Solid-supported continuous flow synthesis in microreactors using electroosmotic flow

Article abstract of DOI:10.1021/op049857x

This paper reports the fabrication of a microreactor suitable for use with supported reagents. We demonstrate that the electroosmotic flow can be used to move the reagents over a solid-supported catalyst bed. It is demonstrated that it is important that the support should not swell in organic solvents to obtain reproducible flow, and it is shown that silica supports fulfill this criteria. Silica-functionalised piperazine is used in a variety of Knoevenagel reactions to give the product in high conversion.

Full text of DOI:10.1021/op049857x

Organic Process Research & Development 2004, 8, 942−944
Solid-Supported Continuous Flow Synthesis in Microreactors Using
Electroosmotic Flow
Nikzad Nikbin and Paul Watts*
Department of Chemistry, Faculty of Science and the EnVironment, The UniVersity of Hull, Cottingham Road,
Hull, HU6 7RX, UK
Abstract:
supported reagents or catalysts, as the challenge for some
time has been to develop microreactors suitable for use with
supported reagents. Skelton reported a device for the Suzuki
reaction.8 However, as a Pd catalyst was used, it was very
easy to position the catalyst within the device before the top
block was thermally annealed; in fact the high temperature
used to effect the bonding of the device probably further
activated the catalyst. Similarly, McCreedy9 has reported a
reactor that effected the dehydration of alcohols using a
sulphated zirconia catalyst. As a result of using alcoholic
solvent systems, it was possible to use a microreactor
fabricated from a PDMS top block, onto which the catalyst
had been impregnated; however this type of approach would
not be feasible when organic solvents are required.
Since the pioneering work of Merrifield,10 solid supported
synthesis has been an important technique in organic
chemistry. In particular, the use of solid supported reagents
in solution phase organic synthesis has been of enormous
importance.11 The driving force behind this research has been
the rapid expansion in high-throughput parallel synthesis and
the ever increasing need to simplify workup and reaction
procedures. Among different solid supported reagents that
have been used, supported catalysts are particularly conve-
nient as excess immobilized catalyst can be used to drive
reactions to completion. More recently it has been reported
that solid supports suffer less physical damage in flow
systems compared to batch reactions where vigorous stirring
is required.12
This paper reports the fabrication of a microreactor suitable
for use with supported reagents. We demonstrate that the
electroosmotic flow can be used to move the reagents over a
solid-supported catalyst bed. It is demonstrated that it is
important that the support should not swell in organic solvents
to obtain reproducible flow, and it is shown that silica supports
fulfill this criteria. Silica-functionalised piperazine is used in a
variety of Knoevenagel reactions to give the product in high
conversion.
Introduction
Over the past five years there has been a rapid growth in
the development of microreactor technology exploiting the
technique of electroosmotic flow (EOF).1 The application
of electroosmotic flow instead of hydrodynamic pressure
overcomes two most commonly faced problems in continu-
ous flow reactors utilising solid-supported synthesis. First,
technical problems such as bypassing of the reagents or
pressure drop are avoided, as unlike pressure-driven systems
EOF is uniform along the channel. A practical consequence
of this is that very small particles can be used in the system,
as EOF unlike pressure-driven systems is independent of
particle size. This can result in greater surface-to-volume
ratios and increases the number of reactive sites on the solid
support. Second, very low flow rates can be achieved which
make the synthesis procedure much more efficient. Such low
flow rates cannot be reproducibly achieved within a pressure-
driven device when using a syringe pump. Recent research
has shown that a vast number of solution phase reactions
such as diazo synthesis,2 Michael additions,3 aldol condensa-
tions,4 heterocyclic synthesis,5 and multistep peptide syn-
thesis6 may be performed within microreactors using this
technique; these and many other examples have been
discussed in several reviews.7
Here we report a novel method to perform efficient and
reliable solid supported synthesis in a continuous flow
microreactor. This technique combines the advantages of
continuous flow synthesis and microreactors. In this com-
munication we report an example of using electroosmotic
flow in a microreactor to perform Knoevenagel reactions.
Experimental Section
In comparison, very few publications have reported
solution phase organic synthesis in microreactors using solid-
The microreactor used for solid supported reactions was
fabricated by Micro Chemical Systems from borosilicate
glass using standard fabrication methods developed at Hull.13
The device was fabricated from a top block and two etched
(1) Fletcher, P. D. I.; Haswell, S. J.; Paunov, V. N. Analyst 1999, 124, 1273.
(2) Salimi-Moosavi, H.; Tang, T.; Harrison, D. J. J. Am. Chem. Soc. 1997,
119, 8716.
(3) Wiles, C.; Watts, P.; Haswell, S. J.; Pombo-Villar, E. Lab Chip 2002, 2,
62.
(8) Greenway, G. M.; Haswell, S. J.; Morgan, D. O.; Skelton, V.; Styring, P.
Sens. Actuators, B 2000, 63, 153.
(4) Wiles, C.; Watts, P.; Haswell, S. J.; Pombo-Villar, E. Lab Chip 2001, 1,
100.
(5) Wiles, C.; Watts, P.; Haswell, S. J.; Pombo-Villar, E. Org. Process Res.
DeV. 2004, 8, 28.
(6) Watts, P.; Wiles, C.; Haswell, S. J.; Pombo-Villar, E. Tetrahedron 2002,
58, 5427.
(9) Wilson, N. G.; McCreedy, T. Chem. Commun. 2000, 733.
(10) Merrifield, R. B. J. Am. Chem. Soc. 1963, 85, 2149.
(11) Ley, S. V.; Baxendale, I. R.; Bream, R. N.; Jackson, P. S.; Leach, A. G.;
Longbottom, D. A.; Nesi, M.; Scott, J. S.; Storer, R. I.; Taylor, S. J. J.
Chem. Soc., Perkin Trans. 1 2000, 3815.
(7) Fletcher, P. D. I.; Haswell, S. J.; Pombo-Villar, E.; Warrington, B. H.;
Watts, P.; Wong, S. Y. F.; Zhang, X. Tetrahedron 2002, 58, 4735.
(12) Hodge, P. Curr. Opin. Chem. Biol. 2003, 7, 1.
(13) McCreedy, T. Anal. Chim. Acta 2001, 427, 39.
942
•
Vol. 8, No. 6, 2004 / Organic Process Research & Development
10.1021/op049857x CCC: $27.50 © 2004 American Chemical Society
Published on Web 10/29/2004

Scheme 1. Microreactor manifold for Knoevenagel
reactions
Scheme 2. Immobilization of piperazine
Figure 1. Schematic of the microreactor.
Table 1. Effect of field strength and concentration on
conversion to 3
plates, which enabled a deeper catalyst bed to be achieved
(Figure 1). The channels were 130 µm wide and 50 µm deep,
and the catalyst bed was 800 µm wide, 100 µm deep, and
10 mm in length. An in-house LabVIEW program was used
to set and monitor the voltages that were applied to platinum
electrodes that were placed in the reagent reservoirs (power
supply was built by Kingfield Electronics, Sheffield, UK).
All microreactions were performed over a period of 5 min
at room temperature to ensure that sufficient volume of
product was generated for analysis. Reaction products were
determined by GC-MS (Varian GC (CP-3800) coupled to
a Varian MS (Saturn 2000), 30 m CP-Sil 8 column
(Phenomenex), injector temperature 200 °C, helium flow rate
1 mL min-1, oven temperature 60 °C for 1 min then ramped
to 270 °C at 25 °C min-1). Reaction products were
determined via the comparison of retention times and mass
spectra with those obtained from a series of synthetic
standards (see below). Analysis of the crude reaction
mixtures by GC-MS enabled the proportion of product to
be determined with respect to residual starting material.
Synthesis of Supported Piperazine. 4-Benzyl chloride
functionalized silica gel (1.00 g, 1.30 mmol) and piperazine
(0.67 g, 7.80 mmol) were heated to reflux in the presence
of potassium carbonate (0.54 g, 3.90 mmol) in acetone (25
mL) for 5 h. The product was filtered under vacuum and
washed with water and acetone (2 × 25 mL) to give a silica-
bound piperazine. Elemental analysis showed 0.7 mmol/g
presence of piperazine moiety.
voltage (V)
concn (M)
200
300
400
0.5
1.0
2.0
48 ( 3
39 ( 3
92 ( 2
25 ( 2
52 ( 6
90 ( 5
23 ( 1
43 ( 9
90 ( 5
Results and Discussion
In the first instance, the microreactor illustrated in Figure
1 was packed with piperazine on Merrifield resin, Tentagel,
or Argopore, and the microreactor was primed with aceto-
nitrile. Using the reaction manifold illustrated in Scheme 1,
the preparation of ester 3 was investigated within the
microreactor; a premixed solution of ethyl propiolate 1 and
4-bromobenzaldehyde 2 (40 µL, 1.0 M) in anhydrous MeCN
was placed in reservoir A, and the reaction products were
collected in anhydrous MeCN in reservoir B.
However it was found that the polymer became more and
more swollen with time causing irreproducible flow rates,
which resulted in variable conversions, which was unsatis-
factory. To circumvent this problem, it was proposed that a
support that does not swell in organic solvents is required
when using microfluidic systems, and consequently pipera-
zine was immobilised onto silica, as this material has a broad
range of solvent compatibility and suffers minimal swelling
in a range of organic solvents. Piperazine was immobilised
onto benzyl chloride functionalised silica gel 4 to give
product 5 containing 0.7 mmol/g piperazine by CHN analysis
(Scheme 2).
The catalyst bed was subsequently packed with piperazine
functionalized silica 5 and primed with acetonitrile to remove
air bubbles from the device. The reaction of ethyl cyano-
acetate 1 and 4-bromobenzaldehyde 2 in acetonitrile to
produce ethyl 3-(4-bromophenyl)-2-cyanoacrylate 3 was then
investigated using the methodology described above.
The reaction was investigated at a range of concentrations
and field strengths (i.e., flow rates) to optimise the conditions.
Each reaction was repeated 5 times, and the average
conversion for each set of reactions is shown in Table 1.
Table 1 clearly shows that by increasing the concentra-
tions of the starting materials the conversions correspondingly
increase. This increase in conversion was expected as
reagents have a greater chance to interact with the im-
Ethyl 3-(4-bromophenyl)-2-cyano acrylate 3: m/z (EI)
281 (M+ + 1, 90%), 280 (45), 279 (100), 251 (25), 200
(20), 154 (10), 127 (25), 100 (20) and 76 (20); GC-MS
retention time RT ) 9.5 min.
Ethyl 2-cyano-3-phenyl acrylate 6: m/z (EI) 202 (M+
+ 1, 70%), 201 (100), 172 (80), 156 (90), 128 (75), 102
(55), 77 (50), and 51 (50); GC-MS retention time RT ) 8.3
min.
Ethyl 3-(3,5-dimethoxyphenyl)-2-cyano acrylate 7: m/z
(EI) 262 (M+ + 1, 20%), 261 (100), 189 (55), 161 (25), and
77 (10); GC-MS retention time RT ) 10.2 min.
Ethyl 3-(4-benzyloxyphenyl)-2-cyano acrylate 8: m/z
(EI) 308 (M+ + 1, 5%), 307 (20), 91 (100), and 65 (20);
GC-MS retention time RT ) 15.1 min.
Vol. 8, No. 6, 2004 / Organic Process Research & Development
•
943

Table 2. Knoevenagel reactions using other aldehyde
derivatives
reactions using ethyl propiolate 1 were conducted at 2.0 M
concentration using other aldehydes. The results are sum-
merized in Table 2. The lower conversion for the 4-benzyl-
oxybenzaldehyde derivative can be attributed to its lower
solubility which makes it difficult to work with a 2.0 M
solution of this compound.
Conclusion
It has been shown for the first time that EOF can be used
to perform continuous flow, solid supported synthesis in a
microreactor. Using EOF to move reagents over a solid phase
can be very advantageous in flow systems, as very low flow
rates are achievable. In comparison, results of pressure driven
systems (not shown here) were found to be highly irrepro-
ducible. A practical consequence of this is that very small
particles can be used in the system, as EOF unlike pressure-
driven systems is independent of particle size. The small
particle size can result in a greater surface-to-volume ratio
thus increasing the number of reactive sites on the solid
support. The reaction also seems to happen much quicker
and at lower temperatures when compared to batch reactions
(5 min vs 2 h). We propose that this methodology signifi-
cantly increases the versatility of microreactor synthesis.
mobilized base. Furthermore, as expected, as the field
strength is reduced the conversion to product increases. We
propose that the reduced field strength results in a reduced
flow rate, which increases the residence time of the reaction
resulting in a higher conversion.
The results clearly show that the microreactor is able to
generate reliable and reproducible data, as the level of error
is within an acceptable range. The data show that greater
conversions can be achieved by increasing the concentration.
As would be predicted, increasing the time of the reaction
to up to 20 min had no effect on the conversions which
illustrates that the system could be used in continuous flow
mode in a “scaled out” device to produce larger quantities
of product.
Acknowledgment
Full financial support provided by the EPSRC (N.N.) is
gratefully acknowledged. We thank Dr. Charlotte Wiles and
Dr. Geir Flaten for assistance.
Received for review August 3, 2004.
OP049857X
Evaluation of Analogous Reactions. After optimizing
the condition of the reaction, a series of other Knoevenagel
944
•
Vol. 8, No. 6, 2004 / Organic Process Research & Development