Plant Process for the Preparation of Cinchona Alkaloid-Based Thiourea Catalysts

Article abstract of DOI:10.1021/acs.oprd.7b00049

Cinchona alkaloid-based thiourea catalysts (1a and 1b) belong to an important class of bifunctional organocatalysts, which has been widely used for a variety of asymmetric reactions. The commercial availability of these catalysts is sporadic, and limited

Full text of DOI:10.1021/acs.oprd.7b00049

Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)
Full Paper
A Plant Process for the Preparation of
Cinchona Alkaloid-Based Thiourea Catalysts
Yi Wang, Karen L. Milkiewicz, Mildred L. Kaufman, Linli He, Nelson G. Landmesser, Daniel V. Levy,
Shawn Allwein, Michael A. Christie, Mark A. Olsen, Christopher J. Neville, and Kannan Muthukumaran
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.7b00049 • Publication Date (Web): 02 Mar 2017
Downloaded from http://pubs.acs.org on March 2, 2017
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 free 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 accessible to all
readers and 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.
Organic Process Research & Development 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 22
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
A Plant Process for the Preparation of Cinchona
AlkaloidꢀBased Thiourea Catalysts
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
Yi Wang*†, Karen L. Milkiewicz†, Mildred L. Kaufman†, Linli He†, Nelson G. Landmesser†,
Daniel V. Levy†#, Shawn P. Allwein †, Michael A. Christie†, Mark A. Olsen§, Christopher J.
Neville§, and Kannan Muthukumaran‡
†Process Research & Development, §Analytical Development, Teva Pharmaceuticals, 383
Phoenixville Pike, Malvern, PA 19355.
‡ Cambrex High Point, 4180 Mendenhall Oaks Pkwy, High Point, NC 27265.
yi.wang@tevapharm.com RECEIVED DATE (to be automatically inserted after your
manuscript is accepted if required according to the journal that you are submitting your
paper to)
ACS Paragon Plus Environment
1

Organic Process Research & Development
Page 2 of 22
1
2
3
4
5
6
7
8
9
R
N
R
OMe
N
OMe
N
N
CF3
CF3
S
N
N
O
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
> 14 kg / batch, ~$2750/kg
Graphic for use in TOC
ACS Paragon Plus Environment
2

Page 3 of 22
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
ABSTRACT Cinchona alkaloidꢀbased thiourea catalysts (1a and 1b) belong to an important
class of bifunctional organocatalysts, which has been widely used for a variety of asymmetric
reactions. The commercial availability of these catalysts is sporadic, and limited to the subꢀ
gram quantities. Herein is described a general, scalable, and practicable process for the
preparation of these catalysts that was used to synthesize more than 14 kg of catalyst per batch.
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 Thiourea Cinchona Alkaloid Scalingꢀup
ACS Paragon Plus Environment
3

Organic Process Research & Development
Page 4 of 22
1
2
3
4
Introduction
5
6
7
8
Bifunctional, modified cinchona alkaloids represent a very important class of chiral
organocatalysts.1 The unique, bulky basic quinuclidine moiety, along with the incorporated
hydrogen bond donors such as hydroxyl or thiourea groups, enables the simultaneous activation
of both the nucleophilic and electrophilic reactants, enhancing the activity and enantioselectivity
of a variety of reactions with a broad substrate scope. Unlike metalꢀbased chiral catalysts,
organocatalysts also have the advantage of removability, recoverability, and moisture and air
stability. These attributes are highly desirable for the development of greener processes in the
pharmaceutical industry.2
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
R = CH2CH3, 1a; R = CHCH2, 1b
Figure 1. Cinchona AlkaloidꢀBased Thiourea Catalysts
With the progress made in one of our clinical programs, it became a high priority to have
multiꢀkilogram quantities of the catalyst (Figure 1, 1a or 1b) at a reasonable price.
Unfortunately, the commercial supply for both catalysts was limited to sporadic availability of
subꢀgram quantities at a premium cost.3 The decision was made to prepare the catalyst inꢀhouse.
Herein, we detail our efforts toward the development of a highly efficient process for the
preparation of catalysts 1a and 1b with significant improvements on the robustness, scalability,
ACS Paragon Plus Environment
4

Page 5 of 22
Organic Process Research & Development
1
2
3
4
5
6
and practicability. The process was used to prepare kilo gram quantities of 1a and 1b in our kilo
labs, and 25 kg of 1b in our pilot plant at a cost of approximately $2750/kg.
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
Original Synthesis
The first synthesis of hydroquinineꢀbased catalyst 1a was reported by Soós’s group in 2005
(Scheme 1).4 The procedure consisted of three steps, including a Mitsunobu reaction, reduction,
and coupling with isothiocyanate 6. The isolation and chromatographic purifications were
required for both intermediate 5a and 1a to furnish the product in over 95% purity. Although the
procedure readily provided grams of product, it was not feasible to carry out on a much larger
scale without modifications.
OMe
OMe
N
OMe
N
N
Ph3P, DIAD, THF
(PhO)2P(O)N3
N
N
Ph3P, H2O
THF
NH2
OH
N3
silica gel
chromatography
N
5a
4a
hydroquinine
CF3
OMe
N
SCN
N
S
CF3
CF3
6
CF3
N
H
THF
N
H
silica gel
chromatography
1a
Scheme 1. Soós’s Synthesis of 1a
Process Development
Process improvement work using hydroquinine as the starting material began with screening
solvents for the first stage Mitsunobu reaction in order to replace the waterꢀmiscible solvent THF
ACS Paragon Plus Environment
5

Organic Process Research & Development
Page 6 of 22
1
2
3
4
5
6
7
8
9
and thus simplify the workup. LCꢀMS analysis of the reaction mixture revealed the major sideꢀ
product was the enamine 7a (Figure 2), generated by the elimination of the secondary hydroxyl
functionality. As shown in Table 1, several nonꢀmiscible solvents were found to provide similar
or better impurity profiles, with CH2Cl2 affording the cleanest reaction (Table 1, Entry 3). Other
benefits of CH2Cl2 included the possible reduction of the solvent volume by 50% (Table 1, Entry
5), and having the organic wash at the bottom layer, simplifying the workup.
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
7a
Figure 2. A possible structure of impurity 7a.
Table1. Solvent Screening for Stage 1 Mitsunobu Reaction
Entry
Solvent
2ꢀMeTHF
THF
Azide 4a (A%)
Enamine 7a (A%)
1
2
3
4
5
85.7
84.1
94.5
87.0
92.1
7.1
10.2
2.5
7.2
3.0
CH2Cl2
toluene
*
CH2Cl2
*The solvent volume was reduced by half.
The stage 2 reduction of azide 4a required 1.8 equivalents of PPh3 for complete conversion.
Generally, the starting material disappeared in 3ꢀ4 hours, but the reaction was held for additional
7 hours to fully consume the intermediate.
ACS Paragon Plus Environment
6

Page 7 of 22
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
The workup started by charging 3N HCl directly into the reaction mixture and allowing for a
phase separation, which was easily achieved using CH2Cl2 as the solvent. Of note, the clear
phaseꢀcut was reached faster than when ethyl acetate was used as in the original process. The
product amine 5a and side product enamine 7a stayed in the upper aqueous layer at acidic pH,
while the unreacted triphenylphosphine and the sideꢀproduct triphenylphosphine oxide remained
in the bottom organic wash. It was found that the pH of the aqueous layer need to be adjusted to
2.0ꢀ2.5 for complete removal of the side products. Lower pH hinder the removal of
triphenylphosphine oxide from the aqueous phase. The resulting acidic aqueous solution was
basified with aqueous ammonia, and backꢀextracted with CH2Cl2. The productꢀcontaining
organic phase was azeotropically dried until the water content was less than 1000 ppm. The
residual solution was used directly in the next stage, coupling with the isothiocyanate 6.
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
The stage 3 coupling reaction proceeded smoothly in methylene chloride with 3ꢀ4 small
impurities (< 2 A% in total by HPLC analysis), including the remaining unreacted amine 5a, and
the carriedꢀover enamine 7a. These very polar impurities were found to be impossible to remove
solely by recrystallization. An acid wash was developed to remove these impurities.
Interestingly, only ethyl acetate could be used as the extraction solvent to facilitate this wash.
Thus, the workup of stage 3 occurred with a solvent swap from methylene chloride to ethyl
acetate, followed by an adjustment to pH 2ꢀ2.5 with 3N HCl. The very polar impurities formed
HCl salts, staying in the bottom aqueous layer. The HCl salt of 1a, however, stayed in the upper
ethyl acetate layer as determined by LCꢀMS analysis. After the phase cut and subsequent
removal of those impurities, the crude product solution was basified with aqueous ammonia and
recrystallized to reach the target purity. It is worth noting that the basification step increased the
amount of a new unknown impurity 8a to 2ꢀ4 A%, as determined by LCMS. Further
ACS Paragon Plus Environment
7

Organic Process Research & Development
Page 8 of 22
1
2
3
4
5
6
optimization of the recrystallization reduced this impurity to less than 0.5 A% in the final
product.
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
8a
Figure 3. A possible structure of impurity 8a.
For the recrystallization of the final product, having the dissolution solvent boiling point
higher than ethyl acetate was ideal to facilitate a complete solvent exchange before the
recrystallization. Based on this, a small set of recrystallization experiments using either a single
solvent or solvent mixtures were performed. The experiments started with dissolving crude 1a at
room temperature or elevated temperature in the selected solvents, then precipitating the product
by either cooling down or by a charge of antiꢀsolvents (Table 2). These experiments identified
two solvent systems that gave a precipitate with good physical properties (Table 2, Entries 4 and
5). Between two solvent systems, acetonitrile provided better purity as it rejected impurity 8a
(Figure 3) more efficiently than iꢀPrOAc/heptanes solvent system. However, none of the
recrystallizations reduced the polar impurities such as enamine 7a or unreacted starting material
amine 5a, which confirmed the need for the acid wash described above. As expected, the
recrystallization of the crude catalyst 1a after the acidꢀbase wash furnished more than 99 A%
purity (Table 2, Entry 8) with preferred ACN system.
ACS Paragon Plus Environment
8

Page 9 of 22
Organic Process Research & Development
1
2
3
4
Table 2. Recrystallization Experiments of 1a
5
6
7
8
9
Entry Initial Purity
(A%)
Solvent System
(vol/vol)
Purity
(A%)
Result
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
2
3
4
5
6
7
8
87.6
87.6
87.6
87.6
87.6
87.6
87.6
95.0
2ꢀMeTHF/heptanes= 1/10 Gum first, then solid 90.0
MTBE/heptanes = 10/50
Gum first, then solid 95.4
iꢀPrOH/H2O =2/4
Gum
89.4
95.6
97.6
NA
iꢀPrOAc/heptanes = 2/20 Solid
CH3CN = 4 Solid
nꢀBuOH/heptanes = 1/10 Solution
PhCH3 = 2
CH3CN = 4
Gum
Solid
NA
99.2
An additional solubility study of 1a in acetonitrile was carried out with Crystalꢀ16® using pure
1a, and the data was worked up with Crystalclear® to generate the solubility curve.5 It was found
that the solubility depended dramatically on the temperature, indicating a simple heating and
cooling sequence should be able to achieve a crystallization with reasonable recovery. Finally,
after further optimization of conditions, the recrystallization was performed in 1.8ꢀ2.0 volumes
of acetonitrile. Typically losses were 7ꢀ10% when the product was collected at 0 °C. However,
it was occasionally found that the slurry was too thick to stir. A few control experiments
determined that residual water was the cause. As a result, the control of the solvent swapping
process was modified to not only monitor the complete replacement of ethyl acetate, but also to
control the water level below 1000 ppm.
When the process was scaled up in the kilo lab, two batches, each starting with 1.2 kg of
hydroquinine, were carried out smoothly. Crude 1a was obtained by concentrating the workup
ACS Paragon Plus Environment
9

Organic Process Research & Development
Page 10 of 22
1
2
3
4
5
6
7
8
9
mixture to dryness on the rotary evaporator. Two batches of crude 1a were combined and
recrystallized. Interestingly, upon the charge of acetonitrile, the product first dissolved and then
precipitated immediately at room temperature, affecting the efficiency of the mixing. Good
mixing could resume after heating the batch to 80 °C, and the recrystallization proceeded as
expected from that point forward. Finally, 3.2 kg of 1a was obtained, representing a 73.3%
overall yield with 98 A% purity (Table 3).
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. Kilo Lab Synthesis of 1a
1a
Scale
(kg)
Amine 5a
(A%)
Batch
Crude Yield Yield Purity
(A%) (kg)
(%)
(A%)
1
2
1.2
1.2
89.6
91.9
95.4
3.2
93.7
73.3
98.3
When planning for further scaleꢀup, it was found that sourcing of highꢀquality hydroquinine at
more than 10 kg scale was surprisingly challenging and only possible through custom synthesis.
On the other hand, quinineꢀbased 1b behaved similarly to 1a in the targeted asymmetric reaction.
The alternative starting material, quinine, was commercially readily available.
The process was further optimized for the preparation of 1b by using ethyl acetate as both the
extraction solvent for amine 5b and the reaction solvent for the stage 3 coupling, eliminating the
second solvent swap. Moreover, the final solvent swap from ethyl acetate to acetonitrile was
performed without concentrating to dryness, eliminating the problem of fast precipitation
ACS Paragon Plus Environment
10

Page 11 of 22
Organic Process Research & Development
1
2
3
4
5
6
observed in the previous kilo lab campaign for the synthesis of 1a, making the new process
suitable for a pilot plant operation (Scheme 2).
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
Ph3P, DIAD, CH2Cl2
Ph3P, H2O
CH2Cl2
(PhO)2P(O)N3
5b
4b
quinine
EtOAc
1)
2)
6
Recrystallization, CH3CN
7b
1b
Scheme 2. Final Plant Process for the Synthesis of 1b
A complete hazard evaluation for the process of 1b was performed before the pilot plant
campaign. The reactants and intermediate streams were analyzed by Thermos Screening Unit
(TSU) and Differential Scanning Calorimetry (DSC), indicating thermal stability up to at least
265 °C. The synthesis of 1b from quinine is a threeꢀstep process containing at least 10 thermal
events. The calorimetry of each thermal event was analyzed independently using the HEL Autoꢀ
MATE software, and results are summarized in Table 4.
In general, the thermal events were not significant because the maximum achievable
temperatures were below the boiling points of the solvents and at least 200 °C below any
potential thermal decomposition points. The only noteworthy event was triphenylphosphine
solution dosing. If there were complete loss of control during the dosing, the maximum
ACS Paragon Plus Environment
11

Organic Process Research & Development
Page 12 of 22
1
2
3
4
5
6
7
8
9
achievable temperature would be sufficient to cause the solvent to boil. However, this boiling
was a cooling event and served as an additional safety factor for the process. In the new process,
the triphenylphosphine is dosed as a solution to provide better control than does portionꢀwise
addition of the solid.
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. Summary of Thermal Events for the Synthesis of 1b
Molar Heat
of Reaction.
Qr
Qr
Process
Step
∆T K
(°C)
Total E in
kJ (Qr)
(kJ/kg of
limiting
reagent)
(kJ/kg of
reaction
mass)
[kJ / mole of
Quinine
(LR)]
DIAD Dosing
5.00
4.69
500
469
51.5
44.9
162.1
34.3
28.0
Diphenyl
Phosphoryl Azide
Solution Dosing
152.0
DI Water Dosing
0.14
9.70
2.37
2.11
2.90
0.74
14
1.12
70.3
14.1
24.4
30.3
6.3
4.4
0.7
Triphenylphosphine
Solution Dosing
970
237
211
29
314.7
76.8
68.5
94.0
23.9
37.0
6.7
3N HCl Dosing
Conc. Aqueous
Ammonia Dosing
5.8
6 Solution Dosing
14.4
3.0
1N HCl Solution
Dosing
74
Conc. Aqueous
Ammonia Dosing
0.41
0.48
41
48
3.3
3.3
13.2
15.6
1.6
1.7
DI Water Dosing
Two batches were performed in the pilot plant at approximately 10 kg scale. The starting
quinine was > 95% pure, and contained less than 5% hydroquinine. The process proceeded as
ACS Paragon Plus Environment
12

Page 13 of 22
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
expected (Table 5), affording a total 26 kg of 1b in two batches. The assay of approximately 96
wt% was dictated by the purity of the starting quinine. Fortunately, the major impurity 1a is a
similarly effective catalyst for the targeted reaction. Hence, 1a as an impurity is not a concern for
the present application.
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. Pilot Plant Synthesis of 1b
1b
Quinine
(kg)
1a
Batch
Yield
(kg)
14.2
12.2
Yield
(%)
70
Assay
(wt%)
95.3
(A%)
1
2
11.0
10.3
3.9
2.8
65
96.4
Conclusion
In summary, we have demonstrated a robust, cost effective scalable process for the synthesis of
hydroquinineꢀ or quinineꢀbased thiourea catalysts 1a and 1b. The process was successfully
demonstrated in the pilot plant, affording over 14 kg of 1b over four stages in a single 100 gal
reactor. Moreover, the new process, consisting of three chemical reactions and one
recrystallization, was performed in a single reactor, significantly improving the practicability and
productivity of the synthesis.
Experimental Section
All materials were purchased from commercial suppliers. Unless otherwise noted, all reagents
and solvents were used as supplied. NMR spectra were obtained using a Bruker 400 MHz
spectrometer in the solvents indicated. HPLC spectra were collected on an Agilent 1200 series
ACS Paragon Plus Environment
13

Organic Process Research & Development
Page 14 of 22
1
2
instrument. Solubility was measured by Crystal 16® and the data was worked up with
Crystalclear®.
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
Kilo lab Preparation of 1a
To a 20 L jacketed glass reactor, equipped with condenser, thermocouple, and nitrogen sweep,
was charged hydroquinine (1.20 kg, 3.68 mole, limiting reagent), and triphenyphosphine (1.16
kg, 4.41 mole, 1.2 equiv.). The batch was purged with nitrogen, dichloromethane (7.2 L, 6.0 V)
was added, and the batch stirred at room temperature for approximately 30 minutes to obtain a
clear solution. After cooling to approximately 0 °C, diisopropyl azodicarboxylate (DIAD, 0.888
kg, 4.41 mole, 1.2 equiv.) was charged through an addition funnel while keeping the batch below
5 °C. The addition funnel was rinsed with an additional dichloromethane (40 mL) and the wash
was combined with the batch. After the batch temperature was stabilized at 0ꢀ5 °C, a solution of
diphenyl phosphoryl azide (DPPA, 1.22 kg, 4.41 mole, 1.2 equiv.) in dichloromethane (1.2 L, 1.0
V) was charged while keeping the batch temperature at approximately 0 °C. The addition line
was rinsed again with dichloromethane (100 mL) and combined with the batch. This addition
took approximately 1 hour. The batch was warmed to approximately 20 °C, and stirring
continued at this temperature for approximately 2 hours before an inꢀprocess control indicated a
complete reaction.
The reduction procedure started with cooling the batch to approximately 15 °C. A total of 1.62
kg (1.35 V) of DI water was charged, followed by a solution of triphenylphosphine (1.74 kg,
6.62 mole, 1.8 equiv.) in dichloromethane (1.8 L, 1.5 V) at approximately 20 °C (Caution: N2
offꢀgassing!). The batch was stirred at approximately 23 °C for 15ꢀ18 hours before quenching
with 3N HCl (2.8 L) to pH 2.2. After partitioning, the product was in the upper aqueous layer,
ACS Paragon Plus Environment
14

Page 15 of 22
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
which was washed twice with 3.12 L (2.6 V) of dichloromethane at ambient temperature. The
dichloromethane layers were discarded. The pH of the aqueous layer was adjusted to > 10 with
28% aqueous ammonia (750 mL, 0.625 V), and extracted with two 3.12 L (2.6 V) portions of
dichloromethane. The dichloromethane layers were combined and concentrated to dryness on
rotary evaporator, affording 1.52 kg of crude 5a as a yellow oil. The residue was dissolved in
dichloromethane (3.0 L, 2.5 V) and concentrated to dryness. The residue was dissolved again in
dichloromethane (3.0 L, 2.5 V), at which point the water content was 328 ppm by KF analysis.
This solution of crude 5a was used in the next step without further manipulation.
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
The coupling reaction commenced with charging the above solution into a 20 L jacketed glass
reactor, equipped with a condenser, thermocouple and nitrogen sweep. The container was rinsed
with dichloromethane (3.0 L, 2.5 V) and the rinse combined with the batch. The reaction mixture
was cooled to approximately 0 °C, and 3,5ꢀbis(trifluoromethyl)phenyl isothiocyanate (6) (898 g,
3.31 mole, 0.9 equiv.) in dichloromethane (2.4 L, 2 V) was charged while keeping the reaction
temperature below 10 °C. The addition took approximately 35 minutes to complete. The
resulting solution was warmed to approximately 20 °C and stirred for 1 hour until inꢀprocess
control indicated a complete reaction. The batch was transferred into a rotary evaporator and
concentrated to dryness under vacuum. The residue was reꢀdissolved in ethyl acetate (9.6 L, 8 V)
and charged back to the reactor. The batch was acidified with 1N HCl (4.0 L, 3.3 V) to
approximately pH 2. The batch was partitioned, and the upper organic layer was washed with of
DI water (4 L, 3.3 V), and then adjusted to pH >10 with 28% aqueous ammonia (950 mL, 0.8 V).
The reaction mixture was partitioned again and the bottom aqueous layer was discarded. The
upper organic layer was washed with DI water (2.4 L, 2 V), and concentrated to dryness on a
ACS Paragon Plus Environment
15

Organic Process Research & Development
Page 16 of 22
1
2
3
4
5
6
7
8
9
rotary evaporator. The residue was azeotropically dried with dichloromethane (39 L, 32.5 V) to
afford a waxy yellow solid, which was further dried under full house vacuum with a nitrogen
sweep at ambient temperature for an additional 2 days to furnish 1.99 kg (90.7% yield) of crude
1a in 95.2 A% purity with 614 ppm of water.
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
The second lot of crude 1a was prepared with 1.20 kg of hydroquinine using the same process,
affording 2.13 kg (97.1% yield) of product in 95.1 A% purity with 461 ppm of water.
The above two lots of crude 1a (4.12 kg) were combined and charged into a 20 L jacketed
glass reactor, equipped with a condenser, thermocouple and nitrogen sweep. The containers
were each rinsed with 2L (0.49 V) of acetonitrile and the rinse combined with the batch. An
additional portion of acetonitrile (3.4 L, 0.83 V) was charged. The product precipitated
immediately, and the stirrer was blocked. The reactor was opened and the solid was broken up
manually with a paddle. After the stirring was resumed, the reactor was reꢀsealed and the
atmosphere rendered inert with nitrogen. The batch was heated to 80 °C to furnish a clear
solution, cooled back to approximately 20 °C, and stirred for 1 hour before cooling further to
approximately 0 °C. The batch was stirred at this temperature for an additional 17 hours then
filtered on a 6 L Büchner funnel. The filter cake was washed twice with 2.0 L (0.49 V) of cold
acetonitrile, and dried in the funnel for approximately 42 hours until no weight loss was observed
over a period of 2 hours. Finally, 3.21 kg of 1a was obtained, representing a 73.3% overall yield
1
in 98.3 A% purity. H NMR (DMSOꢀd6, 400 MHz): δ = 0.75ꢀ0.82 (m, 4H), 1.15ꢀ1.27 (m, 3H),
1.39ꢀ1.46 (m, 2H), 1.55 (brs, 1H), 1.60ꢀ1.63 (m, 1H), 2.40ꢀ2.44 (m, 1H), 2.65ꢀ2.73 (m, 1H), 3.18
(dd, J = 3.6, 13.2 Hz, 1H), 3.26ꢀ3.29 (m, 1H), 3.97 (s, 3H), 6.02 (brs, 1H), 7.45 (dd, J = 2.4, 9.2
ACS Paragon Plus Environment
16

Page 17 of 22
Organic Process Research & Development
1
2
3
4
Hz, 1H), 7.61 (d, J = 4.8 Hz, 1H), 7.69 (s, 1H), 7.94 (brs, 1H), 7.96 (d, J = 9.2 Hz, 1H), 8.15 (s,
5
6
13
2H), 8.74 (d, J = 4.4 Hz, 1H), 8.92 (brs, 1H). 10.22 (brs, 1H). C NMR (100.59 MHz, DMSOꢀ
7
8
9
d6) δ 11.79, 24.73, 25.06, 26.73, 28.04, 36.57, 41.01, 55.55, 56.71, 59.27, 102.98, 119.07,
120.54, 123.14 (d, JCF = 273.6 Hz), 121.21, 127.96, 130.31 (q, JCF = 33.2 Hz), 131.13, 141.67,
143.96, 145.05, 147.58, 157.01, 179.37. HRMSꢀESI (m/z): [M+H]+ Calcd for C29H31F6N4OS:
597.2117; found: 597.2123.
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
Pilot Plant Synthesis of 1b:
A 100 gallon glassꢀlined jacketed reactor was inerted with nitrogen. Quinine (11.0 kg, 33.9
mole, limiting reagent) and triphenylphosphine (10.7 kg, 40.7 mole, 1.2 equiv.) were charged via
EZDock bags. This was followed by the charge of methylene chloride (88.2 kg, 6 V). The batch
was agitated at about 26 °C for approximately 2 hours to obtain a clear solution. After cooling
the batch to approximately 5 °C, a solution of diisopropyl azodicarboxylate (DIAD, 8.20 kg, 40.7
mole, 1.2 equiv.) was charged from an addition tank over approximately 1 hour while
maintaining the temperature below 10 °C. The addition tank and transfer lines were rinsed with
additional methylene chloride (2.70 kg, 1V), and the rinse combined with the batch. This was
followed by the charge of a solution of diphenylphosphoryl azide (DPPA, 11.3 kg, 40.7 mole,
1.2 equiv.) in methylene chloride (14.8 kg, 1 V) over approximately 2.5 hours while maintaining
the batch temperature below 10 °C. Following the addition, the batch was heated to
approximately 20 °C and agitated for an additional 19 hours until HPLC indicated a complete
reaction.
The batch was cooled to approximately 15 °C, and DI water (14.9 kg, 1.35 V) was charged,
followed by a solution of triphenylphosphine (16.0 kg, 61.0 mole, 1.8 equiv.) in methylene
ACS Paragon Plus Environment
17

Organic Process Research & Development
Page 18 of 22
1
2
3
4
5
6
7
8
9
chloride (21.7 kg, 1.5 V) over approximately 50 minutes (Caution: Offꢀgassing!). After the
addition, the batch was warmed to approximately 25 °C and stirred overnight until the reaction
was complete. Workup started with quenching the batch with 3 N HCl over approximately 1
hour until the pH reached 2.5. The agitation was stopped, and the batch was held at ambient
temperature for approximately 1 hour to obtain a clear phase cut. The upper aqueous layer was
washed twice with 38.4 kg of methylene chloride (5.2 V). The batch was adjusted with 28%
aqueous ammonia (15.8 kg) to pH 10.0, extracted with ethyl acetate (98.7 kg, 10 V), and washed
with saturated brine (37.2 kg). The resulting solution was distilled under vacuum (ca. 250
mmHg), while charging ethyl acetate in portions, until KF analysis indicated 410 ppm of water
in the solution. At the end of the distillation, the batch consisted of approximately 8 V of ethyl
acetate.
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
The batch was cooled to approximately 0 °C, and a solution of 3,5ꢀbis(trifluoromethyl)phenyl
isothiocyanate (6) (8.30 kg, 30.5 mole, 0.9 equiv.) in ethyl acetate (19.8 kg, 2V) was charged
into the batch over 30 minutes while maintaining the temperature at approximately 0 °C. After
the addition, the batch was warmed to approximately 20 °C and stirred for an additional 1 hour
until reaction completion. The batch was quenched with 1N HCl (37 L) until the pH reached 2.0ꢀ
2.5, stirred for an additional 20 minutes, and held until a clear phase cut was reached. The lower
aqueous layer was drained and discarded. The upper organic layer was washed with DI water
(37.5 kg, 3.4 V), and the pH was adjusted to over 10 with 28% aqueous ammonia. Additional DI
water (34.1 kg, 3.1 V) was charged, and the batch was agitated for 10 minutes. The agitation
was stopped, and the layers were allowed to separate. The lower aqueous layer was drained and
discarded. The upper organic layer was heated to 40 °C and washed with an additional portion
of DI water (34.4 kg, 3.1 V). Agitation was stopped and the batch was held for approximately 5
ACS Paragon Plus Environment
18

Page 19 of 22
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
hours at 40 °C, allowing for a clear phase cut. The lower aqueous layer was drained and
discarded. The upper organic layer was vacuum distillated (ca. 200 mmHg) while charging
acetonitrile (10 V) in portions, until the KF analysis indicated 339 ppm of water remaining. The
solution contained 43.3 wt% of 1b.
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
The batch was cooled to approximately 0 °C over 1 hour, and agitated at 0 °C for an additional
16 hours. Solids precipitated when the batch reached 31.5 °C. The product was isolated in an
Aurora filter, and the filter cake was washed with approximately 8.80 kg of acetonitrile. The
product was dried under vacuum at 50 °C for 3 days, affording 14.2 kg of catalyst 1b as a white
solid, representing a 70.7 % yield. The HPLC assay was 95.3 wt%, with 3.9 A% of the
hydroquinineꢀbased catalyst 1a as the major impurity. 1H NMR (DMSOꢀd6, 400 MHz): δ = 0.77ꢀ
0.83 (m, 1H), 1.22ꢀ1.28 (m, 1H), 1.56ꢀ1.63 (m, 3H), 2.28 (brm, 1H), 2.67ꢀ2.72 (m, 2H), 3.21 (dd,
J = 10.0, 13.2 Hz, 1H), 3.24ꢀ3.33 (m, 2H), 3.97 (s, 3H), 4.94 (d, J = 10.4 Hz, 1H), 5.00 (d, J =
17.2 Hz, 1H), 5.83 (ddd, J = 7.6, 10.4, 17.2 Hz, 1H), 6.04 (brs, 1H), 7.45 (dd, J = 2.8, 9.2 Hz,
1H), 7.62 (d, J = 4.8 Hz, 1H), 7.70 (S, 1H), 7.94 (S, 1H), 7.96 (d, J = 9.2 Hz, 1H), 8.15 (s, 2H),
13
8.74 (d, J = 4.8 Hz, 1H), 8.96 (d, J = 4.8 Hz, 1H), 10.21 (brs, 1H). C NMR (100.59 MHz,
DMSOꢀd6) δ 25.27, 26.92, 27.34, 40.96, 55.05, 55.61, 59.33, 103.02, 114.22, 115.78, 120.62
(brs), 121.24, 123.18 (q, J = 272.6 Hz), 127.97, 130.36 (q, J = 33.2 Hz), 131.18, 141.68, 141.89,
144.01, 144.91, 147.61, 157.08, 179.43. HRMSꢀESI (m/z): [M+H]+ Calcd for C29H29F6N4OS:
595.1961; found: 595.1975.
ASSOCIATED CONTENT
Supporting Information. HPLC methods used for in process control and final product analysis.
HPLC spectrums of the intermediates and final products 1a and 1b. NMR spectral data of
ACS Paragon Plus Environment
19

Organic Process Research & Development
Page 20 of 22
1
2
3
4
compounds 1a and 1b. Solubility curves of 1a and 1b in acetonitrile. These materials are
5
6
available free of charge via the Internet at http://pubs.acs.org.
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
AUTHOR INFORMATION
Corresponding Author
*Tel: 610ꢀ727ꢀ6419, email: yi.wang@tevapharm.com
ORCID
Yi Wang: 0000ꢀ0001ꢀ5496ꢀ7426.
The authors declare no competing financial interest.
Present Addresses
# Incyte Corporation. 1801 Augstine Cutꢀoff, Wilmington, DE 19803.
Author Contributions
The manuscript was written through contributions of all authors. All authors have given approval
to the final version of the manuscript.
ACKNOWLEDGMENT
We wish to thank Jianzhong Li, Jennifer Van De Rijn, and Raeann Wu for analytical
assistance; Vincent Djuhadi and Shahar Barak for procurement support. We also thank the
operation team including Crispin Cebula, Russell Craig, Robert Cummins, and Steven Mangos
for pilot plant support, as well as helpful discussion with Partha Mudipalli and Jing Fang.
ACS Paragon Plus Environment
20

Page 21 of 22
Organic Process Research & Development
1
2
3
4
REFERENCES
5
6
7
8
1) For reviews on bifunctional cinchona alkaloid catalysis: a) Singh, R. P.; Deng, L.
Cinchona Alkaloid Organocatalysis. In Science of Synthesis, Asymmetric
Organocatalysis; List, B., Maruoka, K., Eds.; Georg Thieme Verlag: Stuttgart, Germany,
2012; vol. 2, pp 41ꢀ117. b) Linus, S.; Filippo, S.; Dixion, D. J. Chem. Sci. 2012, 3, 942ꢀ
958. c) Connon, S. J. Chem. Comm. 2008, 2499ꢀ2510. d) Li, H.; Chen, Y.; Deng, L.
Cinchona Alkaloids. In Privileged Chiral Ligands and Catalysts; Zhou, Q.ꢀL., Eds.;
ThiemeꢀVCH: Weinheim, Germany, 2011, pp 361ꢀ408. For recent examples of thiourea
cinchona alkaloid catalyst promoted asymmetric reaction: e) Fukata, Y.; Asano, K.;
Matsubara, S. J. Am. Chem. Soc. 2013, 135, 12160ꢀ12163. f) Kong, S.; Fan, W.; Wu, G.;
Miao, Z. Angew. Chem. Int. Eng. 2012, 51, 8864ꢀ8867. g) Curti, C.; Rassu, G.;
Zambrano, V.; Pinna, L.; Pelosi, G.; Sartori, A.; Battistini, L.; Zanardi, F.; Casiraghi, G.
Angew. Chem. Int. Eng. 2012, 51, 6200ꢀ6204. h) De. S.; Das, M. K.; Bhunia, S.; Bisai, A.
Org. Lett. 2015, 17, 5922ꢀ5925. i) Engl, O. D.; Fritz, S. P.; Kaslin, A.; Wennemers, H.
Org. Lett. 2014, 16, 5454ꢀ5457.
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
2) For a review, see: a) Hernández, J. G.; Juaristi, E. Chem. Commun., 2012, 48, 5396ꢀ5409.
For examples of the synthesis of (S)ꢀpregabalin using modified cinchona alkaloid
catalysts, see : b) Moccia, M.; Cotigiani, M.; Monasterolo, C.; Torri, F.; Fiandra, C. D.;
Fuller, G.; Kelly, B.; Adama, M. F. A. Org. Process Res. Dev. 2015, 19, 1274ꢀ1281. c)
LeyvaꢀPerez, A.; GarciaꢀGarcia, P.; Corma, A. Angew. Chem, Int. Eng. 2014, 53, 8687ꢀ
8690. d) Park, S. E.; Nam, E. H.; Jang, H. B.; Oh, J. S.; Some, S.; Lee, Y. S.; Song, C. E.
Adv. Synth. Catal. 2010, 352, 2211ꢀ2217. For examples of the synthesis of (+)ꢀbiotin
using modified cinchona alkaloid catalysts, see e) Xiong, F.; Xiong, F.ꢀJ.; Chen, W.ꢀX.;
ACS Paragon Plus Environment
21

Organic Process Research & Development
Page 22 of 22
1
2
3
4
5
6
7
8
9
Jia, H.ꢀQ.; Chen, F.ꢀE. J. Hetero. Chem. 2013, 50, 1078ꢀ1082. f) Dai, H.ꢀF.; Chen, W.ꢀ
X.; Zhao, L.; Xiong, F.; Sheng, H.; Chen, F.ꢀE. Adv. Synth. Catal. 2008, 350, 1635ꢀ
1641. Choi, C.; Tian, S.ꢀK.; Deng, L. Synthesis 2001, 11, 1737ꢀ1741.
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
3) Twenty vendors for 1b were identified using Scifinder®. The list price was
approximately $ 680ꢀ1400/g.
4) Vakulya, B.; Varga, S.; Csámpai, A.; Soo´s, T. Org. Lett. 2005, 7, 1967ꢀ1969.
5) See Supporting Information for the solubility curves of compound 1a and 1b in
acetonitrile.
ACS Paragon Plus Environment
22