀAMB in the literature.5d In the
TGA experiments, the material lost less than 0.03% of its weight, which
indicated that the residual solvent and water content were low.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Recovery of (±)-AMB (9). Chiral resolution is not costꢀeffective as the
maximum possible yield is 50%. The recovery of the unwanted isomer
from the chiral resolution mother liquors was studied herein. Partial
racemization of 1 was observed during the conversion of diastereoisomer
salt 10 to free acid 1 in DCM. Although the spontaneous racemization of
some chiral compounds in polar aprotic solvents has been previously
reported,16 there has been no precedent for such racemization of
(+)ꢀAMB in the absence of an acid. Different solvents were screened for
the racemization of 1, as shown in Table 6. Complete racemization of 1
was observed in EA, MeCN, toluene and pyridine at elevated temperature,
but the HPLC purity was lower when the reaction was conducted in
toluene or pyridine due to partial thermal degradation. MeCN was finally
chosen as the solvent because the racemization time was the shortest
18
ACS Paragon Plus Environment

Page 19 of 34
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
compared with the above solvents (entry 18). Generally, the racemization
of 1 occurred easily in polar aprotic solvents at elevated temperature, as
shown in Table 6. It was unclear whether the racemization could be
catalyzed by a small amount of residual hydrochloric acid from the
preceding steps. This hypothesis was ruled out by the following
experiments. Solid (+)ꢀAMB with high optical purity was dissolved in EA
and washed with an aqueous solution of 5% NaHCO3 and water three
times. Racemic AMB was again obtained when the solution was dried
over Na2SO4 and refluxed for 16 h.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 6. Racemization of (+)-AMB in various solvents and at various
temperatures
entrya
solvent
MeOH
EtOH
T/t (°C/h)
65/16
78/16
82/16
68/16
55/16
98/16
66/16
80/6
HPLCb
99.9
99.9
99.9
99.9
99.8
99.9
99.4
99.1
98.4
95.9
96.3
96.1
83.3
83.1
86.6
58.6
1
2
3
IPA
4
IPE
5
MTBE
heptane
THF
6
7
8
2ꢀMeꢀTHF
acetone
Et3N
9
56/16
90/16
100/16
90/16
39/16
100/16
100/16
100/8
10
11
12
13
14
15
16
dioxane
HOAc
CH2Cl2
NMP
DMSO
DMF
19
ACS Paragon Plus Environment

Organic Process Research & Development
Page 20 of 34
1
2
3
4
5
6
7
8
9
17
18
19
20
21
22
EA
77/16
81/3
50.2c
50.1d
49.4e
51.1f
99.9
MeCN
pyridine
toluene
MeOH/HCl
90/16
100/16
60/16
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
MeOH/MeONa 60/16
99.8
b
aOptical purity is over 99.9% and HPLC purity is over 99.9% for (+)ꢀAMB. Optical
purity. c,dHPLC purity is 99.9%. eHPLC purity is 95.6%. fHPLC purity is 97.1%.
The possible racemization mechanism of (+)ꢀAMB is depicted in
Scheme 4. Although the pyrimidine moiety and the carboxy group could
not form a stable salt, the possible intramolecular hydrogen bonds would
result in a relatively stable bicyclic structure in polar aprotic solvents
such as EA and MeCN. This could in turn enhance the carbonyl’s
tendency to undergo thermal or acidꢀcatalyzed enolization with the
formation of unstable intermediate 1a. Upon tautomerization of the enol
to the carbonyl compound, racemic AMB would be generated. The
racemic mixture was eventually obtained in equilibrium since it was
thermodynamically more stable than the corresponding enantiomers. This
process was further facilitated by the fact that the racemate precipitated
from the solution due to its lower solubility than the enantiomers.
Racemization did not occur in protic solvents such as MeOH (entries 1, 2,
3, 21, and 22) probably because these solvents prevented the formation of
the intramolecular hydrogen bond. The more difficult racemization in
very polar aprotic solvents such as DMSO might be explained by the fact
20
ACS Paragon Plus Environment

Page 21 of 34
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
that the racemate could not precipitate under these conditions (entries 13,
14, 15, and 16). The extremely low solubility of (+)ꢀAMB in nonpolar
solvents such as heptane could be the reason that it did not racemize by
simple heating.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 4. Proposed racemization mechanism of (+)-AMB
The chemical purity of the resolution mother liquors was only
approximately 95%, excluding Sꢀαꢀphenylethylamine on HPLC, and it
was enriched in (ꢀ)ꢀAMB. The free acid was precipitated from the
resolution mother liquors by the treatment with the mixture of dilute
hydrochloric acid and IPA, and the purity of the solid was over 99.5%
after a simple filtration workꢀup procedure. The free acid was refluxed in
MeCN for 3 h and filtered to give racemic 9, which was subjected to the
same resolution process with Sꢀαꢀphenylethylamine to give (+)ꢀAMB in
high purity.
CONCLUSION
In conclusion, we have developed a recoverable, costꢀeffective,
industrially scalable synthetic route to (+)ꢀAMB, an endothelinꢀA
receptor antagonist. Throughout the synthetic route, common and
inexpensive regents, such as the resolving agent Sꢀαꢀphenylethylamine,
21
ACS Paragon Plus Environment

Organic Process Research & Development
Page 22 of 34
1
2
3
4
5
6
7
8
9
were used. Every process step was optimized, and all processꢀrelated
impurities were successfully removed to give the drug substance in high
chemical and chiral purities. The practicable method involved a thermal
racemization of chiral AMB, which was disclosed in the selected polar
aprotic solvents without the aid of an external acid or base. Accordingly,
an efficient recycling resolution process was developed to improve the
overall yield. Furthermore, the process of converting the salt to the free
acid served the dual purpose of dissociating diastereomerically pure salt
10 and recovering free acid 11 from the resolution mother liquors. When
the unwanted isomer from the resolution mother liquors was recycled
only once, the overall process yield was up to 47% in contrast to 18% for
the usual route, as depicted in Scheme 1.5d.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
EXPERIMENTAL SECTION
All solvents and reagents were purchased from suppliers and were used
without further purification. NMR spectra were recorded on a Bruker
Avance III 400 MHz spectrometer in DMSOꢀd6 with Me4Si (TMS) as an
internal standard. Mass spectra (MS) were recorded on an Agilent 6210
series single quadrupole LC/MS. Highꢀresolution mass spectrometry
(HRMS) measurements were recorded on a Bruker Daltonics Solarix 7.0T.
Melting points (MP values) were measured on a WRSꢀ1B apparatus.
Thermogravimetric analysis (TGA) was performed on a TA TGA Q500
analyzer with a heating rate of 10 °C/min under a nitrogen atmosphere.
22
ACS Paragon Plus Environment

Page 23 of 34
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Elemental analysis (EA) was performed on a Thermo Flash 2000
instrument. Specific rotation data were measured on an Anton Paar MCP
500. Powder Xꢀray diffraction (PXRD) data were recorded on a Bruker
D8 Advance. The reactions were monitored by HPLC, and the purities
were calculated from the HPLC peak areas. The HPLC analyses were
recorded using a standard method on a Dionex UltiMate 3000 HPLC
instrument using a Fortis C18 column (250 mm × 4.6 mm, 5 ꢁm), 30 °C,
1 mL/min, 220 nm, 45 min. The mobile phase involved the mixture of 32%
MeCN and 68% aqueous KH2PO4 (0.02 mol/L, pH adjusted to 3.0 by
using phosphoric acid) as phase A and MeCN as phase B. The HPLC
analyses were accomplished with a gradient elution program (time (min)/%
B: 0/0, 10/0, 20/50, 35/50, and 45/0). The chiral HPLC separations were
performed using a CHIRALPAK ADꢀH (250 mm × 4.6 mm, 5 ꢁm,
DAICEL, Shanghai) column with a mobile phase of 95% hexane and 5%
EtOH (0.1% TFA), 30 °C, 0.5 mL/min, 263 nm, and 35 min.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Methyl 3,3-diphenyloxirane-2-carboxylate (3). A 5ꢀL reactor was
charged with 2 (300.0 g, 1.65 mol), NaOMe (156.1 g, 2.89 mol) and THF
(0.6 L) under an atmosphere of nitrogen. ClCH2COOMe (313.6 g, 2.89
mol) was added to the stirring mixture at ꢀ10~ꢀ5 °C over 3 h, and then the
mixture was stirred for another 30 min at ꢀ10~ꢀ5 °C. The reaction was
monitored by HPLC and deemed complete when 2 was less than 3.0%.
The reaction mixture was diluted with MTBE (1.2 L) and water (0.6 L).
23
ACS Paragon Plus Environment

Organic Process Research & Development
Page 24 of 34
1
2
3
4
5
6
7
8
9
The aqueous layer was discarded, and the organic layer was washed twice
with 10% aqueous NaCl until pH = 7~8. Then, the organic solvent was
evaporated and replaced by MeOH (0.3 L) under reduced pressure at
35 °C. The residue was directly used for the next step. When the weight
assay of 3 was performed on HPLC with the external standard method,
the actual yield was 95% (purity > 97.0%). A small analytical sample of
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
epoxide 3 was purified by recrystallization from EA and heptane. H
NMR (400 MHz, DMSOꢀd6) δ 7.35ꢀ7.40 (m, 8H), 7.29ꢀ7.31 (m, 2H),
13
4.33 (s, 1H), 3.48 (s, 3H); C NMR (100 MHz, DMSOꢀd6) δ 166.54,
138.55, 135.42, 128.53, 128.51, 128.28, 128.19, 127.44, 126.83, 66.91,
60.98, 51.90. HRMS m/z [M+Na]+ Calcd for C16H14NaO3: 277.0841;
found: 277.0835.
Methyl 2-hydroxy-3-methoxy-3,3-diphenylpropanoate (4). The
solution from the previous stage was diluted with MeOH (1 L) and
transferred to a 2ꢀL reactor at room temperature. Hydrochloric acid (37%,
4 mL) was added to the above reactor until pH = 2. The exothermic
reaction (35 °C) was complete (<2.0% of 2 by HPLC) after 2 h with the
formation of a white precipitate. The reaction mixture was cooled to
0~5 °C. After stirring for 1 h, the solid product was isolated by filtration,
washed with cold (0~5 °C) MeOH (0.5 L) and then dried in an airꢀdrying
oven at 40 °C to yield the title compound as a white solid (403 g, yield:
1
90%, purity > 99.0%). H NMR (400 MHz, DMSOꢀd6) δ 7.26ꢀ7.33 (m,
24
ACS Paragon Plus Environment

Page 25 of 34
Organic Process Research & Development
1
2
3
4
8H), 7.21ꢀ7.24 (m, 2H), 5.79 (d, J = 3.6 Hz, 1H), 5.20 (d, J = 3.6 Hz, 1H),
5
6
13
3.40 (s, 3H), 3.19 (s, 3H); C NMR (100 MHz, DMSOꢀd6) δ 171.78,
7
8
9
142.65.09, 141.64, 128.37, 127.90, 127.57, 127.10, 126.97, 126.74, 83.76,
74.20, 52.66, 51.07. HRMS m/z [M+Na]+ Calcd for C17H18NaO4:
309.1103; found: 309.1097.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Methyl-2-((4,6-dimethylpyrimidin-2-yl)oxy)-3-methoxy-3,3-diphenyl
propanoate (8). A 10ꢀL reactor was charged with 4 (400.0 g, 1.40 mol), 7
(273.2 g, 1.47 mol), K2CO3 (193.5 g, 1.40 mol) and DMF (480 mL) under
an atmosphere of nitrogen. The stirring mixture was heated at 85 °C for 8
h. The reaction was monitored by HPLC, and 4 was less than 1.0%. The
reaction mixture was cooled to 40~50 °C and diluted with MeOH (1.2 L).
Water (1 L) was added slowly over 30 min. The solid precipitated after 15
min, and more water (2.6 L) was added to the reaction mixture over 30
min while maintaining the mixture at 40~50 °C. The mixture was cooled
to room temperature and stirred for 1 h. The solid was isolated by
filtration, washed with water (0.5 L) three times and then dried in an
airꢀdrying oven at 40 °C. The white powder was suspended in MeOH (1
L) and refluxed to yield a clear solution. Water (0.37 L) was slowly added
to the solution over 40 min, and then the mixture was cooled to 0~5 °C.
After stirring for 1 h, the obtained solid was isolated by filtration, washed
with a mixture of MeOH and H2O (1:1, 0.6 L) twice, and then dried in an
airꢀdrying oven at 40 °C to yield the title compound as a white solid (515
25
ACS Paragon Plus Environment

Organic Process Research & Development
Page 26 of 34
1
2
3
4
1
g, yield: 94%, purity > 99.5%). H NMR (400 MHz, DMSOꢀd6) δ
5
6
7.31ꢀ7.34 (m, 8H), 7.24ꢀ7.29 (m, 2H), 6.96 (s, 1H), 6.09 (s, 1H), 3.37 (s,
7
8
9
13
3H), 2.33 (s, 6H); C NMR (100 MHz, DMSOꢀd6) δ 169.21, 168.33,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
162.89, 142.02, 140.95, 127.93, 127.84, 127.75, 127.33, 127.30, 127.21,
114.98, 83.12, 78.07, 53.15, 51.42, 23.32.
2-((4,6-dimethylpyrimidin-2-yl)oxy)-3-methoxy-3,3-diphenylpropanoi
c acid (9). A 10ꢀL reactor was charged with 8 (500.0 g, 1.27 mol) and
dioxane (1.9 L) under an atmosphere of nitrogen. Aqueous potassium
hydroxide (143.0 g of KOH in water of 0.81 L) was added to the stirring
mixture at room temperature, and then the mixture was heated to 80 °C
for 4 h. The reaction was monitored by HPLC, and 8 was less than 1.0%.
The reaction mixture was cooled to room temperature and diluted with
water (2 L) and DCM (1.5 L). The water layer was separated and diluted
with IPA (2.5 L), and the solution was treated with hydrochloric acid (6 N,
0.25 L) until pH = 6~7. After stirring for 1 h, a large amount of white
solid precipitated, and a second portion of the acid (6 N, 0.25 L) was
slowly added over 1 h until pH = 1~2. The solid product was isolated by
filtration, washed twice with a mixture of IPA and H2O (1:1, 1 L), and
then dried in an airꢀdrying oven at 40 °C to give the title compound as a
1
white solid (461 g, yield: 96%, purity > 99.5%). H NMR (400 MHz,
DMSOꢀd6) δ 12.55 (br, 1H), 7.27ꢀ7.35 (m, 8H), 7.21ꢀ7.25 (m, 2H), 6.95
13
(s, 1H), 6.14 (s, 1H), 3.38 (s, 3H), 2.34 (s, 6H); C NMR (100 MHz,
26
ACS Paragon Plus Environment

Page 27 of 34
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
DMSOꢀd6) δ 169.07, 163.17, 142.66, 141.45, 127.83, 127.73, 127.71,
127.21, 127.00, 114.79, 83.14, 77.61, 53.07, 23.36. HRMS m/z [M+Na]+
Calcd for C22H22N2NaO4: 401.1477; found: 401.1472.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(S)-1-phenylethan-1-aminium (S)-2-((4,6-dimethylpyrimidin-2-yl)oxy
)-3-methoxy-3,3-diphenylpropanoate (10). A 5ꢀL reactor was charged
with 9 (400.0 g, 1.06 mol), Sꢀαꢀphenylethylamine (192.1 g, 1.59 mol),
IPA (0.8 L) and MTBE (3.2 L) under an atmosphere of nitrogen. The
stirring mixture was heated to reflux for 30 min and became clear before
being allowed to naturally cool to 35~38 °C. Crystal seeds (0.2 g) were
introduced to the solution, and the mixture was stirred at 35 °C for 1 h
and then at room temperature for 2 h. The solid was isolated by filtration,
washed with a mixture of MTBE and nꢀheptane (1:1, 1.2 L) and dried in
an airꢀdrying oven at 40 °C to yield crude 10 (238 g, 45%). A 3ꢀL reactor
was charged with the crude product (238 g), IPA (0.48 L) and MTBE (1.9
L). The mixture was refluxed to form a clear solution within 30 min and
then cooled naturally before the crystal seeds (0.2 g) were added to the
solution. After adding the crystal seeds, the mixture was stirred at
47~50 °C for 1 h. The mixture was cooled and stirred at room
temperature for 2 h. The solid was isolated by filtration and washed with
a mixture of MTBE and nꢀheptane (1:1, 0.7 L). The combined mother
liquors were concentrated under reduced pressure, and the solvents were
replaced by IPA to give a solution of 11 for the recovery operation (total
27
ACS Paragon Plus Environment

Organic Process Research & Development
Page 28 of 34
1
2
3
4
5
6
7
8
9
volume: 0.5 L, purity: 95.0%). The diastereomerically pure salt 10 was
dried in an airꢀdrying oven at 40 °C to yield the title compound as a white
solid (212 g, yield: 40%, optical purity > 99.9%, HPLC purity > 99.5%
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
(Sꢀαꢀphenylethylamine was not calculated)). H NMR (400 MHz,
DMSOꢀd6) δ 7.12ꢀ7.39 (m, 15H), 6.84 (s, 1H), 6.15 (s, 1H), 4.09 (q, J =
13
4.4 Hz, 1H), 3.42 (s, 3H), 2.31 (s, 6H), 1.31 (d, J = 4.4 Hz, 3H); C
NMR (100 MHz, DMSOꢀd6) δ 169.31, 168.49, 164.03, 144.47, 143.21,
143.04, 128.37, 127.92, 127.49, 127.39, 127.24, 127.21, 126.44, 126.39,
126.09, 113.80, 83.60, 79.05, 53.07, 49.92, 26.86, 23.41.
(S)-2-((4,6-dimethylpyrimidin-2-yl)oxy)-3-methoxy-3,3-diphenylprop
anoic acid (1). A 2ꢀL reactor was charged with 10 (200.0 g, 0.40 mol)
and IPA (0.4 L) and heated to reflux for 30 min under an atmosphere of
nitrogen. Dilute hydrochloric acid (1.0 N, 0.7 L) was added to the clear
solution at 50 °C, and the mixture was stirred for 1 h before naturally
cooling to room temperature. The precipitate was isolated by filtration
and washed with deionized water (0.3 L) three times and then dried in an
airꢀdrying oven to give crude 1 (144 g). Crude 1 and IPA (0.72 L) were
placed in a 2ꢀL reactor and heated to reflux for 20 min. Deionized water
(0.72 L) was added to the clear solution slowly over 30 min. The crystal
seeds (0.1 g) were added to the solution, and it was stirred at 60 °C for 1
h. The mixture was allowed to cool to room temperature over 2 h. The
solid was filtered and washed with a mixture of IPA and deionized water
28
ACS Paragon Plus Environment

Page 29 of 34
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
(1:1, 0.2 L) and then deionized water (0.2 L) and dried in an airꢀdrying
oven at 40 °C to yield the title compound as a white needleꢀlike
crystalline solid (136 g, yield: 90%, optical purity > 99.9%, HPLC purity >
99.5%). Mp 185.0~185.9 °C. Anal. Calcd for C22H22N2O4: C, 69.83; H,
5.86; N, 7.40. Found: C, 69.72; H, 5.84; N, 7.60. [α]D25 = +168.472 (c 1.0,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
MeOH). H NMR (400 MHz, DMSOꢀd6) δ 12.55 (br, 1H), 7.27ꢀ7.34 (m,
8H), 7.20ꢀ7.25 (m, 2H), 6.95 (s, 1H), 6.13 (s, 1H), 3.38 (s, 3H), 2.34 (s,
13
6H); C NMR (100 MHz, DMSOꢀd6) δ 169.04, 163.16, 142.63, 141.43,
127.81, 127.71, 127.68, 127.19, 126.98, 114.75, 83.14, 77.58, 53.03,
23.33. HRMS m/z [M+Na]+ Calcd for C22H22N2NaO4: 401.1477; found:
401.1472. ESIꢀMS m/z 377.10 [MꢀH]+.
Recovery of 9 from the mother liquors of 11. A 5ꢀL reactor was charged
with IPA and the resolution mother liquors (total volume: 1.3 L)
containing the diastereoisomeric salts (theoretical weight: 318 g) from the
preparation of 10. The stirring solution was heated at reflux to form a
clear solution and then cooled to 30 °C. Then, dilute hydrochloric acid (1
N, 1.1 L) was added to the clear solution at 30 °C over 1 h. The white
precipitate was isolated by filtration, washed with water three times (300
mL×3), dried in an airꢀdrying oven at 40 °C and then transferred to a 2ꢀL
reactor. MeCN (1.2 L) was added to the reactor and heated at reflux
(81 °C) for 3 h. In process control of the racemization was monitored by
29
ACS Paragon Plus Environment