Scalable Continuous Flow Process for the Synthesis of Eflornithine Using Fluoroform as Difluoromethyl Source

Article abstract of DOI:10.1021/acs.oprd.8b00318

The development of a scalable telescoped continuous flow procedure for difluoromethylation of a protected amino acid with fluoroform (CHF₃, R-23) gas and subsequent high temperature deprotection to provide eflornithine, an important Active Pharmaceutical Ingredient (API), is described. Eflornithine is used for the treatment of sleeping sickness and hirsutism, and it is on the World Health Organization's list of essential medicines. Fluoroform is produced in large quantities as a side product in the manufacture of polytetrafluoroethylene (PTFE, Teflon). Fluoroform is an ozone-benign and nontoxic gas, but its release into the environment is forbidden under the Kyoto protocol owing to its high global warming potential. The existing manufacturing route to eflornithine uses chlorodifluoromethane (CHClF₂, R-22) which will be phased out under the Montreal protocol; therefore, the use of the fluoroform presents a viable cost-effective and more sustainable alternative. The process parameters and equipment setup were optimized on laboratory scale for the two reaction steps to improve product yield and scalability. The telescoped flow process utilizing fluoroform gas was operated for 4 h to afford the target molecule in 86% isolated yield over two steps with a throughput of 24 mmol/h.

Full text of DOI:10.1021/acs.oprd.8b00318

Subscriber access provided by The University of Texas at El Paso (UTEP)
Full Paper
Scalable Continuous Flow Process for the Synthesis of
Eflornithine using Fluoroform as Difluoromethyl Source
Manuel Köckinger, Christopher A. Hone, Bernhard Gutmann, Paul
Hanselmann, Michael Bersier, Ana Torvisco, and C. Oliver Kappe
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.8b00318 • Publication Date (Web): 29 Oct 2018
Downloaded from http://pubs.acs.org on October 29, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 26
Organic Process Research & Development
1
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
Scalable Continuous Flow Process for the Synthesis of Eflornithine
using Fluoroform as Difluoromethyl Source
Manuel Köckinger, †,‡ Christopher A. Hone, †,‡ Bernhard Gutmann,§ Paul
Hanselmann,§ Michael Bersier, § Ana Torvisco∥ and C. Oliver Kappe*,†,‡
† Center for Continuous Synthesis and Processing (CC FLOW), Research Center Pharmaceutical
Engineering GmbH (RCPE), Inffeldgasse 13, 8010 Graz, Austria
‡
Institute of Chemistry, University of Graz, NAWI Graz, Heinrichstrasse 28, A-8010 Graz,
Austria
§ Microreactor Technology, Lonza AG, CH-3930 Visp, Switzerland.
∥ Institute of Inorganic Chemistry, Graz University of Technology, Stremayrgasse 9, 8010 Graz,
Austria
____________________
* Corresponding author. E-mail: oliver.kappe@uni-graz.at
ACS Paragon Plus Environment

Organic Process Research & Development
Page 2 of 26
2
1
2
3
4
Graphical contents entry
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
ACS Paragon Plus Environment

Page 3 of 26
Organic Process Research & Development
3
1
2
3
4
Abstract
5
6
7
8
The development of a scalable telescoped continuous flow procedure for difluoromethylation of a
protected amino acid with fluoroform (CHF3, R-23) gas and subsequent high temperature
deprotection to provide eflornithine, an important Active Pharmaceutical Ingredient (API), is
described. Eflornithine is used for the treatment of sleeping sickness and hirsutism, and is on the
World Health Organization’s list of essential medicines. Fluoroform is produced in large
quantities as a side product in the manufacture of polytetrafluoroethylene (PTFE, Teflon®).
Fluoroform is an ozone-benign and nontoxic gas, but its release into the environment is forbidden
under the Kyoto protocol owing to its high global warming potential. The existing manufacturing
route to eflornithine uses chlorodifluoromethane (CHClF2, R-22) which will be phased out under
the Montreal protocol, therefore the use of the fluoroform presents a viable cost effective and
more sustainable alternative. The process parameters and equipment setup were optimized on
laboratory scale for the two reaction steps to improve product yield and scalability. The
telescoped flow process utilizing fluoroform gas was operated for 4 h to afford the target
molecule in 86 % isolated yield over two steps with a throughput of 24 mmol/h.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Keywords: continuous-flow; α-difluoromethylornithine; difluoromethylation; eflornithine;
fluoroform; gas-liquid transformations
ACS Paragon Plus Environment

Organic Process Research & Development
Page 4 of 26
4
1
2
3
4
INTRODUCTION
5
6
7
8
Eflornithine, also known as α-difluoromethylornithine (DFMO, 1), is an Active Pharmaceutical
Ingredient (API) on the World Health Organization’s list of essential medicines. DFMO is used
to treat the second stage of African trypanosomiasis (sleeping sickness). In addition, DFMO is
also used to treat opportunistic infections with Pneumocystis carinii pneumonia, a form of
pneumonia found in people with a weak immune system suffering from conditions such as
acquired immunodeficiency syndrome (AIDS).1 It has also been explored as chemopreventive
agent in cancer therapy with minor success.2 Today, its main use is to treat excessive facial hair
growth on women (hirsutism). The topical cream (Vaniqa®) significantly reduces the
psychological burden of those affected.3
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
CHF2
H2N
O
H2N
OH
difluoromethylornithine (DFMO, )
1
D,L--
Figure 1. Structure of D,L-α-difluoromethylornithine (elfornithine, DFMO, 1) as free base.
Topical administrations use eflornithine hydrochloride monohydrate.3
Eflornithine’s mechanism of action is well understood.4 It acts as a “suicide inhibitor” of
ornithine decarboxylase (ODC), an enzyme responsible for the committed step in polyamine
biosynthesis. Eflornithine irreversibly binds to the active site of ODC, which results in the
suppression of catalytic activity for conversion of the natural substrate. The polyamines produced
by OCD are associated with growth processes and reduced apoptosis.2
Currently, the employed manufacturing strategy for eflornithine 1 (Scheme 1)5 commences by
condensing diethylmalonate and acrylonitrile to afford diethyl 2-(2-cyanoethyl)malonate (2). The
afforded malonate is then difluoromethylated by chlorodifluoromethane (CHClF2, R-22) under
basic conditions.5,6 Subsequent hydrogenolysis of nitrile 3 provides the lactam 4 which is
transformed to the amide 5 through treatment with ammonia gas. The following Hofmann
rearrangement and hydrolysis affords 1. The Hoffmann rearrangement poses safety challenges
caused by the possibility of a run-away reaction.7 The reported overall yield is ~26 % for this
ACS Paragon Plus Environment

Page 5 of 26
Organic Process Research & Development
5
1
2
3
4
5
synthetic route.5 Alternative routes, which still utilize chlorodifluoromethane, reported yields in
the range of 37% to 40%.8
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
1. NaH in DMF
2. CHClF2 (2-3 bar)
6 h, 30 °C
1. Na (1 h, 10 °C)
NC
CO2Et
CO2Et
CO2Et
NC
CO2Et
CO2Et
2.
(1 h, 10 °C)
CN
F2HC
(82%)
(81%)
CO2Et
2
3
1. NaOCl, NaOH
2 h, 80 °C
2. HCl
O
O
CO2H
NH2
H2 (6 bar)
Raney Ni
F2HC
F2HC
EtO2C
(75%)
H2N
(64%)
NH3 (7 bar)
10 h, 40 °C
F2HC
NH
NH
H2NOC
(81%)
3 h, 90 °C
5
4
1
Scheme 1. Industrial synthesis of eflornithine starting from diethylmalonate.5
There are challenges associated with using gases such as NH3, H2 and R-22 in batch reactors, in
particular in terms of mass transfer and safety. Furthermore, R-22 will be completely phased out
under the Montreal protocol because of its ozone depleting potential.9 The Montreal protocol is
an international treaty to ensure protection of the earth´s ozone layer by phasing out the
production of compounds that are responsible for ozone depletion such as
chlorofluorohydrocarbons (CFCs). The phasing out of R-22 will have a significant impact on its
production, availability and cost, particularly in industrialized countries. Therefore, we aimed to
identify a more cost-effective and sustainable difluoromethyl source as an alternative to R-22,
and subsequently design an industrial relevant process for the synthesis of eflornithine utilizing
the new difluoromethyl source.
There is a plethora of difluoromethylation agents available, including CHF2OTf, TMSCF2Br,
(EtO)2POCF2Br and PhCOCF2Cl.10 These reagents cover the needs for small scale laboratory
synthesis very well, but generally are impractical for large scale industrial processes because of
their limited commercial availability, low atom economy and high cost per mole. Fluoroform
(CHF3, R-23), on the other hand, is a large-volume gaseous side product in the production of
polytetrafluoroethylene (PTFE, Teflon®) which can potentially serve as alternative to R-22.
Fluoroform is an ozone-benign and nontoxic gas (bp. = −83 °C). However, the release of
fluoroform into the environment is forbidden under the Kyoto protocol due to its high global
ACS Paragon Plus Environment

Organic Process Research & Development
Page 6 of 26
6
1
2
3
4
5
6
7
8
9
warming potential (GWP). Fluoroform’s global warming potential is 14 800 times higher than
CO2 over a century.11 Consequently, fluoroform needs to be converted to less harmful substances
before release into the environment. Currently, this is achieved mainly through destructive means
such as thermolysis, plasmolysis or catalytic hydrolysis.12,13 The more preferable option would be
to use fluoroform as a feedstock for industrial scale manufacturing of difluoro- or
trifluoromethylated fine chemicals or APIs. Nevertheless, this has been an unyielding endeavor
for a long time, partially caused by the very low reactivity of the gas (pKa = 25–28 in water).13
Recently the first synthetically valuable transformations using fluoroform as CF3-source13-16 and
CHF2-source17,18 have been reported.18b Trifluoromethylation reactions were conducted by
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
−
deprotonating fluoroform with either an electrogenerated14 or a strong base and trapping of CF3
(trifluoromethyl anion) with a suitable acceptor (TMSCl,13 DMF14,15 or Cu(I)16). The formed
intermediates were then reacted with a variety of electrophiles. Most notably, Mikami and co-
workers reported the treatment of a variety of carbonyl compounds with strong lithium bases
such as lithium diisopropylamide (LDA) or lithium bis(trimethylsilyl)amide (LiHMDS) and
fluoroform to afford their corresponding difluoromethylated analogue.17 The mechanism
presumably involves the formation of an electrophilic singlet difluorocarbene formed by rapid α-
−
elimination of fluoride from CF3 .
The described literature protocols using fluoroform gas operate mostly in batch mode. Scaling up
a gas-liquid batch reaction to production quantities, however, poses unique challenges. To
establish the required stoichiometry, a sufficient amount of gas must be dissolved in the reaction
media. The solubility of a gas increases as a function of pressure (Henry’s law), therefore reactors
have to be pressurized to ensure sufficient gas dissolution. However, pressurizing a huge
inventory of gas and reaction media poses significant safety risks. Commercial batch reactors can
typically operate at up to 6 bar, higher pressures require more specialized equipment. Therefore
gas-liquid reactions can benefit massively from a safety and process operation viewpoint by using
continuous processing.19 Historically, the fine chemicals and pharmaceutical industry operated by
using flexible multipurpose batch plants, but recently there has been a paradigm shift from batch
manufacturing to continuous processing.20 This shift has been driven by a need to lower
operating costs, increase safety, achieve higher product quality and lower solvent usage for API
and fine chemical manufacturing. Continuous processing meets those goals by providing
excellent control over process parameters such as residence time, temperature and mixing
ACS Paragon Plus Environment

Page 7 of 26
Organic Process Research & Development
7
1
2
3
4
5
6
because of the small channel dimensions used. The small reactor volume minimizes the risks
associated with handling hazardous chemistry thus making gas-liquid reactions inherently safer
and more controlled.
7
8
9
Recently, efforts have been made to utilize fluoroform gas within continuous flow reactors.
Grushin and co-workers developed a continuous flow protocol for the cupration of fluoroform to
CuCF3, which they demonstrated as a trifluoromethylating agent.21 Ley and coworkers monitored
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
19
fluoroform consumption in real-time by using a benchtop flow F NMR for a continuous-flow
nucleophilic trifluoromethylation.22 Our group has reported using fluoroform in a 3D printed
stainless steel continuous flow reactor for the difluoromethylation of diphenylacetonitrile,23
a
reaction which was originally developed by Mikami and co-workers.17 Most relevant to the
present work, we recently communicated a general continuous flow strategy for the
difluoromethylation of protected amino acids (Scheme 2).24 We achieved this by deprotonation of
benzylidene protected amino acid esters at −30 °C with LiHMDS (2 equiv) and subsequent
reaction with excess of fluoroform (3 equiv) at −15 °C in a continuous flow system. After
warming to room temperature the reaction mixture was quenched with water. The process was
performed at 12 bar to ensure complete dissolution of fluoroform. The obtained protected
difluoromethylated amino acid esters were hydrolyzed in a microwave reactor using superheated
HCl and isolated by precipitation. Although successful, this approach had certain limitations from
a process chemistry perspective, such as the excess of fluoroform employed - necessary to cope
with pressure fluctuations of the system. The highly viscous substrates used were only bench-
stable for ~24 h. Additionally, we were unable to demonstrate scalability of the process due to
precipitation inside the back pressure regulator (BPR) which prevented operation over longer
times.
ACS Paragon Plus Environment

Organic Process Research & Development
Page 8 of 26
8
1
2
3
4
5
6
7
8
9
R1 CO2R2
N
-30 °C
4 min
-15 °C
12 min
25 °C
4 min
250 µL/min
250 µL/min
12 bar
Ph
0.5 M in THF
2 mL
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
6 mL
2 mL
LiHMDS
1 M in THF
R1 CO2H
F2HC
R1 CO2R2
HCl
F2HC
NH2
N
2 equiv

MFC
Ph
8.3 mL/min
3 equiv
Scheme 2. Continuous flow Cα-difluoromethylation with fluoroform (previous work).24
The aim of the research described in this article is to address the challenges associated with our
preliminary difluoromethylation approach described above (Scheme 2), and develop a scalable
continuous difluoromethylation protocol using fluoroform with an attached telescoped high-T/p
deprotection reaction with higher throughput. Furthermore, we were interested in increasing the
sustainability of the process by finding alternatives to the use of THF and CHCl3 as solvents as
they are classified as highly hazardous and problematic, respectively.25
RESULTS AND DISCUSSION
Pressure Stabilization Experiments. One of the main objectives of this work was to reduce the
fluoroform input and, more importantly, the fluoroform remaining at the end of the process,
because its discharge into the environment is strictly regulated. The main reason for using 3 equiv
of fluoroform in our previous report24 was the pressure fluctuation caused by accumulation of
inorganic precipitate inside the Swagelok backpressure regulator (BPR). The pressure fluctuation
resulted in a variation in the fluoroform flow rate over the course of the reaction. We proposed
three approaches to handle the excess use of fluoroform: (a) the use of gas-liquid membrane
separation technology, based on a gas permeable and liquid impermeable membrane,26 to remove
and recycle the surplus fluoroform; (b) improved pressure control in order to prevent fluoroform
flow rate fluctuation; or (c) decrease the fluoroform input to a minimum and react the surplus
with base to more benign compounds. Although intriguing at first, we decided against using a
membrane separation technique because of almost certain membrane fouling caused by the
precipitate and the high solubility of fluoroform in THF.
ACS Paragon Plus Environment

Page 9 of 26
Organic Process Research & Development
9
1
2
3
4
5
6
7
8
9
We commenced our investigation by utilizing our previously reported flow setup (Scheme 2).24
The flow setup comprised of two syringe pumps to introduce a 0.5 M solution of substrate in
THF (Feed 1, 2 mL sample injection loop), and a commercial 1 M solution of LiHMDS in THF
(Feed 2, 4 mL sample injection loop). Feed 1 and 2 were mixed in a Y-shaped connector
submerged within a cooling bath at −30 °C. The substrate was converted to the enolate in a 2 mL
residence time loop (reactor 1). Subsequently, the enolate was reacted with fluoroform, which
was introduced through a second Y-shaped connector. The amount of fluoroform added was
controlled using a calibrated mass flow controller (MFC, Bronkhorst-EL). The combined mixture
then passed through a second cooled residence time loop at −15 °C (reactor 2) and then through a
third residence loop at room temperature (reactor 3). Cooling of reactor 2 was necessary due to
increased solubility of fluoroform at lower temperatures and the delayed precipitation of LiF. The
system pressure was maintained by using an adjustable back pressure regulator (BPR, Swagelok).
At pressures of 5 bar or above and temperatures below 25 °C for reactor 2, the fluoroform
dissolved completely in the liquid feed under the flow rates employed. Higher temperatures at 5
bar pressure resulted in a gas-liquid segmented flow regime. As a result of the large inner volume
(~1 mL) and small aperture of the outlet, the BPR (Swagelok, 0 to 26 bar) clogged periodically
from accumulation of precipitated solids formed in the reaction media, most notably LiF. The
formation of solids resulted in pressure fluctuations of ±2 bar (Table 1, entry 2) and an unsteady
delivery of fluoroform into the reaction mixture, therefore the stoichiometry could not be
carefully controlled. The problem became even more severe when using an Upchurch Scientific
BPR with a fixed back pressure of 100 psi/6.89 bar (Table 1, entry 1), as even the slightest
precipitation stopped the flow through the narrow channels of the BPR entirely. To fulfill our
requirements, a BPR with a small dead volume, precise pressure control and large enough
channel width to ensure a smooth flow without clogging seemed to be most appropriate. The flow
reaction performed most consistently with a Zaiput BPR. The Zaiput BPR is pressurized to the
correct setpoint by filling a gas chamber with a membrane on the bottom. The reaction fluid
needs to work against the compressed gas inside the chamber to flow out on the other side.
Pressure fluctuations were decreased to only ±0.1 bar with no clogging (Table 1, entry 3),
therefore the fluoroform flow rate did not fluctuate to compensate for pressure changes. The
reaction was successfully operated to give the product in 80% yield.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 1. Pressure Fluctuations Using Different Back Pressure Regulators for
Difluoromethylation.
ACS Paragon Plus Environment

Organic Process Research & Development
Page 10 of 26
10
1
2
3
4
5
6
7
8
9
Ph
Feed 1
N
Ph
N
R1
R2
R3
CO2Me
BPR
Ph
6
in THF
Ph
Feed 2
LHMDS
in THF
-30 °C
2 mL
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
N
-15 °C
6 mL
RT
2 mL
N
CO2Me
HF2C
MFC
7
entry
BPR type
fluoroform
pressure
fluctuation [%]
yield [%]a
(equiv)
1
2
3
Upchurch
Swagelok
Zaiput
3
3
3
-
no throughput
±2 bar
±0.1 bar
81
80
Conditions: Feed 1 = 0.5 M solution of 6 in THF, F1 = 0.25 µL/min. Feed 2 = 1 M solution of LiHMDS in
THF, F2 = 0.25 µL/min. a Analyzed by 19F NMR (α, α, α,-trifluorotoluene as internal standard)
Process Optimization Experiments. With a more reliable BPR in hand, we investigated the
influence of flow rate, residence time and fluoroform equivalents on the product yield. To ensure
full conversion to the corresponding lithium enolate at higher flow rates, reactor 1 was changed
from a 2 mL to a 4 mL coil to allow access to a longer residence time. Furthermore, to enable
improved mixing we exchanged the Y-shaped connector after reactor 1 for a 2 mL borosilicate
static glass mixer plate (Uniqsis). An increase of flow rates and shorter residence time resulted in
a considerable drop in yield to just 16% of product 7 (Table 2, entry 3). We could improve the
yield back to acceptable levels (87%) without changing the setup by increasing the amount of
base from two to three equivalents (Table 2, entries 5, 6 and 8). Although a dilution of the 0.5 M
solution of 6 to 0.25 M gave full conversion to the desired product, we opted for the higher
concentrated substrate solution because an acceptable conversion and double space time yield
could be achieved. During the course of this study, we observed that cooling of the second loop
was unnecessary at higher flow rates, because the reaction medium was still sufficiently cold
(0 °C) to ensure complete dissolution of the gas, and to achieve a similar yield. Subsequently, the
reaction was also investigated by using 2-methyltetrahydrofuran (2-MeTHF) as the solvent for
the substrate feed. The substrate displayed slightly better solubility in 2-MeTHF. Using 2-
MeTHF also helped with phase separation during the aqueous workup owing to its limited water
miscibility. Although currently more costly than THF, prices for 2-MeTHF are predicted to fall
ACS Paragon Plus Environment

Page 11 of 26
Organic Process Research & Development
11
1
2
3
4
5
6
7
8
9
considerably over time as production has been greatly expanded and more suppliers are offering
this solvent.27 In addition, the overall sustainability of the process benefits as 2-MeTHF can be
derived from renewable resources, such as corncobs and bagasse. In addition, 2-MeTHF is dried
more easily than THF due to the formation of an azeotrope rich in water.27 Notably, in contrast to
THF, 2-MeTHF is not classified as CMR-substance (carcinogenic, mutagenic and toxic for
reproduction).
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. Process Throughput Optimization for Difluoromethylation using Fluoroform.
Ph
N
Feed 1
CO2Me
Ph
N
R1
R2
R3
6
T [°C]
25 °C
-30 °C
12 bar
Feed 2
Ph
LiHMDS
1.0 M in THF
N
4 mL
2 mL
Ph
N
CO2
Me
HF2C
8 mL
MFC
7
3 equiv
entry
6 feed conc
[mol/L]
0.5
feed 1
feed 2
[mL/min]
0.25
0.5
temperature R2
yield [%]b
[mL/min]
0.25
0.5
[°C]
−15
−15
−15
−15
−15
25
1
2
3
4
5
80
62
16
1
87
85
0.5
0.5
0.5
0.5
1
1.33
0.8
1
0.66
1.2
6
0.5
0.8
1.2
7
8a
0.25
0.5
0.8
0.8
1.2
1.2
25
25
quantitative
86
a
b
Conditions: Back pressure = 12 bar. Feed 1 solution of 6 in 2-MeTHF, Volume of R2+R3 = 16 mL.
Analyzed by 19F NMR (α, α, α,-trifluorotoluene as internal standard)
Base Screening. The use of LiHMDS clearly contributes most to the overall cost of the process,
therefore we tried to replace it with a more cost effective alternative. Unfortunately all our
attempts employing n-BuLi, LDA, KOtBu and LiOtBu were largely unsuccessful and yielded
only negligible amounts of product (1-2% in the case of LDA). The outcome of those
ACS Paragon Plus Environment

Organic Process Research & Development
Page 12 of 26
12
1
2
3
4
5
6
7
8
9
experiments is very much in line with the published literature on these transformations.17
Notably, Mikami and coworkers proposed a SN2-type mechanism to explain the need for this
specific base.28 Best results apart from LiHMDS could be obtained by using 5 equiv of
LiN(cyHex)2 which afforded compound 7 in 10% yield (Table S1).29
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
Fluoroform Input. We next investigated the influence of decreasing the amount of fluoroform
on product yield. For these experiments the substrate solution and LiHMDS were introduced to
the reactor via sample injection loops (Table 3). To measure quantitatively how much fluoroform
was still dissolved immediately after the difluoromethylation, we decreased the residence time
through decreasing the volume of reactor 2 (R2) to 4 mL which resulted in lower yield but a
cooled (~0 °C) output solution stream which kept virtually all the fluoroform in solution.
Immediately after quenching the cold output stream with cold water, the organic phase was
transferred into an NMR tube, which was sealed to avoid loss of fluoroform through gassing out,
19
and subsequently analyzed by F NMR. The gassing out of fluoroform in case of 2 equiv and
3 equiv (Table 3) could not be completely avoided. Gratifyingly, decreasing the amount of
fluoroform to equimolar amounts only had a small effect on yield (Table 3). The use of 1.25
equiv of fluoroform afforded the highest yield with the product formed in 67% yield.
Table 3. Optimization of Fluoroform Input Using 19F NMR with α, α, α,-Trifluorotoluene as
Internal Standard.
Ph
N
Feed 1
CO2Me
Ph
N
R1
R2
6
-30 °C
12 bar
Ph
Feed 2
Ph
LiHMDS
in THF
N
4 mL
N
CO2Me
HF2C
4 mL
MFC
7
ACS Paragon Plus Environment

Page 13 of 26
Organic Process Research & Development
13
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
Conditions: Feed 1 = 0.5 M solution of 6 in 2-MeTHF, F1 = 0.8 mL/min. Feed 2 = 1 M solution of
LiHMDS in THF, F2 = 1.2 mL/min. Residence time 4 min, back pressure 12 bar. Output stream had a
temperature of 0 °C and was quenched with cold water. A sample of 200 µL was diluted to 700µL with
CDCl3 and analyzed by 19F NMR (α, α, α,-trifluorotoluene as internal standard) at 282 MHz.
Protection Group Screening. Bisimine 6 is an unstable and highly viscous liquid, therefore we
decided to screen for a more suitable protecting group for the amino residue. The optimal
substrate would be bench stable, crystalline, easy to synthesize and soluble in 2-MeTHF. For this
reason a series of substituted benzaldehydes was tested to form the corresponding bisimine
derivatives of ornithine methyl ester through standard condensation reaction. Only four of the 13
tested benzaldehydes produced solid products (Table S2). The most promising candidate was
methyl (S)-2,5-bis(((E)-4-chlorobenzylidene)amino)pentanoate (8) formed by condensation of
ornithine methyl ester with 4-chlorobenzaldehyde in anhydrous CH2Cl2 in the presence of
triethylamine. 4-Chlorobenzaldehyde was also the most cost efficient aromatic aldehyde (apart
from benzaldehyde) in this series of aromatic aldehydes. Slow addition of trimethylamine base
(~1 h) at 0 °C played a crucial role in obtaining a pure sample of 8, as deviating from these
conditions resulted in the formation of significant quantities of lactam 9. Lactam 9 was the only
side product formed during the condensation reaction in about 2-4% yield (Scheme 3). It should
be noted that the solvent used in the condensation reaction was changed from chloroform24 to
dichloromethane. Although dichloromethane is not a green solvent, it is preferable to other
ACS Paragon Plus Environment

Organic Process Research & Development
Page 14 of 26
14
1
2
3
4
5
6
7
8
9
chlorinated solvents as it is not classified as a cancerogen.25 Structural details of 8 and 9 were
elucidated by X-ray crystallography (see Supporting Information for details). Notably, bisimine 8
could be stored for prolonged periods of time under argon at low temperatures (−30 °C) without
degradation (Figure S2) and was bench stable at room temperature for at least 2 weeks. The water
content of a sample of 8 was determined by Karl Fischer titration (Table S3). A concentration of
1475 ppm H2O was found, which could in theory hydrolyze 1.48% of 8.
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
O
O
, Et3N
MeO
N
O
Cl
N
Cl
MeO
NH2
2HCl
8
CH2Cl2
NH2
Difluoromethylation
Cl
O
CF2H
N
Cl
MeO
N
O
N
Cl
HN
10
9
2-4%
87% assay yield
Cl
Scheme 3. Formation of 8 and byproduct 9 and subsequent transformation to 10 with optimized
difluoromethylation conditions.
Bisimine 8 was converted to its difluoromethylated analogue 10 using the conditions described in
19
Table 2, entry 8 in 87% assay yield (based on an average of 3 measurements, determined by F
NMR) over a 60 min run time. The collected mixture was used for the optimization experiments
for the deprotection procedure. The mixture was diluted with toluene and stored under argon to
avoid formation of a gel-like precipitate.
Fluoride Determination. The inorganic fluoride content of the reaction mixture after
difluoromethylation was determined, since the formation of HF during the acidic hydrolysis is a
potential safety risk. HF also contributes towards corrosion of certain reactor materials, such as
ACS Paragon Plus Environment

Page 15 of 26
Organic Process Research & Development
15
1
2
3
4
5
6
7
8
9
stainless steel or glass. A 3 mL sample, corresponding to approximately 2 mmol of fluoride, was
extracted with H2O (3 × 25 mL). Ion chromatography was used to quantify the amount of
fluoride dissolved in the aqueous phase. The fluoride amount measured by ion chromatography
was 1.5 mmol in the aqueous analyte. This value corresponds to a maximum concentration of HF
in the reaction mixture of <1 wt% after extraction into the aqueous phase, which poses no risk to
health and safety but may contribute to corrosion of steel and glass.
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
Microwave-Assisted Deprotection. After preparing a sufficient amount of the protected
intermediate 10, we proceeded to optimize the envisioned deprotection approach with aqueous
HCl. We opted for the acidic hydrolysis of compound 10 as simultaneous deprotection of the
imine and ester functionality is possible, although ester hydrolysis is considerably slower. The
deprotection was performed in a high-T/p range employing sealed vessel microwave heating
technology. Optimized microwave batch conditions can subsequently be conveniently translated
to (conventionally heated) continuous-flow processes to provide scalability (“microwave-to-
flow” paradigm).30 The optimization was mainly conducted to minimize the hydrolysis time and
achieve high conversion. The translation of the previously published protocol (6 M aqueous HCl
at 150 °C for 45 min)24 to a continuous flow approach would have required a large internal
volume for the heated coil (135 mL for a total flow rate of 3 mL/min), and operation at high
pressure and temperature was deemed a safety risk. Starting from our previously published
approach, we attempted to shorten the deprotection time without loss of conversion (Table 4,
entry 1-3) but were unsuccessful employing 6 M HCl. We thus opted for concentrated HCl
(12 M) instead and could achieve full conversion to the dihydrochloride salt of DFMO (1) in
10 min at 160 °C with 12 bar internal pressure (Table 4, entry 11).
Table 4. Microwave Optimization of Acidic Hydrolysis Conditions.
ACS Paragon Plus Environment

Organic Process Research & Development
Page 16 of 26
16
1
2
3
Cl
4
5
6
7
8
NH2
NH2
CO2
H+
H+
2HCl
H2N
H2N
CO2Me
HF2C
Cl

HF2C
H
N
11
9
1
N
CO2Me
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
HF2C
10
entry
HCl
concentration
6 M
temperature
[°C]
140
time
[min]
45
45
30
30
20
30
20
pressure
[bar]
5
conversion [%]a
1
2
3
4
5
6
7
8
9
>99
>99
85
>99
82
>99
>99
95
>99
>99
6 M
6 M
150
150
140
140
150
150
150
160
160
8
8
5
5
8
8
8
11
11
12 M
12 M
12 M
12 M
12 M
12 M
12 M
10
30
10
11
Conditions: 3 mL of difluoromethylation reaction mixture (0.15 M solution of 10 in
2-MeTHF/THF/toluene) and 3 mL of HCl were heated in sealed 25 mL microwave vessels. a Analyzed
by 1H NMR by integrating base line separated CHF2-peaks of 10 and 1
Continuous Flow Deprotection Experiments. After having identified optimal conditions for the
acidic hydrolysis of the protected intermediate 10, a continuous flow reactor was assembled. The
flow setup consisted of two syringe pumps (Syrris Asia and HighTec Zang Syrdos) with one
2 mL sample injection loop (SL) to introduce a sample of difluoromethylated reaction mixture
and a second stream to introduce 12 M HCl directly through the pump. The two feeds were
mixed in a T-shaped connector. The mixing of the two feeds results in precipitation, which
subsequently dissolved within a few seconds. Thus, a short 2 mL premixing coil was used before
introducing the stream into a 10 mL perfluoroalkoxy alkanes (PFA) coil which was submerged
within a stirred heated oil bath. The back pressure was adjusted depending on temperature to
avoid formation of gas segments. PFA was used as small amounts of hydrofluoric acid might
form during the addition of HCl which corrodes steel coils over time. A safety evaluation was
ACS Paragon Plus Environment

Page 17 of 26
Organic Process Research & Development
17
1
2
3
4
5
6
conducted to ensure that the PFA tubing used (1/8 in. o.d.; 0.8 mm i.d) has sufficient wall
thickness to withstand the high temperatures and pressure applied.31
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
Table 5. Acidic Hydrolysis Optimization in Continuous Flow Mode.
Cl
F1 = 0.5 mL/min
Cl
10
N
heated coil
10 mL
pre mixing
NH2
CO2H
SL
N
BPR H2N
CO2Me
HF2C
HF2C
2HCl
1
HCl conc.
20-30°C
2 mL
F2 = 0.25 mL/min
entry temperature
[°C]
back pressure
conversion [%]a
1
2
3
140
150
160
7
9
11
88
93
>99
Conditions: Feed 1 = 0.5 M solution of 0.15 M solution of 10 in 2-MeTHF/THF/toluene,
F1 = 0.5 mL/min. Feed 2 = conc. aq. HCl, F2 = 0.25 mL/min. 16 min residence time was used. aAnalyzed
by 1H NMR by integrating base line separated CHF2-peaks of 10 and 1.
Telescoped Continuous Flow experiments. To achieve our ultimate goal, a scalable telescoped
difluoromethylation procedure with subsequent deprotection in continuous flow, we connected
the two continuous flow setups described above. The previously used 2 mL static glass mixer was
exchanged for a smaller 1 mL static glass mixer with narrower channels which could be
submerged within a cooled water bath to ensure an improved temperature control. The heated coil
was exchanged for a 40 mL PFA coil and submerged within a heated stirred oil bath, so that a
higher total flow rate (3 mL/min) could be used in order to provide a higher throughput.
Precipitation after mixing the difluoromethylation feed and HCl feed (feed 3) was minimized due
to faster mixing from the higher overall flow rates used. First, two runs using sample injection
loops were performed to proof feasibility of the telescoped reaction (Table 6, entry 1). An amount
ACS Paragon Plus Environment

Organic Process Research & Development
Page 18 of 26
18
1
2
3
19
of 1.05 equiv of fluoroform was used and fully consumed (determined by F NMR of aqueous
and organic layer). To assess scalability we conducted a 4 h long run experiment with a
throughput of 24 mmol/h (using conditions from Table 6, entry 2). The substrate solution (0.5 M
in 2-MeTHF) and LiHMDS (1.0 M in THF) were directly pumped through the syringe pumps
and the injection loops were not used. The employed back pressure rose from 12 to 15 bar. This
pressure increase might stem from warming of the BPR and expansion of the gas inside the gas
chamber which would result in a higher back pressure. From this protocol the target compound
eflornithine hydrochloride monohydrate (DFMO, 1) was prepared in 86% (19.48 g) isolated yield
after purification.
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
Table 6. Telescoped Continuous Flow Reaction.
Cl
Cl
Cl
10
N
F1 = 0.8 mL/min
N
CO2Me
Cl
8
HF2C
N
N
-30 °C
CO2Me
0° C
in MeTHF
12 bar
H2N
LiHMDS
in THF
4 mL
MFC
25 °C
14 mL
160°C
40 mL
1 mL
NH2
CO2H
F2 = 1.2 mL/min
HF2C
1
HCl conc.
F3 = 1.0 mL/min
entry temperature
[°C]
back pressure
conversion [%]
[bar]
12
12
1
2
150
160
95
<99
Conditions: Feed 1 = 0.5 M solution of 6 in 2-MeTHF, F1 = 0.8 mL/min. Feed 2 = 1 M solution of
LiHMDS in THF, F2 = 1.2 mL/min. Feed 3 = 12 M HCl in H2O, F3 = 1 mL/min. Residence time 23 min,
back pressure 12 bar (1.05 equiv fluoroform). Conversion from 10 to 1 measured by 1H NMR via
integration of base line separated CHF2-peaks of 10 and 1 of dried residue.
CONCLUSION
ACS Paragon Plus Environment

Page 19 of 26
Organic Process Research & Development
19
1
2
3
4
5
6
7
8
9
A telescoped continuous-flow protocol was developed for the preparation of eflornithine.
Fluoroform, a large scale industrial by-product of Teflon®, was used as a cost effective reagent.
In search of an optimal starting material for the synthesis, we identified the bisimine 8 of
commercially available methyl ester of ornithine dihydrochloride and 4-chlorobenzalehyde as a
bench stable, crystalline and easy to handle compound. Furthermore, compared to our original
preliminary report,24 the fluoroform input was greatly reduced from 3.00 equiv to 1.05 equiv,
without affecting yield. Full consumption of fluoroform was facilitated by using an excess of
base. After optimization of deprotection conditions with HCl in a microwave reactor, these
conditions were then translated into a continuous flow process. The difluoromethylation and
deprotection were connected together as a telescoped process to achieve a throughput of
24 mmol/h. Eflornithine hydrochloride monohydrate was isolated in 86% yield, which is
significantly higher than previously reported yields for the less desirable process based on
chlorodifluoromethane (37% to 40%).14,15 The overall processing time was decreased from 23 h
in the case of the industrial process to a mere 23.5 min in continuous flow, thus potentially
minimizing manufacturing cost.
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
1
General. H NMR spectra were recorded on a Bruker 300 MHz or 500 MHz instrument.
19
13C NMR spectra were recorded on the 300 MHz instrument at 75 MHz. F NMR spectra were
recorded on the 300 MHz instrument at 282 MHz. Chemical shifts (δ) are expressed in ppm
downfield from TMS as internal standard. The letters s, d, dd, t, q, and m are used to indicate
singlet, doublet, doublet of doublets, triplet, quadruplet, and multiplet. α, α, α,-Trifluorotoluene
19
1
was used as internal standard for F NMR, nitromethane was used for H NMR to determine
purity of starting material and yield of reactions. The purity and supplier for each chemical were:
(S)-methyl 2,5-diaminopentanoate dihydrochloride (95% purity, Fluorochem), 4-
chlorobenzaldehyde (97% purity, Sigma Aldrich), lithium bis(trimethylsilyl)amide (1 M solution
in THF, Sigma Aldrich), fluoroform (99.995% purity, Messer), 2-methyltetrahydrofuran
anhydrous, (≥99% purity, inhibitor free, Acros Organics), tetrahydrofuran anhydrous (≥99%,
Acros Organics) concentrated HCl (35% in water, VWR), dichloromethane (>98% purity,
VWR), chloroform (99.2% purity, VWR) and triethylamine (≥99%, Riedel de Haën). Microwave
reactions were carried out in a Biotage Initiator+ single mode microwave instrument. Reaction
ACS Paragon Plus Environment

Organic Process Research & Development
Page 20 of 26
20
1
2
3
4
5
6
times refer to hold times at the temperatures indicated, not to total irradiation times. The
temperature was measured with an IR sensor on the outside of the reaction vessel.
7
8
9
Preparation of 4-Chlorobenzaldehyde Protected Ornithine Methyl Ester 10. The protection
protocol was adapted from previously described procedures.24,32 A dry 2 L round bottom flask
with magnetic stirring bar was charged with (S)-methyl 2,5-diaminopentanoate dihydrochloride
(25.0 g, 114 mmol) and 4-chlorobenzaldehyde (24.2 g, 240 mmol), sealed and flushed with argon
three times. Anhydrous dichloromethane (750 mL) was added and then stirred. The colorless
reaction mixture was cooled to 0 °C. After addition of triethylamine (33.2 mL, 240 mmol) over
1 h, the reaction mixture was allowed to warm to room temperature and stirred for 18 h. The
solvent was removed under reduced pressure. The off-white residue was treated with Et2O to
precipitate Et3N·HCl. The formed colorless precipitate was filtered off and the obtained filtrate
was concentrated in vacuo. The obtained off-white residue was recrystallized from petrol
ether/EtOAc to provide the desired product 10 as colorless crystalline solid (41.1 g, 105 mmol,
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
92% yield) mp. 89–92 °C. H NMR (500 MHz, CDCl3) δ 8.26 (dd, J = 11.6, 5.6 Hz, 2H), 7.75–
7.70 (m, 2H), 7.67–7.63 (m, 2H), 7.43–7.35 (m, 4H), 4.06 (dd, J = 8.4, 5.3 Hz, 1H), 3.76 (s, 3H),
3.65 (t, J = 6.7 Hz, 2H), 2.16 – 2.07 (m, 1H), 2.06 – 1.94 (m, 1H), 1.81 – 1.66 (m, 2H) ppm. 13C
NMR (75 MHz, CDCl3) δ 172.4, 162.2, 159.9, 137.3, 136.6, 134.7, 134.2, 129.8, 129.3, 129.0,
128.9, 73.3, 61.2, 52.3, 31.3, 27.4 ppm.
Optimized Continuous Flow Procedure for Difluoromethylation (Table 2, Entry 8). The
flow setup consisted of two syringe pumps (Asia Syrris) to introduce a solution of substrate 6
(391 mg, 1 mmol) and internal standard (146 mg, 1 mmol) in 2 mL of 2-MeTHF (Feed 1), and a
commercial solution of LiHMDS (1.0 M in THF, Sigma-Aldrich) (Feed 2). Injection loops
(perfluoroalkoxy alkanes, PFA, 0.8 mm i.d., 1.59 mm o.d.; internal volume: 2.0 mL, Feed 1, and
4 mL, Feed 2) were used to deliver the two feeds. To start the experiment, the complete reactor
setup was flushed by pumping dry 2-MeTHF with flow rates of Feed 1 = 800 μL/min and Feed 2
= 1200 μL/min. Fluoroform was introduced into the reactor with a flow rate of 9.23 mL/min
(1.05 equiv) using a calibrated Bronkhorst MFC. The internal pressure of the reactor reached the
target pressure of 12 bar after approximately 10 min. Substrate 6 (391 mg, 1.00 mmol) and
α, α, α,-trifluorotoluene (146 mg, 1 mmol) was dissolved in neat 2-MeTHF and diluted to
2.00 mL in a volumetric flask with 2-MeTHF (Feed 1). A LiHMDS solution (1.0 M, 4 mL) in
THF was used as Feed 2. Both solutions were loaded into their corresponding injection loops.
ACS Paragon Plus Environment

Page 21 of 26
Organic Process Research & Development
21
1
2
3
4
5
6
7
8
9
Feed 1 and feed 2 were pumped from the injection loops in a Y-shaped connector (Y Assembly
PEEK 1/4-28 0.040 in) in a cooling bath (−30 °C). The combined mixture passed through a first
residence loop at −30 °C (1/8 in. o.d.; 0.8 mm i.d.; residence volume V1 = 4.0 mL), before the
mixture was combined with fluoroform in a Uniqsis glass static mixer (UQ-5102, 1.8 mL internal
volume). The combined mixture then passed through a second residence time loop (1/8 in. o.d.;
0.8 mm i.d.; residence volume V2 = 14 mL) and left the flow system through an adjustable back
pressure regulator (Zaiput BPR-10). The effluent was collected in a glass vial and an aliquot
taken for determination of yield by 19F NMR. For the 1 h experiment (Scheme 3) fluoroform
input was reduced to 9.23 mL/min (1.05 equiv) the substrate solution and LiHMDS were directly
pumped through the syringe pumps and the injection loops were not used.
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
Telescoped Continuous Flow Procedure – Long Run (Table 6, entry 2). The flow setup
consisted of three continuous syringe pumps (2 × Asia Syrris, 1 × HighTec Zang Syrdos) to
introduce a solution of substrate in 2-MeTHF (Feed 1), a commercial solution of LiHMDS (1.0
M in THF, Sigma-Aldrich) (Feed 2) and concentrated HCl (35% in water) (Feed 3). Feed 1, 2 and
3 were directly pumped through the syringe pumps. To start the experiment, the complete reactor
setup was flushed by pumping dry 2-MeTHF with flow rates of Feed 1 = 800 μL/min and Feed 2
= 1200 μL/min. HCl was introduced with Feed 3 = 1000 μL/min. Fluoroform was introduced into
the reactor with a flow rate of 9.23 mL/min (1.05 equiv) using a calibrated Bronkhorst MFC. The
internal pressure of the reactor reached the target pressure of 12 bar after approximately 10 min.
Substrate 10 (100 mmol) was dissolved in neat 2-MeTHF and diluted to 200 mL in two
volumetric flasks with 2-MeTHF (Feed 1). A LiHMDS solution (1.0 M, 800 mL) in THF was
used as Feed 2. Feed 1 and Feed 2 pumped directly through the syringe pumps were combined in
a Y-shaped connector (Y Assembly PEEK 1/4-28 0.040in) in a cooling bath (−30 °C). The
combined mixture went through a first residence loop at −30 °C (1/8 in. o.d.; 0.8 mm i.d.;
residence volume V1 = 4.0 mL), before the mixture was combined with fluoroform in a Syrris
Asia glass static mixer (1 mL internal volume). The mixture then went through a second
residence loop at 25 °C (1/8 in. o.d.; 0.8 mm i.d.; residence volume V2 = 14 mL) and were mixed
with feed 3 in a T-shaped connector (T Assembly PTFE 1/4-28 0.040in). The combined stream
then went through a third residence loop at 160 °C (1/8 in. o.d.; 0.8 mm i.d.; residence volume
V3 = 40 mL) and left the flow system through an adjustable back pressure regulator (Zaiput
BPR-10). The biphasic mixture was collected for 3.5 h (84 mmol) and the aqueous phase was
ACS Paragon Plus Environment

Organic Process Research & Development
Page 22 of 26
22
1
2
3
4
5
6
7
8
9
washed with Et2O and concentrated under reduced pressure. The crude product was dissolved in a
small amount of H2O and the pH was adjusted to 4 with Et3N. The resulting slurry was filtered
and washed with cold EtOH abs. and CHCl3. The residue was recrystallized from EtOH/H2O to
give product 1 hydrochloride monohydrate as colourless powder. (17.05 g, 72.3 mmol, 86%
1
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
yield). Mp. 228 °C; H NMR (300.36 MHz, D2O): δ = 6.46 (t, JHF = 52.8 Hz, 1H), 3.05
(t,3JHH = 7.6 Hz, 2H), 2.25-1.97 (m, 2H), 1.96–1.79 (m, 1H), 1.76–1.59 (m, 1H) ppm.13C NMR
3
1
1
(75 MHz, D2O): δ = 167.8 (d, JCF = 6.4 Hz), 114.0 (dd, JCF = 249.7 Hz, JCF = 247.0 Hz), 64.5
2
2
3
4
(dd, JCF = 20.4 Hz, JCF = 18.7 Hz), 38.8 (d, JCF = 7.3 Hz), 31.6 (d, JCF = 3.2 Hz), 20.8 ppm.
2
2
19F NMR (282 MHz, D2O): δ = −126.28 (dd, JFF = 283.5 Hz, JHF = 52.4 Hz), −131.76 (dd,
2JFF = 283.5 Hz, 2JHF = 52.4 Hz) ppm.
ACKNOWLEDGEMENT
The CCFLOW project (Austrian Research Promotion Agency FFG No. 862766) is funded
through the Austrian COMET Program by the Austrian Federal Ministry of Transport, Innovation
and Technology (BMVIT), the Austrian Federal Ministry of Digital and Economic Affairs
(BMDW) and by the State of Styria (Styrian Funding Agency SFG).
Supporting Information
Description and images of the continuous flow set-up, 1H NMR, 13C NMR and 19F NMR spectra
of all isolated products and X-ray crystallographic data for compound 8 and 9. This material is
available free of charge via the Internet at http://pubs.acs.org.
References
(1)
(a) Bacchi, C. J.; Goldberg, B.; Garofalo-Hannan, J.; Rattendi, D.; Lyte, P.; Yarlett, N.
Fate of Soluble Methionine in African Trypanosomes: Effects of Metabolic Inhibitors.
Biochem. J. 1995, 309, 737–743. (b) Bari, C. Di; Pastore, G.; Roscigno, G.; Schechter, P.
J.; Sjoerdsma, A. Late-Stage African Trypanosomiasis and Eflornithine. Ann. Intern. Med.
1986, 105, 803. (c) Gilman, T. M.; Paulson, Y. J.; Boylen, C. T.; Heseltine, P. N. R.;
Sharma, O. P. Eflornithine Treatment of Pneumocystis Carinii Pneumonia in AIDS. J. Am.
Med. Assoc. 1986, 256, 2197.
(2)
(a) Raul, F. Revival of 2-(Difluoromethyl)ornithine (DFMO), an Inhibitor of Polyamine
Biosynthesis, as a Cancer Chemopreventive Agent. Biochem. Soc. Trans. 2007, 35, 353–
355. (b) Gerner, E. W.; Meyskens, F. L. Polyamines and Cancer: Old Molecules, New
Understanding. Nat. Rev. Cancer 2004, 4, 781–792. (c) Gunaratna, P. C.; Wilson, G. S.;
ACS Paragon Plus Environment

Page 23 of 26
Organic Process Research & Development
23
1
2
3
4
5
Slavik, M. Pharmacokinetic Studies of α-Difluoromethylornithine in Rabbits Using an
Enzyme-Linked Immunosorbent Assay. J. Pharm. Biomed. Anal. 1994, 12, 1249–1257.
6
7
8
9
(3)
(4)
Barman Balfour, J. A.; McClellan, K. Topical Eflornithine. Am. J. Clin. Dermatol. 2001,
2, 197–201.
(a) Phillips, M. A.; Coffinot, P.; Wang, C. C. Cloning and Sequencing of the Ornithine
Decarboxylase Gene from Trypanosoma Brucei Implications For Enzyme Turnover and
Selective Difluoromethylornithine Inhibition. J. Biol. Chem. 1987, 262, 8721–8727. (b)
Bacchi, C. J.; Garofalo, J.; Mockenhaupt, D.; McCann, P. P.; Diekema, K. A.; Pegg, A. E.;
Nathan, H. C.; Mullaney, E. A.; Chunosoff, L.; Sjoerdsma, A.; et al. In Vivo Effects of α-
DL-Difluoromethylornithine on the Metabolism and Morphology of Trypanosoma Brucei
Brucei. Mol. Biochem. Parasitol. 1983, 7, 209–225. (c) Vincent, I. M.; Creek, D.; Watson,
D. G.; Kamleh, M. A.; Woods, D. J.; Wong, P. E.; Burchmore, R. J. S.; Barrett, M. P. A
Molecular Mechanism for Eflornithine Resistance in African Trypanosomes. PLoS
Pathog. 2010, 6, e11001204.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(5)
(6)
Mettler, H.; Greth, E., 2-Difluormethyl-2-cyanethylmalonsäureniedrigalkylester, Lonza
AG CH672124A5, 1989.
For a continuous flow process of converting compound 2 into 3 (Scheme 1), see:
Gutmann, B.; Hanselmann, P.; Bersier, M.; Roberge, D.; Kappe, C. O. Continuous-Flow
Difluoromethylation with Chlorodifluoromethane under Biphasic Conditions. J. Flow
Chem. 2017, 7, 46–51.
(7)
(8)
Zhu, J.; Chadwick, S. T.; Price, B.; Zhao, S.; Costello, C. A.; Vemishetti, P., Processes for
the Production of Alpha-Difluoromethyl ornithine (DFMO), WO03020209A2, 2002.
(a) Seki, M.; Suzuki, M.; Matsumoto, K. Convenient Synthesis of α-
Difluoromethylornithine. Biosci. Biotechnol. Biochem. 1993, 57, 1024–1025. (b) Bey P.;
Jung M.; Merrell, Toraude et Compagnie, 2-(Difluoromethyl)-2,5-diaminopentanoic acid,
US 4413141, 1983.
(9)
Ozone Secretariat; United Nations Environment Programme. The Montreal Protocol on
Substances That Deplete the Ozone Layer; Secretariat for The Vienna Convention for the
Protection of the Ozone Layer & The Montreal Protocol on Substances that Deplete the
Ozone Layer, 2000.
(10) (a) Ni, C.; Hu, J. Recent Advances in the Synthetic Application of Difluorocarbene.
Synthesis 2014, 46, 842–863. (b) Yerien, D. E.; Barata-Vallejo, S.; Postigo, A.
Difluoromethylation Reactions of Organic Compounds. Chem. - A Eur. J. 2017, 23,
14676–14701.
(11) Forster, P.; V. Ramaswamy; P. Artaxo; T. Berntsen; R. Betts; D.W. Fahey; J. Haywood; J.
Lean; D.C. Lowe; G. Myhre; et al. Changes in Atmospheric Constituents and in Radiative
Forcing. In: Climate Change 2007: The Physical Science Basis. Contribution of Working
Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate
Change; Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M.Tignor
and, H.L. Miller, Eds.; Cambridge University Press, Cambridge, United Kingdom and
New York, NY, USA, 2007.
ACS Paragon Plus Environment

Organic Process Research & Development
Page 24 of 26
24
1
2
3
4
5
(12) Han, W.; Li, Y.; Tang, H.; Liu, H. Treatment of the Potent Greenhouse Gas, CHF3- An
Overview. J. Fluor. Chem. 2012, 140, 7–16.
6
7
8
9
(13) Prakash, G. K. S.; Jog, P. V; Batamack, P. T. D.; Olah, G. A. Taming of Fluoroformꢀ:
Direct Nucleophilic Trifluoromethylation of Si, B, S, and C Centers. Science 2012, 338,
1324–1327.
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) (a) Shono, T.; Ishifune, M.; Okada, T.; Kashimura, S. A Novel Trifluoromethylation of
Aldehydes and Ketones Promoted by an Electrogenerated Base. J. Org. Chem. 1991, 56,
2–4. (b) Barhdadi, R.; Troupel, M.; Périchon, J. Coupling of Fluoroform with Aldehydes
Using an Electrogenerated Base. Chem. Commun. 1998, 0, 1251–1252.
(15) (a) Folléas, B.; Marek, I.; Normant, J.-F.; Saint-Jalmes, L. Fluoroform: An Efficient
Precursor for the Trifluoromethylation of Aldehydes. Tetrahedron 2000, 56, 275–283. (b)
Langlois, B. R.; Billard, T. Fluoroform, Fluoral, Trifluoroacetic, and
Trifluoromethanesulfinic Acids Derivatives as New Reagents for the Introduction of
Polyfluorinated Moieties; ACS Symposium Series, Vol. 911 2005; 57–86. (c) Folléas, B.;
Marek, I.; Normant, J.-F.; Jalmes, L. Saint. Fluoroform: An Efficient Precursor for the
Trifluoromethylation of Aldehydes. Tetrahedron Lett. 1998, 39, 2973–2976. (d) Langlois,
B. R.; Billard, T. Some Recent Results in Nucleophilic Trifluoromethylation and
Introduction of Fluorinated Moieties. Synthesis 2003, 2, 185–194. (e) Large, S.; Roques,
N.; Langlois, B. R. Nucleophilic Trifluoromethylation of Carbonyl Compounds and
Disulfides with Trifluoromethane and Silicon-Containing Bases. J. Org. Chem. 2000, 65,
8848–8856. (f) Billard, T.; Bruns, S.; Langlois, B. R. New Stable Reagents for the
Nucleophilic Trifluoromethylation. 1. Trifluoromethylation of Carbonyl Compounds with
N-Formylmorpholine Derivatives. Org. Lett. 2000, 2, 2101–2103.
(16) (a) Zanardi, A.; Novikov, M. A.; Martin, E.; Benet-Buchholz, J.; Grushin, V. V. Direct
Cupration of Fluoroform. J. Am. Chem. Soc. 2011, 133, 20901–20913. (b) Novák, P.;
Lishchynskyi, A.; Grushin, V. V. Trifluoromethylation of α-Haloketones. J. Am. Chem.
Soc. 2012, 134, 16167–16170. (c) Novák, P.; Lishchynskyi, A.; Grushin, V. V.
Fluoroform-Derived CuCF
for Low-Cost, Simple, Efficient, and Safe
3
Trifluoromethylation of Aryl Boronic Acids in Air. Angew. Chemie Int. Ed. 2012, 51,
7767–7770.
(17) (a) Iida, T.; Hashimoto, R.; Aikawa, K.; Ito, S.; Mikami, K. Umpolung of Fluoroform by
C-F Bond Activation: Direct Difluoromethylation of Lithium Enolates. Angew. Chemie -
Int. Ed. 2012, 51, 9535–9538. (b) Aikawa, K.; Maruyama, K.; Honda, K.; Mikami, K. α-
Difluoromethylation on sp3 Carbon of Nitriles Using Fluoroform and Ruppert–Prakash
Reagent. Org. Lett. 2015, 17, 4882–4885. (c) Aikawa, K.; Maruyama, K.; Nitta, J.;
Hashimoto, R.; Mikami, K. Siladifluoromethylation and Difluoromethylation onto C(sp3),
C(sp2 ), and C(sp) Centers Using Ruppert–Prakash Reagent and Fluoroform. Org. Lett.
2016, 18, 3354–3357.
(18) (a) Okusu, S.; Tokunaga, E.; Shibata, N. Difluoromethylation of Terminal Alkynes by
Fluoroform. Org. Lett. 2015, 17, 3802–3805; for review on the application of fluorofom in
synthesis, see: (b) Zhang, C. Application of Fluoroform in Trifluoromethylation and
Difluoromethylation Reactions. Arkivoc 2017, part i, 67–83.
ACS Paragon Plus Environment

Page 25 of 26
Organic Process Research & Development
25
1
2
3
4
5
6
7
8
9
(19) (a) Mallia, C. J.; Baxendale, I. R. The Use of Gases in Flow Synthesis. Org. Process Res.
Dev. 2016, 20, 327–360. (b) Gemoets, H. P.; Su, Y.; Shang, M.; Hessel, V.; Luque, R.;
Noël T. Liquid Phase Oxidation Chemistry in Continuous-flow Microreactors. Chem. Soc.
Rev. 2016, 45, 83-117. (c) Gavriilidis, A.; Constantinou, A.; Hellgardt, K.; Hii, K. K.;
Hutchings, G. J.; Brett, G. L.; Kuhn, S.; Marsden, S. P. Aerobic Oxidations in Flow:
Opportunities for the Fine Chemicals and Pharmaceuticals Industries. React. Chem. Eng.
2016, 1, 595–612.
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
(20) (a) Movsisyan, M.; Delbeke, E. I. P.; Berton, J. K. E. T.; Battilocchio, C.; Ley S. V.;
Stevens, C. V. Taming Hazardous Chemistry by Continuous Flow Technology. Chem.
Soc. Rev. 2016, 45, 4892-4928. (b) Jiménez-González, C.; Poechlauer, P.; Broxterman, Q.
B.; Yang, B.-S.; am Ende, D.; Baird, J.; Bertsch, C.; Hannah, R. E.; Dell’Orco, P.;
Noorman, H.; et al. Key Green Engineering Research Areas for Sustainable
Manufacturing: A Perspective from Pharmaceutical and Fine Chemicals Manufacturers.
Org. Process Res. Dev. 2011, 15, 900–911. (c) Gutmann, B.; Cantillo, D.; Kappe, C. O.
Continuous-Flow Technology - A Tool for the Safe Manufacturing of Active
Pharmaceutical Ingredients. Angew. Chemie - Int. Ed. 2015, 54, 6688–6728. (d)
Plutschack, M. B.; Pieber, B.; Gilmore, K.; Seeberger, P. H. The Hitchhiker’s Guide to
Flow Chemistry. Chem. Rev. 2017 117, 11796-11893. (e) May, S. A. Flow Chemistry,
Continuous Processing, and Continuous Manufacturing: A Pharmaceutical Perspective. J.
Flow Chem. 2017, 7 137–145. (f) Pieber, B.; Gilmore, K.; Seeberger, P. H. Integrated
Flow Processing - Challenges in Continuous Multistep Synthesis. J. Flow Chem. 2017, 7,
129–136. (g) Gérardy, R.; Emmanuel, N.; Toupy, T.; Kassin, V. E.; Tshibalonza, N. N.;
Schmitz, M.; Monbaliu, J. C. M. Continuous Flow Organic Chemistry: Successes and
Pitfalls at the Interface with Current Societal Challenges. Eur. J. Org. Chem. 2018. (h)
Jensen, K. F. Flow Chemistry-Microreaction Technology Comes of Age. AIChE J. 2017,
63, 858–869.
(21) Mazloomi, Z.; Bansode, A.; Benavente, P.; Lishchynskyi, A.; Urakawa, A.; Grushin, V. V.
Continuous Process for Production of CuCF via Direct Cupration of Fluoroform. Org.
3
Process Res. Dev. 2014, 18, 1020–1026.
(22) Musio, B.; Gala, E.; Ley, S. V. Real-Time Spectroscopic Analysis Enabling Quantitative
and Safe Consumption of Fluoroform during Nucleophilic Trifluoromethylation in Flow.
ACS Sustain. Chem. Eng. 2018, 6, 1489–1495.
(23) Gutmann, B.; Köckinger, M.; Glotz, G.; Ciaglia, T.; Slama, E.; Zadravec, M.; Pfanner, S.;
Maier, M. C.; Gruber-Wölfler, H.; Kappe, C. O. Design and 3D Printing of a Stainless
Steel Reactor for Continuous Difluoromethylations Using Fluoroform. React. Chem. Eng.
2017, 2, 919–927.
(24) Köckinger, M.; Ciaglia, T.; Bersier, M.; Hanselmann, P.; Gutmann, B.; Kappe, C. O.
Utilization of Fluoroform for Difluoromethylation in Continuous Flow: A Concise
Synthesis of α-Difluoromethyl-Amino Acids. Green Chem. 2018, 20, 108–112.
(25) Prat, D.; Wells, A.; Hayler, J.; Sneddon, H.; McElroy, C. R.; Abou-Shehada, S.; Dunn, P.
J. CHEM21 Selection Guide of Classical- and Less Classical-Solvents. Green Chem. 2015,
18, 288–296.
(26) Brzozowski, M.; O’Brien, M.; Ley, S. V.; Polyzos, A. Flow Chemistry: Intelligent
ACS Paragon Plus Environment

Organic Process Research & Development
Page 26 of 26
26
1
2
3
4
5
Processing of Gas–Liquid Transformations Using a Tube-in-Tube Reactor. Acc. Chem.
Res. 2015, 48, 349–362.
6
7
8
9
(27) (a) Bryan, M. C.; Dillon, B.; Hamann, L. G.; Hughes, G. J.; Kopach, M. E.; Peterson, E.
A.; Pourashraf, M.; Raheem, I.; Richardson, P.; Richter, D.; et al. Sustainable Practices in
Medicinal Chemistry: Current State and Future Directions. 2013. (b) Supporting the
Advancement of Chemistry through Sound Environmental, Social
&
Fiscal
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
Responsibilities https://www.sigmaaldrich.com/content/dam/sigma-
aldrich/docs/Sigma/Brochure/greener_solvent_alternatives.pdf?utm_source=redirect&utm
_medium=promotional&utm_campaign=insite_greener_solvent_alternatives (accessed Jul
25, 2018).
(28) Honda, K.; Harris, T. V.; Hatanaka, M.; Morokuma, K.; Mikami, K. Computational SN2-
Type Mechanism for the Difluoromethylation of Lithium Enolate with Fluoroform through
Bimetallic C−F Bond Dual Activation. Chem. - A Eur. J. 2016, 22, 8796–8800.
(29) We were unable to reproduce the difluoromethylation protocols reported by Dolbier and
co-workers utilizing KOH as base, see: (a) Thomoson, C. S.; Dolbier, W. R. Use of
Fluoroform as a Source of Difluorocarbene in the Synthesis of Difluoromethoxy- and
Difluorothiomethoxyarenes. J. Org. Chem. 2013, 78, 8904–8908. (b) Thomoson, C. S.;
Wang, L.; Dolbier, W. R. Use of Fluoroform as a Source of Difluorocarbene in the
Synthesis of N-CF2H Heterocycles and Difluoromethoxypyridines. J. Fluor. Chem. 2014,
168, 34–39.
(30) Glasnov, T. N.; Kappe, C. O. The Microwave-to-Flow Paradigm: Translating High-
Temperature Batch Microwave Chemistry to Scalable Continuous-Flow Processes. Chem.
Eur. J. 2011, 17, 11956–11968.
(31) https://www.adtech.co.uk/products/fluoroplastic-tubing-and-rod/pfa-tube.php, accessed on
18.10.2018.
(32) (a) Bey, P.; Vevert, J.; Dorsselaer, V. Van; Kolb, M. Direct Synthesis of A-
Halogenomethyl--Amino Acids from the Parent -Amino Acids. J. Am. Chem. Soc.
1979, 44, 2732–2742. (b) Kalyanam, N.; Rapole, K. R.; Rajendran, R.; Majeed, M. A
Novel Reaction Cascade Leading to a Spontaneous Intramolecular Dipolar Cycloaddition
through Simultaneous Generation of 1,3-Dipole and Dipolarophile. Tetrahedron Lett.
2013, 54, 5155–5158.
ACS Paragon Plus Environment