Continuous Flow Conditions for High Temperature Formation of a Benzodioxan Pharmaceutical Intermediate: Rapid Scaleup for Early Phase Material Delivery

Article abstract of DOI:10.1021/acs.oprd.9b00499

We have developed a continuous flow method to enable rapid scaleup of an enantioenriched benzodioxan intermediate. We propose that the reaction proceeds through an intramolecular SNAr cyclization. Interestingly, traditional SNAr conditions resulted in impurity formation. When the starting material 2 was treated with alkali base in polar aprotic solvents, an undesired product regioisomer was observed. The formation of this regioisomer impurity could be suppressed by using a less polar solvent, an organic base, and high temperature. By employing continuous flow technology, these high temperature conditions could be scaled up to produce 540 g of the desired intermediate. The continuous flow reactor allowed for rapid thermal equilibration, which minimized problematic product decomposition by reducing the time that the product was exposed to high temperature.

Full text of DOI:10.1021/acs.oprd.9b00499

Subscriber access provided by University of Otago Library
Full Paper
Continuous flow conditions for high temperature formation of a benzodioxan
pharmaceutical intermediate: Rapid scaleup for early phase material delivery
Pengbin Li, Shaohui Yang, Ruiheng Zhu, Baoquan Sun, Zhongbo Li,
Ping Huang, Jonas Buser, Jose Miguel Minguez, and Sarah Jean Ryan
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.9b00499 • Publication Date (Web): 21 Feb 2020
Downloaded from pubs.acs.org on February 22, 2020
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 41
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
Continuous flow conditions for high temperature
formation of a benzodioxan pharmaceutical
intermediate: Rapid scaleup for early phase material
delivery
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 NAMES
Pengbin Li,i Shaohui Yang,i Ruiheng Zhu,i Baoquan Sun,i Zhongbo Li,i Ping Huang,i
Jonas Y. Buser,ii Jose Miguel Minguez,iii Sarah J. Ryan*ii
AUTHOR ADDRESS
i) Shanghai SynTheAll Pharmaceutical Co., Ltd. (“STA”), No. 7 Building, #90 Delin Rd,
WaiGaoQiao Free Trade Zone, Shanghai, 200131, People’s Republic of China
ii)
Eli Lilly and Company, Small Molecule Design and Development (SMDD), Lilly
Technology Center North, 1223 W. Morris St, Indianapolis, IN, 46221
ACS Paragon Plus Environment
1

Organic Process Research & Development
Page 2 of 41
1
2
3
4
5
iii)
Eli Lilly and Company, Discovery Chemistry Research and Technologies (DCRT),
6
7
8
Avda. de la Industria 30, 28108 Alcobendas, Madrid, Spain.
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
For Table of Contents Only
ACS Paragon Plus Environment
2

Page 3 of 41
Organic Process Research & Development
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
ABSTRACT
We have developed a continuous flow method to enable rapid scaleup of an
enantioenriched benzodioxan intermediate. We propose that the reaction proceeds
through an intramolecular SNAr cyclization. Interestingly, traditional SNAr conditions
resulted in impurity formation. When the starting material 2 was treated with alkali base
in polar aprotic solvents an undesired product regioisomer was observed. The formation
ACS Paragon Plus Environment
3

Organic Process Research & Development
Page 4 of 41
1
2
3
4
5
6
7
8
9
of this regioisomer impurity could be suppressed by using less polar solvent, organic base
and high temperature. By employing continuous flow technology, these high temperature
conditions could be scaled up to produce 540 g of the desired intermediate. The
continuous flow reactor allowed for rapid thermal equilibration, which minimized
problematic product decomposition by reducing the time that the product was exposed to
high temperature.
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
Continuous flow, high temperature, benzodioxan
ACS Paragon Plus Environment
4

Page 5 of 41
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
INTRODUCTION
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
One of the biggest challenges faced by the pharmaceutical industry is the low rate of
clinical success. A recent study analyzed the data of over 21,000 compounds from
January 2000 to October 2015.1 The authors report that only 13.8% of drug programs
actually make it from Phase I to approval. Alarmingly, this percentage is significantly lower
for small molecule chemical entities than it is for biologics.2,3
To address this challenge, the pharmaceutical industry needs to implement new
strategies to achieve clinical proof of concept. Chemists responsible for delivering
molecules in the early Phase portfolio are being tasked with reducing the time and cost
of providing material to the clinic. Our goal is to commit minimal investment, while still
enabling rapid execution of the key clinical studies required to progress a molecule
through development. These studies should provide meaningful data that facilitates the
decision to either rapidly terminate or rapidly advance to later Phase development.
Advancement from Phase I to Phase II significantly increases the likelihood of achieving
ACS Paragon Plus Environment
5

Organic Process Research & Development
Page 6 of 41
1
2
3
4
5
6
7
8
9
approval (35.1%).1 One way to deliver a molecule quickly is by utilizing technology that
facilitates scaleup of literature routes, avoiding the time and monetary investment
required to develop a novel synthetic approach.
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
We were interested in targeting the enantiopure benzodioxan 1 as an intermediate for the
synthesis of an active pharmaceutical ingredient in preclinical development (Equation 1).
Several different methods to prepare enantiopure benzodioxan molecules are reported in
the literature. Catechol starting materials can be used to dispilace enantiopure
electrophiles (Scheme 1, A).4 In 2016, Tang demonstrated that achiral monoallylated
catechols could undergo enantioselective aryloxyarylation of alkenes to furnish
enantiopure benzodioxans, albeit with limited substrate scope (Scheme 1, B).5 In 2001,
Buchwald reported that enantiopure alcohols could substitute aryl-bromides via Pd-
catalyzed intramolecular C—O etherification (Scheme 1, C).6 Cai and coworkers
expanded on this work, disclosing that the intramolecular C—O etherification can also be
employed in the desymmetrization of diols using Pd or Cu in combination with chiral
ligands.7 More recently, Zhang and Senanayake have reported the enantioselective
ACS Paragon Plus Environment
6

Page 7 of 41
Organic Process Research & Development
1
2
3
4
5
6
7
8
hydrogenation of substituted benzo[1,4]dioxines using Rh or Ir and chiral phosphine
ligands (Scheme 1, D).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
Equation 1. Benzodioxan pharmaceutical intermediate
O
O
CN
O
1
Cl
Cl
Scheme 1. Methods of preparation of benzodioxans: A) Catechol preparation method.4
B) Enantioselective allylic aryloxyarylation.5 C) Intramolecular C—O etherification.6 D)
Enantioselective benzo[1,4]dioxine hydrogenation.8
A. Catachol Preparation of (S)-Piperoxan4(b)
B. Enantioselective Pd-Catalyzed Alkene Aryloxyarylation5
1. K2CO3, acetone 
2. H2, Pd, MeOH
Pd2(dba)3, Ligand,
NaOtBu, C6F6
3. Tf2O, TEA, DCM,
HO
O
HO
O
O
O
O
piperidine
+
PhBr
+
BnO
Br
N
Br
HO
C. Pd-Catalyzed C O Etherification6,7
D. Rh or Ir Catatlyzed Enantioselective Hydrogenation8(c)
Pd(OAc)2, TrixiePhos,
Cs2CO3, toluene, 70 °C
Ru or Ir, Ligand
Hydrogen
OH
R1
Br
O
O
O
O
O
O
R1
R1
R1
O
benzo[1,4]dioxine
ACS Paragon Plus Environment
7

Organic Process Research & Development
Page 8 of 41
1
2
3
4
5
6
7
8
9
The best method for the preparation of 1 may involve one of these reported methods, or
it could involve a completely novel, undisclosed method of preparation. However, due to
the potential regioselectivity challenges of a catechol approach,9,10 in combination with
the urgent need for material to fund preclinical toxicology studies, these synthetic
approaches were not investigated for this early-phase material delivery. Fortunately, the
synthesis of 1 was previously reported in the literature.11 The reported synthetic method
involves the enantiopure alcohol 2 (Scheme 2), which was subjected to high temperature
conditions in the presence of NaH for SNAr cyclization to the benzodioxan 1. According
to the literature report for the preparation of 1, the SNAr reaction required immediate
cooling. We speculated that the product might not be stable under the reaction conditions
over an extended period. In our hands, we observed lower isolated product yields when
the reaction was conducted at larger scale in batch mode, presumably due to the
extended cooldown time that is required as the reactor volume increased (see below:
Table 2, Entries 2 and 3). To address the poor yield associated with product
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
ACS Paragon Plus Environment
8

Page 9 of 41
Organic Process Research & Development
1
2
3
4
5
decomposition, we became interested in applying continuous flow to the production of 1
6
7
8
(Scheme 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
Scheme 2. This work: Continuous flow SNAr Cyclization
T
Backpressure: 750 psi
OH
F
O
O
CN
O
4m 1/8
HC tubing
104m 1/8
HC tubing
O
2
CN
High Pressure
Syringe Pump
O
1
220 °C GC oven
250.8 mL
Cl
In 10 vol Diglyme
with 3.2 equiv. DBU
Cl
Cl
Cl
Continuous reactors have improved the ability of chemists to explore extreme reaction
conditions in a safe and controlled manner.12 The inherent nature of continuous flow
processing allows high product throughput using a small reactor size. This small reactor
size permits a wide temperature range with rapid heat-up and cooldown time. Many plug
flow reactors (PFRs) can withstand high pressure, allowing high temperature reactions to
occur in the liquid phase even when operating well above the boiling point of the solvent.
ACS Paragon Plus Environment
9

Organic Process Research & Development
Page 10 of 41
1
2
3
4
5
6
7
8
9
We initially became interested in continuous flow technology for its safety and control
aspects.13,14 Continuous flow can also allow rapid scaleup of literature conditions that
might be difficult to achieve in batch mode.15 This should aid in reducing the lead time
and development costs to deliver material for early phase clinical trials.
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
Herein we report the results of continuous flow production of enantioenriched benzodioxin
1 to enable rapid material delivery for preclinical studies (Scheme 2). The reaction can be
safely and effectively completed above the boiling point of the solvent. Furthermore, the
use of continuous flow technology addresses problematic product decomposition.
RESULTS AND DISCUSSION
To investigate the production of 1 we began by targeting the enantiopure alcohol 2. The
reported method for the preparation of 2 began with acetophenone 3, which was treated
with pyrrolidone hydrogen tribromide to selectively monobrominate the methyl group
(Scheme 3).11 The resulting bromide 4 was next reacted with phenol 5 to afford the
ether 6. The ketone functionality of 6 was reduced under Corey-Bakshi-Shibata (CBS)
ACS Paragon Plus Environment
10

Page 11 of 41
Organic Process Research & Development
1
2
3
4
5
6
7
8
reduction conditions with (S)-CBS catalyst and the borane diethylaniline complex to
produce the alcohol 2, which was elaborated to more complex enantiopure products.16
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
Scheme 3. Literature Reported Synthesis of Compound 2.11
STEP 2
HO
F
CN
STEP 1
O
5
O
2-pyrrolidone HBr3
MeOH:DCM (2:1), rt
Br
K2CO3, acetone, rt
O
O
4
3
Cl
Cl
Cl
Cl
STEP 3
(S)-CBS (1M in toluene)
PhNEt2•BH3, THF, rt
OH
F
O
F
O
O
O
O
CN
6
2
CN
Cl
Cl
Cl
Cl
The preparation of ketone 6 was achieved with modifications to the methods reported in
the literature (Scheme 4). The bromination reagent in the Step 1 was changed from
pyrrolidone hydrogen tribromide to readily available N-bromosuccinimide. The modified
reaction employed a catalytic amount of toluenesulfonic acid (TsOH) to favor an ionic
pathway, affording good conversion of starting material and high selectivity for the
ACS Paragon Plus Environment
11

Organic Process Research & Development
Page 12 of 41
1
2
3
4
5
6
7
8
9
monobromination product. The Step 2 substitution reaction solvent was modified from
acetone to DMF. The higher dielectric point of DMF compared to acetone (36.7 vs 20.7)
allowed better solubility of the base and permitted a faster rate of reaction. The use of
DMF as a solvent also simplified the workup procedure. The product could be directly
precipitated and isolated from the reaction mixture by the addition of water and
isopropanol.
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
With ketone 6 in hand we investigated the Step 3 reduction. Unfortunately, in our hands
the enantioselectivity of the Step 3 CBS reduction to provide the alcohol 2 was poor
(79.1% ee, Scheme 4).16 Alternate reaction conditions for Step 3 were quickly developed
and the CBS reduction was modified to a transition metal catalyzed, asymmetric transfer
hydrogenation.Ruthenium systems are well established for the enantioselective reduction
of aromatic ketones.17 As such, we investigated several ruthenium catalysts with formic
acid: trimethylamine complex as the hydride source (Table 1). Due to the high ee of the
product and good commercial availability, RuCl[(R,R)-Tsdpen](p-cymene) was selected
for further scaleup (Table 1, Entry 1). The use of this ruthenium catalyst resulted in high
ACS Paragon Plus Environment
12

Page 13 of 41
Organic Process Research & Development
1
2
3
4
5
quality alcohol 2 with high ee on 963 g scale (95.8% ee, Scheme 4). The alcohol 2 was
6
7
8
isolated by crystallization to investigate the Step 4 SNAr cyclization.
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
Scheme 4: Optimization to compound 2
STEP 2
HO
F
CN
STEP 1
5
NBS (1.1 eq),
K2CO3 (1.5 eq), 5 (1.0
eq), DMF (6 vol),
10-15 °C, 3h
TsOH (30 mol%),
EtOAc (10 mL/g),
20-30 °C, 4h
O
O
Br
O
O
4
3
Cl
Cl
Cl
Cl
STEP 3
OH
F
O
F
O
Conditionsa or b
O
O
O
CN
6
2
CN
Cl
Cl
Cl
Cl
a(S)-CBS reduction condtions
11 g scale, 79.2% ee, 60% isolated yield
b transfer hydrogenation conditions
963 g scale, 95.8% ee, 93% isolated yield
aCBS conditions: (S)-CBS catalyst (1M in toluene, 1.05 eq), PhNEt2•BH3 (2 eq), THF, -20
°C
b Transfer hydrogenation conditions: RuCl[(R,R)-Tsdpen] (p-cymene) (0.7 mol%), formic
acid (4.0 eq), triethylamine (1.6 eq), IPA, 55 °C
ACS Paragon Plus Environment
13

Organic Process Research & Development
Page 14 of 41
1
2
3
4
5
6
7
8
9
Table 1: Transfer Hydrogenation conditions for the formation of 2
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
Entry1 Catalyst
2
2
(area%) (ee%)
1.
RuCl[(R,R)-
94.4
97.9
Tsdpen]
cymene)
(p-
(p-
2.
3.
RuCl[(R,R)-
96.2
96.7
97.2
96.1
Msdenb]
RuCl[(R,R)-
Fsdpen]
cymene)
4.
RuCl[(R,R)-
96.3
96.6
97.1
96.9
Tsdeneb]
5.2
RuCl[(R,R)-
Tsdeneb]
1All reactions were completed on a 20 mmol scale using 2 mol% Ru catalyst, 2 mL/g
formic acid:TEA (5:2) and 10 mL/g toluene at 50 °C for 18 h. 2MeCN was used in place
of toluene as a solvent
ACS Paragon Plus Environment
14

Page 15 of 41
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
With alcohol 2 in hand, we focused our attention on the SNAr cyclization. The conditions
reported in the literature required high temperature (140–150 °C) and utilized sodium
hydride as a base. We began by repeating these conditions. When the reaction was
performed at 140 ºC with NaH (60% in mineral oil) in diglyme, we were able to obtain 54%
isolated product yield (Table 2, Entry 1). In order to avoid the use of NaH, we turned to
milder bases. When DBU was used as a base the reaction required 20 h to achieve
complete conversion of starting material when completed on a 50 g scale (Table 2, Entry
3). While analysis of a HPLC chromatogram indicated high area% purity, the isolated
product yield was only 41%. We suspected that this was due to product decomposition
over an extended period at high temperature, resulting in poor mass balance (Table 2,
Entry 2 vs Entry 3). To address this product decomposition, we first investigated the use
of alternate bases that are known to mediate SNAr cyclization at lower temperature. When
K2CO3 was used in place of DBU in diglyme there was no conversion of starting material
(Table 2, Entry 4). Stronger bases such as tBuOK led to decomposition of starting
material, resulting in a significant number of impurities (Table 2, Entry 5). These impurities
were not characterized further. We were initially excited to observe complete conversion
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
ACS Paragon Plus Environment
15

Organic Process Research & Development
Page 16 of 41
1
2
3
4
5
6
7
8
9
of starting material in the presence of K2CO3 or Cs2CO3 in DMF at 80 °C (Table 2, Entries
6 and 7). Unfortunately, closer inspection of the products by NMR spectroscopy revealed
that a mixture of 1 and its regioisomer 1b had formed in a 3:2 ratio.18 Presumably, 1b is
generated via substitution of the C—O bond prior to cyclization (see below, Scheme 5).
The enantiomeric purity of 1b was not determined.
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 2. Initial reaction exploration for the formation of 1
OH
F
O
O
CN
O
O
O
Conditions
CN
O
+
O
O
CN
2
1
1b
Cl
Cl
Cl
Cl
Cl
Cl
Entry Solvent
Base (eq.)
Temperature (2) area% (1) area% (isolated yield)
1
2
3
Diglyme NaH (1.4)
Diglyme DBU (1.2)
Diglyme DBU (1.4)
140 °C (3 h)
140 °C
0
54
NA
NA (67%) 10 g scale
140 °C (3 h) 33
140 °C (20 h) 0
91
54
85 (41%) 50 g scale
4
5
6
“
K2CO3 ( 2.0) 100 °C
3
Toluene tBuOK ( 2.0) 25 °C
20
0
0
DMF
K2CO3 (2.0)
80 °C
78 (3:2 mixture of 1 and
1b)
7
DMF
Cs2CO3 (2.0) 80 °C
0
78 (3:2 mixture of 1 and
1b)
ACS Paragon Plus Environment
16

Page 17 of 41
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
We propose that the formation of regioisomer 1b begins with epoxide formation to liberate
the phenoxide (Scheme 5). This phenoxide can then attack either of the electrophilic
carbons of the epoxide to provide 2 and its regioisomer 2b. The formation of regioisomers
as a result of epoxide opening under similar reaction conditions has been reported previously.19
Cyclization of 2b would result in formation of the undesired product regioisomer 1b,
cyclization of 2 would lead to the desired product 1. This pathway is proposed to be
operational in the presence of alkali bases and coordinating solvents due to the
stabilization of ionic intermediates. A similar pathway involving formation of the epoxide
followed by coordination of the Lewis acidic potassium or cesium could also result in the
formation of intermediate 2b.
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 5. Proposed mechanism for the generation of the regioisomer 1b
ACS Paragon Plus Environment
17

Organic Process Research & Development
Page 18 of 41
1
2
3
4
5
6
7
8
F
H
O
F
O
O
H
NC
O
O
O
O
+
F
OH
cyclization
epoxide formation
CN
O
O bond substitution
C
+
O
O
O
CN
2
1b
CN
2b
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
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
The use of DBU in diglyme at high temperature had resulted in low isolated yield (41%,
Table 2, Entry 2). However, we believe that this is due to product decomposition over an
extended period at high temperature. To minimize product exposure to the forcing
reaction conditions we turned to continuous flow preparation of 1. Due to its small
footprint, continuous flow technology permits rapid thermal equilibration minimizing heat-
up and cooldown time. Based on our previous experience with high temperature, high
pressure flow reactions,20 we sought to apply a PFR to the preparation of 1.21 When the
use of a PFR is suggested, it is useful to conduct small-scale batch screening reactions
to rapidly test discrete variables in parallel. These are readily performed in 0.5–1 mL
volume pressure reactors constructed of Swagelok or similar steel or Hastelloy parts.22,
23 Building on our previous results with DBU in diglyme (Table 2), we screened a range
of organic bases in diglyme using these small 1 mL pressure vessels (Table 3, Entry 1).
ACS Paragon Plus Environment
18

Page 19 of 41
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
TEA, DIPEA, pyridine, lutidine, N-methyl morpholine, and diazabicyclo non-5-ene (DBN)
did not provide significant conversion of starting material at 190 °C after 30 m (Table 3,
Entries 1a–f). DBU remained our preferred base for this reaction (Table 3, Entry 1g). We
next investigated the effect of temperature and base equivalents (Table 3, Entry 2).
Increasing the temperature to 240 °C significantly increased the conversion of starting
material from 38% to 89 % (Table 3, Entry 1g vs Entry 2a). This was improved further by
increasing the equivalents of base to 2.6 eq., giving 99% conversion of starting material
and 86 area% of the desired product (Table 3, Entry 2b). The temperature could be
lowered to 220 °C with no reduction in product area% (Table 3, entry 2c). Maintaining the
equivalents of base between 2.6 and 7.8 provided similar conversion and product area%
(Table 3, Entries 2c–e).
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
We were interested in the effect of solvent on the reaction outcome (Table 3, Entry 3).
THF and dioxane did not result in any improvements to the process that utilized diglyme.
However, diglyme did provide higher isolated product purity than THF or dioxane and was
therefore preferred for early phase material delivery. The polar, aprotic solvent DMAc
ACS Paragon Plus Environment
19

Organic Process Research & Development
Page 20 of 41
1
2
3
4
5
6
7
8
9
initially appeared to provide promising results by HPLC analysis (Table 3, Entry 3b).
However, subsequent analysis by chiral chromatography revealed that the product ee
was low under these conditions (only 50.0 % ee). As such, we selected DBU as the base
(between 2.6 and 7.8 eq.) in diglyme to investigate in a PFR (Table 3, Entries 2c–e). The
undesired impurity 1b was not observed under these reaction conditions (See Table 2
and Scheme 5 for the structure 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
Table 3. Small-scale batch reaction to investigate continuous flow conditions
OH
F
O
O
CN
Conditions
O
O
O
CN
2
Cl
1
Cl
Cl
Cl
Entry Base (eq.)
Solvent
Tempb
(2)
area%c
(1) area%c (in situ
enantiomeric purity)d
1a
b
c
TEA (1.3)
Diglyme 190 °C 97
2.2
1.8
1.6
2.0
2.3
7.5
33
DIPEA (1.3)
97
96
97
96
35
62
N-Me morpholine (1.3)
Pyridine (1.3)
Lutidine (1.3)
DBN (1.3)
d
e
f
g
DBU (1.3)
ACS Paragon Plus Environment
20

Page 21 of 41
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
2a
b
DBU (1.3)
(2.6)
Diglyme 240 °C 11
79
0.8
220 °C 1.2
0.3
86 (94% ee)
c
DBU (2.6)
(5.2)
86
d
87
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
e
(7.8)
0.0
86
3a
b
DBU (5.2)
THF
220 °C 2.3
85
DMAc
Dioxane
0.0
11
89 (50% ee)
c
73
aAll reactions were completed in a 1 mL swagelok port connector. Reactions were carried
out on a 0.4 mL scale, with 16 mL/g solvent for 30 min. bReactions were heated in a GC
c
oven fitted with a rotating reactor holder to allow agitation. area% was determined by
HPLC analysis using MeCN as an eluant. dchiral purity was determined by HPLC analysis
using MeOH as an eluant.
A 16.4 m length Hastelloy tubular PFR was selected to test the continuous flow conditions.
The reactor contained a 0.125 inch outer diameter and 0.028 inch wall thickness. The
reactor was placed in a GC oven to allow for rapid heating. The back-pressure was set to
750 psi using a back-pressure regulator. After the GC oven had reached the desired
temperature, a solution of the starting material 2 and DBU in diglyme was pumped into
ACS Paragon Plus Environment
21

Organic Process Research & Development
Page 22 of 41
1
2
3
4
5
6
7
8
9
the reactor using a HPLC pump. The reaction mixture eluted from the GC oven into an
additional 1 m of Hastelloy tubing. This allowed cooling of the reaction mixture with the
ambient atmosphere prior to product collection. A thermal detector was installed before
the back-pressure regulator to ensure that the temperature of the solution had cooled to
room temperature prior to collection. The rate of heat transfer of the Hastelloy tubing used
in the reactor was tested. We discovered that it can be heated to the desired temperature
(220 °C) within 1 minute when placed in the preheated GC oven and can be cooled back
to room temperature within 1 min under ambient conditions after eluting from the oven.
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
Initial PFR runs were conducted with 15 g of compound 2 to evaluate the chemistry (Table
4). The approximate residence time was calculated by dividing the reactor volume by the
flow rate (V/Q). This calculation provides the time necessary to process one volume of
reactor fluid at the entrance conditions. It does not account for thermal expansion of the
solvent in the high temperature reactor and cannot be assumed to provide the actual
mean residence time (t). The residence time distribution can be calculated using tracer
techniques that measure the concentration of a tracer moving through the reactor under
ACS Paragon Plus Environment
22

Page 23 of 41
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
the desired conditions.24 This can provide useful information about the fluid dynamics of
a plug-flow system. However, it was not considered necessary for this early phase
material delivery. The V/Q was set to 30 min. Each experiment was run for more than 1.5
h before sampling the solution to allow approximately 3 reactor volume turnovers. The
starting material 2 that was used for each of these reactions was 95.2% ee.
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 PFR study began by repeating the conditions identified in batch. The reaction utilized
3.2 eq of DBU, in 15 mL/g of diglyme at 220 °C (Table 4, Entry 1). To test that steady
state was achieved after 3 reactor volume turnovers we began sampling the reactor after
1.5 h reaction time. Three aliquots were taken every 5 minutes and analyzed by HPLC.
The conversion of starting material and product area% were consistent throughout the
reaction progress suggesting that steady state had been achieved. We next investigated
the effect of solvent volumes on the reaction outcome. When 10 mL/g of solvent was used
in place of 15 mL/g with a V/Q of 30 min. the conversion of starting material improved
slightly from 95% to 99% (Table 4, Entries 1 and 2). The in situ chiral purity was 92% ee
when either 10 or 15 mL/g were used. The isolated yield was also very similar for both
ACS Paragon Plus Environment
23

Organic Process Research & Development
Page 24 of 41
1
2
3
4
5
6
7
8
9
experiments (62 or 63% respectively). However, the purity of the isolated product was
superior when 10 mL/g was used in place of 15 mL/g, presumably due to better
conversion of starting material (Table 4, Entries 1 and 2). 10 mL/g was preferred due to
the improved product purity, and reduced waste that accompanies the use of lower
solvent volumes. This did not affect our ability to use a PFR for the SNAr cyclization as
the starting material and product remained soluble under these conditions. Increasing the
temperature from 220 °C to 240 °C did not significantly alter the conversion of starting
material or area% of product. However, the in situ chiral purity was lower at higher
temperature (88% ee vs 92% ee). This chiral purity remained low when the V/Q was
reduced from 30 to 20 minutes at 240 °C. While the chiral purity could be upgraded during
product isolation, the lower in situ ee resulted in a slight reduction of the isolated yield
(58%) (Table 4, Entry 4). We decided to move ahead with the scaleup using 10 mL/g
diglyme at 220 °C with a V/Q of 30 min.25
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. Continuous flow conditions for the generation of 1
ACS Paragon Plus Environment
24

Page 25 of 41
Organic Process Research & Development
1
2
3
4
5
6
OH
F
O
O
CN
Conditions
O
O
O
CN
2
Cl
1
Cl
Cl
7
Cl
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
Entry Solvent
volumes
V/Q
Temperature
220 °C
(2)
(1) area% (in
situ ee)
76
area%
4.5
1
15
30 min (1.58 h)
(1.66 h)
4.5
76
(1.75 h)
4.7
76 (93% ee)
Isolated yield, purity and ee
62% yield, 97% purity, 95% ee
2
10
30 min (1.58 h)
(1.66 h)
220 °C
1.0
1.0
1.0
80
80
(1.75 h)
81 (92% ee)
Isolated yield, purity and ee
63% yield, 99% purity, 95% ee
3
10
30 min (1.58 h)
(1.66 h)
240 °C
0.1
0.1
0.1
79
80
(1.75 h)
79 (88% ee)
Isolated yield, purity and ee
60% yield, 99% purity, 94% ee
4
10
20 min (1.58 h)
(1.66 h)
240 °C
0.6
0.6
0.6
82
81
(1.75 h)
81 (91% ee)
Isolated yield, purity and ee
aReactions were completed on a 15 g scale with 3.2 eq. of DBU
58% yield, 98% purity, 97% ee
The SNAr cyclization was completed using 850 g of 2 (after potency correction) in a PFR.
The reaction progress was monitored over time by HPLC (Graph 1). Sample analysis
ACS Paragon Plus Environment
25

Organic Process Research & Development
Page 26 of 41
1
2
3
4
5
6
7
8
9
began after 3 reactor volume turnovers to assure steady state had been achieved . The
product area% remained between 80 and 85 area% during the 24 h run. The starting
material amount was also consistently between 0 and 2 area% throughout the reaction.
The entire reaction solution was collected after the product was pushed out of the reactor
with diglyme, and the in situ yield was determined to be 64.4%. After work up the product
was isolated in 61.8% yield with 99% purity and 95.1% ee. The product loss in the mother
liquor was 5%. The mass balance of the isolated product was higher than expected 104%,
suggesting a slight error in the analysis of the crude reaction solution.
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
Graph 1. Reaction progress over time for the generation of 1
3.2 equiv. DBU
10 mL/g diglyme
220-225 °C, 30 min V/Q
OH
F
O
O
CN
O
O
O
CN
2
Cl
1
Cl
Cl
Cl
ACS Paragon Plus Environment
26

Page 27 of 41
Organic Process Research & Development
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
CONCLUSION
In conclusion, we have developed a continuous flow method to permit rapid scaleup of
benzodioxan intermediate 1 at high temperature. The use of continuous flow technology
allowed the literature conditions to be employed with only minor modification. This
avoided the need for a route redesign and route scouting investigations. The starting
alcohol 2 was produced with high ee in good purity via a ruthenium catalyzed transfer
hydrogenation approach. Subjecting 2 to high temperatures with DBU as a base in
ACS Paragon Plus Environment
27

Organic Process Research & Development
Page 28 of 41
1
2
3
4
5
6
7
8
9
diglyme prevented formation of an undesired regioisomer 1b which was observed under
more traditional SNAr conditions (alkali base in DMF). The desired product 1 was
produced in 62% isolated yield, allowing production of 540 g of product with 95 wt% purity
(99% area% purity) in 30 h. The reactor was set up employing a back-pressure regulator
to achieve high pressures. These high pressures kept the reaction mixture in the liquid
phase at high temperature. The V/Q was set to 30 minutes to allow rapid elution of the
product and avoid product decomposition that was associated with product instability
during prolonged reaction time under the high temperature conditions.
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
General Experimental
Purchased starting materials and reagents were used without further purification. HPLC analysis
(IPC and purity testing) were performed on Shimadzu LC-20A instrument using a Waters Xbridge
C18 (75 mm* 4.6 mm, 3.5 μm) column. The analytical method utilized 2 mobile phases; mobile
phase A consisted of 0.05% TFA in H2O (v/v) and mobile phase B consisted of 0.05% TFA in
ACN (v/v). The gradient elution method was completed as follows: 0.01 min, 15% of mobile phase
B; 15.00 min, 95% of mobile phase B; 18.01 min, 15% of mobile phase B. Eluting material was
ACS Paragon Plus Environment
28

Page 29 of 41
Organic Process Research & Development
1
2
3
4
5
6
7
8
9
detected using a UV detector set at 220 nm. Samples were prepared as follows: for compound 3,
0.25 mg/mL in ACN; for compound 4, 0.5 mg/mL in ACN; for compound 5, 0.4 mg/mL in ACN;
for compound 6, 2 & 1, 0.25 mg / mL in ACN. HPLC analysis (chiral purity) was performed on
a Shimadzu LC-20A Instrument; Analysis method for compound 2: column, Chiralpak IC (250
mm* 4.6 mm, 5 μm); mobile phase, IPA: Hexane (20 : 80, v/ v); Sample preparation: compound
2 was diluted to 5.0 mg/mL by EtOH; Analysis method for compound 1: column, Chiralpak IA
(250 mm* 4.6 mm, 5 μm); mobile phase, IPA: Hexane (15: 85, v/ v); sample preparation,
compound 1 was diluted to 1.0 mg/mL for chiral purity analysis. LCMS analysis was performed
on Agilent HPLC 1200 with 6120 Quadrupole LC/MS detector. 1H NMR (400 MHz) and 13 C (100
MHz) spectra was recorded on Bruker ultrashield 400, AVANCE II 400.
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-(4-((3,4-dichlorobenzyl)oxy)phenyl)ethan-1-one (3)
To a round bottom flask was added 4-hydroxyacetophenone (100 g, 1.0 eq., 0.73 mol). The flask
was charged with DMF (300 mL), then K2CO3 (153 g, 1.5 eq., 1.11 mol) and 3,4-dichlorobenzyl
chloride (158 g, 1.1 eq., 0.81 mol) were added. The mixture was heated to 60–65 °C and stirred
for 18–20 h at this temperature. After complete conversion of 4-hydroxyacetophenone (as
monitored by HPLC analysis), the mixture was allowed to cool to 30–35 °C. Then EtOAc (1 L)
and water (1.2 L) were added. The biphasic mixture was stirred and then the layers were separated.
The organic layer was washed with 0.5 L water. The organic layer was concentrated to 100–200
mL under vacuum. n-Heptane (300 mL) was then charged and the mixture was concentrated to
100–200 mL. An additional 100 mL of EtOAc was charged. The obtained solution was heated to
60–65 °C with stirring. n-heptane (500 mL) was next added via dropwise addition while
ACS Paragon Plus Environment
29