Microwaves under pressure for the continuous production of quinoline from glycerol

Article abstract of DOI:10.1016/j.cattod.2014.10.050

Abstract Microwave heating is an interesting technology for chemical engineering, since it can provide effective volumetric heating of the reaction medium and reduce energy costs. Many commercially available laboratory-scale microwave reactors have already been used to carry out chemical reactions on a small scale (a few milliliters), and at high temperatures and pressures. Some research has been undertaken to scale-up microwave processes and make them suitable for a larger scale production. Indeed, combining wave propagation through the walls of a reactor with resistance toward high pressure and temperature as well, is not an easy task. For these reasons, this work focuses on the development of a pilot scale microwave apparatus used for the heating of larger reaction volumes under pressure, and under controlled conditions. The specially designed microwave apparatus allows chemical reactions in batch or continuous mode. The applicator operates in single mode enabling a uniform electromagnetic field, and well controlled operating conditions. The main advantage of the setup is the quite large reactor volume that permits either relatively long residence times or relatively high mass flowrates (up to 1 kg/h). The developed microwave apparatus was then used for quinoline synthesis from glycerol via a modified Skraup reaction. The major advantage of our system is the ability to carry out continuous chemical synthesis, at a large pilot scale, and high temperatures (200-220 °C), while ensuring a better control of the pressure (max. 19 bar) through the control of the power absorbed by the reaction medium.

Full text of DOI:10.1016/j.cattod.2014.10.050

G Model
CATTOD-9351; No. of Pages9
ARTICLE IN PRESS
Catalysis Today xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Microwaves under pressure for the continuous production of
quinoline from glycerol
H. Saggadia,b, I. Polaerta,∗, D. Luartb, C. Lenb,c, L. Estela
a LSPC – Laboratoire de Sécurité des Procédés Chimiques, EA 4704 – INSA Rouen, Avenue de l‘Université, Saint-Étienne-du-Rouvray Cedex 76801, France
b TIMR – Transformations Intégrées de la Matière Renouvelable, EA 4297 UTC – ESCOM, Compiègne 60200, France
c Department of Chemistry, University of Hull, Hull, HU6 7RX, England, United Kingdom
a r t i c l e i n f o
a b s t r a c t
Article history:
Microwave heating is an interesting technology for chemical engineering, since it can provide effec-
tive volumetric heating of the reaction medium and reduce energy costs. Many commercially available
laboratory-scale microwave reactors have already been used to carry out chemical reactions on a small
scale (a few milliliters), and at high temperatures and pressures. Some research has been undertaken to
scale-up microwave processes and make them suitable for a larger scale production.
Received 28 August 2014
Received in revised form 6 October 2014
Accepted 18 October 2014
Available online xxx
Indeed, combining wave propagation through the walls of a reactor with resistance toward high pres-
sure and temperature as well, is not an easy task. For these reasons, this work focuses on the development
of a pilot scale microwave apparatus used for the heating of larger reaction volumes under pressure, and
under controlled conditions. The specially designed microwave apparatus allows chemical reactions in
batch or continuous mode. The applicator operates in single mode enabling a uniform electromagnetic
field, and well controlled operating conditions. The main advantage of the setup is the quite large reactor
volume that permits either relatively long residence times or relatively high mass flowrates (up to 1 kg/h).
The developed microwave apparatus was then used for quinoline synthesis from glycerol via a modified
Skraup reaction. The major advantage of our system is the ability to carry out continuous chemical syn-
thesis, at a large pilot scale, and high temperatures (200–220 ◦C), while ensuring a better control of the
pressure (max. 19 bar) through the control of the power absorbed by the reaction medium.
© 2014 Elsevier B.V. All rights reserved.
Keywords:
Microwaves
Pressurized reactor
Continuous flow
Green process
Intensification
Glycerol
1. Introduction
such as in the drying process of fruits, food, and pharmaceutical
powders as well as for polymer chemistry. In medicine, microwaves
The application of alternative forms of energy sources for con-
ducting chemical reactions is at the heart of process intensification
approach. Among these energy sources, microwave heating is
recognized as a very promising technique to intensify chemical pro-
cesses [1,2]. One can notice the rapid growth of microwave assisted
procedures in organic synthesis since the publication of the pio-
neering papers by Gedye et al. [3] and Giguere et al. [4] in 1986.
Rapid microwave protocols can be developed for most chemical
transformations requiring heat. The activity in this new technique
has experienced exponential growth and has been reviewed in the
areas of material science, food technology, environmental tech-
nology, organic synthesis, polymer synthesis, etc. [5,6]. Indeed,
microwave technology is already used in some industrial sectors
are used in resonance therapy and thermography [7,8]. Microwave
irradiation offers many advantages compared to conventional heat-
ing. Electromagnetic energy is actually transferred directly into
the material via radiation without loss by means of heat conduc-
tion and convection. Consequently, an instantaneous and uniform
volumetric heating is obtained for microwave-absorbing material,
and heating occurs therefore from the inside to the outside of
the medium. Easy automation and incident power control are also
achieved. In numerous chemical syntheses, rapid heating, selective
heating and enhancements of yield and purity are obtained [5,8,9].
Many laboratory-scale microwave reactors have been widely
used for conducting different chemical reactions. The commercially
available microwave devices are divided into two groups. The first
category consists in multimode ovens. They are often equipped
with a small tubular reactor inside (Teflon coil, or glass or quartz
tube) [10], of only a few milliliters, in which different organic syn-
theses have been carried out. Some of them have been performed
under high pressure (up to 120 bar) and at temperatures of up to
∗
Corresponding author. Tel.: +33 232956668.
E-mail addresses: hanen.saggadi@insa-rouen.fr (H. Saggadi),
isabelle.polaert@insa-rouen.fr (I. Polaert).
http://dx.doi.org/10.1016/j.cattod.2014.10.050
0920-5861/© 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: H. Saggadi, et al., Microwaves under pressure for the continuous production of quinoline from glycerol,
Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.10.050

G Model
CATTOD-9351; No. of Pages9
ARTICLE IN PRESS
2
H. Saggadi et al. / Catalysis Today xxx (2014) xxx–xxx
250 ◦C [11,12]. Most of the devices have been optimized in order to
achieve better reaction control, namely temperature and pressure
control. These developments have obviously led to investigating
the scalability of microwave heating, either by using larger reaction
volumes in batch mode, or by multiplying the number of small vials
in the ovens [12–15], or by using the continuous mode [16–18].
Despite being able to control the main parameters (temperature,
pressure, and reaction medium stirring), as well as the automation
of the device, an important parameter remains uncontrolled: the
electromagnetic field distribution. While most of the commercially
available multimode devices involve a static microwave diffuser
to homogenize the wave distribution, the electromagnetic field
distribution can never be controlled and its homogeneity cannot
be ensured. This leads to a lack of temperature uniformity in the
reaction medium and hot spots can occur, especially if the cavity
possesses multi-reaction vials.
The second category consists in more advanced microwave reac-
tors, involving a waveguide to create a standing wave. Single-mode
applicators are used, allowing the propagation of a quasi-uniform
field on the reactor zone. The reaction can then be conducted under
well controlled operating conditions. The electric field densities in
single-mode cavities are higher, allowing absorbing mixtures to be
rapidly heated. Microwave input and reflected powers can be mea-
sured, the absorbed energy estimated, and the interactions of the
wave with the medium can be identified. The most commonly used
batch devices (e.g., Biotage, CEM, Anton Paar) permit, in some cases,
interesting conditions for chemical reactions (temperatures up to
300 ◦C and pressures up to 30 bar). However, in these commercially
available single-mode batch reactors, the reaction volume is very
small (∼50 mL) and not adapted to production [19,20].
with resistance toward pressure is a hard task. Indeed, con-
trolling experimental conditions has to be managed, including
pressure, temperature, time, homogenization of the medium,
reagents/products flow (in the case of continuous mode reaction),
and post-reaction cooling. Our approach requires scale-up, solv-
ing the technical issues, in terms of technical tools for temperature
and pressure measurement and control, as well as reactor design
and implementation of elements allowing batch and continuous
mode operations for larger quantities. It is also imperative to ensure
that the apparatus functions safely, and to plan power control and
anticipate intervention procedures in the case of any emergency.
The developed microwave pilot has been tested on quinoline
production from glycerol using a modified Skraup reaction. This
reaction is of great interest for the production of a pharmaceuti-
cal intermediate (quinoline) from a biosourced material, glycerol.
In a previous study [26], we demonstrated the feasibility of the
reaction under microwave irradiation in green conditions, using
water as a solvent and sulphuric acid as a catalyst, without any
additional oxidant or toxic reactant. High temperature (220 ◦C) and
pressure (15–25 bar) are required for the reaction. Acrolein is pro-
duced as an intermediate, and reacts in situ with aniline to produce
quinoline, therefore avoiding handling with its toxicity and insta-
bility. The developed microwave pilot was operated for quinoline
production in both batch and continuous flow modes and temper-
ature/pressure/power control were ensured.
2. Batch mode
Our team has been investigating microwave synthesis of
quinolines for several years, using a laboratory scale apparatus
(monowave 300 from Anton-Paar), where a few milliliters (30 mL)
of reagents were used [25,26]. In this small apparatus, the Skraup
reaction was performed starting from glycerol (30 mmol), aniline
(10 mmol) in the presence of sulphuric acid (30 mmol) in 10 mL
of water, under microwave irradiation, at 220 ◦C (25 bar pressure)
for 10 min, giving quinoline with a 44% yield. We decided first, to
increase fivefold the reactor volume in order to check the feasibility
of the reaction at a larger scale, focusing on the good propagation
and absorption of the electromagnetic energy in the larger volume.
We also wanted to develop a reactor volume more compatible with
a 1 kg/h continuous production of quinoline.
As for multimode systems, some studies have focused on the
scalability of single-mode pressurized devices, either by allow-
ing larger volume reactions (12 bar, 500 mL)[21], or by developing
continuous mode tubular reactors [19,22,23]. Although these
developments have led to conducting chemical reactions at high
temperatures (up to 300 ◦C) and under pressure (up to 50 bar) [19],
once again, the volumes do not exceed 50 mL. This lays down typ-
ically short reaction times (in the order of a few minutes), and
small flow rates. Despite these improvements, there are many
remaining difficulties to scale up microwave pressurized reactors.
One of the main limitations is the restricted penetration depth of
microwave irradiation into absorbing materials, which is generally
in the order of just a few centimeters. When using larger batch
reactors, reagents in the center of the reaction vessel are mostly
heated by convection and not by direct microwave dielectric heat-
ing. The heating rate of the mixture is therefore affected, leading to
extended processing times [20,24]. This physical limitation is one of
the main reasons for the development of continuous flow reactors.
Another important issue for the scalability of the continuous
pressurized microwave reactions is the necessity to use reactors
that are transparent to microwave. Only materials like quartz glass,
some polymers and some ceramics can be used [19]. However,
combining these materials with resistance toward chemicals, high
pressure and temperature is not easy to carry out. Thus, scaling-
up microwaves must mainly fulfill the following criteria: (1) safe
and reliable equipment, (2) high reproducibility, (3) good reaction
control, (4) fast heating time, (5) reasonable volume, (6) accept-
able energy balance, (7) comparable or higher yield to conventional
systems, and (8) acceptable cost [19].
2.1. Microwave apparatus
The cylindrical PTFE reactor was designed and customized,
allowing reaction volumes up to about 135 mL, that is to say, five
time more than the reacting volume used for our first investigations
[25,26]. The temperature inside the reactor was measured using
an optical fiber immersed in the reaction medium through a glass
tube. The complete batch experimental diagram is shown in Fig. 1.
Our system consisted of a Sairem (GMP 20K/SM) microwave high-
voltage generator providing power up to 1900 W, coupled with a
magnetron delivering an electromagnetic wave at 2.45 GHz. Only
incident and reflected powers were measured with power sensors
and the absorbed power was deduced from the difference. The
magnetron was connected to a WR340 waveguide allowing sta-
tionary waves propagation in a TE10 mode. A tuning piston at the
end of the waveguide was used to optimize the electromagnetic
energy absorption by varying the guide length. The waveguide was
equipped with an applicator and a chimney in which the reactor
was inserted. Pressure control was assured by elements installed
downstream to the reactor: a pressure sensor, a safety valve (cali-
brated at 25 bar), a pressure regulator equipped with a gauge and a
valve. Computer control and supervision of the set-up were carried
out using Labview software.
Keeping these issues in mind, the challenge of this work is to
design a continuous microwave reactor dedicated to organic syn-
thesis at high temperatures and pressures. A chemical production
of about 1 kg/h is targeted, for relatively slow reactions requiring
quite long residence times (10–20 min). This device has to operate
reliably and safely with volatile compounds including organic sol-
vents. Nevertheless, developing a microwave-transparent reactor
Please cite this article in press as: H. Saggadi, et al., Microwaves under pressure for the continuous production of quinoline from glycerol,
Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.10.050

G Model
CATTOD-9351; No. of Pages9
ARTICLE IN PRESS
H. Saggadi et al. / Catalysis Today xxx (2014) xxx–xxx
3
Fig. 1. Batch mode microwave system.
In order to check its resistance toward pressure and tempera-
ture, the apparatus was subjected to several safety tests before use.
The first test was performed at ambient temperature under 40 bar
of nitrogen pressure. The second test was performed at 190 ◦C in
an oil bath under 10 bar, and the last one, under continuous con-
ditions at room temperature, connected to a pump and reaching a
pressure up to 25 bar. The reactor was also tested with water under
microwave irradiation which allowed us not only to check resis-
tance toward pressure, temperature and microwaves, but also to set
a microwave protocol for better power and pressure/temperature
control.
experiment, the irradiation was stopped and the system was cooled
to ambient conditions (Zone C).
Quantification was carried out by high performance liquid
chromatography (HPLC) coupled to
a UV detector (wave-
length = 220 nm). Analysis was made with an isocratic elution of
acetonitrile (A) and water (B) with (A/B): 40:60. Acquisition and
data processing were done with Azur software.
2.3. Results and discussion
Referring to experiments performed in the small scale reactor
[25,26], as well as the saturation data of water and water–glycerol
mixtures, we decided to conduct new experiments, in the new
reactor of a larger volume (135 mL) at 14 bar (pressure correspond-
ing to about 210 ◦C). The opening value of the pressure regulator
was pre-set at 15 bar to avoid any pressure increase in the sys-
tem. A constant value of the absorbed power was applied until the
approximately desired pressure (≈13.8–14 bar) was reached. After
2–3 min, the liquid started boiling and the pressure started increas-
ing. By conducting this way the experiment, the mixture remained
at the liquid–vapor equilibrium, meaning that all the time, the liq-
uid phase was at equilibrium with the gas phase. Pressure increased
continuously due to the heating of the medium and its partial
vaporization. When the targeted value of pressure was reached
(after 40 min), automatic control was then immediately switched-
on, in such a way that the magnetron delivered only the required
power to maintain this pressure value during the stabilization time.
For the conducted experiments, an increase of the reflected
power (and therefore of the incident power, trying to maintain the
same value of the absorbed power) was systematically noticed as
soon as the pressure started to rise after about 2–3 min. This indi-
cated a change in the dielectric properties of the medium, due to
the reaction mixture boiling. In the absence of magnetic stirring,
we supposed that the reactive mixture was homogenized by bulk
motions and convective heat transfer. At the end of the experiment,
a homogeneous dark brown liquid was recovered.
2.2. Procedure
In our pilot scale microwave apparatus, the Skraup reaction
was conducted starting with 50 mmol of aniline, 150 mmol of
glycerol and 150 mmol of sulphuric acid in 50 mL of water as a
solvent (corresponding to a 75 mL reacting volume). The mixture
was stirred with a spatula in order to homogenize the medium
(the mixture is perfectly homogeneous at ambient conditions). The
absence of leaks was checked at ambient temperature under 30 bar
of nitrogen. The reactor was then depressurized, placed in the
chimneys and connected to the rest of the installation. The appa-
ratus control was carried out by an automatic PID regulation of
the absorbed power with respect to the pressure. Pressure regu-
lation was chosen rather than temperature since the medium was
not stirred, which could leads to a heterogeneous temperature dis-
tribution within the reactor. Therefore, the measured temperature
was less reliable than the pressure. This allowed us to prevent any
sudden increase in pressure and, especially, to ensure safer and
repeatable experimental conditions. The same microwave protocol
was applied for all experiments, and thus had better repeatabil-
ity. The adopted method is illustrated in Fig. 2, and consisted
of three main steps: allowing pressure to rise up to the desired
value using a fixed absorbed power (Zone A); once the desired
pressure was reached, this value was kept constant during the
Thus, our system allowed carrying out the Skraup reaction at
adequate temperature and pressure conditions, giving quinoline
with a 41% yield, similar to the result obtained in the small scale
experiment by an automatic regulation of the absorbed power with
regard to the pressure set-up value (Zone B); at the end of the
Please cite this article in press as: H. Saggadi, et al., Microwaves under pressure for the continuous production of quinoline from glycerol,
Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.10.050

G Model
CATTOD-9351; No. of Pages9
ARTICLE IN PRESS
4
H. Saggadi et al. / Catalysis Today xxx (2014) xxx–xxx
Zone A
Zone C
Zone B
500
450
400
350
300
250
200
150
100
50
Incident power
Reflected power
absorbed power
0
0
10
20
30
40
50
60
Time (min)
(a)
250
200
150
100
50
10
9
8
7
6
5
4
3
2
1
0
Temperature
Pressure
Zone B
Zone A
Zone C
0
0
10
20
30
40
50
60
Time (min)
(b)
Fig. 2. Evolution of the experimental parameters during an experiment: (a) microwave incident power, reflected power and absorbed power evolution; (b) temperature and
pressure evolution.
reactor. In addition, reaction duration was studied in these same
conditions of temperature and pressure (P = 14 bar and T = 210 ◦C):
three independent experiments were carried out maintaining a
pressure/temperature rising time of 20 min, and a reacting time at
constant pressure of 5 min, 10 min or 20 min. In these conditions,
the reaction time did not affect the quinoline yield. Full aniline con-
version and maximum yield were obtained after 5 min of reaction
time indicating that the reactions were relatively fast as soon as the
adequate pressure and temperature conditions were reached.
For the experiments that were carried out, a drop in the reflected
power at a pressure of about 8 bar was always noticed, which seems
particularly interesting (Fig. 2a). This means, that at this time, the
system is absorbing the major part of the supplied power, probably
due to the medium dielectric property changes. If the experiment
control was carried out with respect to the temperature, this could
cause thermal runaway and possibly, overpressure in the reac-
tor, since the response time of the measured temperature is much
higher than the one of the pressure. The superiority of system con-
trol by pressure regulation is now demonstrated.
pressure influence on the quinoline yield more closely. Further-
more, it is always preferable to reduce operating pressure for safety
reasons.
Results reported in Table 1 showed that Skraup quinoline syn-
thesis started at about 7 bar and kept on going to 9 bar (Table 1,
entries 1 and 2). At 9 bar, a stable reaction time of 40 min was
required to consume almost all of the aniline (Table 1, entry 4
vs. entry 3). Consequently, these conditions were not suitable for
continuous mode experiments, since they would have led to too
long residence time. Experiments led at 12 bar and 14 bar (Table 1,
Table 1
Pressure influence on the quinoline synthesis.
Entry Microwave protocol
T (◦C) Aniline
conversion (%)
Quinoline
yield (%)
1
2
3
4
5
6
Pressure rising: in 14.7 min
182
193
190
190
204
210
17
2
5
Irradiation stopped at 7.5 bar
Pressure rising: in 12 min
Irradiation stopped at 9.5 bar
Pressure rising: in 19.7 min
10 min at 9 bar
Pressure rising: in 19.6 min
40 min at 9 bar
Pressure rising: in 19.5 min
10 min at 12 bar
26
61
21
42
42
41
The change in dielectric properties, at this stage of the experi-
ment, may be due to the evolution of the temperature and pressure
conditions in the reactor, as well as a change in the composition
of the medium as a consequence of the beginning of the reaction.
The second possibility seems to be more plausible, considering that
this pressure (8 bar) corresponds to a temperature of about 185 ◦C
(Fig. 2). To confirm this hypothesis, we decided to examine the
95
98
Pressure rising: in 22 min
10 min at 14 bar
100
Please cite this article in press as: H. Saggadi, et al., Microwaves under pressure for the continuous production of quinoline from glycerol,
Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.10.050

G Model
CATTOD-9351; No. of Pages9
ARTICLE IN PRESS
H. Saggadi et al. / Catalysis Today xxx (2014) xxx–xxx
5
250
to allow inlet/outlet flow, fluid cooling at the reactor outlet, as
well as various security equipment required for safe process run-
ning.
Conventional heating
Microwave heating
200
150
100
50
Nevertheless, the strong acidic conditions of the Skraup reaction
were not compatible with the 316L stainless steel pipes and tubing
of our pilot, which were not resistant to corrosion. To overcome
this problem, we decided to change the mixture concentration. A
reaction with a diluted mixture (0.1 mol/L of aniline; 3 eq of glyc-
erol and 3 eq of sulphuric acid) was carried out. Despite the fact
that this choice was not suitable for obtaining good aniline conver-
sion and quinoline yields, we decided to go further in the design
and development of the continuous mode pressurized microwave
apparatus, which will be suitable for other organic synthesis.
0
0
50
100
Time (min)
150
200
250
Fig. 3. Temperature profiles using two heating modes: microwaves vs. conventional
heating.
3.1. Apparatus
To operate the microwave pressurized reactor in continuous
mode, the following elements were added to the batch apparatus
(Fig. 4):
entries 5 and 6) gave similar results. Finally, experiments in the
newly designed and five times larger apparatus gave the same
quinoline yields as in the small scale experiments (performed in the
monowave 300), but at a reduced reaction temperature (204 ◦C vs.
220 ◦C), and at a significantly reduced operating pressure (12 bar
vs. 25 bar). Furthermore, better experimental parameters control
was obtained.
•
a feed tank, wherein the starting mixture was stored and mechan-
ically stirred
•
a Dosapro Milton Roy metering pump (Pmax = 450 bar). The pump
was connected upstream to a pressure regulator equipped with
a pressure gauge, working as a safety element to face any sudden
pressure increase
In order to compare the energy efficiency of microwave heat-
ing to thermal heating, a Skraup quinoline synthesis was done in
a conventional pressurized reactor (4560 Mini Bench from Parr),
allowing conditions up to 350 ◦C and 200 bar. A 160 mL volume
Hastelloy reactor was used, equipped with a glass vessel of 105 mL.
It was heated by applying a PID control with respect to the temper-
ature measured at the outer walls of the reactor and the reaction
temperature was deduced from a preliminary calibration. The reac-
tion was carried out in the same conditions as under microwave
heating (0.1 mol/L of aniline; 3 eq of glycerol and 3 eq of sulphuric
acid), respecting the maximum volume allowed by the reactor. To
ensure a temperature of 200 ◦C, we applied a set point of 325 ◦C
to the oven, and achieved the target temperature within 112 min.
This temperature was then held for 10 min. Finally, the furnace was
removed and the reactor cooled down to ambient temperature. The
obtained quinoline yield was similar to the one obtained under
microwaves (37% vs. 42% in microwave, at 100% aniline conver-
sion).
•
a valve connected upstream to the reactor to purge air from the
tubing and to check the inlet liquid flow
•
a stream feed located at 1 cm from the top of the reactor, the
outlet was 1 cm from the bottom of the reactor
•
a temperature measurement provided by a PT100 sensor, con-
nected downstream to the reactor. Batch system pressure control
and safety items were kept (pressure sensor, safety valve, pres-
sure regulator equipped with a gauge, and the valve)
•
a condenser connected downstream to the pressure regulator
Since the pressure was now managed by the pump, the auto-
matic control of the absorbed power was carried out by PID action
with respect to the temperature.
3.2. Results and discussion
Considering the recorded temperature profiles (Fig. 3) and the
comparison summarized in Table 2, the use of our microwave
device considerably reduces energy consumption. The experiment
is carried out in only 32 min with microwave heating, while 122 min
are required using conventional heating, applying a furnace tem-
perature of 325 ◦C. The microwaves device handles about 40%
more for a similar quinoline yield and the energy consumption
(in kJ/mol of quinoline produced) is much higher when thermal
heating is used. With the conventional method, it is necessary to
heat the whole mass of the stainless steel reactor by conduction,
and deal with important thermal losses. The rapid heating fea-
ture of microwaves and the efficient use of the energy by direct
absorption are the main benefits of our pilot scale microwave pro-
cess.
An example of the experimental parameters evolution is shown
in Fig. 5. The experiment was carried out starting with a 0.1 mol/L
aniline solution in water, with 3 eq of glycerol and 3 eq of sulphuric
acid.
Zone A represents the transient regime. The required reaction
pressure was controlled by the pump and the pressure regulator
(14–15 bar). Microwaves were switched on at t = 0 allowing the
temperature to increase until measuring 140 ◦C at the outlet of
the reactor. At this stage, some pressure fluctuations as well as the
reflected power (and therefore the incident and absorbed power)
occurred, indicating that the system had reached its boiling point
(210 ◦C inside the reactor at this pressure). Stabilization of the sys-
tem was achieved, allowing some pressure oscillations of 0.5 bar.
The pressure was then slightly increased (15–16 bar) by acting on
the pressure regulator in order to maintain the medium in a liquid
state. PID automatic control of the absorbed power was switched
on to maintain a required temperature value of 133 ◦C at the outlet
of the reactor. The steady state (Zone B) was rapidly reached, and all
the operating parameters were stabilized. The experiment ended
by switching off the microwaves. The pump was stopped and the
reactor cooled to ambient conditions.
3. Continuous mode
With the purpose of producing about 1 kg/h of organic prod-
ucts under microwave and under high temperature and pressure
conditions, we studied the possibility of designing a continuous
pilot apparatus from the developed microwave batch apparatus.
The same PTFE reactor was used in order to achieve the temper-
ature/pressure conditions required for quinoline synthesis. Our
approach consisted in making changes to the batch apparatus
The continuous mode experiments allowed us to study
two parameters: the flowrate/residence time ꢀ, and pres-
sure/temperature (Toutlet). Results are summarized in Table 3. On
Please cite this article in press as: H. Saggadi, et al., Microwaves under pressure for the continuous production of quinoline from glycerol,
Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.10.050

G Model
CATTOD-9351; No. of Pages9
ARTICLE IN PRESS
6
H. Saggadi et al. / Catalysis Today xxx (2014) xxx–xxx
Table 2
Comparison of the two heating processes showing advantages of the microwave pressurized reactor.
Heating process
Reactants mass (g)
Temperature and pressure
Required time to attain
temperature (min)
Quinoline yield (%)
Consumed energy (kJ/mol
of quinoline produced)
Microwaves
Conventional
83
60
204 ◦C; 12 bar
22
112
41
37
11,014
452,200
200 ◦C (furnace at 325 ◦C); 16 bar
Fig. 4. Continuous mode microwave pilot reactor.
450
400
350
300
250
200
150
100
50
Incident power
Reflected power
Absorbed power
Zone C
Zone C
Zone A
Zone B
0
-50
0
50
100
150
200
250
300
Time (min)
(a)
160
140
120
100
80
20
15
10
5
Temperature
Relative pressure
60
40
20
Zone B
Zone A
0
0
-50
0
50
100
150
200
250
300
Time(min)
(b)
Fig. 5. Experimental parameters evolution during the continuous quinoline synthesis: (a) power evolution over time; (b) temperature and pressure evolution over reaction
time.
Please cite this article in press as: H. Saggadi, et al., Microwaves under pressure for the continuous production of quinoline from glycerol,
Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.10.050

G Model
CATTOD-9351; No. of Pages9
ARTICLE IN PRESS
H. Saggadi et al. / Catalysis Today xxx (2014) xxx–xxx
7
Table 3
Microwave continuous quinoline synthesis starting from a 0.1 mol/L aniline solution in water, with 3 eq of glycerol and 3 eq of sulphuric acid.
Entry
Mass flow rate (g min−1
)
ꢀ (min)
Toutlet of the reactor (◦C)
P (bar)
Conversion (%)
Yields (%)
1
2
3
6.9
16.6
16.6
14
6.13
6.13
131–133 (∼200 inside)
131–133 (∼200 inside)
151–152 (∼218 inside)
15–16
15–16
19–20
8
2.2
5.3
1
0.51
4
the one hand, the results obtained showed that increasing the res-
idence time (by reducing the flowrate, Table 3, entry 2 vs. entry 1),
promoted aniline reactivity and quinoline formation. On the other
hand, increasing temperature/pressure conditions (Table 3, entry 3
vs. entry 2) improved quinoline yield. Quinoline yields and aniline
conversion were very modest, because they were limited by the
low acidic concentration of the medium due to chemical resistance
constraints.
These experiments prove that the quinoline continuous pro-
duction under microwave irradiation and at high temperature
(200 ◦C) and 15 bar pressure is possible. The developed pilot
microwave reactor makes continuous production possible for about
1 kg/h of organic products at high temperature and pressure (up
to 19 bar and 220 ◦C). Although conversions and yields are low
and corrosion problems caused by concentrated reaction medium
have to be solved, a microwave pilot continuously operating
under pressurized conditions is now available for green chem-
istry.
3.3. Energetic balance for the continuous mode microwave
reactor
Fig. 6. Energy exchanges involved during a continuous mode experiment.
•
•
•
Ideal continuous stirred tank reactor (CSTR).
Steady state is established and the reaction volume is constant.
Because energy consumption is at the heart of process inten-
sification, our objective is also to design a continuous microwave
process which is as cost effective as possible. We used the power
data collected during our experiments to estimate the part of the
efficiently used energy input for the process and have since identi-
fied the improvements to be made.
The mass flow rate is the same at the inlet and at the outlet of the
reactor (Qe = Qs = Q).
The determined enthalpy of the reaction (ꢁHr = −247 kJ/mol of
aniline) includes the contribution of all the intermediate reac-
tions in the Skraup synthesis.
•
An energetic balance on the microwave pilot reactor allowed
us to assess the energy efficiency of the system. First, the enthalpy
of the global Skraup reaction was estimated by differential scan-
ning calorimetry (DSC) measurements. The apparatus used in this
study was a Q1000 from TA Instruments, equipped with Universal
Analysis 2000 software. Analyses were carried out in sealed vessels
ensuring their chemical resistance to the acidic reaction medium
and allowing the elevated pressure and temperature conditions of
the reaction. The performed tests consisted in heating our reaction
mixture at a rate of 10 ◦C/min from 20 ◦C to 300 ◦C and following the
heat flow recorded by the DSC. Analyses were made using a typical
reaction mixture of glycerol (3 eq), sulphuric acid (3 eq) and 1 eq of
aniline in water (with a concentration of 1 mol of aniline per liter of
water). After homogenization, a 5.48 mg sample was placed in the
crucible and subjected to the DSC temperature increase. The anal-
ysis showed an exothermal peak of 147.6 J/g of mixture. The top of
the peak was detected at 210 ◦C, which corresponds to our experi-
ments temperature. The enthalpy of the reaction was estimated at
−247 kJ/mol of aniline.
•
The real temperature value of the outlet flow corresponds to the
estimated one (about 200 ◦C inside).
The incident, reflected and absorbed powers PI (W), PR (W) and
Pabs (W) respectively were given in Fig. 5.
Considering the partial mass balance for aniline (which is the
limiting reactant), the conversion XA can be written as:
FAs − FAe
XA
=
FAe
where FAe and FAs are the aniline molar flows at the inlet and at the
outlet of the reactor (mol s−1).
The energy balance can now be formulated as follows:
˙
˙
Pabs + He = Hs + PꢁH + Ploss
r
˙
˙
Here, He(W) and Hs(W) are the enthalpy fluxes at the inlet and
the outlet of the reactor, respectively. Pabs represents the absorbed
microwave power, which is equal to 135 W as shown in Fig. 5. Ploss is
the energy loss (W) of the system, which can be therefore expressed
by the following equation:
In this study, we considered the example corresponding to entry
1 of Table 3, for which the main powers involved are shown in
Fig. 6. Q, H and T are respectively the mass flow rate (g s−1), the
enthalpy (J/g) and the temperature (K) of the mixture. Subscripts e
and s mention the inlet and outlet of the reactor. The enthalpy of
the reaction determined by DSC (ꢁHr = −247 kJ/mol of aniline) was
calculated per gram of converted aniline as it is the limiting reagent
in Skraup synthesis, ꢁHr = −2.66 kJ/g of converted aniline.
Before establishing the heat balance, the following hypotheses
were considered:
˙
˙
˙
˙
Ploss = Pabs − PꢂH + He − Hs = Pabs − (XAFAeꢁHr) + He − Hs
r
˙
˙
= Pabs − (XACAeQꢁHr) − (Hs − He)
CAe: aniline’s molar concentration in the starting mixture
(CAe = 1 × 10−4 mol g−1).
Q: mass flowrate (Q = 6.9 g min−1 = 0.115 g s−1).
Please cite this article in press as: H. Saggadi, et al., Microwaves under pressure for the continuous production of quinoline from glycerol,
Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.10.050

G Model
CATTOD-9351; No. of Pages9
ARTICLE IN PRESS
8
H. Saggadi et al. / Catalysis Today xxx (2014) xxx–xxx
The difference in enthalpy fluxes can be calculated considering
the specific heat of the mixture at the inlet and at the outlet of the
reactor (Cpme and Cpms) and the mass flow rate (Qe = Qs = Q).
In the second part of this paper, we demonstrated the possible
continuous operation of the microwave reactor under pressure for
chemical synthesis. An experimental protocol for adequate and safe
experiments was established. However, the quinoline synthesis in
continuous mode was limited by the mixture’s acidic concentra-
tion and by the chemical resistance of different elements of the
pilot. The energy balance of the device was established, showing
reasonable heat losses, and a microwave energy absorption that
could even be improved by changing the microwave geometry and
by an automatic tuning of the cavity. A pilot scale continuous reac-
tor, intensified by microwaves, working at pressure conditions up
to 19 bar and temperatures up to 220 ◦C is now available for the
continuous production of chemicals.
˙
˙
Hs − He = Q(CpmsTs − CpmeTe)
The calculation of the specific heat of the mixture starting from
mass fractions of the components and their individual specific heat
shows that these values are very similar to the specific heat of water
at the corresponding temperature (at 20 ◦C for the inlet and 210 ◦C
for the outlet).
Considering all this data, the heat losses can be evaluated as
follows:
Ploss = Pabs − QCpe(Ts − Te) − (XAQCAeꢁHr)
References
Ploss = 135 − 111.32 + 0.28 = 24.5 W
[1] R. Cecilia, U. Kunz, T. Turek, Possibilities of process intensification using
microwaves applied to catalytic microreactors, Chem. Eng. Process.: Process
Intensif. 46 (9) (2007) 870–881.
[2] R. Cherban´ ski, E. Molga, Intensification of desorption processes by use of
microwaves—an overview of possible applications and industrial perspectives,
Chem. Eng. Process.: Process Intensif. 48 (1) (2009) 48–58.
[3] R. Gedye, F. Smith, K. Westaway, H. Ali, L. Baldisera, L. Laberge, J. Rousell, The
use of microwave ovens for rapid organic synthesis, Tetrahedron Lett. 27 (3)
(1986) 279–282.
[4] R.J. Giguere, T.L. Bray, S.M. Duncan, Application of commercial microwave ovens
to organic synthesis, Tetrahedron Lett. 27 (41) (1986) 4945–4948.
[5] M.A. Surati, S. Jauhari, K.R. Desai, A brief review: microwave assisted organic
reaction, Arch. Appl. Sci. Res. 4 (1) (2012) 645–661.
[6] I. Polaert, L. Estel, A. Ledoux, Microwave-assisted remediation of phe-
nol wastewater on activated charcoal, Chem. Eng. Sci. 60 (22) (2005)
6354–6359.
[7] S. Ravichandran, E. Karthikeyan, Microwave synthesis—a potential tool for
green chemistry, Int. J. ChemTech Res. 3 (1) (2011) 466–470.
[8] B. Toukoniitty, J.-P. Mikkola, D.Y. Murzin, T. Salmi, Utilization of electromag-
netic and acoustic irradiation in enhancing heterogeneous catalytic reactions,
Appl. Catal. A: Gen. 279 (1) (2005) 1–22.
[9] B.A. Roberts, C.R. Strauss, Toward rapid, ‘green’, predictable microwave-
assisted synthesis, Acc. Chem. Res. 38 (8) (2005) 653–661.
[10] J. Marquié, G. Salmoria, M. Poux, A. Laporterie, J. Dubac, N. Roques, Acylation
and related reactions under microwaves. 5. Development to large laboratory
scale with a continuous-flow process 1, Ind. Eng. Chem. Res. 40 (21) (2001)
4485–4490.
In our reaction conditions, the power value generated by the
enthalpy of the reaction (XAQCAeꢁHr = 0.28 W) is very low and has
an insignificant contribution in the energy balance. Furthermore,
heat losses represent 18% of the microwave absorbed power. This
value illustrates the energy transferred from the fluid to the walls
of the reactor, the metallic assembly block and then to the sur-
rounding atmosphere. The majority of the absorbed power (Pabs) is
mainly used to increase the liquid temperature rising and to com-
pensate heat losses. Although the value of heat losses is not high,
it could even be reduced by changing the geometry of the reactor
and by a better insulation of the reactor, which could reduce heat
losses dissipated to the surrounding air.
Regarding the power curves, we can notice that the efficiency
of our microwave device is approximately 45% by calculating the
ratio of the absorbed power to the incident power delivered by
the magnetron. This value is governed by the geometry of the sys-
tem, notably by the adjustment of the piston’s position in order
to minimize the reflected power. This adjustment depends on
the interaction of the microwaves with the medium, and there-
fore the dielectric properties. With manual tuning of the piston,
as in our case, the optimal adaptation is sometimes impossi-
ble.
[11] A. Stadler, C.O. Kappe, Microwave-mediated Biginelli reactions revisited. On
the nature of rate and yield enhancements, J. Chem. Soc., Perkin Trans. 2 (7)
(2000) 1363–1368.
As a conclusion for this part, the major efforts have still to
be made on the microwave geometry, that is to say, the wave-
transmission system, the adaptation tuning, and the reactor shape
in order to better limit wave reflections and ensure better energy
absorption. Heat losses could also be reduced by insulation of the
reactor to be even more economical.
[12] C.J. Abeywickrama, Y.P. Timilsena, S.K. Rakshit, L. Chrusciel, N. Brosse,
Rapid optimization of typha grass organosolv pretreatments using parallel
microwave reactors for ethanol production, Ind. Eng. Chem. Res. 52 (4) (2013)
1691–1697.
[13] A. Stadler, C.O. Kappe, High-speed couplings and cleavages in microwave-
heated, solid-phase reactions at high temperatures, Eur. J. Org. Chem. 2001
(5) (2001) 919–925.
[14] A. Stadler, B.H. Yousefi, D. Dallinger, P. Walla, E.V. der Eycken, N. Kaval,
C.O. Kappe, Scalability of microwave-assisted organic synthesis. From single-
mode to multimode parallel batch reactors, Org. Process Res. Dev. 7 (2003)
707–716.
4. Conclusion
[15] M. Treu, T. Karner, R. Kousek, H. Berger, M. Mayer, D.B. McConnell, A. Stadler,
Microwave-assisted parallel synthesis of fused heterocycles in a novel parallel
multimode reactor, J. Comb. Chem. 10 (6) (2008) 863–868.
[16] G. Pipus, I. Plazl, T. Koloini, Esterification of benzoic acid in microwave tubular
flow reactor, Chem. Eng. J. 76 (3) (2000) 239–245.
[17] T. Cablewski, A.F. Faux, C.R. Strauss, Development and application of a con-
tinuous microwave reactor for organic synthesis, J. Org. Chem. 59 (12) (1994)
3408–3412.
The undertaken work led to the development of a microwave
pilot reactor, allowing high temperature and pressure required for
the Skraup quinoline synthesis, for a reaction volume multiplied
by 5 compared to what is usually used in many organic synthesis
laboratories.
Firstly, the batch pilot scaled up by five was designed and
optimized. Pressure/temperature conditions were controlled. The
regulation of the absorbed power with respect to the measured
pressure was demonstrated to be very advantageous, allowing sta-
ble, repeatable and safe operation. Batch quinoline synthesis gave
similar yields compared to those obtained in the small scale experi-
ments (42% yield), although under reduced temperature conditions
(200 ◦C vs. 220 ◦C) and a significantly reduced pressure in the sys-
tem (12 bar vs. 25 bar).
The comparison between our microwave device and con-
ventional heating clearly showed the advantages of the
microwave reactor, by reducing the reaction time and energy
consumption.
[18] I. Braun, G. Schulz-Ekloff, D. Wöhrle, W. Lautenschläger, Synthesis of AlPO4–5
in a microwave-heated, continuous-flow, high-pressure tube reactor, Micro-
porous Mesoporous Mater. 23 (1) (1998) 79–81.
[19] R. Morschhäuser, M. Krull, C. Kayser, C. Boberski, R. Bierbaum, P.A. Püschner,
T.N. Glasnov, C.O. Kappe, Microwave-assisted continuous flow synthesis on
industrial scale, Green Process. Synth. 1 (2012) 281–290.
[20] J.D. Moseley, C.O. Kappe, A critical assessment of the greenness and energy
efficiency of microwave-assisted organic synthesis, Green Chem. 13 (4) (2011)
794.
[21] G. Pipusˇ, I. Plazl, T. Koloini, Esterification of benzoic acid with 2-ethylhexanol
in
a microwave stirred-tank reactor, Ind. Eng. Chem. Res. 41 (5) (2002)
1129–1134.
[22] S. Horikoshi, S. Matsuzaki, T. Mitani, N. Serpone, Microwave frequency effects
on dielectric properties of some common solvents and on microwave-assisted
syntheses: 2-Allylphenol and the C12 C2 C12 Gemini surfactant, Radiat.
Phys. Chem. 81 (12) (2012) 1885–1895.
Please cite this article in press as: H. Saggadi, et al., Microwaves under pressure for the continuous production of quinoline from glycerol,
Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.10.050

G Model
CATTOD-9351; No. of Pages9
ARTICLE IN PRESS
H. Saggadi et al. / Catalysis Today xxx (2014) xxx–xxx
9
[23] M. Nishioka, M. Miyakawa, Y. Daino, H. Kataoka, H. Koda, K. Sato, T.M. Suzuki,
Single-mode microwave reactor used for continuous flow reactions under ele-
vated pressure, Ind. Eng. Chem. Res. 52 (12) (2013) 4683–4687.
[25] H. Saggadi, D. Luart, N. Thiebault, I. Polaert, L. Estel, C. Len, Toward the synthesis
of 6-hydroxyquinoline starting from glycerol via improved microwave-assisted
modified Skraup reaction, Catal. Commun. 44 (2014) 15–18.
[24] T.N. Glasnov, C.O. Kappe, The microwave-to-flow paradigm: translating
high-temperature batch microwave chemistry to scalable continuous-flow
processes, Chemistry 17 (43) (2011) 11956–11968.
[26] H. Saggadi, N. Thiebault, I. Polaert, L. Estel, C. Len, Quinoline and phenanthroline
preparation starting from glycerol via improved microwave-assisted modified
Skraup reaction, RSC Adv. 4 (2014) 21456.
Please cite this article in press as: H. Saggadi, et al., Microwaves under pressure for the continuous production of quinoline from glycerol,
Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.10.050