Continuous flow upgrading of glycerol toward oxiranes and active pharmaceutical ingredients thereof

Article abstract of DOI:10.1039/c9gc01819k

A robust continuous flow procedure for the transformation of bio-based glycerol into high value-added oxiranes (epichlorohydrin and glycidol) is presented. The flow procedure features a central hydrochlorination/dechlorination sequence and provides economically and environmentally favorable conditions involving an organocatalyst and aqueous solutions of hydrochloric acid and sodium hydroxide. Pimelic acid (10 mol%) shows an exceptional catalytic activity (>99% conversion of glycerol, a high selectivity toward 1,3-dichloro-2-propanol and 81% cumulated yield toward intermediate chlorohydrins) for the hydrochlorination of glycerol (140 °C) with 36 wt% aqueous HCl. These conditions are validated on a sample of crude bio-based glycerol. The dechlorination step is effective (quantitative conversion based on glycerol) with concentrated aqueous sodium hydroxide (20 °C) and can be directly concatenated to the hydrochlorination step, hence providing a ca. 2:3 separable mixture of glycidol and epichlorohydrin (74% cumulated yield). An in-line membrane separation unit is included downstream, providing usable streams of epichlorohydrin (in MTBE, with an optional concentrator) and glycidol (in water). The scalability of the dechlorination step is then assessed in a commercial pilot-scale continuous flow reactor. Next, bio-based epichlorohydrin is further utilized for the continuous flow preparation of β-amino alcohol active pharmaceutical ingredients including propranolol (hypertension, WHO essential), naftopidil (prostatic hyperplasia) and alprenolol (angina pectoris) within a concatenable two-step procedure using a FDA class 3 solvent (DMSO). This work provides the first example of direct upgrading of bio-based glycerol into high value-added pharmaceuticals under continuous flow conditions.

Full text of DOI:10.1039/c9gc01819k

View Article Online
View Journal
Green
Chemistry
Cutting-edge research for a greener sustainable future
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: R. Morodo, R.
Gérardy, G. Petit and J. M. Monbaliu, Green Chem., 2019, DOI: 10.1039/C9GC01819K.
Volume 20
Number
May 2018
Pages 1919-2160
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
9
7
Green
Chemistry
Cutting-edge research for a greener sustainable future
rsc.li/greenchem
Accepted Manuscripts are published online shortly after acceptance,
before technical editing, formatting and proof reading. Using this free
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
shall the Royal Society of Chemistry be held responsible for any errors
or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
ISSN 1463-9262
PAPER
Paul T. Anastas et al.
The Green ChemisTREE: 20 years after taking root with the 12
principles
rsc.li/greenchem

Page 1 of 13
Please Gd roe en no tC ha ed mj u iss tt r my argins
View Article Online
DOI: 10.1039/C9GC01819K
Green Chemistry
ARTICLE
Continuous flow upgrading of glycerol toward oxiranes and active
pharmaceutical ingredients thereof
<
Received 00th January 20xx,
a
a
a
a,*
Accepted 00th January 20xx
DOI: 10.1039/x0xx00000x
www.rsc.org/
Romain Morodo, Romaric Gérardy, Guillaume Petit and Jean-Christophe M. Monbaliu
A robust continuous flow procedure for the transformation of biobased glycerol into high value-added oxiranes
epichlorohydrin and glycidol) is presented. The flow procedure features a central hydrochlorination/dechlorination
(
sequence and provides economically and environmentally favorable conditions involving an organocatalyst and aqueous
solutions of hydrochloric acid and sodium hydroxide. Pimelic acid (10 mol%) shows an exceptional catalytic activity (99%
conversion of glycerol, a high selectivity toward 1,3-dichloro-2-propanol and 81% cumulated yield toward intermediate
chlorohydrins) for the hydrochlorination of glycerol (140 °C) with 36 wt.-% aqueous HCl. These conditions are validated on
a sample of crude biobased glycerol. The dechlorination step is effective (quantitative conversion based on glycerol) with
concentrated aqueous sodium hydroxide (20 °C) and can be directly concatenated to the hydrochlorination step, hence
providing a ca. 2:3 separable mixture of glycidol and epichlorohydrin (74% cumulated yield). An in-line membrane
separation unit is included downstream, providing usable streams of epichlorohydrin (in MTBE, with an optional
concentrator) and glycidol (in water). The scalability of the dechlorination step is then assessed in a commercial pilot-scale
continuous flow reactor. Next, biobased epichlorohydrin is further utilized for the continuous flow preparation of β-amino
alcohol active pharmaceutical ingredients including propranolol (hypertension, WHO essential), naftopidil (prostatic
hyperplasia) and alprenolol (angina pectoris) within a concatenated two-step procedure using a FDA class 3 solvent
(DMSO). This work provides the first example of direct upgrading of biobased glycerol into high value-added
pharmaceuticals under continuous flow conditions.
petrosourced and derived from propene, it is now mainly
produced as by-product of the
saponification/transesterification of animal or vegetable
Introduction
a
Increasing environmental awareness is conductive to a
transition from conventional, exclusively petrobased chemical
processes toward more sustainable alternatives. An emerging
and viable option consists in sourcing industrially relevant
building blocks from biomass. In particular, the chemical or
biochemical transformation of biobased platform molecules is
a rapidly increasing area of research with already industrial
success stories.1
7
–9
triglycerides. One of the largest sectors concerned with the
production of 1 is biodiesel manufacturing with an estimated
6
11
amount of 4  10 tons by 2024. Once purified, glycerol is
used as an additive in the food and beverage industry (additive
E422) or for pharmaceutical and personal care
1
2,13
preparations.
Emerging markets for glycerol also include
–4
electronic cigarettes.
Defined as “molecules with multiple functional groups that
possess the potential to be transformed into new families of
5
,6
useful molecules“ by the US Department of Energy, biobased
platform molecules are typically low complexity, low molecular
weight oxygenated molecules (Figure 1). Among these
platform molecules, propane-1,2,3-triol (also known as
glycerol, 1) is the smallest polyol that bears a significant
7
–10
industrial potential.
If glycerol was once exclusively
a. Center for Integrated Technology and Organic Synthesis, Research Unit MolSys,
University of Liège, B-4000 Liège (Sart Tilman), Belgium
jc.monbaliu@uliege.be www.citos.uliege.be
Electronic Supplementary Information (ESI) available: fluidic components,
experimental procedures, details of the fluidic setup and procedures (micro- and
mesoscale), off-line analysis with structural assignments (NMR). See
DOI: 10.1039/x0xx00000x
Fig. 1. Representative examples of biobased platform molecules.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

Please Gd roe en no tC ha ed mj u iss tt r my argins
Page 2 of 13
ARTICLE
Journal Name
Despite its apparent structural simplicity, 1 is a particularly eventually leads to the formation of dichlorohydrins 3a,b. This
View Article Online
DOI: 10.1039/C9GC01819K
useful C3 platform that can be upgraded into a large variety of sequence of equilibria greatly affects the overall reaction
7
–10
chemical building blocks that are otherwise petrobased.
Its output, and advanced optimization aims at maximizing the
chemical versatility in conjunction with its ever decreasing formation of 1,3-dichloropropan-2-ol (3b), the reactivity of
3
2
price (240 USD per ton) are major driving forces for the which is greater than 3a for the Williamson step. The
development of processes thereof. For instance, 1 has been hydrochlorination step is typically seen as rate determining for
successfully upgraded through oxidation, reduction, the preparation of 4,5. The epoxide formation typically relies
3
3
34
dehydration, deoxydehydration reactions or a combination of on an inorganic base, such as NaOH,
Ca(OH)
2
or
1
4
15
16
20
24
35
the aforementioned to allyl alcohol, acrolein, acrylic acid,
hydrotalcite-derived basic oxides. The ratio of compounds
3a,b/2a,b after the hydrochlorination step is an important
factor for the epichlorohydrin to glycidol 5/4 ratio after
1
7
18
19
divinyl glycol, dihydroxyacetone, acetol, pyruvaldehyde,
acrylonitrile,21 propene, lower alcohols and aromatics.
22
23
Various carbonates,2 esters26 and ethers27 derived from 1 treatment with a base, although 4 can be formed from 5 under
5
have also a promising industrial forecast.
alkaline conditions.
Glycidol (4) and epichlorohydrin (5) are valuable C3 oxirane
Besides their use in the polymer Industry, the unique
building blocks still mostly petrobased with applications deeply structural features of oxiranes 4,5, i.e. the combination of
rooted in the polymer Industry (epoxy resins, paints and multiple electrophilic centers on a C3-backbone, make them
coatings). Glycidol (4) is typically manufactured from allyl highly desirable building blocks for the preparation of fine
3
6
alcohol or indirectly from 5, while epichlorohydrin (5) is chemicals. In particular, 5 and derivatives have been
typically manufactured from propylene (Figure 2a). Solvay extensively studied as central building blocks for the
(
Epicerol process) and Dow (Glycerol to Epichlorohydrin, GTE preparation of active pharmaceutical ingredients (APIs)
4
36–40
process) developed biobased industrial processes starting bearing a -amino alcohol scaffold (Figure 3).
from 1 that cover about ¼ of the global production of (5) with
major production plants in China and Thailand.
Fig. 3. Examples of -amino alcohol APIs and precursor synthesized from
epichlorohydrin 5.
The assets of continuous flow micro/mesofluidic processes
are now well established for improving chemical performance
9
,41
while reducing the overall impact of a process. In addition,
the possibility to concatenate steps within uninterrupted flow
Fig. 2. a. General petrobased process for the preparation of oxirane 5. b.
Details of the biobased process for the preparation of oxirane 5 starting
from glycerol (1). The chlorination step illustrated here relies on HCl. Red reactor networks is particularly relevant for processes
arrows indicate the equilibria related to the hydrochlorination step. Blue
involving toxic intermediates such as monohydrins 2a,b,
arrows indicate reactions triggered under alkaline conditions (dechlorination
step). Homogeneous catalysts such as carboxylic acids 6 are typically used
for step 1.
42
dichlorohydrins 3a,b, and oxiranes 4,5.
Building upon our expertise in the development of
4
3,44
innovative flow processes for pharmaceutical production
The production of oxiranes 4,5 from 1 proceeds through a and for the upgrading of biobased molecules,14,45–47 we report
preliminary hydrochlorination step yielding a mixture of herein a continuous procedure for the upgrading of glycerol
monochlorohydrins (2a,b) and dichlorohydrins (3a,b), followed into APIs through the intermediate formation of
with an intramolecular Williamson reaction (Figure 2b). The epichlorohydrin following a multi module continuous flow
hydrochlorination step of 1 has attracted considerable strategy. The first module is dedicated to the transformation
attention; various chlorination agents such as gaseous or and valorization of glycerol into oxiranes 4,5. It features a
aqueous HCl,2 SOCl
8,29
30
or trimethylchlorosilane have been hydrochlorination/dechlorination sequence that provides a
31
2
successfully employed for the chlorination of 1, and various separable mixture of glycidol and epichlorohydrin. The
3
2
catalysts were assessed. The reaction proceeds first through sequence to 4,5 relies on economically and environmentally
the formation of 2-chloro-1,3-propanediol (2a) and 3-chloro- favorable conditions involving an organocatalyst and aqueous
1
,2-propanediol (2b), and
a
second chlorination then solutions of hydrochloric acid and sodium hydroxide. Pimelic
2
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 3 of 13
Please Gd roe en no tC ha ed mj u iss tt r my argins
Journal Name
ARTICLE
acid (10 mol%) shows an exceptional catalytic activity for the Scientific). Feed and collection lines consisted oVfiePwFAArtictleuObnilninge
hydrochlorination of glycerol at 140 °C with concentrated (1.58 mm outer diameter, 800 μm internDalOdI:i1a0m.1e03te9/rC)9eGqCu0ip18p1e9dK
aqueous HCl. The dechlorination step (Williamson reaction) is with PEEK/ETFE connectors and ferrules (IDEX/Upchurch
very efficient (>99% conversion based on glycerol) with Scientific). Liquid feeds were handled with high force Chemyx
concentrated aqueous sodium hydroxide (20 °C) and is directly
concatenated to the hydrochlorination step. An in-line
membrane separation unit was included downstream,
providing usable streams of epichlorohydrin (in MTBE) and
glycidol (in water). These conditions were validated on a
sample of unrefined biobased glycerol produced from
vegetable oils. The scalability of the critical dechlorination was
assessed in a commercial silicon carbide mesofluidic pilot
reactor. The second module is dedicated to the use of
epichlorohydrin obtained upstream for the continuous flow
preparation of various β-amino alcohol APIs, including
propranolol (hypertension, WHO essential), naftopidil
Fusion 6000 syringe pumps equipped with SS syringes and
Dupont Kalrez O-rings or with HPLC pumps (ThalesNano micro
HPLC). Downstream pressures were regulated with back
pressure regulators from Zaiput Flow Technologies (BPR-10) or
from IDEX/Upchurch Scientific. In-line liquid-liquid separation
was handled with a SEP-10 membrane separator from Zaiput
Flow Technologies.
Mesofluidic setup. Mesofluidic experiments (dechlorination
step) were carried out with a Corning® Advanced-Flow™ G1 SiC
Reactor (6 fluidic modules connected in series, 60 mL total
internal volume). Feed and collection lines consisted of PFA
tubing (1/4” o.d.) equipped with PFA or SS Swagelok
connectors and ferrules. Liquid feeds were handled with
Corning dosing lines (HMP gear pump and FLOM PTFE
intelligent pump). The reactor temperature was managed with
a LAUDA Integral XT 280 thermostat using LAUDA Therm 180
silicone oil. Downstream pressure was regulated with a back
pressure regulator from Zaiput Flow Technologies (BPR-1000).
(prostatic hyperplasia) and alprenolol (angina pectoris). This
work provides the first example of direct upgrading of
biobased glycerol into high value-added pharmaceuticals
under continuous flow conditions.
Experimental section
General information
Conversion and yield were determined by Gas Typical runs
Chromatography coupled to Flame Ionization Detection (GC-
Continuous flow generation of chlorohydrins 2a,b-3a,b from
biobased glycerol (1). Pimelic acid (6k, 10 mol% with respect to
) was solubilized in aqueous HCl (36 wt.-%) and loaded in a
PFA injection loop. The pumps used to deliver the solution of
HCl/catalyst and neat 1 were set to 68 µL min and 8.4 µL min
1,
respectively. Both streams were mixed through a PEEK T-
mixer and reacted in a PFA capillary coil (1.5 mL internal
volume, estimated 20 min residence time) at 140 °C under 8
bar of counterpressure. The reactor effluent was collected at
steady state, neutralized, diluted with water and ethanol, and
analyzed by GC-FID (>99% conversion, 81% cumulated yield,
FID) or by High Performance Liquid Chromatography coupled
to Diode-Array Detection (HPLC-DAD). Selectivity is defined as
the ratio between the yield and the conversion. Cumulated
yield is defined as the sum of all individual yields for a series of
related molecules. GC and HPLC conversions and yields were
determined using external calibration curves established with
commercial standards of glycerol, 3-chloropropane-1,2-diol,
1
-1
-
2
,3-dichloropropan-1-ol, 1,3-dichloropropan-2-ol, glycidol,
epichlorohydrin, 1-naphthol, 2-allylphenol, propranolol,
naftopidil and synthesized samples of naphthyl glycidyl ether,
2
3
-allylphenyl glycidyl ether, alprenolol and 1-(2-allylphenoxy)-
-[4-(2-methoxyphenyl)-1-piperazinyl]-2-propanol
(see see Table 2, entry 13).
Supporting Information for experimental procedures).
Structural identity was confirmed by 1H and C NMR Concatenated process toward epichlorohydrin (5) and glycidol (4)
13
spectroscopy (400 MHz Bruker Avance spectrometer) in CDCl
3
from biobased glycerol (1). Pimelic acid (6k, 10 mol% with
(Supporting Information). Glycerol, HCl 36 wt.-%, formic acid, respect to 1) was solubilized in aqueous HCl (36 wt.-%) and
acetic acid, glycolic acid, succinic acid, malic acid, L-(+)-tartaric
acid, oxalic acid, malonic acid, glutaric acid, adipic acid, pimelic
acid, γ-butyrolactone, ε-caprolactone, 3-chloropropane-1,2-
diol, 2-chloropropane-1,3-diol, NaOH, 2,3-dichloropropan-1-ol,
loaded in a PFA injection loop. The pumps used to deliver the
-1
solution of HCl/catalyst and neat 1 were set to 68 µL min and
-1
8
.4 µL min , respectively. Both streams were mixed through a
PEEK T-mixer and reacted in a PFA capillary coil (1.5 mL
internal volume, estimated 20 min residence time) at 140 °C
under 10 bar of counterpressure. The reactor effluent was
next mixed with aqueous NaOH (4 M, 1.5 equiv. with respect
1
2
,3-dichloropropan-2-ol, glycidol, epichlorohydrin, 1-naphthol,
-allylphenol, isopropylamine, (2-methoxyphenyl)piperazine,
4 2 2
propranolol, naftopidil, Zn(ClO ) and Zn(OTf) were obtained
from commercial sources.
-1
to HCl; 247.5 µL min ) through a PEEK T-mixer and reacted in a
PFA capillary coil (3.2 mL internal volume, estimated 10 min
residence time) at room temperature under 5.2 bar of
counterpressure. The reactor effluent from the second loop
was next extracted with methyl tert-butyl ether (MTBE, 320 µL
Experimental setup
Microfluidic setups. Microfluidic reactors consisted of modular
continuous flow assemblies constructed with PFA capillaries
(1.58 mm outer diameter, 800 μm internal diameter) equipped
-1
min ) through a PEEK T-mixer and a packed-bed column filled
with PEEK/ETFE connectors and ferrules (IDEX/Upchurch
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please Gd roe en no tC ha ed mj u iss tt r my argins
Page 4 of 13
ARTICLE
Journal Name
with glass beads (150-212 µm diameter). The biphasic solution coil (1.2 mL internal volume, estimated 10 min resViideweAnrcticeletOimnliene)
was next processed through a Zaiput Flow Technologies liquid- at 140 °C under 7 bar of counterpressure fl ent
D
.
O
T
I:
h
1
e
0.
r
1
e
03
a
9
c
/
t
C
o
9
r
G
e
C
f
0
1
u
8
19
K
liquid separator (SEP-10) equipped with a hydrophobic was collected at steady state, diluted with ethanol and
membrane (0.5 µm pore size). The aqueous outlet of the analyzed by GC-FID or/and by HPLC-DAD (10a: 91% conversion,
separator was connected to an adjustable BPR to allow 75% yield, 10b: >99% conversion, 95% yield, 10c: 88%
efficient liquid separation. The reactor effluents were collected conversion, 74% yield, 10d: >99% conversion, >99% yield, see
at steady state, neutralized, diluted with water and/or ethanol Table 6, entries 12 and 16-18). Purification procedures for
and analyzed by GC-FID (>99% conv., 74% cumulated yield compounds 10a-d are available in the Supporting Information
toward 4 and 5, for partition coefficients see Table 3, entry 5).
(section 2.2.12).
Continuous flow dechlorination of monochlorohydrin (2b) and Concatenated catalyst-free process for the direct preparation
dichlorohydrin (3b) toward glycidol (4) and epichlorohydrin (5) of APIs 10a,b from epichlorohydrin (5). A 0.18 M solution of
(
mesofluidic setup). The pump used to deliver the solution 7a (1 equiv.) in DMSO containing 1.1 equiv. of KOtBu and a
containing 3-chloropropane-1,2-diol (2b, 0.60 M), 1,3- stream of neat epichlorohydrin (5, 4 equiv.) were injected at
-
1
-1
dichloropropan-2-ol (3b, 0.77 M) and pimelic acid (6k, 0.17 M) in flow rates of 82.5 µL min and 5 µL min , respectively. Both
-1
HCl 36 wt.-% was set to 4.7 mL min (3 min residence time) or to streams were mixed through a PEEK T-mixer and reacted in a
.4 mL min (1.5 min residence time). The pump used to deliver the PFA capillary coil (1.75 mL internal volume, 20 min residence
-1
9
solution of NaOH 4 M was set to 15.3 mL min-1 (3 min residence time) at 70 °C. The reactor effluent was next mixed with neat
time) or to 30.6 mL min-1 (1.5 min residence time). Both streams amine 9a or 9b (6 equiv., 10.3 or 15.7 µL min , respectively)
were mixed and reacted in a Corning® Advanced-Flow™ G1 SiC through a PEEK T-mixer and reacted in a PFA capillary coil
Reactor (60 mL internal volume) operated at 21 °C under 5 bar of (2 mL internal volume, estimated 20 min residence time) at
counterpressure. The reactor effluent was collected at steady 140 °C under 7 bar of counterpressure. The effluent from the
state in a phosphate buffer solution (pH 6.2), diluted with second loop was collected at steady state, diluted with ethanol
-1
ethanol, and analyzed by GC-FID (see Table 4).
and analyzed by GC-FID and/or by HPLC-DAD (10a: 47% yield,
0b: 60% yield).
1
Continuous flow catalyst-free Williamson synthesis toward
glycidyl ether 8a,b. A 0.18 M solution of 7a or 7b (1 equiv.) in
DMSO containing 1.1 equiv. of KOtBu and a stream of neat
Results and Discussion
Continuous flow synthesis of mono- (2a,b) and dichlorohydrins
-1
epichlorohydrin were injected at flow rates of 165 µL min and
1
0 µL min-1
,
respectively. Both streams were mixed through a (3a,b) from biobased glycerol (1)
PEEK T-mixer and reacted in a PFA capillary coil (3.5 mL
internal volume, 20 min residence time) at 70 °C under 5 bar of
counterpressure. The reactor effluent was collected at steady
state, diluted with ethanol, and analyzed by GC-FID (8a: 97%
conversion and 90% yield; 8b: 99% conversion and 68% yield,
see Table 5, entry 6 and 12).
First reports on the chlorination of alcohols under continuous
flow conditions date back to the late 80s. Tundo and co-
workers reported a solvent-free chlorination of primary
alcohols using a Lewis acid or a phosphonium salt catalytic bed
48,49
and aqueous HCl at 170 °C.
More recently, Kappe et al.
described the conversion of several primary alcohols into the
corresponding chlorides with concentrated aqueous HCl (30
Catalyst-free continuous flow aminolysis toward propranolol 10a. wt.-%) at 160-180 °C (20 bar) within 15 min of residence time
5
0
The pumps used to deliver 8a (0.18 M in DMSO) and neat in a 1 mL reactor chip. Hessel and coworkers also reported
-
1
-1
isopropylamine (9a) were set to 55 µL min and 2.6 µL min , the preparation of a library of chlorides starting from primary
and secondary alcohols and concentrated aqueous HCl (36 wt.-
respectively. Both streams were mixed through a PEEK T-mixer
and reacted in a PFA capillary coil (1.2 mL internal volume,
%
) under flow conditions for the preparation of APIs including
5
1
antiemetic agents and APIs from the antihistamine family.
2
0 min residence time) at 140 °C under 7 bar of
This process was later adapted with gaseous HCl for the
counterpressure. The reactor effluent was collected at steady
state, diluted with ethanol and analyzed by GC-FID (91%
conversion, 75% yield, see Table 6, entry 7).
5
2
preparation of a precursor of anticonvulsant Rufinamide. de
Araujo Filho and colleagues studied the fluid dynamics related
to the continuous hydrochlorination of 1. A liquid feed of 1
containing acetic acid as catalyst and a gaseous feed of HCl
were reacted in a bubble column reactor of 1.62 cm internal
Catalyzed continuous flow aminolysis toward APIs 10a-c and
analog 10d. The pump used to deliver 8a or 8b (0.18 M) was set
5
3
diameter.
-1
to 110 µL min . Neat amine 9a or 9b was delivered at a flow
The chlorination of glycerol (1) under meso-/microfluidic
conditions and utilizing aqueous HCl has not been reported in
the literature so far. The data available from the prior Art
indicate that the performance of the reaction is greatly
enhanced in the presence of an acid catalyst. Most of the
-1
rate of 5.1 or 10.3 µL min , respectively. Aqueous Zn(ClO
4 2
)
(
0.38 M, 10 mol% with respect to 8a or 8b) was delivered at a
-1
flow rate of 5.1 µL min . The three streams were mixed
through a PEEK cross-junction and reacted in a PFA capillary
4
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 5 of 13
Please Gd roe en no tC ha ed mj u iss tt r my argins
Journal Name
papers dealing with the chlorination of 1 report the use of
ARTICLE
View Article Online
Table 1. Preliminary optimization of the hydrochlorination of glycerol toward
5
4,55
DOI: 10.1039/C9GC01819K
homogeneous carboxylic acid catalysts,
with only a few
chlorohydrins 2 and 3 under continuous flow conditions.a
exceptions.29 In 2014, Santacesaria and coworkers reviewed a
list of carboxylic acids and derivatives used as catalysts for the
chlorination of 1.32 Their survey indicated that alicyclic
carboxylic acids (3 mol%) such as acetic, hexanoic and
heptanoic acids gave excellent results (93-94% conv., 87-91%
selec. for 3b) within 4 h at 110 °C. Widely available lactones
such as γ-butyrolactone and ε-caprolactone were reported
with excellent catalytic activities as well (96% conv., 94% selec.
for 3b). In 2005, Gilbeau patented a process for the production
of dichloropropanols 3a,b from 1 with HCl under batch
conditions at 120-130 °C (20 h) in the presence of adipic acid
Res.
Entry Time
(min)
2b
3b
Cumulated
yield
T
°C)
Pressure Conv.
(bar)
yield yield
(%)b
(
(%)b
(%)b
(%)
1
2
20
20
20
20
20
20
20
30
90
2.5
2.5
2.5
8
3
3
/
/
3
100
110
120
130
140
150
140
15
5
5
3
17
10
34
48
73
30
37
/
10
38
53
88
56
57
4
44
4
5
8
54
5
6
7c
8c
8
88
16
26
20
11
11
>99
>99
5
4
as a catalyst (99% conv. and 82% selec. for 3a,b). Catalytic
efficiency is typically related to lack of steric hindrance in the
vicinity of the carboxylic acid function rather than to the
aCatalyst: acetic acid (10 mol%). bDetermined by GC-FID. Significant amounts of
solid residues were observed inside the reactor.
c
4
,56
a
pK .
According to an extensive mechanistic study by
monochlorhydrin 2a were detected (<1%). 1,2,3-
trichloropropane 3c, which is formed as a very toxic trace
contaminant in the petrobased process, was not detected
during the experiments.
The reaction temperature had a critical impact on the
reaction: in the 90-110 °C range, low conversions of glycerol
and low yields of 2b were observed (entries 1-3, Table 1).
Increasing the temperature up to 120 °C triggered the
formation of dichlorohydrin 3b along with an increased
amount of 2b and thus an overall increase of the conversion of
Santacesaria,28 the catalysis of the hydrochlorination reaction
involves a transient esterification of catalyst 6 by 1, and the
subsequent formation of an oxonium reactive intermediate
through an intramolecular displacement of the carboxylate,
hence regenerating the catalyst. The oxonium reactive
N
intermediate then reacts with a chloride anion through a S 2
reaction. Accordingly, the formation of monochlorohydrin 2b
is about 15 times faster than 2a, and the second chlorination
toward 3b is generally 90 to 100 times faster than the
formation of 3a.28
1
. At 140 °C, a maximum cumulated yield of 88% was attained,
Starting from the data available in the prior Art,
concentrated aqueous HCl was selected as a convenient
hydrochlorination reagent rather than gaseous HCl for safety
although 2b remained the major product (entries 4-6). Higher
process temperatures or longer residence times gave
quantitative conversion and better selectivities for
dichlorohydrin 3b; however, significant side-reactions toward
the formation of humins led to the accumulation of solid
materials in the reactor, hence reducing the overall yield
3
2
concerns. Acetic acid was selected as a model catalyst, and
preliminary optimization was undertaken under continuous
flow conditions for the production of chlorohydrins 2a,b and
3
a,b from 1 (Scheme 1).
(entries 7-8) and eventually clogging the reactor. The
continuous flow procedure showed great promises with high
conversion and cumulated yield within only 20 min of
residence time, while conventional batch procedures typically
require several hours and gaseous HCl to reach similar
performance. Additionally, the sealed nature of the flow
system alleviates catalyst loss by evaporation, which is a well-
3
2
known issue for the hydrochlorination of 1 in batch.
According to this preliminary screening, the best conditions
involved a reaction temperature of 140 °C with a residence
time of 20 min under 8 bar of counterpressure.
Next, a library of homogeneous carboxylic acids 6a-m was
screened (Figure 4). Despite overall good performance, acetic
the presence of a homogeneous carboxylic acid catalyst 6 under continuous acid (6b) led to a poor selectivity toward dichlorohydrins 3a,b.
Scheme 1. Hydrochlorination of glycerol with concentrated aqueous HCl in
flow conditions (see also Tables 1-2).
The various catalysts envisioned are regrouped according to
four different classes: class A includes formic (6a), acetic (6b)
and glycolic (6c) acids; class B regroups -hydroxylated
dicarboxylic acids derived from the succinic acid backbone
The catalyst (10 mol%) was dissolved in the feed of
aqueous HCl, and the resulting solution was mixed with neat 1
through a T-mixer and reacted at 90-150 °C in a coil reactor
(6d), including malic (6e) and L-tartaric acid (6f); class C gathers
(Scheme 1). The pressure was regulated downstream to
aliphatic dicarboxylic acids with increasing chain length,
including oxalic (6g), malonic (6h), glutaric (6i), adipic (6j) and
pimelic (6k) acids; group D clusters two common lactones,
enable superheated conditions while maintaining the reagents
in the liquid state. Samples were collected at steady state and
analyzed by off-line GC-FID (Table 1). Only traces of
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Please Gd roe en no tC ha ed mj u iss tt r my argins
Page 6 of 13
ARTICLE
Journal Name
namely, γ-butyrolactone (6l) and ε-caprolactone (6m, Figure 4
and Table 2). The same continuous flow setup as described in
Scheme 1 was used for the screening of catalysts.
Table 2. Catalyst study for the continuous flow generation of VchielworoAhrtyicdlerinOsn2line
and 3 starting from glycerol.a
DOI: 10.1039/C9GC01819K
2
b
3a
3b
Cumulated
yield
In the absence of a catalyst, poor conversions and only
traces of 2b were obtained. In addition, the formation of
significant amounts of solid material was observed (entry 1,
Table 2). Compared to acid acetic (entry 2), formic (6a) and
glycolic (6c) acids gave low conversion and yield (entries 3-4).
Catalyst
Conv.
(%)b
Entry
Class
yield
%)b
yield
(%)b
yield
(%)b
(mol%)
(
(%)
1c
2
3c
/
/
26
88
32
39
84
45
29
8
4
0
0
0
0
1
0
0
0
0
0
0
0
2
0
1
0
16
<1
0
4
A
A
A
B
B
B
C
C
C
C
C
C
D
D
6b (10)
6a (10)
6c (10)
6d (10)
6e (10)
6f (10)
6g (10)
6h (10)
6i (10)
6j (5)
73
17
15
60
32
8
88
17
15
71
32
8
5
7
58
Thermal instability of 6a and 6c is likely affecting their
catalytic efficiency compared to 6b. The catalysts from class A
clearly favor the formation of monochlorohydrin 2a, with only
low amounts (16% with 6b) or traces (6a,c) of dichlorohydrin.
The catalytic efficiency of acids 6d-f in group B clearly
illustrates the importance of the steric hindrance associated
with the -hydroxyl groups (entries 5-7). Succinic acid (6d)
gave good conversion (84%) and cumulated yield (71%), with a
greater selectivity for monochlorohydrin 2b (entry 5).
4
5
10
<1
0
6
7c
8
9c
7
0
7
90
51
63
95
>99
95
89
43
34
38
47
35
58
60
7
50
38
42
77
81
77
80
1
0
4
1
1d
4
1
2
3
4
5
6k (5)
30
44
19
19
1
6k (10)
6l (10)
6l (10)
1
1
a
b
Conditions (see Scheme 1): 140 °C, 20 min residence time, 8 bar. Determined
by GC-FID. cSignificant amounts of solid residues were observed inside the
reactor. Catalyst solubility was too low for a 10 mol% loading.
d
of catalyst). These results are in sharp contrast with the
selectivity reported for conventional carboxylic acid catalysts
such as acetic acid (6b). Group D catalysts 6l,m gave high
conversion and yield with
a
high selectivity for
monochlorohydrin 2b (entries 14,15). To conclude the
screening of a library of homogeneous carboxylic acid
catalysts, pimelic acid (6k) gave the best results with a
quantitative conversion and
a
high selectivity for
dichlorohydrin 3b, and was thus selected for the next steps of
this study. Other parameters such as the HCl/1 ratio, the
introduction of a co-solvent or the use of organic solutions of
HCl were explored as well, but the reaction could not be
further improved (Supporting Information).
Fig. 4. Library of homogeneous catalysts for the hydrochlorination step (see
also Scheme 1 and Table 2).
Adding one (6e) or two (6f) -hydroxyl groups on the
Biobased crude (unrefined) 1 obtained from vegetable oils
succinic acid backbone drastically affected the conversion (see Supporting Information for specifications, Section 2.1,
45% with 6e and 29% with 6f), and only low yields of Table S3) was next evaluated for the hydrochlorination toward
(
monochlorohydrin 2b (entries 6-7) were obtained. Significant chlorohydrins with the optimized conditions (Table 2, entry
amounts of solid material were observed in the reactor with 13). Quantitative conversion of glycerol was observed and 41%
tartaric acid (6f). Class C catalysts 6g-k were next evaluated (2b), 1% (3a) and 19% (3b) yields were obtained. Even in
(entries 8-13). Oxalic acid 6g led to poor conversion and presence of various contaminants (water, salts, organics,
cumulated yield (entry 8). Malonic acid (6h) led to a higher MeOH, etc.), 61% selectivity toward chlorohydrins is observed,
conversion (entry 9); however, 6h also promoted side- highlighting the robustness of the process and its potential
reactions since it led to a reduced selectivity and solid direct application on unrefined glycerol.
materials were observed in the reactor. Glutaric (6i) and adipic
(
6j) acids performed reasonably well with 51 and 63%
Continuous flow hydrochlorination/dechlorination sequence of 1
toward oxiranes 4,5
conversion, respectively, with high selectivity toward
monochlorohydrin 2b (entries 10-11). It should be emphasized
that the loading of 6j was reduced to 5 mol% due to its limited
solubility, yet giving good conversion. Excellent results were
obtained using pimelic acid 6k (entries 12-13): quantitative
conversion, high cumulated yield and, most importantly, a
higher selectivity toward dichlorohydrin 3b (44% with 10 mol%
The dechlorination step from 3b to 5 was investigated in a
millireactor by de Araujo Filho to provide insight on the
reaction kinetics with aqueous NaOH. The authors concluded
that the intramolecular Williamson reaction was fast (near
complete conversion in 30 s at 60-70 °C), and could be even
6
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 7 of 13
Please Gd roe en no tC ha ed mj u iss tt r my argins
Journal Name
ARTICLE
further accelerated with higher temperatures and an excess of
sodium hydroxide.33 Pérez-Ramírez reported a continuous
View Article Online
DOI: 10.1039/C9GC01819K
Table 3. Continuous flow in-line liquid-liquid separation of glycidol (4) and
epichlorohydrin (5)a
process involving a solid base packed in a fixed-bed reactor for
the gas-phase dechlorination of 3b into 5. The yield dropped
from ~60% to ~15% during the first hour of continuous
operation.35 Based on the data from the prior Art, and our
study of the hydrochlorination of 1 in flow, decision was made
to directly concatenate the dechlorination step with the
upstream hydrochlorination step, thus enabling a fully
telescoped process with no handling or isolation of toxic
chlorohydrins. The reactor effluent was mixed with aqueous
sodium hydroxide through a T-mixer and further reacted in a
second reaction coil at room temperature (Scheme 2).
Given that an excess of HCl was utilized upstream in the
hydrochlorination step, an excess of sodium hydroxide was
required to effect both neutralization and epoxidation. First
trials using aqueous NaOH 12.5 M failed because of reactor
clogging (precipitation of NaCl). The concentration of NaOH
was decreased to 4 M, which led to 74% cumulated yield for
oxiranes 4,5 in a ca. 2:3 ratio (10 min residence time at 20 °C).
c
Partition coefficient (%)
Extraction Aq./Org.
solvent
Entry
ratiob
Aqueous
phase
Organic
phase
1
2
3
4
5
6
7
8
9
MeTHF
MeTHF
MeTHF
MTBE
MTBE
MTBE
DBE
2:1
3:2
1:1
3:2
1:1
1:2
1:1
1:1
1:1
4 (90), 5 (12)
4 (89), 5 (11)
4 (82), 5 (1)
4 (98), 5 (9)
4 (10), 5 (88)
4 (11), 5 (89)
4 (18), 5 (99)
4 (2), 5 (91)
4 (2), 5 (99)
4 (6), 5 (99)
4 (2), 5 (94)
4 (25), 5 (99)
4 (<1), 5 (99)
4 (98), 5 (1)
4 (94), 5 (1)
4 (98), 5 (6)
4 (75), 5 (1)
4 (>99), 5 (1)
MEK
BA
MeTHF = 2-methyltetrahydrofuran, MTBE = methyl tert-butyl ether, DBE =
dibutyl ether, MEK = methyl ethyl ketone, BA = n-butyl acetate. For general
process conditions, see Scheme 2. bVolumetric ratio between the aqueous
a
c
reactor effluent and the organic extraction solvent. Determined by GC-FID.
Compared to the yields obtained for the hydrochlorination mixed with an extraction solvent through a PEEK T-mixer and
step alone (Table 2, entry 13), the results obtained for the the resulting biphasic mixture was passed through a packed-
concatenated system suggest that the dechlorination step is bed reactor filled with glass beads before membrane
highly selective toward both 4 and 5.
separation. Several biobased and/or green extraction solvents
according to the main solvent selection guides, including
5
9–62
CHEM21, GSK and Sanofi solvent guides were considered.
Extraction solvents including 2-methyltetrahydrofuran
MeTHF), methyl tert-butyl ether (MTBE), dibutyl ether (DBE),
(
methyl ethyl ketone (MEK) and n-butyl acetate (BA) were
evaluated for the selective extraction of 5 out of the reactor
effluent.
Good extraction performance was obtained with MeTHF,
which allowed to retrieve ca. 90% of 5 from the aqueous crude
reactor effluent. The best extraction results were obtained
when the aqueous/organic phase ratio was greater than 1
(Table 3, entries 1-2). With a 1:1 ratio, 99% of 5 was extracted
in MeTHF along with a significant amount of 4 (18%). MTBE
performed extremely well: with a 1:1 aqueous/organic phase
ratio, 99% of 5 was effectively extracted in the organic phase,
while retaining 98% of 4 in the aqueous effluent. Under these
conditions, an effective and highly selective separation of
oxiranes 4,5 was possible. Additionally, the low boiling point
(55 °C) of MTBE was definitively seen as an asset for further
downstream processing of 5. No further improvements of the
extraction process were noticed with DBE or MEK as organic
solvents (entries 7-8). BA gave the best extraction
performance, providing after membrane separation an
Scheme 2. Concatenated process for the upgrading of glycerol (1) into
oxiranes 4,5 including a downstream in-line selective extraction unit. >99%
conversion (based on glycerol) and 74% cumulated yield for oxiranes 4,5 aqueous stream containing >99% of 4 and 1% of 5, while the
(
yield for 4: 30%; yield for 5: 44%). The packed-bed column is not
organic stream contained 99% of 5 and less than 1% of 4.
However, the much higher boiling point of BA (126 °C)
compared to MTBE was seen as a drawback for further
downstream applications. Upon optimization, a very effective
system for the upgrading of glycerol into valuable oxiranes 4,5
was thus devised. The successful extraction based on MTBE
enabled to recover a usable organic stream of 5, as well as an
aqueous solution of 4. An optional concentrator was also
implemented for the direct concentration of 5 and recycling of
represented for clarity (Supporting Information for the detailed setup). An
optional concentrator can be included downstream for the semi-continuous
distillation and recycling of MTBE (Supporting Information) to recover
concentrated biobased 5 (see also Table 3).
The implementation of a downstream in-line extraction
unit was considered for the purification of the reactor effluent.
The extraction unit relied on a membrane separator equipped
with a hydrophobic membrane. The reactor effluent was
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

Please Gd roe en no tC ha ed mj u iss tt r my argins
Page 8 of 13
ARTICLE
Journal Name
MTBE by distillation (see Supporting information). Within the Application of biobased epichlorohydrin (5) for the pVrieewpAarrtiaclteioOnnlionef
high value-added β-amino-alcohol APIs uDnOdIe: 1r0.c1o0n39ti/nCu9oGuCs01f8l1o9wK
conditions
context of this project, the study went further on the
scalability of the dechlorination step and the use of reactive
epichlorohydrin (5) for the preparation of APIs.
With the optimization of the upgrading of 1 for the
preparation of oxiranes 4,5 and the development of a selective
extraction procedure, further valorization of biobased
epichlorohydrin was sought. The preparation of high value-
added β-amino alcohol APIs from 5 such as propranolol (10a, a
medication against hypertension listed as an essential drug by
the WHO), naftopidil (10b, a medication for treating prostatic
hyperplasia) and alprenolol (10c, a medication for treating
angina pectoris) was envisioned under continuous flow
conditions. A common strategy starting from 5 generally
requires two steps: (a) a Williamson glycidyl ether synthesis
followed by (b) a final aminolysis of the oxirane ring (Figure 6).
Scalability of the continuous flow synthesis of glycidol (4) and
epichlorohydrin (5) from monochlorohydrin (2b) and
dichlorohydrin (3b)
Based on our optimized microfluidic conditions for the
hydrochlorination/dechlorination sequence of 1, the scalability of
the dechlorination step was evaluated. A model solution mimicking
the results of the hydrochlorination of 1 (Table 2, entry 13)
containing 2b, 3b and residual catalyst 6k in aqueous hydrochloric
acid was reacted with a stream of aqueous NaOH (4 M) in a
commercial mesofluidic pilot scale reactor (Corning® Advanced-
Flow™ G1 SiC reactor, 60 mL internal volume) equipped with 6
fluidic modules (Figure 5). The highly exothermic neutralization and
the subsequent epoxidation reaction were conveniently handled in
the mesofluidic reactor (Table 4). A short residence time of 1.5 min
led to quantitative conversion of monochlorohydrin 2b and 97% of
conversion for dichlorohydrin 3b (entry 1). High yields were
obtained for oxiranes 4 (81%, based on 2b) and 5 (75%, based on
Fig. 6. General two-step strategy for the preparation of β-amino alcohol
APIs.
3
b), highlighting the efficiency of the mesofluidic system at room
Concerning the first step (glycidyl ether synthesis),
Seeberger38 and McCluskey recently reported continuous
flow procedures for the generation of glycidyl ethers starting
from 5. However, their procedures relied on a phase-transfer
temperature and using a reduced residence time. The process
allowed a production of 6.50 mol per day of 4 and 7.75 mol per day
of 5. Increasing the residence time (entry 2), and thus decreasing
the total flow rate, led to a significant decrease of conversion of 2b
39
3
8
38
catalyst, long residence times (up to 60 min), the use of a
(53%) and yield toward 4 (46%). This phenomenon most likely
3
9
large excess of 5 (50 equiv.) and a FDA class 2 solvent
reflects the decrease in mixing efficiency at lower flow rates, hence
affecting the neutralization and dechlorination processes.
3
9
(
DMF). In this study, an alternative catalyst-free procedure
6
3
relying on biobased 5 and a FDA class 3 solvent (DMSO) was
developed. KOtBu was selected as a base since it is both an
inexpensive reagent and has a low toxicity profile. The
DMSO/KOtBu combination was reported to provide both
strong basicity and high solubility, which are important
characteristics for this continuous flow procedure. The
continuous flow setup devised for the preparation of
intermediate glycidyl ethers 8 combined simplicity and
robustness, yet ensuring high versatility (Scheme 3). Keeping
6
4
Fig. 5. Simplified flow chart for the dechlorination toward oxiranes 4,5. The
insert on the right shows the details of a silicon carbide fluidic module (FM)
from a Corning® Advanced-Flow™ G1 SiC reactor. A = aqueous solution of the same reactor configuration while changing the feed
NaOH (4 M); B = solution of 2b, 3b in aqueous HCl (36 wt.-%) in the presence solution that contained aryl alcohols 7a,b enabled the
of pimelic acid (6k, 10 mol%). BPR: back pressure regulator.
preparation of two different glycidyl ether intermediates 8a,b.
Table 4. Continuous flow dechlorination of 2b and 3b toward oxiranes 4,5
under mesofluidic conditionsa
Residence
time
Conv.
2bb
Conv.
3bb
Yield 4b
%)
Yield 5b
Entry
(
(%)
(min)
(%)
(%)
1
2
1.5
3
>99
53
81
46
97
93
75
77
a
General conditions: 2b (0.60 M), 3b (0.77 M) and 6k (0.17 M) in HCl 36 wt.-%.,
Scheme 3. Continuous flow preparation of glycidyl ether intermediates 8a,b
from biobased 5 (see also Table 5). General conditions: 4 equiv. of 5, 1
equiv. of aryl alcohol 7a,b, 1.1 equiv. of KOtBu (see also Table 5).
NaOH 4 M (1.5 equiv. with respect to HCl), 21 °C, 5 bar, 60 mL of internal
volume. Determined by GC-FID.
b
8
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 9 of 13
Please Gd roe en no tC ha ed mj u iss tt r my argins
Journal Name
ARTICLE
The optimization began with 1-naphthol (7a) for the 0.18 M solution of glycidyl intermediate 8a with a stream of
View Article Online
DOI: 10.1039/C9GC01819K
preparation of a common precursor 8a to propranolol (10a) neat isopropylamine 9a.
and naftopidil (10b). The temperature was screened with a
defined residence time of 20 min (Table 5, entries 1-6). The
best results were obtained at 70 °C with excellent conversion
(97%) and yield (90%) (entry 6). At 80 °C, extensive solid
formation led to reactor clogging (entry 7-8). Decreasing the
residence time down to 10 min enabled higher temperature
without clogging issues with good conversion (97%) and yield
(81%) as well (entry 9). The conditions were next transposed
to 2-allylphenol (8b) as a direct precursor to alprenolol (10c)
and analog (10d) (entries 11-12). At 70 °C, 99% of conversion
was obtained with 68% of yield for intermediate 8b (entry 12),
therefore the selectivity was lower than for 8a.
The second step aimed at the aminolysis of glycidyl ether
intermediates 8a,b. Previous literature reports for the
preparation of β-amino alcohols by aminolysis of oxiranes
under continuous flow conditions were already reported.
Jamison described a microfluidic system under superheated
conditions for the efficient preparation of amino alcohols in
ethanol and more recently a seven-step continuous flow
system for the preparation of the antibacterial drug linezolid,
which included an aminolysis step toward an amino alcohol
intermediate.3 Seeberger and McCluskey reported a similar
approach with a toluene/ethanol solvent mixture for the
preparation of various β-amino alcohol intermediates of
7,40
Scheme 4. Continuous flow aminolysis reaction toward β-amino alcohol APIs
and analog 10a-d. General conditions: 1 equiv. of 8, 1.2-6 equiv. of amine 9
and optionally 5-10 mol% ZnX (see also Table 6). Yields (GC-FID or HPLC-
2
3
8,39
pharmaceutical interest.
Lebel also reported the ring
opening of an aryl epoxide using an excess of ammonium
hydroxide.65
DAD, see Table 6) for compounds 10a-d are indicated (isolated yields are
indicated in parentheses).
Both catalyst-free66 and Zn-catalyzed67 conditions were
Since the upstream preparation of glycidyl ether
intermediates 8a,b was carried out in DMSO, decision was considered. The optimization started in the absence of a
made to utilize DMSO as a reaction medium for the aminolysis catalyst, and the impact of the temperature was assessed
reaction toward β-amino alcohol APIs and analog 10a-d (Table 6, entries 1-6) with a residence time of 20 min. Process
(
Scheme 4). The optimization started with the preparation of temperatures of 140 °C or higher led to quantitative
propranolol 10a. Typical conditions included, as reactant, a conversion with high yield of propranolol 10a (75%, entries 5-
). The excess of amine 9a was also optimized at 140 °C
entries 7-9). Decreasing the excess of 9a (3 equiv.) slightly
6
(
Table 5. Optimization for the preparation of glycidyl ether intermediates 8a,b.a
affected the conversion while maintaining a similar yield (entry
7). Further decrease of the excess led to a global loss of
efficiency with both conversion and yield dropping to 53 and
Aryl
alcohol
7
Residence
time
T
Product
Conv.b
(%)
Yieldb
(%)
Entry
(
°C)
8
(min)
3
9%, respectively (entries 8 and 9). Comparable outcomes
1
2
7a
7a
7a
7a
7a
7a
7a
7a
7a
7a
7b
7b
20
30
40
50
60
70
80
80
80
90
60
70
20
20
20
20
20
20
20
15
10
10
20
20
8a
8a
8a
8a
8a
8a
8a
8a
8a
8a
8b
8b
23
46
62
92
97
97
/
23
39
62
86
87
90
/
were obtained when decreasing to 10 min of residence time
(entry 10).
Aminolysis reactions are known to be improved in the
presence of zinc-based catalyst, and in particular zinc
perchlorate was reported as a very effective homogeneous
3
4
5
6
7
6
catalyst. The catalytic activity of Zn(ClO
explored for the preparation of 10a. To this end, a third feed
consisting of an aqueous solution of Zn(ClO was added to
the continuous flow setup (Scheme 4 and Table 6, entries 11-
15). The activity of Zn(ClO enabled a significant reduction in
4 2
) was therefore
7c
8c
9
/
/
4 2
)
97
96
99
99
81
78
42
68
1
1
1
0
1
2
4 2
)
residence time (10 min) while keeping similar performance as
for the non-catalyzed process. With 10 mol% of catalyst,
propranolol (10a) was obtained in 75% yield within 10 min of
residence time at 140 °C (7 bar). Other Zn-based catalysts such
a
General conditions (see Scheme 3): 5 (4 equiv.), 7a,b (1 equiv.) and KOtBu (1.1
equiv.). bDetermined by GC-FID. Important amounts of solid residues leading
to reactor clogging.
c
2
as Zn(OTf) or higher process temperatures and residence
times failed to further increase the yield (entries 13-15). The
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 9
Please do not adjust margins

Please Gd roe en no tC ha ed mj u iss tt r my argins
Page 10 of 13
ARTICLE
Journal Name
yields of 47% and 60% (with respect to 7a) toward 10a and
View Article Online
DOI: 10.1039/C9GC01819K
Table 6. Optimization of the aminolysis reaction toward β-amino alcohol APIs
1
0b, respectively, were obtained. As observed for the
and analog 10a-d.a
individual steps, amine 9b led to a higher yield compared to
a. An excess of reagent 5 for the Williamson step is likely
9
Res.
Glycidyl
Entry
Amine
T
Conv.b
(%)
Yieldb
(%)
time
Product
responsible for the decrease in overall yield in the
concatenated version of the process. However, this procedure
allows the efficient valorization of biobased 5 toward high-
value added APIs in a single microfluidic network under
catalyst-free conditions and with 40 min of residence time.
ether
(equiv.)
(°C)
(min)
1
2
3
4
5
6
7
8
9
8a
8a
8a
8a
8a
8a
8a
8a
8a
8a
8a
8a
8a
8a
8a
8a
8b
8b
9a (6)
9a (6)
9a (6)
9a (6)
9a (6)
9a (6)
9a (3)
9a (2)
9a (1.2)
9a (3)
9a (3)
9a (3)
9a (3)
9a (3)
9a (3)
9b (3)
9a (3)
9b (3)
80
20
20
20
20
20
20
20
20
20
10
10
10
10
10
15
10
10
10
10a
10a
10a
10a
10a
10a
10a
10a
10a
10a
10a
10a
10a
10a
10a
10b
10c
10d
14
49
14
36
59
73
75
75
75
57
39
48
64
75
66
63
73
95f
74
>99f
100
120
130
140
150
140
140
140
140
140
140
140
150
140
140
140
140
83
95
>99
>99
91
74
53
1
0
78
1
1c
94
1
1
1
1
1
1
1
2d
3e
91
92
Scheme 5. Catalyst-free concatenated system including a Williamson and an
aminolysis step for the production of APIs 10a,b starting from
epichlorohydrin (5). General conditions: 1 equiv. of 7a, 4 equiv. of 5 and 6
equiv. of amine 9a,b. Yields for compounds 10a,b are indicated (GC-FID or
HPLC-DAD).
4d
99
5d
6d
7d
8d
99
>99
88
>99
Conclusion
a
b
c
For general conditions, see Scheme 4. Determined by GC-FID. Third feed of
d
e
aqueous Zn(ClO
feed of aqueous Zn(OTf)
4
)
2
(2 mol%). Third feed of aqueous Zn(ClO
4
)
2
(10 mol%). Third
This work illustrates the first direct continuous flow upgrading
of biobased glycerol toward high value-added oxirane building
blocks glycidol and epichlorohydrin under continuous flow
f
2
(10 mol%). Determined by HPLC-DAD.
best conditions for the aminolysis of 8a with isopropylamine conditions, and the subsequent utilization of epichlorohydrin
9a) thus involved 140 °C under 7 bar of counterpressure in the for the preparation of important active pharmaceutical
(
presence of Zn(ClO
time.
)
4 2
(10 mol%) within 10 min of residence ingredients. The upgrading of glycerol toward oxiranes 4,5
relies on a two-step continuous flow setup. The first step is
The optimized conditions for the preparation of dedicated to the hydrochlorination reaction, and allows the
propranolol 10a (Table 6, entry 12) were then applied for the rapid generation of chlorohydrins using economically and
continuous flow preparation of 10b-d (Scheme 4) starting from environmentally favorable conditions with concentrated
either glycidyl ether intermediates 8a,b in combination with aqueous hydrochloric acid. Pimelic acid is highlighted for its
amines 9a,b. Upon reacting naphthyl glycidyl ether (8a) with excellent catalytic efficiency with high selectivity toward
neat (2-methoxyphenyl)piperazine (9b), a 95% yield was dichlorohydrins. Hydrochlorination of an unrefined glycerin
obtained toward naftopidil (10b). The higher yield emphasizes sample was also performed, leading to an overall good
that higher selectivity is obtained with secondary amines such selectivity toward chlorohydrins despite the presence of
as 9b, since secondary amines cannot undergo a bis-alkylation, various contaminants. The second step aims at the
6
8
by contrast to primary amines. Combining 2-allylphenyl dechlorination of the chlorohydrins and relies on
a
glycidyl ether (8b) with neat isopropylamine (9a) gave a high concentrated aqueous solution of sodium hydroxide. Both
yield of 74% for alprenolol (10c). The setup was also amenable steps are directly concatenated and are further improved with
to the preparation of a novel analog of 10b and 10c starting the addition of an extraction unit, hence providing usable
from glycidyl ether 8b and amine 9b, which was obtained in flows of glycidol (in water) and epichlorohydrin (in MTBE, with
quantitative yield. Compounds 10a-d were isolated (63-85% an optional concentrator) through in-line membrane
yields) by liquid/liquid extraction and either recrystallization or separation. The scalability of the dechlorination step was also
column chromatography on silica gel (see Supporting investigated starting from chlorohydrins in concentrated
Information, section 2.2.12)
aqueous hydrochloric acid and led to the production of
Next, the concatenation of the synthesis of glycidyl ether oxiranes 4,5 in good yields while conveniently handling the
a and of its aminolysis with 9a,b was envisioned for the highly exothermic neutralization and the subsequent
8
preparation of target APIs 10a,b (Scheme 5). Decision was epoxidation reaction at pilot-scale. Biobased epichlorohydrin is
made to opt for a catalyst-free process avoiding the presence next further used for the preparation of valuable β-amino
of a metal catalyst while keeping a reasonable total residence alcohol APIs according to two additional reaction steps,
time of 40 min for two chemical steps. Moderate to good including a glycidyl ether synthesis and a final aminolysis. The
1
0 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 11 of 13
Please Gd roe en no tC ha ed mj u iss tt r my argins
Journal Name
ARTICLE
concatenable procedure is amenable to the preparation of a
library of three known APIs and one analog, including
propranolol (hypertension, WHO essential), naftopidil
2017, 19, 3186–3213.
N. Sotto, C. Cazorla, C. Villette, M. BDilOlaIm: 1b0o.1z03a9n/dC9CG. CLe0n18, 19K
ACS Sustain. Chem. Eng., 2016, 4, 6996–7003.
View Article Online
1
1
1
2
2
7
8
9
0
1
G. M. Lari, C. Mondelli and J. Pérez-Ramírez, ACS Catal.,
(prostatic hyperplasia) and alprenolol (angina pectoris).
2
015, 5, 1453–1461.
M. Velasquez, A. Santamaria and C. Batiot-Dupeyrat, Appl.
Catal. B Environ., 2014, 160–161, 606–613.
Conflicts of interest
There are no conflicts to declare.
G. M. Lari, R. García-Muelas, C. Mondelli, N. López and J.
Pérez-Ramírez, Green Chem., 2016, 18, 4682–4692.
M. O. Guerrero-Pérez and M. A. Bañares, ChemSusChem,
2
008, 1, 511–513.
Acknowledgments
22
C. J. A. Mota, V. L. C. Gonçalves, J. E. Mellizo, A. M. Rocco,
J. C. Fadigas and R. Gambetta, J. Mol. Catal. A Chem., 2016,
This work was supported by the University of Liège (Welcome
Grant WG-13/03, JCMM) and by the “Fonds de la Recherche
Scientifique de Belgique” (F.R.S.-FNRS) under Grant No.
J011719F (Fluidic upgrading of biobased platforms, JCMM). Dr.
4
22, 158–164.
2
3
M. H. Haider, N. F. Dummer, D. W. Knight, R. L. Jenkins, M.
Howard, J. Moulijn, S. H. Taylor and G. J. Hutchings, Nat.
Chem., 2015, 7, 1028–1032.
Stéphane Biltresse (Cargill) is gratefully acknowledged for the 24
samples of unrefined and refined biobased glycerol.
F. Wang, W. Xiao, L. Gao and G. Xiao, RSC Adv., 2016, 6,
42984–42993.
M. O. Sonnati, S. Amigoni, E. P. Taffin de Givenchy, T.
Darmanin, O. Choulet and F. Guittard, Green Chem., 2013,
15, 283–306.
P. S. Kong, M. K. Aroua, W. M. A. W. Daud, H. V. Lee, P.
Cognet and Y. Pérès, RSC Adv., 2016, 6, 68885–68905.
M. Sutter, E. Da Silva, N. Duguet, Y. Raoul, E. Métay and M.
Lemaire, Chem. Rev., 2015, 115, 8609–8651.
R. Tesser, E. Santacesaria, M. Di Serios, G. Di Nuzzi and V.
Fiandra, Ind. Eng. Chem. Res., 2007, 46, 6456–6465.
S. H. Lee, D. R. Park, H. Kim, J. Lee, J. C. Jung, S. Y. Woo, W.
S. Song, M. S. Kwon and I. K. Song, Catal. Commun., 2008,
2
5
Notes and references
2
2
2
2
6
7
8
9
1
2
3
4
A. T. Tullo, Chem. Eng. News, 2016, 94, 13.
M. McCoy, Chem. Eng. News, 2018, 96, 14–15.
M. Bomgardner, Chem. Eng. News, 2019, 97, 12.
B. M. Bell, J. R. Briggs, R. M. Campbell, S. M. Chambers, P.
D. Gaarenstroom, J. G. Hippler, B. D. Hook, K. Kearns, J. M.
Kenney, W. J. Kruper, D. James Schreck, C. N. Theriault and
C. P. Wolfe, Clean, 2008, 36, 657–661.
5
US Department of Energy, Top Value Added Chemicals
from Biomass Volume I - Results of Screening for Potential
Candidates from Sugars and Synthesis Gas, 2004.
J. J. Bozell and G. R. Petersen, Green Chem., 2010, 12, 539–
9
, 1920–1923.
P. Carré and P. Mauclère, Comptes Rendus Acad. Sci., 1931,
92, 1567–1569.
3
3
3
3
0
1
2
3
1
6
7
8
9
D. Giomi, M. Malavolti, O. Piccolo, A. Salvini and A. Brandi,
RSC Adv., 2014, 4, 46319–46326.
E. Santacesaria, R. Vitiello, R. Tesser, V. Russo, R. Turco and
M. Di Serio, Ind. Eng. Chem. Res., 2014, 53, 8939–8962.
C. A. de Araujo Filho, S. Heredia, K. Eränen and T. Salmi,
Chem. Eng. Sci., 2016, 149, 35–41.
G. Braun, Org. Synth., 1936, 16, 30.
G. M. Lari, G. Pastore, C. Mondelli and J. Pérez-Ramírez,
Green Chem., 2018, 20, 148–159.
5
54.
S. Bagheri, N. M. Julkapli and W. A. Yehye, Renew. Sustain.
Energy Rev., 2015, 41, 113–127.
M. Anitha, S. K. Kamarudin and N. T. Kofli, Chem. Eng. J.,
2
016, 295, 119–130.
R. Gérardy, N. Emmanuel, T. Toupy, V.-E. Kassin, N. N.
Tshibalonza, M. Schmitz and J.-C. M. Monbaliu, Eur. J. Org.
Chem., 2018, 2301–2351.
3
3
4
5
1
1
0
1
R. Gérardy, R. Morodo, J. Estager, P. Luis, D. P. Debecker
and J.-C. M. Monbaliu, Top. Curr. Chem., 2019, 377, 1.
K. Ahuja and S. Singh, Glycerol Market Size, Global Market
Insights Inc., 2018, https://www.gminsights.com/industry-
analysis/glycerol-market-size (Accessed May 4, 2019).
A. Mortensen, F. Aguilar, R. Crebelli, A. Di Domenico, B.
Dusemund, M. J. Frutos, P. Galtier, D. Gott, U.
Gundert-Remy, J. Leblanc, O. Lindtner, P. Moldeus, P.
Mosesso, D. Parent-Massin, A. Oskarsson, I. Stankovic, I.
Waalkens-Berendsen, R. A. Woutersen, M. Wright, M.
Younes, P. Boon, D. Chrysafidis, R. Gürtler, P. Tobback, A.
M. Rincon, A. Tard and C. Lambré, EFSA J., 2017, 15,
e04720.
3
3
3
3
4
4
4
6
7
8
9
0
1
2
G. S. Singh, K. Mollet, M. D’Hooghe and N. De Kimpe,
Chem. Rev., 2013, 113, 1441–1498.
M. W. Bedore, N. Zaborenko, K. F. Jensen and T. F.
Jamison, Org. Process Res. Dev., 2010, 14, 432–440.
T. Nobuta, G. Xiao, D. Ghislieri, K. Gilmore and P. H.
Seeberger, Chem. Commun., 2015, 51, 15133–15136.
P. J. Cossar, J. R. Baker, N. Cain and A. McCluskey, R. Soc.
Open Sci., 2018, 5, 171190.
1
2
M. G. Russell and T. F. Jamison, Angew. Chem. Int. Ed.,
2
019, 58, 7678–7681.
L. Rogers and K. F. Jensen, Green Chem., 2019, 21, 3481–
498.
3
R. Gérardy and J.-C. M. Monbaliu, Multistep Continuous-
Flow Processes for the Preparation of Heterocyclic Active
Pharmaceuticals Ingredients in Topics in Heterocyclic
Chemistry: Flow Chemistry for the Synthesis of
Heterocycles, eds. U. K. Sharma and E. V. Van der Eycken,
Springer International Publishing, Cham, 2018, pp. 1–102.
R. Gérardy, M. Winter, A. Vizza and J.-C. M. Monbaliu,
React. Chem. Eng., 2017, 2, 149–158.
1
1
1
1
3
4
5
6
M. Pagliaro and M. Rossi, The Future of Glycerol, The Royal
Society of Chemistry, 2010.
N. Ntumba Tshibalonza and J.-C. M. Monbaliu, Green
Chem., 2017, 19, 3006–3013.
B. Katryniok, S. Paul and F. Dumeignil, ACS Catal., 2013, 3,
1
819–1834.
4
3
D. Sun, Y. Yamada, S. Sato and W. Ueda, Green Chem.,
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 11
Please do not adjust margins

Please Gd roe en no tC ha ed mj u iss tt r my argins
Page 12 of 13
ARTICLE
Journal Name
4
4
V.-E. H. Kassin, R. Gérardy, T. Toupy, D. Collin, E. Salvadeo,
F. Toussaint, K. Van Hecke and J.-C. M. Monbaliu, Green
Chem., 2019, 21, 2952–2966.
View Article Online
DOI: 10.1039/C9GC01819K
4
5
R. Gérardy, M. Winter, C. R. Horn, A. Vizza, K. Van Hecke
and J.-C. M. Monbaliu, Org. Process Res. Dev., 2017, 21,
2
012-2017.
4
4
4
4
5
5
5
5
6
7
8
9
0
1
2
3
N. N. Tshibalonza, R. Gérardy, Z. Alsafra, G. Eppe and J.-C.
M. Monbaliu, Green Chem., 2018, 20, 5147–5157.
Z. Wang, R. Gérardy, G. Gauron, C. Damblon and J.-C. M.
Monbaliu, React. Chem. Eng., 2019, 4, 17–26.
P. Tundo, P. Venturello and E. Angeletti, J. Chem. Soc.
Perkin Trans. I, 1987, 2157–2158.
P. Tundo, G. Moraglio and F. Trotta, Ind. Eng. Chem. Res.,
1
989, 28, 881–890.
B. Reichart, G. Tekautz and C. O. Kappe, Org. Process Res.
Dev., 2013, 17, 152–157.
S. Borukhova, T. Noël and V. Hessel, ChemSusChem, 2016,
9
, 67–74.
S. Borukhova, T. Noël, B. Metten, E. de Vos and V. Hessel,
Green Chem., 2016, 18, 4947–4953.
C. A. de Araujo Filho, D. Mondal, S. Haase, J. Wärnå, K.
Eränen, J.-P. Mikkola and T. Salmi, Chem. Eng. Sci., 2016,
1
49, 277–295.
P. Kraft, P. Gilbeau, B. Gosselin and S. Claessens, World Pat.
005054167, 2005.
5
5
4
5
2
D. J. Schreck, W. J. J. Kruper, R. D. Varjian, E. Jones, Mark,
R. M. Campbell, K. Kearns, B. D. Hook, J. R. Briggs and J. G.
Hippler, World Pat. 2006020234, 2006.
5
6
X. Hou, Y. Fu, X. Zhu, H. Yin and A. Wang, Asia-Pacific J.
Chem. Eng., 2015, 10, 626–632.
5
5
7
8
G. H. Coleman and D. Craig, Org. Synth., 1932, 12, 80.
R. J. Lewis, Hawley’s Condensed Chemical Dictionnary, John
Wiley & Sons, Inc., New York, 15th Ed., 2007.
5
9
R. K. Henderson, C. Jiménez-González, D. J. C. Constable, S.
R. Alston, G. G. A. Inglis, G. Fisher, J. Sherwood, S. P. Binks
and A. D. Curzons, Green Chem., 2011, 13, 854–862.
D. Prat, O. Pardigon, H.-W. Flemming, S. Letestu, V.
Ducandas, P. Isnard, E. Guntrum, T. Senac, S. Ruisseau, P.
Cruciani and P. Hosek, Org. Process Res. Dev., 2013, 17,
6
0
1
517–1525.
6
6
6
1
2
3
F. P. Byrne, S. Jin, G. Paggiola, T. H. M. Petchey, J. H. Clark,
T. J. Farmer, A. J. Hunt, C. R. McElroy and J. Sherwood,
Sustain. Chem. Process., 2016, 4, 7.
D. Prat, A. Wells, J. Hayler, H. Sneddon, C. R. McElroy, S.
Abou-Shehada and P. J. Dunn, Green Chem., 2016, 18, 288–
2
96.
U.S. Food and Drug Administration, Q3C — Tables and List
Guidance for Industry (Revision 3), 2017,
https://www.fda.gov/regulatory-information/search-fda-
guidance-documents/q3c-tables-and-list-rev-3 (Accessed
May 4, 2019).
6
6
6
6
6
4
5
6
7
8
D. E. Pearson and C. A. Buehler, Chem. Rev., 1974, 74, 45–
8
6.
C. Audubert, A. Bouchard, G. Mathieu and H. Lebel, J. Org.
Chem., 2018, 83, 14203–14209.
J. R. Lizza and G. Moura-Letts, Synthesis, 2017, 49, 1231–
1
242.
Shivani, B. Pujala and A. K. Chakraborti, J. Org. Chem.,
007, 72, 3713–3722.
2
H. Desai, B. D’Souza, D. Foether, B. Johnson and H. Lindsay,
Synthesis, 2007, 2007, 902–910.
1
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 13 of 13
Please Gd roe en no tC ha ed mj u iss tt r my argins
Journal Name
ARTICLE
View Article Online
DOI: 10.1039/C9GC01819K
Table of Content
A robust continuous flow procedure for the transformation of
biobased glycerol into high value-added β-amino alcohol
pharmaceutical active ingredients.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 13
Please do not adjust margins