Continuous-Flow Synthesis of Phenothiazine Antipsychotics: A Feasibility Study

Article abstract of DOI:10.1002/ejoc.201801689

The continuous flow synthesis of a model phenothiazine antipsychotic is presented herein, using 3-chloropropionyl chloride as a central building block. The basic phenothiazine-derived scaffold is (atom)-efficiently and mildly synthesized with the aim to present continuous flow technology as a contributor to fast and efficient syntheses of challenging APIs, that are nowadays experiencing supply disruptions and global shortages.

Full text of DOI:10.1002/ejoc.201801689

A Journal of
Accepted Article
Title: Continuous flow synthesis of the phenothiazine antipsychotics: a
feasibility study
Authors: Marine Movsisyan, Laurens De Coen, Thomas Heugebaert,
Arno Verlee, Bart Roman, and Chris Victor Stevens
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.201801689
Link to VoR: http://dx.doi.org/10.1002/ejoc.201801689
Supported by

10.1002/ejoc.201801689
European Journal of Organic Chemistry
COMMUNICATION
Continuous flow synthesis of the phenothiazine antipsychotics: a
feasibility study
Marine Movsisyan,[a][b] Laurens M. De Coen,[a][b] Thomas S. A. Heugebaert,[a] Arno Verlee,[b] Bart I.
Roman[a] and Christian V. Stevens*[a]
Abstract: The continuous flow synthesis of a model phenothiazine
antipsychotic is presented herein, using 3-chloropropionyl chloride as
a central building block. The basic phenothiazine-derived scaffold is
(atom)-efficiently and mildly synthesized with the aim to present
continuous flow technology as a contributor to fast and efficient
syntheses of challenging APIs, that are nowadays experiencing
supply disruptions and global shortages.
an adaptation of a recently in-house developed flow process to
the production of antipsychotic agents of the phenothiazine class.
Phenothiazine-derived drugs form the largest of the 5 main
classes of antipsychotic drugs. They contain a phenothiazine and
a nitrogen-containing substituent attached to the termini of a
three-carbon spacer. Pipothiazine 1 is a typical antipsychotic of
the phenothiazine class for the treatment of chronic schizophrenia.
Other commercially available injection drugs such as pipothiazine
palmitate 2 and fluphenazine decanoate 3 and APIs such as
Perphenazine 4, Prochlorperazine 5 and Trifluoperazine 6 are
likewise used for the maintenance treatment of schizophrenia and
paranoid psychoses (Figure 1).16
Introduction
The production of active pharmaceutical ingredients (APIs) is
mainly conducted via labour-intensive and time-consuming batch
processes. The pharmaceutical industry has been slow to adopt
continuous processing technology, a field that has been of great
interest to the petrochemical industry in the continuous production
of high volumes of simple organic molecules.1 Regulatory
agencies, such as the US Food and Drug Administration (FDA)
encourage endeavours to make the switch from batch to
continuous manufacturing given the long-term savings in
resources such as time, money and space. Additional advantages
include the continuous - instead of post-production - monitoring of
drug quality and the enhanced process reliability and flexibility.2, 3
Big pharma companies such as GSK, Sanofi, Lonza, Eli Lilly and
Novartis have already put large efforts in flow chemistry
research.1 The industry, however, bases its decision on a switch
from batch to continuous manufacturing not only on regulatory
incentives and long-term technology prospects, but also on the
current technology readiness level (maturation). It is therefore
clear that additional technology de-risking in academia remains
necessary before an actual switch to continuous production will
be considered by pharma-manufacturing industries. There are
multiple research groups already focussing on facilitating the use
of continuous flow technology in the field op pharmaceuticals.4-14
More examples of the effectivity and reliability of continuous flow
technology for the production of APIs are required, preferably on
production scale.13, 15
Figure 1. Phenothiazine-derived drugs: pipothiazine (1), pipothiazine palmitate
(2), fluphenazine decanoate (3), Perphenazine (4), Prochlorperazine (5) and
Trifluoperazine (6).
The choice of the phenothiazine class of APIs in this illustrative
study stems from recent issues in the batch manufacturing of
these agents and ensuing important drug shortages. Overall,
more than 200 cases of drug shortages were reported by the FDA
during 2011-2014. The main causes of these shortages can be
In this study, we wish to contribute to such de-risking of
continuous flow manufacturing in a pharmaceutical context with
rooted back to mainly quality or manufacturing concerns, leading
17
to disruptions in manufacturing.13,
In the class of the
phenothiazine-derived drugs, pipothiazine palmitate
2 was
discontinued in 2015 by Sanofi due to global shortage of its active
pharmaceutical ingredient (Figure 1).18, 19 Just recently, Sanofi
also reported the permanent discontinuation of the supply of its
injectable fluphenazine decanoate 3 by the end of 2018, due to
manufacturing issues of the active pharmaceutical ingredient.20, 21
Both of these phenothiazines are structurally related to
pipothiazine 1 (Figure 1). In this study, we describe a continuous
way to synthesise the basic scaffold of pipothiazine 1 using an
alternative source of raw material, 3-chloropropionyl chloride.
[a]
Corresponding author: Prof. Dr. ir. Christian Stevens
Dr. ir. M. Movsisyan, ir. L. M. De Coen, Dr. ir. T. S. A. Heugebaert,
A. Verlee, and Dr. B. I. Roman
Department of Green Chemistry and Technology
Synthesis, Bioresources and Bioorganic Chemistry Research Group
Coupure Links 653, Ghent – Belgium
E-mail: chris.stevens@Ugent.be
[b]
Equal contribution.
Supporting information for this article is given via a link at the end of
the document.
The original reports on the synthesis and biological activity of a
series of phenothiazine-derived compounds were published by
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201801689
European Journal of Organic Chemistry
COMMUNICATION
Yale, Sowinski and Bernstein.22, 23 In their synthetic route, the N-
alkylated phenothiazine 11 was prepared from 1-bromo-3-
chloropropane 8 in two steps in the presence of a strong base
(Scheme 1 (a)).22-24 More recently, Moglioni et al. developed an
alternative procedure for synthesising phenothiazine compounds,
e.g. 16 from 3-chloropropionyl chloride 13 (Scheme 1 (b)).25-27 In
both methods, two purification steps are necessary. Route (a) and
(b) provide the end product 11 and 16 in an overall yield of 45%
and 31%, respectively. Furthermore, long reaction times, elevated
temperatures and excessive amounts of starting products are
needed in both procedures.
Scheme 2. The suggested synthesis route of phenothiazine-derived
compounds.
Results and discussion
Our flow strategy to model compound 16 (Scheme 3) consists of
two steps starting from 3-chloropropionyl chloride 13. The first
step involves the formation of the amide 14 by reacting the acid
Scheme 1. (a) The synthesis of a phenothiazine-derived compound 11 as
reported by Yale, Sowinski and Bernstein and (b) the synthesis of
phenothiazine-derived compound 16 as reported by Moglioni et al.
chloride 13 with phenothiazine 12. Next,
a nucleophilic
substitution using 4-piperidineethanol 15 furnishes the desired
phenothiazine-derived precursor 16. Both amide formation and
(subsequent) nucleophilic substitutions have been reported by
other groups as well, for example in the flow synthesis of
Gleevec.4
In this proof-of-principle study, we use 3-chloropropionyl chloride
as a three carbon building block for the continuous flow production
of phenothiazine-based precursor 16 (Scheme 2), the basic
scaffold of this class of antipsychotics. The envisioned procedure
(Scheme 2, indicated in green) is an extension of our previous
work on the continuous, safe and efficient synthesis of 3-
chloropropionyl chloride 13 from the cheap and readily available
acrylic acid 17 (Scheme 2, indicated in blue).28 The other building
blocks in the synthesis are the readily available and cheap
phenothiazine 12 (Scheme 1) and 4-piperidineethanol 15
(Scheme 1). The goal is to evaluate the two-step introduction of
the C3-linker 13 and to improve yield, atom economy and reaction
parameters such as temperature and reaction time of these two
key steps through enhanced process control. The reduction and
further esterification of 16 will not be studied here, as these are
reported to proceed smoothly in a batch set-up.25
Step one was optimized prior to telescoping the two-step
procedure. A solution of the acid chloride 13 (3 M, DMF) and a
solution of 12 (2 M, DMF) were pumped at equimolar ratios
through a tubular continuous flow reactor (Table 1). Within 20
minutes at room temperature, 66% of 3-chloropropionyl chloride
13 was converted into the desired amide 14 (Entry 1). Using 1
equivalent of triethylamine in the phenothiazine stream to
increase the conversion and further reduce reaction time, resulted
in a hampering of the flow in the reactor because of insoluble
triethylammonium chloride salts (Entry 2). At 60 °C, a slightly
higher conversion of acid chloride 13 was reached (75%), in the
absence of NEt3 (Entry 3). Small percentages of undesired
elimination product 19 were also present. Increasing the
temperature to 135 °C to increase conversion and reaction rate
allowed a full conversion of 3-chloropropionyl chloride 13 (Entry
6). Unsurprisingly, 15% of the α,β-unsaturated product 19 was
formed at this higher temperature. An intermediate temperature
of 100 °C gave almost full conversion of the acid chloride 13 (95-
98%), within 5 minutes, with only small amounts of elimination
product (5-6%) (Entry 4). Doubling the reaction time to 10 minutes
(entry 5) only had a minor effect on the conversion. The optimal
conditions for the initial formation of 14 were thus chosen by
balancing residence time and throughput, side-product formation
and temperature. These were set at 100 °C with a residence time
of 5 minutes with equimolar amounts of starting products 12 and
13 and base-free (Entry 4).
Table 1. Results of the continuous flow synthesis of amide 14.
Entry
Molar ratio
12:13
T
tr
Conv.[a]
[%]
Ratio[a]
14:19
100:0
(°C)
20
(min.)
20
1
1:1
66
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201801689
European Journal of Organic Chemistry
COMMUNICATION
into 16 but led to more side-product formation, as analyzed
through LC-MS (Entry 10). As such, an overall conversion of 73%
into the desired compound 16 was reached, using equimolar
amounts of starting materials 12, 13 and 15. It is noteworthy to
state that 1 equivalent of base proved to be sufficient, despite the
liberation of 2 equivalents of HCl. The crude product was
extracted, precipitated as its HCl salts and again extracted to yield
the final free amine product in 73%. In an operating time of 5
minutes, we were able to purify up to 1.0 grams of 16, giving a
hourly throughput of 12.4 grams.
2[b]
3
1:1
1:1
1:1
1:1
1:1
20
20
10
5
Clogging
85
60
94:6
4
100
100
135
97
93:7
5
10
99
90:10
6
5
100
84:16
[a] based on 1H-NMR analysis. [b] 1 eq. NEt3 in solution of 12.
Table 2. Results of the continuous flow synthesis of a phenothiazine-derived
precursor 16.
Entry
Additive
Eq.
15
1
T
tr
Observation[a]
(°C)
20
(min.)
1.85
3.7
2.6
3.7
2.6
3.5
3.5
5.2
5
1
/
No conversion
No conversion
No conversion
No conversion
No conversion
Clogging
2
/
1
100
100
130
130
100
100
100
100
3
/
2
4
/
1
5
/
2
6
K2CO3
K2CO3 + Sand
NEt3
1
Scheme 3. The continuous flow two-step procedure for the synthesis of a
phenothiazine-derived precursor 16.
7
1
Clogging
8
1
Clogging
9[b]
NEt3 in
H2O:MeOH
NEt3 in
H2O:MeOH
1
73% 16 – 7% 14
20% 19
–
–
Next, the displacement of the remaining chloride with 4-
piperidineethanol 15 was performed directly on the output of step
1, consisting of approximately 90% of 3-chloro-1-(10H-
phenothiazin-10-yl)-1-propanone 14, 5% side-product 19 and 5%
phenothiazine 12 (Scheme 3, Table 2). This avoids handling or
purification of the intermediate 14. The substitution was first
investigated with a solution of 4-piperidineethanol 15 (1.5 M,
DMF). The experiments included different temperatures (20-
130 °C), and investigated effects of altering the amounts of 4-
piperidineethanol 15 (1-2 equivalents) and varying the residence
times (1.85-3.7 minutes) (Entries 1-5, Table 2). In none of these
base free trials, the desired compound 16 was formed. As a
second approach and in line with the batch procedure reported by
Moglioni and coworkers25, an omnifit® glass column was packed
with K2CO3 (Entry 6) to ensure alkaline conditions for the
substitution reaction. After mixing the output of reactor 1 with the
solution of 4-piperidineethanol 15 (1 equivalent, 1.5 M, DMF), the
reaction mixture was pumped through the K2CO3-packed column
at 100 °C.29 Unfortunately, high pressure build-up did not allow us
to analyze the result of this procedure. Packing a combination of
K2CO3 and sand as an inert material again proved to result in a
high pressure build-up (Entry 7). Finally, instead of using solid
K2CO3 as base, 1 equivalent NEt3 was added in the solution of 4-
piperidineethanol 15 (1 equivalent, 1.5 M, DMF). This trial resulted
in clogging, most likely due to the presence of the
triethylammonium chloride salt (Entry 8). To counter these effects,
lowering the concentration and applying an aqueous system to
solubilize the generated salt would be necessary.
10[b]
1
100
10
70% 16 – 5% 14
25% 19
[a] Product ratio in reactor outlet, based on LC-UV/DAD analysis, 220 nm.
[b] no BPR used, viscosity of the reaction mixture provided enough system
pressure to prevent boil-out.
We were able to conduct the two-step synthesis of 16 in a
telescoped procedure starting from the C3-linker 3-
chloropropionyl chloride 13, without the need of intermediate
purification. The overall yield (73%) herein was significantly
higher than reported for the batch procedure25 (31%, after two
purification steps). Other improvements included reaction time
and temperature. Via the procedure in the continuous flow setup,
a total reaction time of 10 minutes and a temperature of 100 °C
were sufficient to reach 73% conversion towards the desired
phenothiazine product 16. In the batch procedures, a total
reaction time of two days and elevated temperatures over 130 °C
were required.25 Furthermore, in the herein disclosed method
equimolar amounts of starting products 12, 13 and 15 and of NEt3
are utilized, while the batch procedure requires excessive
amounts, e.g. 2.4 equivalents of 13 and 3.8 equivalents of K2CO3.
To further obtain pipothiazine derivative 18, the high yielding
borane-reduction, as previously reported by Moglioni et al., and
further esterification can be performed.25
Conclusion
A 0.3 M solution of 4-piperidineethanol 15 (1 equivalent) and NEt3
(1 equivalent) in H2O:MeOH (1:1) was linked to the output of step
1 (Scheme 3). The reaction was conducted at 100 °C, as
crystallization of the product occurred at lower temperatures. Use
of this solution of 4-piperidineethanol 15 at elevated temperatures
resulted in formation of the desired product 16 within 5 minutes
(Entry 9). The intermediate product 14 and the elimination product
19 were also present in the reaction mixture. Prolonging the
residence time to 10 minutes, did not result in higher conversion
In conclusion, the continuous flow synthesis of a model
phenothiazine antipsychotic 16 was described. The procedure
uses 3-chloropropionyl chloride 13, which can readily be obtained
from acrylic acid, as a central building block.28 This C3-linker was
evaluated to be an effective intermediate to synthesize the basic
phenothiazine-derived scaffold (atom)-efficiently; in terms of
stoichiometry of starting material, and mildly; in terms of reaction
time and temperature. This study presents continuous flow
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201801689
European Journal of Organic Chemistry
COMMUNICATION
technology as a contributor to fast and efficient syntheses of
challenging APIs, which are nowadays subject to supply
disruptions and global shortages. This proof-of-concept can be
extended to analogous syntheses of phenothiazine-derived APIs
such as Fluphenazine 3, Perphenazine 4, Prochlorperazine 5 and
Trifluoperazine 6.
10.
11.
C. M. Griffiths-Jones, M. D. Hopkin, D. Jönsson, S. V. Ley, D. J.
Tapolczay, E. Vickerstaffe and M. Ladlow, J. Comb. Chem., 2007, 9,
422-430.
L. F. Raveglia and G. A. Giardina, Future Medicinal Chemistry, 2009, 1,
1019-1023.
12.
13.
C. Wiles and P. Watts, Future Medicinal Chemistry, 2009, 1, 1593-1612.
A. Adamo, R. L. Beingessner, M. Behnam, J. Chen, T. F. Jamison, K. F.
Jensen, J.-C. M. Monbaliu, A. S. Myerson, E. M. Revalor, D. R. Snead,
T. Stelzer, N. Weeranoppanant, S. Y. Wong and P. Zhang, Science,
2016, 352, 61-67.
Experimental section
14.
15.
R. E. Ziegler, B. K. Desai, J.-A. Jee, B. F. Gupton, T. D. Roper and T. F.
Jamison, Angew. Chem. Int. Ed., 2018, 57, 7181-7185.
K. P. Cole, J. M. Groh, M. D. Johnson, C. L. Burcham, B. M. Campbell,
W. D. Diseroad, M. R. Heller, J. R. Howell, N. J. Kallman, T. M. Koenig,
S. A. May, R. D. Miller, D. Mitchell, D. P. Myers, S. S. Myers, J. L. Phillips,
C. S. Polster, T. D. White, J. Cashman, D. Hurley, R. Moylan, P. Sheehan,
R. D. Spencer, K. Desmond, P. Desmond and O. Gowran, Science, 2017,
356, 1144-1150.
The continuous flow synthesis of phenothiazine-derived compound 16 is
described (Scheme 3). A first stream containing a 3 M solution of 3-
chloropropionyl chloride 13 in DMF and a second stream containing a 2 M
solution of phenothiazine 12 in DMF are continuously pumped in a tubular
continuous flow reactor using syringe pumps (Chemyx). The tubular
reactor is constructed by means of PFA tubing with an internal diameter of
0.020” (0.50 mm) and an outer diameter of 1/16”. The reactor has a total
internal volume of 3 mL (IDEX), allowing a residence time of 5 minutes.
The output of the first reactor is (by means of a T-junction, IDEX)
immediately coupled to a third reagent stream which is pumped using a
reciprocating pump (ReaXus) and contains a 0.3 M solution of 4-
piperidineethanol 15 and triethylamine (1 equivalent) in H2O:MeOH (1:1).
The combined streams are pumped through a second tubular flow reactor.
The tubular reactor for step 2 is PFA-polymer with an internal diameter of
1 mm and an outer diameter of 1.6 mm with a total volume of 15 mL
(BOLA), allowing a residence time of 5 minutes. The entire system is
heated to a temperature of 100 °C in a bath containing silicon oil. Samples
are collected at the outlet of reactor two. After basic extraction, the desired
product 16 was obtained by precipitation in its corresponding HCl salt and
neutralisation to yield a pale yellow oil in 73%.
16.
17.
A. K. Ghose, V. N. Viswanadhan and J. J. Wendoloski, Journal of
Combinatorial Chemistry, 1999, 1, 55-68.
Food and Drug Administration, Strategic Plan for Preventing and
Mitigating
Drug
Shortages
https://www.fda.gov/downloads/Drugs/DrugSafety/DrugShortages/UCM
372566.pdf).
18.
19.
M. Toms, 2014.
P. Haddad, M. Taylor, M. Patel and D. Taylor, British Journal of
Psychiatry, 2015, 206, 521-521.
20.
21.
Medicine Optimisation Team, 2017.
Discontinuation
of
two
different
depot
antipsychotics,
https://holeousia.com/2017/06/22/discontinuation-of-two-different-
depot-antipsychotics/).
22.
23.
24.
H. L. Yale, F. Sowinski and J. Bernstein, Journal of the American
Chemical Society, 1957, 79, 4375-4379.
H. L. Yale and F. Sowinski, Journal of the American Chemical Society,
1960, 82, 2039-2042.
Acknowledgements
P. C. Bossle, C. P. Ferguson, W. E. Sultan, W. J. Lennox, G. E. Dudley,
T. H. Rea and J. I. Miller, Journal of Medicinal Chemistry, 1976, 19, 370-
373.
Financial support from the Agency for Innovation by Science and
Technology in Flanders (IWT), 3M and Fund for Scientific
Research-Flandres (FWO) is gratefully acknowledged.
25.
26.
G. P. Sarmiento, G. Y. Moltrasio and A. G. Moglioni, Arkivoc, 2009, 2009,
33-41
G. P. Sarmiento, R. G. Vitale, J. Afeltra, G. Y. Moltrasio and A. G.
Moglioni, European Journal of Medicinal Chemistry, 2011, 46, 101-105.
Argentine Pat., 2006.
Keywords: Microreactors • API • Antipsychotic
27.
28.
29.
1.
2.
D. Obermayer and T. Glasnov, Chimica Oggi - Chemistry Today, 2017,
35, 16-18.
3M Innovative Properties Company, EP17208982.3, 2018.
A. Baker, M. Graz, R. Saunders, G. J. S. Evans, I. Pitotti and T. Wirth,
Journal of Flow Chemistry, 2015, 5, 65-68.
C. Cao, Flow Chemistry: Pathway for Continuous API Manufacturing,
https://www.pharmasalmanac.com/articles/flow-chemistry-pathway-for-
continuous-api-manufacturing).
3.
Z. Brennan, FDA calls on manufacturers to begin switch from batch to
continuous
production,
https://www.in-
pharmatechnologist.com/Article/2015/05/01/FDA-calls-on-
manufacturers-to-begin-switch-from-batch-to-continuous-
production?utm_source=copyright&utm_medium=OnSite&utm_campai
gn=copyright, (accessed 01 May 2015).
4.
5.
6.
7.
8.
9.
M. D. Hopkin, I. R. Baxendale and S. V. Ley, Chem. Commun., 2010, 46,
2450-2452.
B. Gutmann, D. Cantillo and C. O. Kappe, Angew. Chem. Int. Ed., 2015,
54, 6688-6728.
N. S. Suveges, R. O. M. A. de Souza, B. Gutmann and C. O. Kappe, Eur.
J. Org. Chem., 2017, 2017, 6511-6517.
I. R. Baxendale and S. V. Ley, Bioorg. Med. Chem. Lett., 2000, 10, 1983-
1986.
A. Kirschning, H. Monenschein and R. Wittenberg, Angew. Chem. Int.
Ed., 2001, 40, 650-679.
H. Lange, C. F. Carter, M. D. Hopkin, A. Burke, J. G. Goode, I. R.
Baxendale and S. V. Ley, Chemical Science, 2011, 2, 765-769.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.201801689
European Journal of Organic Chemistry
COMMUNICATION
COMMUNICATION
The continuous flow synthesis of a model phenothiazine antipsychotic is presented
in this work, using 3-chloropropionyl chloride as a central building block. The basic
phenothiazine-derived scaffold is (atom-)efficiently and mildly synthesized with the
aim to present continuous flow technology as a contributor to fast and efficient
syntheses of challenging APIs, that are nowadays experiencing supply disruptions
and global shortages.
Marine Movsisyan, Laurens M. De
Coen, Thomas S. A. Heugebaert, Arno
Verlee, Bart I. Roman and Christian V.
Stevens*
Page No. – Page No.
Continuous flow synthesis of the
phenothiazine antipsychotics: a
feasibility study
[a]
Corresponding author: Prof. Dr. ir. Christian Stevens
Dr. ir. M. Movsisyan, ir. L. M. De Coen, Dr. ir. T. S. A. Heu
A. Verlee, and Dr. B. I. Roman
Department of Green Chemistry and Technology
Synthesis, Bioresources and Bioorganic Chemistry Resea
Coupure Links 653, Ghent – Belgium
This article is protected by copyright. All rights reserved.