Novel Application of Polymer Networks Carrying Tertiary Amines as a Catalyst Inside Microflow Reactors Used for Knoevenagel Reactions

Article abstract of DOI:10.1002/ejoc.202000978

A novel application is described for utilizing hydrogel dots as organocatalyst carriers inside microfluidic reactors. Tertiary amines were covalently immobilized in the hydrogel dots. Due to the diffusion of reactants within the swollen hydrogel dots, the

Full text of DOI:10.1002/ejoc.202000978

European Journal of Organic Chemistry
Accepted Article
Accepted Article
Title: Novel application of polymer networks carrying tertiary amines as
catalyst inside microflow reactors used for Knoevenagel reactions
Authors: Patrik Berg, Franziska Obst, David Simon, Andreas Richter,
Dietmar Appelhans, and Dirk Kuckling
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Eur. J. Org. Chem. 10.1002/ejoc.202000978
Link to VoR: https://doi.org/10.1002/ejoc.202000978
01/2020

10.1002/ejoc.202000978
European Journal of Organic Chemistry
RESEARCH ARTICLE
Novel application of polymer networks carrying tertiary amines as
catalyst inside microflow reactors used for Knoevenagel
reactions
Patrik Berg,[a] Franziska Obst,[b] David Simon,[b] Andreas Richter,[c] Dietmar Appelhans,[b] Dirk
Kuckling*[a]
Key Topic: Microfuidic Synthesis
Abstract: A novel application is described for utilizing hydrogel dots
as organocatalyst carriers inside microfluidic reactors. Tertiary
amines were covalently immobilized in the hydrogel dots. Due to the
diffusion of reactants within the swollen hydrogel dots, the accessible
amount of catalysts inside a microfluidic reactor chamber can be
increased compared to the accessible amount of surface-bound
catalysts. To perform fast Knoevenagel reactions, important flow
parameters had to be validated to optimize the reactor performance
while keeping the dimensions of the reactor chamber constant; e. g.
the height of the hydrogel dots had to be adjusted to the invariable
dimensions of the reactor chamber, or an adjustment of
organocatalysts in the hydrogel dots had to be validated to achieve
the highest conversion rate during a certain residence time. To
characterize the conversion, nuclear magnetic resonance (NMR) and
UV-Vis-spectroscopy were utilized as an offline and online method,
respectively. With suitable hydrogel dots, the influence of different
flow parameters (e.g., operating flow rate and reactant concentration)
on the selected model reactions in the microfluidic reactor was
investigated. Finally, a variety of reactants were screened with the
optimized flow parameters. With these results, the turnover frequency
was determined for the Knoevenagel reactions in a microfluidic
reactor, and the results were compared with published data that were
determined by other synthetic approaches.
(supplier information). In combination with imprinting methods for
polydimethylsiloxane[1], 3D printing[2] or high-resolution milling
cutter[3], all the necessary components to build MFRs are
available.
In general, flow reactors combines the benefits of continuous
operation[4], online detection[5] and reaction parameter adjustment
during synthesis[6] to optimize the reactor performance. In smaller
scaled flow reactors, i.e., microfluidic reactors, further benefits are
observed. This includes excellent heat transfer, laminar flow and
good control of reaction conditions (temperature, etc.) and several
more, which makes the reactions more economic and safer.[7]
With the previously discussed benefits, a MFR enables good
handling and fast screening of different reactions. Finally, different
online detection methods, such as NMR[8], HPLC[9], IR[10], mass
spectrometry[11] or UV/Vis[12], have been successfully used. In
addition, MFRs are safer because only small amounts of
chemicals are used, and the in situ use of highly reactive or toxic
reagents is easily possible.[13] All these benefits are important
aspects in modern chemistry because of the ongoing importance
of “green chemistry”.[14,15]
Supported organocatalysts are an important research topic
because the catalytic activity of organocatalysts can be combined
with the physical properties of carrier material to utilize reusable
catalysts,[16,17] increase catalytic activity[17] or long-term use.[18]
Therefore, this principle is often applied in flow reactors. Already
in the 1970s, a fully automated reactor setup was described that
used organocatalysts inside a MFR.[19] A surface-modified
alumina bed was used as a carrier for organocatalysts inside the
reactor chamber.
Introduction
In modern chemistry, microfluidic reactors (MFRs) have become
increasingly important. In the past, batch reactors were easier to
assemble than flow reactors, but today, with modern technical
possibilities, there is more hardware that is commercially
available to assemble flow reactor systems. Microfluidic pumps
with microstepping motors enable flow rates of less than 5 pL/min
[a]
P. Berg, Prof. Dr. D. Kuckling
Department of Chemistry, Faculty of Science
Paderborn University
Warburger Str. 100, 33098 Paderborn (Germany)
E-mail: dirk.kuckling@uni-paderborn.de
https://chemie.uni-paderborn.de/arbeitskreise/organische-
chemie/kuckling
[b]
[c]
F. Obst, Dr. D. Simon, Dr. D. Appelhans
Institute for Polymer Research
Prof. Dr. A Richter
Institute of Semiconductors and Microsystems
Technische Universität Dresden
Helmholtzstr. 10, 01062 Dresden (Germany)
Figure 1: Schematic illustration of monoliths (top way) or gels (bottom way) used
as carrier for organocatalysts within reactor chamber and the resulting
inaccessible volume (blue) within the reactor chamber (grey box) of each
approach.
Since the 2000s numerous reactions were performed under flow
conditions including the prominent Knoevenagel reaction. Here,
Supporting information for this article is given via a link at the end of
the document
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000978
European Journal of Organic Chemistry
RESEARCH ARTICLE
mostly inorganic-based materials are used as carrier for
catalysts.[20–24] This approach can easily been used, however,
forces an inaccessible volume within the reactor chamber due to
the solid carrier material (Figure 1: blue area). To overcome the
limitations given by inaccessible volume swollen polymer
networks (gels) were used as carriers.[18,25–27] In contrast to
monoliths, gels result in a sponge-like structure which avoid any
inaccessible volume due to the diffusion of reactants inside the
gels (Figure 1). Hence, gels are more effective than monoliths.[18]
By different processes of gel synthesis it is possible to prepare a
homogenous distribution of immobilized catalysts from gel
surface to bulk volume.[28] This facilitates to perform the reaction
at each location with equal effectivity and without any inactive
volume within the reactor chamber. However, compared to
inorganic-based materials, polymers as carrier were less often
used as support material to perform the Knoevenagel
reaction.[29,30]
method of integration prevents the bleeding of catalysts. A wide
range of low molecular weight catalysts are known that can be
applied as monomers.[32] Here, the application of hydrogel dots as
organocatalyst carrier enables a high variability in the assembled
MFR with respect to the performed reaction and the required
conditions.
Results and Discussion
To realize our approach, the assembly of the MFR follows the
schematic circuit diagram of Figure 2 A (and Figure S1). Different
steps must be accomplished to build the reactor chamber (Figures
2 B and S2).
In general, inorganic materials or polymers as carriers are mostly
used within packed-bed flow reactors.[18,20–23,25–27,29,30] This type of
reactors often uses column-like reactor chambers which require a
large amount of catalytically active material[29,30] and/or a complex
set up with high pressure pumps to enable the flow rate due to the
tight packing.[18,25] However, high pressure is tough to apply
properly due to collapsing and deactivation of poorly cross-linked
swollen gels during the flow reaction.[26] Beside this poorly cross-
linked gels shows weak mechanical stability during processing.[31]
To face this problem a certain amount of carrier material must be
filled into the reactor chamber to be preswollen there.[18,25,29,30] It
must be mentioned that gels are adaptable structures. Their
swelling strongly depends on the surrounding conditions (i.e.
solvent). Hence, an important aspect for packed-bed flow reactors
is the volumetric limitation of the reactor chamber. If there is a
mismatch between the dimensions of the reactor chamber and
the volume of isotropically swollen gels, the restriction of the
chamber dimensions limits the degree of swelling to a lower level.
It is estimated that this forced lower degree of swelling reduces
the reactor efficiency.[26] It would be expected that the adjustment
of gel size to reactor chamber volume is inevitable to enable
highest reactor efficiency.
Micro-structuring of gels via photolithography is a powerful tool to
handle this adjustment. This method offers
a variety of
advantages in contrast to packed-bed flow reactors. For example,
with photopatterning the arrangement of the gels could be
adjusted by the used masks. As a consequence of the less dense
gel structure, and gaps between the gels, a low and controllable
backpressure can be realized. Beside this, the size of the gels
could be controlled by patterning conditions as well. Due to these
facts a MFR was built to utilize gels as carrier for organocatalysts
in the Knoevenagel reaction in combination with physical
advantages of microfluidic set-ups.
Recently, this principle was applied to hydrogel dots with high
cross-linker content to be utilized as a cage for enzymes inside
microfluidic reactors.[12] It was possible to synthesize hydrogel
dots with incorporated enzymes via photolithography on glass
slides and to successfully use these structures for organic
reactions in MFR experiments. Here, diffusion within the hydrogel
dots was genetical for performing the reaction. Finally, our
Figure 2: A – Schematic illustration of the MFR system. B – Sequence for
assembling the reactor (red circle in A); left – preparation of hydrogel dots on
glass surface; middle – covering hydrogel dots with imprinted PTFE cover layer
to form reactor chamber; right – alumina frame for fixing and connecting
sandwich structure to flow system. C – Structure of the monomers used for the
preparation of the hydrogel dots by photolithography; black box = N,N-
dimethylacrylamide (DMAAm) used for adjusting the catalyst content, red box =
methylene(bis)acrylamide (BMA) used as the cross-linker, blue box = N-[3-
(dimethyl)amino] propylacrylamide (3DMAPAAm) as organocatalyst. D – Image
of the arranged position of the as-prepared hydrogel dots. E – Schematic
illustration of the imprinted structures on the PTFE layer prepared by a high-
resolution milling cutter (channel depth of 140 µm). F – Image of covered
approach combines hydrogel dots as
a matrix and the
immobilization of covalently integrated organocatalysts. This
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000978
European Journal of Organic Chemistry
RESEARCH ARTICLE
hydrogel dots during an MFR experiment. G – Image of the assembled MFR
system from the laboratory; the red circle marks the MFR.
dots to the height of the imprinted structures to create structures
that completely filled the reactor chamber. The swelling in the z-
direction of the surface-attached hydrogel dots is calculable from
the free swelling results in accordance with Rühe.[34] Proper
polymerization conditions involve the monomer concentration,
irradiation time and irradiation intensity. The influence of these
parameters on hydrogel dot height was investigated via confocal
microscopy (Figure S3 – S6). The final polymer preparation was
performed with a monomer concentration of 3.7 M, an irradiation
time of 9.4 sec and an intensity of 428 mW with an 8 cm distance
between the light source and sample. To place the dots properly
in the reactor chamber, a suitable mask with a structure of 158
dots in a diamond-shaped position was applied for the imprinted
reactor (Figure 2 D).
Full swelling was ensured by using the maximal chamber height,
and the degree of swelling of gels had to be determined. Hence,
the degree of swelling in water and DMSO depending on the
cross-linker content was investigated for the unattached gels
within the volume of the samples (SI equation (1) and Figure S7).
Both solvents were suitable for the selected reaction and in
agreement with “green chemistry”.[14,15] A cross-linker content of
1 mol% provides the highest degree of swelling and therefore
enables the use of an immobilized catalyst in full capacity (Figure
S8). Finally, the Knoevenagel reaction was carried out in a solvent
mixture of isopropanol:DMSO (1:1) to reduce the surface tension
and back pressure of the flow system. From the degree of swelling
of the bulk gels in the final solvent mixture, the extension along
the z-axis of the swollen surface-attached gels depending on the
catalyst content could be calculated according to Rühe (see SI
equation (2)).[34] The results are summarized in Table 1. With the
calculated z-axis degree of swelling, the polymer height in the dry
state could be used to adjust the size of hydrogel dots. Therefore,
the height of the polymer dots for a polymer composition of 1
mol% cross-linker and 10, 50 or 90 mol% catalyst loading was
measured via confocal microscopy (compare Figure S9). Six
randomly selected dots on two separately prepared glass slides
were measured. To calculate the height in the swollen state, the
extension along the z-axis was used in combination with the
determined height in the dry state (Figure S10).
First, the photopolymerization of the desired ratios of monomers
must be optimized (Figure 2 C). In this work, tertiary amines were
used to catalyze a number of fast Knoevenagel reactions as
reported by Diaz.[16] These reactions were selected, since the flow
rates in the range of µL/min and the given volume of reactor
chamber restrict the residence time, which can affect the reaction
time to the range of a few minutes. For this work, acrylamide
compounds were used as catalyst, cross-linker and gel forming
agent (Figure 2 C). These monomers were commercially
available and provide a good degree of swelling in suitable
solvents for the selected Knoevenagel reaction. However, the
variability of solvents offers a high flexibility of the MFR used for
different reactants. Further, the acrylamides enable the fast
preparation of the hydrogel dots via photolithography. The
resulting hydrogel dots (Figure 2 D) were arranged in a diamond-
shaped position for enhancing the residence time of reactants in
the reaction chamber. Beside this, this design was selected due
to the reduced flow velocity and shear force within the reactor
chamber.[1] These hydrogel dots were covered with an imprinted
PTFE layer (Figure 2 E). This cover layer was imprinted by a high-
resolution milling cutter and included a meander-like channel for
mixing the reactants because it was shown that this mixer type is
most reliable[33] as well as the reactor chamber fitting with the
diamond champed position of the hydrogel dots (Figure 2 E). The
channels were fabricated with a depth of 140 µm, which yields a
reactor chamber volume of 31 µL. Figure 2 F shows the
assembled reactor chamber through a transparent bottom of the
alumina frame during an experiment. The entire assembled MFR
system is shown in Figure 2 G, where the circle indicates the
assembled reactor, which is also marked in Figure 2 A.
With this reactor setup, the influence of the different parameters
on the reactor conversion was investigated. For example, the
influence of the hydrogel dot composition, flow rate or reactant
concentration was varied to optimize the reactor performance of
the selected Knoevenagel reaction. To study the influences of the
different parameters on the Knoevenagel reaction, a model
system was selected because of the given reactor chamber
volume a fast reaction is mandatory.[16] For example, it was
reported that p-anisaldehyde (pAnis) and malononitrile (MDN)
can undergo high conversion within 5 min in the presence of a
catalyst (Figure 3).[9]
Table 1: Degree of swelling in isopropanol: DMSO (1:1) for unattached
polymers with 1 mol% cross-linker content and calculated z-axis degree
of swelling for attached polymers.
degree of swelling [-]
unattached polymer
networks
calculated extension along
z-axis [-] attached
hydrogel dots
catalyst loading
[mol%][a]
10
50
90
8.6 (±0.6)
7.9 (±0.7)
7.0 (±1.1)
3.3
3.2
2.9
Figure 3: The model reaction of pAnis and MDN as a sample system to indicate
reactor efficiency depending on different parameters.
To enable the most efficient use of the reactor chamber, hydrogel
dots were used. First, attention must be paid to the physical
properties of flexible networks, e.g. the degree of swelling. The
reactor itself was assembled by covering the surface-attached
hydrogel dots with an imprinted PTFE layer, and everything was
fixed in an alumina frame to connect the reactor with the flow
system via capillary tubes (Figure 2). The height of the imprinted
structures was set to 140 µm. To efficiently use the reactor
chamber, it was necessary to adjust the height of the hydrogel
[a]polymers prepared from a total monomer concentration of 3.7 M with
428 mW UV-light intensity and 1 mol% cross linker, differences were
adjusted with N,N-dimethylacrylamide.
As seen from Table 2, hydrogel dots with a catalyst loading of 90
mol% fit the swollen state to the height of the reactor chamber.
Thus, this polymer composition was determined in accordance
with reaching the highest conversion of the MFR. To investigate
the reactor performance, it was first necessary to determine the
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000978
European Journal of Organic Chemistry
RESEARCH ARTICLE
conversion of the selected reactions in the absence of a catalyst.
This is an important aspect for the offline determination of the
conversion, since the reactants are mixed in the inlet and outlet
phase and might further react while accumulating. Further, the
concentration of the reactants must be optimized to meet the
needs of different analysis methods.
the influence of the blank conversion on the reactant conversion
due to residence time in the collection vessel (see Table S1).
Hence, the samples were collected over 1 h and then were frozen
immediately to stop the reaction. The samples were warmed just
before measuring the NMR spectra.
With the model reaction of pAnis and MDN (Figure 3), the MFR
experiments were performed to compare both the detection
methods. Figure 4 shows a similar behavior of the conversion vs.
time curves for the shape of curves, maximum conversion and
long-term conversion (≥ 25 h) of the continuous and discontinuous
detection methods. From Figure 4, a decrease in the reactor
conversion was observed. This effect can also be tracked visually
by the appearance of the gel dots. Due to the transparent bottom
of the MFR, the dots inside the reactor chamber could be
observed. The reaction of 3,5-dichloro-2-hydroxybenzaldehyde
and MDN was selected due to their highest reaction rates of all
the performed Knoevenagel reactions (see Table S1). The
product of this reaction has a pink color, which indicates
conversion inside the MFR chamber (Figure 5). During the 8 h
experiment, the color of the gel dots changed from colorless to
brown, which is assumed to be due to the enrichment of MDN (the
MDN that was purchased from the supplier was brown colored)
and product within the dots.
Table 2: Determined height of hydrogel dots in a dried state and the
calculated height in swollen state by approximation.
calculated
height of swollen
state [µm]
filling of
reactor
chamber [%][b]
catalyst loading
[mol%][a]
height of dry
state [µm]
10
50
90
32 (± 2.9)
40 (± 2.2)
44 (± 4.7)
109
128
130
80
90
90
[a]polymers were prepared from total monomer concentration of 3.7 M with
428 mW UV-light intensity and 1 mol% cross linker, differences were
adjusted with N,N-dimethylacrylamide; [b]calculated with the set height of
imprinted structures of 140 µm.
The amount of synthesized product was limited in the µL/min
range by the flow rate and the reactant concentration up to solvent
saturation. The results of the blank Knoevenagel reaction with
different aldehydes and two CH-active compounds (MDN and
ethyl cyanoacetate (ECAc)) are summarized in Table S1. For all
the reactions, a concentration was used so that the reactants and
products were always dissolved. As observed, it is not possible to
store the reactants together before the experiment and to collect
a product solution over a long period until analysis of conversion
is performed due to blank reaction. To directly measure the
conversion after the reaction, a continuous analysis via a flow
cuvette (size of flow cell 6.2 µL) was used in combination with
UV/Vis spectroscopy. For this method, calibration curves of pure
substances must be determined to transform the absorbance into
conversion (SI, chapter 2.4). In addition to UV/Vis spectroscopy,
A
B
it was also possible to determine the conversion in
discontinuous procedure by NMR.
a
Figure 5: Images of polymer gel dots during the MFR reaction of 3,5-dichloro-2-
hydroxybenzaldehyde with MDN after A - 1 h and B - 8 h by using a polymer
composition of 90 mol% 3DMAPAAm, 9 mol% DMAAm and 1 mol% BMA. The
flow direction is from left to right along the blue arrow.
70
60
50
40
30
20
10
0
The diffusion of the reactants into the hydrogel dots and the
reactions occurring inside the dots influence the efficiency of the
hydrogel dots in a complex manner, e.g. degree of swelling and
chemical composition. When this MFR experiment was repeated,
comparable results were observed with respect to the average
conversion and the shape of the conversion-time diagram (Figure
S13). For four different MFR experiments, an average conversion
of 55 (± 7) % over the first 8 h was calculated. This result shows
that this system is robust against external influences during the
assembly or running process.
0
2
4
6
8
10 12 14 16 18 20 22 24 26
In general, the online detection (UV/Vis) of the reactor
performance is most beneficial for a continuous flow reactor
because of the ability to make feasible online adjustments to
optimize reactor performance. However, NMR is the most suitable
method to analyze the reaction mixture with respect to all the
included compounds (e.g., reactants, products, and side
reactions).
time / h
Figure 4: Comparison of conversion-time-diagram of the continuous (black dots)
analysis via UV/Vis spectroscopy and the discontinuous (red dots) analysis via
NMR spectroscopy of the model reaction with pAnis and MDN (1 M each) under
an operating flow of 2 µL/min and a polymer composition of 90 mol% catalyst,
9 mol% DMAAm and 1 mol% BMA.
For this method, it was necessary, especially for the MDN, to
divide the sample collection into smaller time periods to reduce
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000978
European Journal of Organic Chemistry
RESEARCH ARTICLE
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
Following these results, the reactant concentration in the MFR
was varied to characterize the MFR efficiency. Different
concentrations of starting material, up to the saturation of MDN
(in a 2 M stock solution) for the model reaction (Figure 3), were
tested.
0 mol%
10 mol%
50 mol%
90 mol%
A
0,7
0,1 mmol mL-1
0,5 mmol mL-1
0,6
1,0 mmol mL-1
2,0 mmol mL-1
0,5
0,4
0,3
0,2
0,1
0,0
0
1
2
3
4
5
6
7
8
time / h
Figure 6: Reactor conversion vs. running time as determined via NMR for the
model reaction of pAnis and MDN (1 M) and polymer samples with 0 (black
dots), 10 (red dots), 50 (green dots) and 90 mol% (blue dots) catalyst loading
and 1 mol% BMA with a flow rate of 2 µL/min. Data of 10 and 50 mol% at 2 h,
50 and 90 mol% at 7 h, as well as 0 and 10 mol% at 8 h overlaps at the same
value.
0
1
2
3
4
5
6
7
8
time / h
0.6
0.5
0.4
0.3
0.2
0.1
0.0
B
An important parameter of the flow system was the catalyst
loading of the hydrogel dots. To identify the best catalyst loading,
flow experiments were performed with 0, 10, 50 and 90 mol%
catalyst-loaded hydrogel dots, and the results were analyzed with
NMR (Figure 6) for the model reaction (Figure 3). From the NMR
spectra, the average conversion was calculated during the 1 h
time periods (Figure S14).
0 h
1 h
3 h
6 h
Table 3: Average conversion of the model reaction of pAnis and MDN (1 M
each) after 8 h with a flow rate of 2 µL/min. Bracketed values describes the
highest determined conversion during 8 h MFR running time.
catalyst loading [mol%]
average conversion [%]
0
10
50
90
0.0
0.5
1.0
1.5
2.0
2 (4)
10 (28)
30 (53)
31 (57)
ceduct / mmol mL-1
Figure 7: A – Screening of the reactor efficiency for the model reaction (Figure
3) depending on the reactant concentration at 0.1 (blue dots), 0.5 (green dots),
1.0 (red dots) and 2.0 M (black dots) reactant stock solution via 1H-NMR
Without a catalyst, only a small amount of product was detected
as shown in Table S1, while an increase in the product
concentration was observed for catalyst loading from 10 to 50
mol%. In contrast, at 90 mol% catalyst loading, the conversion
leveled off, and a similar behavior was determined for catalyst
loading at 50 mol% (Table 3). The two curves show only a small
difference in the product concentration during the first 8 h.
However, the long-term activity (more than 25 h) revealed a
product concentration of 0.05 M for 90 mol% catalyst loading and
0.01 M for 50 mol% catalyst loading (Figure S14). Hence, the best
polymer network composition was 90 mol% catalyst, 9 mol%
DMAAm and 1 mol% BMA because after a running time of >22 h,
50 mol% catalyst loading produces only slightly more product than
the catalyst loading of 0 and 10 mol% (Figure S14). The similar
effectivity of 50 and 90 mol% during first 8 h running time can be
explained by the weak dependence of gel composition and
degree of swelling. Thereby, a catalysts content of 90 mol% within
the gel forces a higher reactant conversion in comparison to
catalysts content of 50 mol% due to the equally swollen state of
gels at each composition. It is assumed that in case of other
monomers showing a stronger dependence of gel composition
and degree of swelling the most beneficial composition must be
found and will be an equilibrium between the catalytic activity of
the gel and the swelling behavior.
spectroscopy.
B – Product concentration depending on the reactant
concentration in the MFR for 0 h (black dots), 1 h (red dots), 3 h (green dots)
and 6 h (blue dots) with the running time detected via NMR using a gel
composition of 90 mol% catalyst and 1 mol% cross-linker at a flow rate of 2.0
µL/min.
Figure 7 (A) shows an increase in the reactor efficiency depending
on the concentration following the second-order kinetics of the
model reaction. This behavior is also indicated by the exponential
slope shown in Figure 7 (B). The results shown in Figure 7 show
that the best MFR performance was reached with a 2 M reactant
concentration. The last parameter considered to influence the
MFR performance was the flow rate. The standard conditions with
an operating flow rate of 2.0 µL/min were compared with flow
rates of 4.0 and 8.0 µL/min. Table 4 lists the residence times
inside the reactor chamber and the average conversion of the
model reaction (Figure 3) with a concentration of 2 M. An
increasing flow rate resulted in a decrease in the conversion in
the reactor chamber (Table 4). This increase in flow rate is
responsible for the reduced residence time inside the reactor
chamber. Thus, the most beneficial flow rate was 2 µL/min. In
Table 4, the highest total amount of product generated within 8 h
was obtained for a flow rate of 8 µL/min. However, it was
generated with a four times higher amount of reactant (from 0.96
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000978
European Journal of Organic Chemistry
RESEARCH ARTICLE
mmol for 2 µL/min to 3.84 mmol for 8 µL/min) that was pumped
through the reactor by using a reactant stock solution of 2 M.
diameter, which was determined by confocal microscopy, the
volume of the dried hydrogel dots was calculated (Figure S9). The
mass of hydrogel dots was calculated via the density that was
identified for the hydrogels (0.90 (±0.06) g/mL), which is in
accordance with the literature for DMAAm gels.[35] The maximum
mass of hydrogel dots was determined from the total irradiated
volume of the 3.7 M monomer mixture in a certain composition
and correlates to a 45 % yield. With this result, the amount of
catalyst inside the attached hydrogel dots (with composition of 90
mol% 3DMAPAAm, 9 mol% DMAAm and 1 mol% BMA) was 2.2
µmol for a reactor chamber of 31 µL and 158 dots. With an
operating flow rate of 2.0 µL/min, a 0.96 mL reactant solution was
pumped through the reactor for 8 h. This indicates that 2.2 µmol
catalyst was used for the 0.96 mL reactant mixture. This was a
catalyst content of approximately 0.25 mol% with respect to
reactants. This batch reaction was run for 15 min, corresponding
to the residence time of the reactant mixture inside the reactor
chamber with an operating flowrate of 2.0 µL/min.
Table 4: The average conversion for the model reaction of pAnis and MDN
(2 M) depending on the operating flow rate and residence time inside the
MFR chamber.
flow rate
[µL/min]
residence time
[min][a]
average
conversion [%][b]
product
[mmol][c]
2.0
4.0
8.0
18
9
44
19
12
0.42
0.36
0.46
5
[a]residence time calculated with a MFR chamber volume of 31 µL; [b]polymer
composition of 90 mol% catalyst, 1 mol% cross linker and 9 mol% DMAAm,
prepared from 3.7 M monomer concentration; [c]calculated from the reactant
concentration which is pumped through the reactor chamber over 8 h.
60
2,0 µL min-1
4,0 µL min-1
Table 5: MFR conversions for different aldehyde reactants with MDN.
50
8,0 µL min-1
reference
flow
experiment
[%][b,c]
flow
experiment
[%][b]
batch
reference
[%][b,d]
40
30
20
10
0
reaction
aldehyde[a]
1
2
3
4
5
6
7
27
44
32
22
41
29
25
5
3
1
1
7
0
0
43
16
39
9
0
1
2
3
4
5
6
7
8
time / h
Figure 8: Conversion-time-diagram of the model reaction between pAnis and
MDN (2 M) with polymer dots prepared from 90 mol% catalyst-loading and 1
mol% cross-linker. The different flow rates were screened at 2.0 µL/min (black
dots), 4.0 µL/min (red dots) and 8 µL/min (green dots).
These results show that a high flow rate is not economical due to
the high amount of unreacted compounds. A reduction in flow rate
to less than 2 µL/min was also not useful due to the crystallization
of reactants from the reaction mixture at the outlet capillary tube.
This crystallization is caused by the evaporation of volatile
solvents, resulting in a blocking of the MFR. A closer look at the
conversion-time diagrams of the model reaction for the different
flow rates (Figure 8) revealed that the conversion decreased more
rapidly with increasing flow rate. Under a higher flow rate, a higher
amount of reactants were pumped through the reactor in a certain
time, and there was less time remaining for the diffusion of
reactants within the dots. With the results shown in Figures 6 – 8,
the best polymer network composition, reactant concentration
and flow rate were found. With these parameters, a reactant
screening was performed (Figure S15 – S21). Therefore, two
reference measurements were done. First, the flow experiment
was compared with a reference experiment under beneficial flow
conditions by using hydrogel dots without a catalyst (composition
of 99 mol% DMAAm and 1 mol% BMA). Second, the flow
experiment was compared with a batch experiment, in which an
equal amount of low molecular weight catalyst (3DMAPAAm) and
residence time was applied. Thus, it was necessary to calculate
the amount of catalyst inside the 158 gel dots. With the height and
38
24
49
[a]Aldehyde and MDN stock solution concentration = 2 M; [b]Calculated from
1H-NMR spectra of reaction solution in respect to aldehyde; [c]Reference
was measured with a gel composition of 99 mol% DMAAm and 1 mol%
cross-linker; [d]batch conditions were 0.25 mol% of low molecular weight
catalyst and reaction time of 15 min.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000978
European Journal of Organic Chemistry
RESEARCH ARTICLE
The comparison of the flow experiment with the catalyst and the
batch reference values showed that the flow experiment could be
more effective (Table 5, reactions 2, 4, 5, and 6). An advantage
of flow reactors is the continuous synthesis of products. It was
observed that 2.2 µmol catalyst can be used for over 6 d for a
Knoevenagel reaction. For the reaction of pAnis and MDN, an
average conversion of approximately 25 % was observed (Figure
S22). Here, in total, 16.8 mL of reactant solution was pumped
through the reactor chamber. For a comparable batch reaction, a
reactant solution with as low as 0.01 mol% of catalyst with respect
to reactants must be used. It was also shown that an almost
constant conversion could be obtained for 5 – 6 d, indicating that
the running time of the reactor can be prolonged to more than 6 d
without a further decrease in conversion. Comparable results for
the reaction of pAnis and ECAc were observed (Figure S22).
ECAc was used in addition to MDN as a CH-active compound.
With ECAc, a more complex product structure was created, in
particular, E/Z isomers of the product. In addition, a long-term
experiment was also performed (Figure S22). The same
conversion behavior was observed for pAnis and MDN. ECAc is
less active than MDN, which results in a lower conversion
(compare Table S1). Hence, it was possible to determine the
reactor conversion with only one averaged 1H-NMR sample. The
results are summarized in Table 6. Lower conversions were
obtained as found for the reaction of MDN with different aldehydes
(Table 5), which can undoubtedly be explained by reactant
reactivity (Table S1). Nevertheless, there was almost no
conversion obtained under batch conditions. Furthermore, E/Z-
ratios of almost 100:0 were determined for the conversion of
ECAc due to steric hindrance between ester functions and
aromatic or aliphatic residues, and these results were in
accordance with literature reports.[16] To compare the MFR
efficiency with other approaches, the turnover frequency of the
catalyst was calculated. The respected values are summarized in
Table 7.
29
5
6
7
0
0
0
0
0
2
(100/0)
21
(100/0)
17
(100/0)
[a]Aldehyde and MDN stock solution concentration = 2 M; [b]Calculated from
1H-NMR spectra of the reaction solution in respect to aldehyde; [c]Bracketed
values describe the E/Z ratio; [d]Reference values were measured with a gel
composition of 99 mol% DMAAm and 1 mol% cross-linker; [e]batch
conditions were 0.25 mol% of low molecular weight catalyst and a reaction
time of 15 min; [f]The configuration was confirmed via the characteristic
trans-CN-H coupling constant of reference substances.
Several different catalysts used in batch reactions were described
for the reaction of benzaldehyde with MDN. According to the
literature, immobilized primary amines enable TOF from 50 h-1[36]
up to 577 h-1.[37] However, these values are hardly comparable,
since the TOF depends on reactants and on reaction parameters.
Table 7: Calculated TOF of different reactions. TOF was calculated in
respect to a residence time of 15 min. Here, 2.2·10-3 mmol of catalyst
(immobilized on surface), a 1 M reactant solution and the average
conversion over 8 h running time of the MFR were used for calculations.
TOF of reaction
with MDN [h-1]
TOF of reaction
with ECAc [h-1]
reaction
aldehyde[a]
1
2
3
4
5
6
7
15
24
17
12
22
16
14
11
13
13
3
Table 6: MFR conversion of different aldehyde reactants with ECAc.
reference
flow
experiment
[%][b,c]
batch
reference
[%][b,e]
flow
experiment
[%][b,d]
reaction
aldehyde[a]
20
1
2
3
4
0
0
0
0
2
0
1
0
(100/0)
24
(100/0)[f]
16
11
9
24
(100/0)
5
(100/0)
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000978
European Journal of Organic Chemistry
RESEARCH ARTICLE
reactants for initiator synthesis, dimethylphenylphosphonite (98 %) was
purchased from Sigma-Aldrich, while 2,4,6-trimethylbenzoylchlorid
(98+ %) and 2-butanon (99 %) were purchased from Alfa Aesar. 3-
(trichlorosilyl)propylmethacrylate (> 90 %) used for surface modification
and was purchased from Sigma-Aldrich. The used solvents for flow
experiments, dimethyl sulfoxide (DMSO) analytical grade, and iso-
propanol (technical grade) were purchased from Grüssing GmbH. The
polydimethylsiloxane (PDMS), to fabricate the imprinted PDMS layer, was
purchased from VWR (“Sylgard® 184 Siliconelastomer Kit”). For flow
reactions different aldehydes and β-dicabonyl compounds were used.
Malononitrile (99 %) was purchased from Janssen Chemica and
cyanoaceticacidethylester (98 %) was purchased from Ferak.
Benzaldehyde (98 %) was purchased from TCI and isovalerraldehyde
(97 %), isobutyraldehyde (98 %), 4-methoxybenzaldehyde (98 %) were
purchased from Sigma Aldrich. The substituted benzaldehydes, 2-
methoxybenzaldehyde (98 %), 2-hydroxy-3-methoxybenzaldehyde (96 %),
Therefore, TOFs were calculated for similar systems as reported
by Diaz.[16] 4-methoxybenzaldehyde and malononitrile reach a
TOF of 44 h-1 and an ECAc of 20 h-1. In comparison with the
results summarized in Table 7, the catalysts reported in the
literature were two times more active. These differences can be
explained by the catalyst activity, with the primary amines being
more active than the tertiary amines. Another example using
tertiary amines shows that the reaction of pAnis with MDN or
ECAc could be performed with a high yield and a TOF of 7 h-1.[38]
This result shows that the reported MFR is more active. Tertiary
amines applied in our approach are two times more active for the
reaction of pAnis with ECAc and three times more active for the
reaction with MDN than these same reactions in a comparable
system.
3,4-dihydroxybenzaldehyde
(97 %)
and
2-hydroxy-3,5-
dichlorobenzaldehyde (98 %) were purchased from Lancaster. 2,3-
dimethylbenzaldehyde (97 %) was purchased from Alfa Aesar.
Conclusion
Analysis methods: The confocal microscope Keyence “VK-9700” was
used for height determination of polymer dots. Images were taken with the
software “VK Viewer 2.4.0.1” and the analysis of polymer height was done
with the software “VK Analyzer 3.4.0.1”. The 1H-NMR spectra for
conversion determination were measured on a Bruker “Avance 300”
spectrometer. Product spectra of synthesised products were measured on
a Bruker “Avance 500” spectrometer for 1H-, 13C- and 31P-NMR spectra.
UV/Vis spectra were measured with a “Specord50plus” photometer, from
Jena Analytik. For time dependent conversion measurements, the kinetic
program of “ASpectUV 1.2.3” software was used in combination with a flow
through cuvette, from HellmaAnalytics, with 6.2 µL flow cell volume. The
ESI-MS measurements were done with a “Synapt-G2 HDMS” mass
spectrometer, from Waters. The spectrometer is used in combination with
a time-of-flight analyser.
The successful use of hydrogel dots as carriers for
organocatalysts within a microfluidic reactor was described.
Initially, the height of the hydrogel dots was adjusted with respect
to catalyst loading. Therefore, the proper conditions for the initial
photopolymerization were found. The conditions were used to
adjust the height in the dry state because the height in the swollen
state was estimated by relation from the literature. The
immobilized catalyst was used for a Knoevenagel reaction. Here,
the flow parameters for this type of reaction were modified to
optimize the reactor performance. To analyze the reactor
performance two different methods were established. It was
shown that an online detection (UV/Vis) and offline method (NMR)
could be used comparably. Additionally, this method was
observed to be robust against external influences, such as
differences during the assembly or operation process. By using
the NMR detection method, influences on the reactor conversion
were investigated by changing generic parameters, such as flow
rate, concentration and catalyst loading. With the optimized
parameters, the MFR was successfully used for screening
different reactants. Here, the steric claim of the substituents and
the reactant reactivity (MDN and ECAc) were changed. The
screening of different reactants was compared with a reference
flow reaction (without catalysts) and with a comparable batch
reaction. Here, it was seen that the benefits of flow reactors do
not clearly occur for each example of highly reactive reactants.
However, for the less reactive reactions, the MFR was several
times more effective than a comparable batch reaction. By using
ECAc as the reactant, the E/Z ratio was also investigated. Here,
it was observed that the results were in accordance with those in
the literature. The described system can be used for long-term
experiments (over 140 h). The conversion-time diagram shows a
plateau of conversion, which indicates that an increase in the
MFR time can be expanded without any restrictions.
Initiator synthesis: The initiator synthesis was done according to Majima
et al..[1] Under argon atmosphere dimethylphenylphosphonite (1.6 mL,
1.56 g, 11.3 mmol) was put inside a 50 mL round bottom flask. 2,4,6-
trimethyl-benzoylchloride (1.5 mL, 1.65 g, 9.0 mmol) was added dropwise
by a dropping funnel. After complete addition the reaction was stirred 24 h
in darkness at room temperature. Then potassium bromide (3.2 g),
dissolved in 2-butanone (50 mL), were
added. The mixture was heated to 50 °C
for 10 min, and then stirred for further 4 h
in darkness at room temperature. The
formed solid was filtered by Büchner
funnel and washed two times with 2-
butanone (100 mL). The final product was
isolated with 66 % yield.
1H-NMR in D2O (500 MHz): δ (ppm) = 2.01 (6H; 2-CH3; s); 2.16 (3H; 4-
CH3; s); 6,80 (2H; 3; s); 7.42 (2H; 3’; dt (3JHH = 7.7 Hz, 3JHH = 3.1 Hz)); 7.49
(1H; 4’; t (3JHH = 7.8 Hz)); 7.73 (2H; 2’; dd (3JHH = 7.3 Hz; 3JHH = 11.4 Hz));
13C-NMR in D2O (125 MHz): δ (ppm) = 18.7 (2-CH3); 20.3 (4-CH3); 128.1
(CH-4’); 128.4/128.5 (CH-3’); 132.1 (CH-3); 132.3/132.4 (CH-2’); 132.5 (C-
1’); 133.5 (C-4); 133.7 (C-1); 138.0/138.3 (C-2); 238.8 (1-CO); 31P-NMR in
D2O (202 MHz): δ (ppm) = 12.31 (s); ESI-MS; m/z (relative intensity):
Calculated for [C16O3PH16Li+H]+ 295.1070 found 295.1055 (75 %);
Calculated for [C16O3PH16Na+H]+ 311.0808 found 311.0795 (100 %);
Calculated for [C16O3PH16Li+Na]+ 317.0889 found 317.0880 (25 %)
Experimental Section
General procedure for reaction under batch conditions: 2 mL of
aldehyde (2 M) in isopropanol:DMSO (v:v = 1:1) stock solution was mixed
with desired ration (0.25 or 5 mol%) of catalyst. Afterwards, 2 mL of
malononitrile (2 M) or ethyl cyanoacetate (2 M) in isopropanol:DMSO (v:v
Chemicals: The monomers N,N-dimethylacrylamide (> 99 %) (DMAAm)
and N-[3-(dimethylamino)-propyl]acrylamide (98 %) (3DMAPAAm) were
purchased from Tokyo Chemical Industries (TCI). The cross-linker N,N-
bismethylacrylamide (99+ %) were purchased from Acros Organics. The
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000978
European Journal of Organic Chemistry
RESEARCH ARTICLE
Determination of degree of swelling: The degree of swelling was
determined by measuring length and diameter of prepared polymer
samples. Therefore, 0.5 mL the monomer mixture in desired ration (see
chapter 1.5 – Photopolymerization) was inserted into 1 mL plastic-pipette.
The solution was irradiated with UV-ligth for 9.4 sec with an intensity of
40 % (compare chapter 1.5 – Photopolymerization). Afterwards, the
plastic-pipette was cut into pieces to extract the cylindric polymer gel. For
1 day a lixiviation with iso-propanol was done, before the sample was cut
into smaller pieces. At the end the samples were dried at 50 °C for several
day until a constant mass was observed. With light microscope the length
and diameter of the samples were measured by a microscopic ruler.
Following, the samples were swollen in the desired solvent overnight. In
the end the sample were also measured with microscopic ruler.
= 1:1) stock solution were added. The mixture was stirred at room
temperature for 8 h. For high active malononitrile each hour 100 µl sample
for NMR measurement was taken, while reaction with ECAc a sample after
8 h was taken. For purification the reaction mixture was mixed with 10 mL
water. The suspension was stirred with 70 ml diethyl ether for at least 4 h.
The organic phase was isolated by filtration through a phase separation
filter (Macherey-Nagel “MN 617 WA”). The solvent was evaporated, and
the product was dried in vacuo.
Surface modification: The polymers were bound to glass slides (7.6 cm
x 2.6 cm), purchased from Carl Roth. The microscope slides were washed
with isopropanol under supersonic treatment at 70 °C for 10 min.
Afterwards, also under supersonic treatment at 70 °C, the microscopy
slides were cleaned with water and at last with ethanol. After drying in
nitrogen flow these substrates were modified by gas deposition. 20 of the
dried slides were placed in a desiccator with 300 µL of 3-(trichlorosilyl)-
propylmethacrylate with 50 mbar vacuum for at least 2 h. A successful
modification could be proofed by the hydrophobic effect of glass surface.
With the volume in dry and swollen state the degree of swelling was
calculated after following equation (1):
ꢀ = V
equation (1)
ꢁꢂꢃꢄꢄꢅꢆ
ꢇ
ꢈꢉꢊ
Photopolymerization: The photopolymerization is carried out via an
“Omnicure® S1500” UV-Lamp, from Lumen Dynamics. UV irradiation for
photopolymerization was done with an intensity of 428 mW (intensity =
20 %) at the end of the lighting cable, with 8 cm distance of light source to
substrate. 15 mmol of monomers was mixed in the desired ratios and were
dissolved in 1.7 mL deionized water with an initiator concentration of
17.5 mg/mL. 195 µL of monomer solution were put into an incubation
chamber gasket, from ThermoFisher. The incubation chamber gasket was
covered by a modified microscopy slide and the photopolymerization mask
(diamond shaped position of 158 dots) on the back side. This sandwich
was centered below UV light for 9.4 s. Afterwards the incubation chamber
was removed, and the polymer structures were washed in isopropanol for
18 h. The substrate with polymer dots were dried in air and stored at room
temperature.
d
Degree of swelling [-]
Volume of swollen gel [mm³]
Volume of dry gel [mm³]
Vswollen
Vdry
Acknowledgements
We acknowledge the Paderborn University Faculty of Engineering
for providing the confocal microscope.
Keywords: microfluidic reactor • hydrogel dots • immobilized
organocatalysts • organocatalytic reaction • green chemistry
Reactor set up: The flow system was set up according to the following
schematic illustration (figure S1). The flow rate was generated by a
“Legato 200” syringe pump, from KDScientific. As reservoirs two “Hamilton
1000 series” syringes were used. All tubes are PTFE capillary tubes with
iD = 0.2 mm, from Fisher Scientific. The reactor itself is a self-made
alumina holder (figure S2) to connect imprinted PDMS or PTFE layer with
the flow system. The reactor design is described by Simon et al..[2]
Polymer dots on microscopy slides were inserted into the reactor by
covering the polymer structure with an imprinted PDMS or PTFE layer. The
channel height on these imprinted covers were approximately 140 µm.
This sandwich was inserted into the alumina holder (figure S2) and fixed
by screws with torque of 8 cN/m. The PDMS or PTFE layer could be
connected to the alumina holder with special screw threads and drill holes
to capillary tubes.
References
[1]
[2]
[3]
F. Obst, D. Simon, P. J. Mehner, J. W. Neubauer, A. Beck, O. Stroyuk,
A. Richter, B. Voit, D. Appelhans, Beilstein J. Org. Chem. 2019, 31, 9684.
C. M. B. Ho, S. H. Ng, K. H. H. Li, Y.-J. Yoon, Lab on a chip 2015, 15,
3627.
D. J. Guckenberger, T. E. de Groot, A. M. D. Wan, D. J. Beebe, E. W. K.
Young, Lab on a chip 2015, 15, 2364.
[4]
[5]
G. Jas, A. Kirschning, Chemistry 2003, 9, 5708.
C. J. Welch, X. Gong, J. Cuff, S. Dolman, J. Nyrop, F. Lin, H. Rogers,
Org. Process Res. Dev. 2009, 13, 1022.
[6]
[7]
D. C. Fabry, E. Sugiono, M. Rueping, React. Chem. Eng. 2016, 1, 129.
a) K. Jähnisch, V. Hessel, H. Löwe, M. Baerns, Angew. Chem. 2004, 116,
410; b) X. Yao, Y. Zhang, L. Du, J. Liu, J. Yao, Renewable and
Sustainable Energy Rev. 2015, 47, 519; c) K. S. Elvira, X. Casadevall i
Solvas, R. C. R. Wootton, A. J. deMello, Nat. Chem. 2013, 5, 905; d) J.
Wegner, S. Ceylan, A. Kirschning, Chem. Commun. 2011, 47, 4583; e)
V. Hessel, H. Löwe, Chem. Eng. Technol. 2005, 28, 267; f) G. Veser,
Chem. Eng. Sci. 2001, 56, 1265.
Flow reaction: First the polymer dots were pre-swollen with
isopropanol:DMSO (v:v = 1:1) mixture for at least 2 h. Therefore, the
“Legato 200 Syringe” pump was loaded with two syringes (Braun 20 mL
syringe with iD = 20.1 mm), filled up with pure solvent mixture and the flow
rate were set to 4.0 µL/min. Afterwards the syringe pump was loaded with
reactant syringes (Hamilton 1000 series). Flow rate was set on 2.0 µL/min.
Finally, the conversion of resulting product was determined.
[8]
[9]
D. A. Foley, E. Bez, A. Codina, K. L. Colson, M. Fey, R. Krull, D. Piroli,
M. T. Zell, B. L. Marquez, Anal. Chem. 2014, 86, 12008.
J. P. McMullen, M. T. Stone, S. L. Buchwald, K. F. Jensen, Angew. Chem.
Int. Ed. 2010, 49, 7076.
Determination of conversion: The conversion was determined by NMR
spectroscopy or UV/Vis spectroscopy. NMR spectroscopy enables an
offline determination of conversion and a fast screening of different
reactions, while UV/Vis enables an online determination with very high
sensitivity during the whole reactor run. For NMR spectroscopy the outlet
flow was collected into a vial with 400 µL DMSO-d6 and was frozen
immediately after 1 h of sample collection with liquid nitrogen. The sample
was warmed up just before the tube is inserted into the NMR spectrometer
[10] J. S. Moore, K. F. Jensen, Org. Process Res. Dev. 2012, 16, 1409.
[11] B. P. Loren, M. Wleklinski, A. Koswara, K. Yammine, Y. Hu, Z. K. Nagy,
D. H. Thompson, R. G. Cooks, Chem. Sci. 2017, 50, 10083.
[12] D. Simon, F. Obst, S. Haefner, T. Heroldt, M. Peiter, F. Simon, A. Richter,
B. Voit, D. Appelhans, React. Chem. Eng. 2019, 4, 67.
[13] a) M. O'Brien, I. R. Baxendale, S. V. Ley, Org. Lett. 2010, 12, 1596; b) C.
B. McPake, C. B. Murray, G. Sandford, Tetrahedron Lett. 2009, 50, 1674.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000978
European Journal of Organic Chemistry
RESEARCH ARTICLE
[14] J. H. Clark, Green Chem. 1999, 1, 1.
[15] P. T. Anastas, J. C. Warner, Green chemistry. Theory and practice,
Oxford Univ. Press, Oxford, 2000.
[16] D. Kühbeck, G. Saidulu, K. R. Reddy, D. D. Díaz, Green Chem. 2012,
14, 378.
[17] H. A. Zayas, A. Lu, D. Valade, F. Amir, Z. Jia, R. K. O’Reilly, M. J.
Monteiro, ACS Macro. Lett. 2013, 2, 327.
[18] P. Llanes, C. Rodriguez-Escrich, S. Sayalero, M. A. Pericas, Org. Lett.
2016, 18, 6292.
[19] R. W. Coughlin, M. Aizawa, B. F. Alexander, M. Charles, Biotechnol.
Bioeng. 1975, 17, 515.
[20] S. M. Lai, C. P. Ng, R. Martin-Aranda, K. L. Yeung, Microporous
Mesoporous Mat. 2003, 66, 239.
[21] P. Watts, C. Wiles, Chem. Eng. Technol. 2007, 30, 329.
[22] C. Wiles, P. Watts, S. J. Haswell, Tetrahedron 2004, 60, 8421.
[23] G. Zhang, T. Zhang, X. Zhang, K. L. Yeung, Catal. Commun. 2015, 68,
93.
[24] X. Zhang, E. S. Man Lai, R. Martin-Aranda, K. L. Yeung, Applied
Catalysis A: General 2004, 261, 109.
[25] S. B. Ötvös, I. M. Mándity, F. Fülöp, J. Catal. 2012, 295, 179.
[26] X. C. Cambeiro, R. Martín-Rapún, P. O. Miranda, S. Sayalero, E. Alza,
P. Llanes, M. A. Pericàs, Beilstein J. Org. Chem. 2011, 7, 1486.
[27] M. Planchestainer, M. L. Contente, J. Cassidy, F. Molinari, L. Tamborini,
F. Paradisi, Green Chem. 2017, 19, 372.
[28] a) M. Benaglia, A. Puglisi, F. Cozzi, Chem. Rev. 2003, 103, 3401; b) J.
Lu, P. H. Toy, Chem. Rev. 2009, 109, 815.
[29] H. Ishitani, Y. Saito, Y. Nakamura, W.-J. Yoo, S. Kobayashi, Asian J. Org.
Chem. 2018, 7, 2061.
[30] D.-Z. Xu, S. Shi, Y. Wang, RSC Adv. 2013, 3, 23075.
[31] K. Haraguchi, H.-J. Li, Macromolecules 2006, 39, 1898.
[32] a) X. Yu, A. Herberg, D. Kuckling, Eur. Polym. J. 2019, 120, 109207; b)
P. Riente, J. Yadav, M. A. Pericàs, Org. Lett. 2012, 14, 3668; c) T. E.
Kristensen, T. Hansen, Eur. J. Org. Chem. 2010, 2010, 3179.
[33] P. J. Mehner, F. Obst, D. Simon, J. Tang, A. Beck, D. Gruner, M. Busek,
D. Appelhans, U. Marschner, B. Voit et al. in Volume 1: Development and
Characterization.
[34] R. Toomey, D. Freidank, J. Rühe, Macromol. 2004, 37, 882.
[35] J. Chen, H. Park, K. Park, J. Biomed. Mater. Res. 1999, 44, 53.
[36] Y. Luan, Y. Qi, H. Gao, R. S. Andriamitantsoa, N. Zheng, G. Wang, J.
Mater. Chem. A 2015, 3, 17320.
[37] K. Isobe, T. Hoshi, T. Suzuki, H. Hagiwara, Mol. Divers. 2005, 9, 317.
[38] G. Li, J. Xiao, W. Zhang, Green Chem. 2011, 13, 1828.
This article is protected by copyright. All rights reserved.

10.1002/ejoc.202000978
European Journal of Organic Chemistry
RESEARCH ARTICLE
Layout 1:
RESEARCH ARTICLE
Gels in flow microflow reactor:
Novel application for polymer
networks as carrier for
organocatalysts inside microflow
reactors combines benefits of flow
reactors with supported catalysts.
Patrik Berg, Franziska Obst, David
Simon, Prof. Dr. Andreas Richter, Dr.
Dietmar Appelhans, Prof. Dr. Dirk
Kuckling*
Page No. – Page No.
Novel application of polymer
networks carrying tertiary amines as
catalyst inside microfluidic flow
reactors used for Knoevenagel
reactions
This article is protected by copyright. All rights reserved.