Functionalized inorganic monolithic microreactors for high productivity in fine chemicals catalytic synthesis

Full text of DOI:10.1002/anie.200805580

Angewandte
Chemie
DOI: 10.1002/anie.200805580
Microreactors
Functionalized Inorganic Monolithic Microreactors for High
Productivity in Fine Chemicals Catalytic Synthesis**
Abdelkrim El Kadib, Ricardo Chimenton, Alexander Sachse, Franꢀois Fajula, Anne Galarneau,
and Bernard Coq*
Continuous-flow catalytic microreactors offer safe, eco-
friendly, and intensified processes for the long-term produc-
tion of fine chemicals.[1] In many cases, flexible production
with higher yields in much shorter time can be achieved than
in batch reactions. One can identify three kinds of flow
microreactors, namely parallel open microchannels, packed
beds, and monoliths. The use of monolithic structured
catalysts overcomes some physico-chemical drawbacks, such
as fluid dynamics, pressure drop, heat and mass transfer, low
contacting efficiency, and mechanical attrition, associated
with conventional packed-bed reactors.[2] However, the
monolithic structured catalysts will not solve any issue related
to chemical deactivation of the catalyst. Monolith reactors are
composed of 1–5 mm parallel channels or foams with
interconnected 0.1–1.0 mm flow-through pores, and are
mainly used in gas phase reactions or catalytic combustion.[3]
For the synthesis of fine chemicals, there is a need for
smaller convection pores and skeleton thickness (struts)
ranging from 5 to 50 mm to process bulky molecules efficiently
with fast diffusion to and from the active sites within the
mesopores. The required specifications for these materials
would be: 1) monolith diameter of about 0.5–10 mm to
perform nano- and microsynthesis; 2) narrow size distribution
of flow-through and diffusion pores, which should guarantee a
flat flow profile and a uniform residence time for reactants
through the monolith and is an improvement over randomly
packed-bed microreactors in which uncontrolled fluid dynam-
ics may generate stagnation and broader residence time
distribution; 3) high thermal (200–3008C) and mechanical
stability; and 4) easy generation of a variety of catalytic active
sites at the surface.
these polymeric materials may cause pressure drops along the
microchannel resulting in variations in the residence time for
pressure-driven microreactors.[1c,5] Associated drawbacks are
the typically limited mechanical resistance, lower thermal
stability, chemical inertness, and non-uniformity of radial
permeability. Furthermore, a complex synthesis of the
functionalized active polymers is required, limiting the ease
of tenability. To circumvent, at least in part, some of these
disadvantages, composite materials consisting of inorganic
carrier materials with incorporated polymers were design-
ed.[1h] A highly promising alternative to these materials are
the inorganic carriers. Such materials have recently received a
great deal of interest, with flow-through pores at the micro-
meter scale and a second network of homogeneously
distributed mesopores within the struts.[6] These materials
were essentially designed for HPLC applications with col-
umns featuring flow-through pores smaller than 2 mm.
Although heterogeneous catalysis extensively uses silica-
based materials, to the best of our knowledge, these materials
have not yet been evaluated in catalytic reactions in flow
conditions. The versatility of inorganic silica supports is the
availability of surface silanols that are prone to react allowing
the grafting of a wide variety of active organic functional
groups (e.g. acidic, basic, and chelating ligands). Functional-
ized ordered mesoporous silica have received considerable
interest for broad applications including catalysis.[7] The
knowledge gained in this field appears very promising for
extrapolation of the design concept of functionalized silica to
their monolithic version for catalysis, which has to date not
been reported.
Herein two main innovations are proposed. 1) The on-line
grafting of catalytic functions on monolith silica matrices by a
continuous cyclic process; 2) the first examples of catalytic
reactions carried on shaped and functionalized inorganic
monolith microreactors featuring flow-through pores of 9 mm
and diffusion pores of 13 nm. These materials demonstrate
unique hydrodynamic behavior in liquid-phase reaction
processes.
Monolith samples, hereafter called MonoSil (6.3 mm
diameter, ca. 0.08 g silica per centimeter length; Figure 1a),
featuring reproducible textural properties were synthesized
according to a previous report.[6e] They exhibited the follow-
ing characteristics:[8] average cavities of 9 mm connected
through 4.3 mm windows (mercury porosimetry); skeleton
thicknesses of 3 mm (SEM, Figure 1d,e); diffusive mesopores
of about 13 nm (nitrogen sorption); volume of flow-through
and diffusion pores of circa 1.95 mLgÀ1 (mercury porosim-
etry), and 1.25 mLgÀ1 (nitrogen sorption), respectively. The
total porosity and the surface area of the monoliths are 0.86
Polymer-based monolithic supports first received great
attention with regards to catalytic applications in fine-
chemicals synthesis.[1h,4] However, the possible swelling of
[*] Dr. A. El Kadib, Dr. R. Chimenton, A. Sachse, Dr. F. Fajula,
Dr. A. Galarneau, Dr. B. Coq
Institut Charles Gerhardt Montpellier—UMR 5253 CNRS/ENSCM/
UM2/UM1, 8, rue de l’Ecole Normale
34296 Montpellier cedex 5 (France)
E-mail: bernard.coq@enscm.fr
[**] We thank Christine Biolley, Daniel Brunel, Jerome Babin for
experimental support, and Patrice Creux for 3D X-ray tomography
examination. This work was supported by an ANR grant Institut
Carnot “Chimie, Environnement et Dꢀveloppement Durable”
(07 CARN 00301), and by 7PCRD EU funds from the Marie-Curie
initial Training Network NANO-HOST (grant agreement ITN
215193-2).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200805580.
Angew. Chem. Int. Ed. 2009, 48, 4969 –4972
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4969

Communications
Scheme 1. The Knovenagel reaction between benzaldehyde and ethyl
cyanoacetate.
Scheme 2. Transesterification of triacetine by methanol.
Figure 1. a) MonoSil as prepared; b) MonoSil microreactor; c) X-rays
tomography of a cross section; SEM images of d) flow-though pores
and e) struts.
noethyl acetate (Scheme 1); 2) the acid transesterification of
triacetine with methanol (Scheme 2). These model reactions
may be poisoned by their co-products, water (Scheme 1) and
glycerol (Scheme 2). The former constitutes an efficient
[9]
À
strategy for the creation of C C bond. The latter is
and 650 m2 gÀ1, respectively. The open-cell structure of
MonoSil ressembles that of ceramic foams,[3b] but 100 times
smaller. The MonoSil exhibits a high mechanical strength as
shown by mercury porosimetry.[8] Two striking features of
MonoSil are:
1) A uniform radial distribution of voids and struts, as shown
by X-ray tomography (Figure 1c), and a very narrow size
distribution of windows and diffusion pores.[8] These
characteristics guarantee an extremely flat flow profile
and the same residence time for all reactant molecules.
The flow resistance through the MonoSil can be compared
to that of a microreactor packed with irregularly shaped
particles of about 50 mm.[8]
considered as a model for the transesterification of triacyl-
glycerol. MonoSils proved very active and selective (> 99%)
in these two kinds of reactions. In a continuous process, the
long-term stability of the catalyst activity is a key point. For
the two reactions, MonoSils did not show any deactivation
over several hours, as examplified for the Knovenagel
reaction using NH2-MonoSil, which maintained 84% cya-
noethyl acetate conversion for at least 28 h on stream
(Figure 2), exhibiting a TON (turnover number is the
2) A very short diffusion path of 3 mm in the struts.
Compared to a microreactor packed with 50 mm particles
(same flow resistance), the typical diffusion time scale t is
then 200 times smaller in MonoSil (t = l2/2D, l is the
diffusion length and D is the diffusion coefficient). This
smaller time scale is an outstanding characteristic of the
microreactor that processes bulky molecules in fine
chemical synthesis.
The targeted catalysts, basic NH2-MonoSil and acid-
ic HSO3-MonoSil, were prepared by grafting aminopropyl-
Figure 2. Conversion efficiency (Ceff) as a function of time and flow
triethoxysilane
or
2-(4-chlorosulfonylphenyl)ethyltri-
~
~
*
*
conditions in monolith ( , ) and packed-bed ( , ), for the
methoxysilane through a one-step cyclic flow process, without
the addition of a catalyst to avoid self-polymerization of the
species to be grafted (see the Experimental Section). Nitro-
gen sorption isotherms of hybrid monoliths exhibit the same
profile as those of parent MonoSil, with a decrease in surface
area, total pore volume, and pore diameter as expected for
grafted silica.[8] 29Si CP MAS displays T-type resonances,
demonstrating that the organosilanes are covalently bonded
to the MonoSil matrix.[8] To the best of our knowledge, this is
the first example of catalytic activity in post-grafted silica
performed under flow conditions. The amount of grafted
species was found to be 1.41 mmolgÀ1 for NH2-MonoSil and
0.95 mmolgÀ1 for HSO3-MonoSil. The behavior of the
catalysts was evaluated in two reactions: 1) The base-cata-
lyzed Knovenagel reaction between benzaldehyde and cya-
~
~
*
*
Knovenagel ( , ) and transesterification ( , ) reaction.
number of reactant moles converted per mole of catalyst) of
2280. This is an outstanding performance, as silica may suffer
some textural and/or structural alterations from interaction
with water.[10] The stability of the catalytic properties of NH2-
MonoSil is most likely due to the continuous exclusion of the
formed water from the monolith. Fine chemicals syntheses
using heterogeneous catalysts are commonly carried out in
batch reactors and more recently in packed-bed microreac-
tors.[9c] Therefore, the performances of the MonoSil catalysts
(1.9 cm length, 0.160 g) were compared with a closed stirred-
tank reactor (batch) and a packed-bed microreactor (4.2 mm
4970 www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 4969 –4972

Angewandte
Chemie
inner diameter), both containing 0.160 g of a sieved fraction
(50–100 mm) of crushed MonoSil. Batch and flow experiments
have been compared in terms of conversion and efficiency
(Figure 2 and Figure 3).
the distribution of flow-through pores in the 5–50 mm range
and diffusion pores (ordered or disordered) within 3 to 10 nm
in size and strut thicknesses of 1–10 mm. These materials can
be made catalytically active using the grafting or anchoring of
acid, base, and redox functionalities. A new area will open up
in the field of heterogeneous catalysis for the synthesis of fine
chemicals (and microsynthesis), where mass transfer of bulky
molecules is of utmost importance. In spite of their promising
potentials, the loss of the catalyst’s effectiveness and selec-
tivity owing to pore clogging arising from mass transfer and
site inhibition is the main limitation for the development of
porous solid catalysts (including zeolites and structured
mesoporous materials) for the synthesis of fine chemicals.[11]
The implementation of the MonoSil reactor may circumvent
these phenomena. Another potential advantage of the
MonoSil type reactor compared to a batch reactor is the
higher solid/liquid ratio in the reaction volume, which
minimizes the risk of homogeneous side reactions. Therefore,
a great future lies ahead for the use of nanostructured
inorganic monolith microreactors (silica, titania, zirconia, and
others) in the field of catalysis.
Figure 3. Productivity of NH2- and HSO3-MonoSil in batch (powder),
packed-bed (powder), and monolith reactors. The productivity in the
batch reactor was compared at a similar extent of conversion as in the
monolith (i.e. 84% for the reaction in Scheme 1 and 79% for the
reaction in Scheme 2).[8]
Experimental Section
To generate a monolith microreactor, the MonoSil rod (6.3 mm
diameter, 1.9 cm length) was inserted in a heat-shrinkable PTFE tube
together with two glass tubes (4 mm inner diameter) at each end,
It is clear in the flow mode that the MonoSil catalysts are
significantly more efficient than in batch mode for the two
reactions (Figure 3). HSO3-MonoSil was 18 times more
efficient in the monolith than in the batch reactor. NH2-
MonoSil was 13 times more effective in monolith than in the
batch mode. Several physical and chemical factors, alone or in
combination, allowed for a faster molecular flow to and from
the active sites and can explain the higher efficiency of the
MonoSil microreactors. These factors are:
1) A larger contact area between the reaction medium and
the catalyst, owing to the large surface area of the solid in
contact with the flow (convective mass transfer).
2) A much shorter diffusion path for the molecules through
the struts (3 mm) compared to that through grains, which
are ten times larger in size.[8] This is helpful for the
reactant molecules and for the product, which will be
scavenged out of the pore network faster.
3) The non-accumulation of co-products, which strongly
adsorb on active sites, in the reservoir containing the
reactants prevents the inhibition effect by water and
glycerol in case of the model reactions. In contrast, the
inhibition becomes more and more sensitive as the
conversion proceeds in the batch reactor. This point is
certainly of great importance.[8]
which was then placed within
a heat shrinkable PTFE tube
(Figure 1b). NH2-MonoSil and HSO3-MonoSil were prepared on
the monolith that had been activated overnight at 423 K using a
solution containing the precursor functionalities: aminopropyl tri-
ethoxysilane to yield NH2-MonoSil, and 2-(4-chlorosulfonylphenyl)-
ethyltrimethoxysilane to yield HSO3-MonoSil, both in 60 mL of
ethanol. Precursors were allowed to flow at a rate of 2.5 mLminÀ1
(343 K for 10 h). The amount of catalytically active precursors was
about 5 molecules per nm2 silica. NH2- and HSO3-MonoSil were then
washed several times with ethanol, methanol/water (50:50), and
acetone to recover the catalyst. The monoliths were successively dried
for 2 h at RT, 2 h at 323 K, and 8 h at 273 K. The materials were
characterized by chemical analyses, 3D X-ray tomography, SEM,
TGA, solid-state NMR, mercury porosimetry, and nitrogen sorption.
The test reactions were carried out in liquid phase and flow
conditions, feeding the MonoSil and packed-bed reactors with a
HPLC micropump (0.5 mLminÀ1, 0.43 mmsÀ1, pressure drop ca.
300 kPa). Reaction conditions for a) Knovenagel reaction: cya-
noethyl acetate (0.68 molLÀ1), benzaldehyde (0.80 molLÀ1), solvent
DMSO, RT; b) transesterification reaction: triacetine (0.70 molLÀ1),
triacetine/methanol = 1:6, temperature 333 K. The concentrations of
the reactants and products were determined by sampling periodically
on line by using GC. For the reaction in batch, a closed stirred-tank
reactor (100 mL, Autoclave Engineer, 1000 rpm) was used, fed with
the reactant solution (60 mL) at the same concentration as in the flow
experiments. The reaction in Scheme 1 was processed with 150 mg
catalyst, reaction in Scheme 2 with 780 mg catalyst. Under these
conditions, the conversion was not greatly affected by external and
internal mass transfer. Solution aliquots were periodically withdrawn
for GC analysis.
It is also noticeable that the MonoSil is more effective
than the packed-bed reactor. Non-accumulation of co-prod-
ucts is also observed in case of packed-bed reactor, and thus
the reasons for lower efficiency might be due to the first two
explanations.
Herein the potential of an inorganic monolith micro-
reactor with micrometer flow-through pores is demonstrated.
With the knowledge gained in the preparation of silica
monoliths,[6e] breakthroughs in catalysis can be contemplated
by using monoliths designed with an independent control of
Received: November 14, 2008
Published online: May 28, 2009
Keywords: heterogeneous catalysis · mesoporous materials ·
.
microreactors · silica monoliths
Angew. Chem. Int. Ed. 2009, 48, 4969 –4972
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4971

Communications
Non-Cryst. Solids 2006, 352, 4003 – 4007; h) T. Amatani, K.
Nakanishi, K. Hirao, T. Kodaira, Chem. Mater. 2005, 17, 2114 –
2119; i) Z.-G. Shi, Y.-Q. Feng, L. Xu, S.-L. Da, Y.-Y. Ren,
Microporous Mesoporous Mater. 2004, 68, 55 – 59; j) J. El Has-
kouri, et al., Chem. Commun. 2002, 330 – 331; k) S. A. El-Safty,
T. Hanaoka, F. Mizukami, Chem. Mater. 2005, 17, 3137 – 3145;
l) D. Brandhuber, V. Torma, C. Raab, H. Peterlik, A. Kulak, N.
Husing, Chem. Mater. 2005, 17, 4262 – 4271; m) P. Y. Feng, X. Bu,
G. D. Stucky, D. J. Pine, J. Am. Chem. Soc. 2000, 122, 994 – 995;
n) K. Kawakami, Y. Sera, S. Sakai, T. Ono, H. Ijima, Ind. Eng.
Chem. Res. 2005, 44, 236 – 240; o) S.-F. Kim, C.-K. Shin, C.-H.
Ahn, J. Porous Mater. 2006, 13, 201 – 205.
[1] a) M. A. Mason, K. E. Price, J. L. Steinbacher, A. R. Bogdan,
D. T. McQuade, Chem. Rev. 2007, 107, 2300 – 2318; b) P. Watts,
C. Wiles, Chem. Commun. 2007, 443 – 467; c) B. Ahmed-Omer,
J. C. Brandt, T. Wirth, Org. Biomol. Chem. 2007, 5, 733 – 740;
d) K. Geyer, J. D. C. Codee, P. H. Seeberger, Chem. Eur. J. 2006,
12, 8434 – 8442; e) M. Brivio, W. Verboom, D. N. Reinhoudt, Lap
Chip 2006, 6, 329 – 344; f) P. Watts, S. J. Haswell, Chem. Soc. Rev.
2005, 34, 235 – 246; g) K. Jꢀhnish, V. Hessel, H. Lowe, M. Bearns,
Angew. Chem. 2004, 116, 410 – 451; Angew. Chem. Int. Ed. 2004,
43, 406 – 446; h) A. Kirschning, C. Altwicker, G. Drager, J.
Harders, N. Hoffmann, U. Hoffmann, H. Shonfeld, W. Sodo-
lenko, U. Kunz, Angew. Chem. 2001, 113, 4118 – 4120; Angew.
Chem. Int. Ed. 2001, 40, 3995 – 3997.
[7] a) A. Corma, H. Garcia, Adv. Synth. Catal. 2006, 348, 1391 –
1412; b) J. H. Clark, D. J. Macquarie, S. J. Tavener, Dalton Trans.
2006, 4297 – 4309; c) J. A. Melero, R. van Grieken, G. Morales,
Chem. Rev. 2006, 106, 3790 – 3812; d) T. Mallat, A. Baiker,
Chem. Rev. 2004, 104, 3037 – 3058; e) Z. L. Lu, E. Lindner, H. A.
Mayer, Chem. Rev. 2002, 102, 3543 – 3578; f) A. P. Wight, M. E.
Davis, Chem. Rev. 2002, 102, 3589 – 3614; g) D. Brunel, A. C.
Blanc, A. Galarneau, F. Fajula, Catal. Today 2002, 73, 139 – 152;
h) K. K. Sharma, T. Asefa, Angew. Chem. 2007, 119, 2937 – 2940;
Angew. Chem. Int. Ed. 2007, 46, 2879 – 2882; i) E. Mathieu, A.
Walcarius, Talanta 2003, 59, 1173 – 1188; j) D. Brunel, Micro-
porous Mesoporous Mater. 1999, 27, 329 – 344.
[2] J. A. Moulijn, Ir. M. Makkee, A. E van Dieper, Chemical
Process Technology, Wiley, New York, 2001.
[3] a) A. Cybulski, J. A. Moulijn, Catal. Rev. Sci. Eng. 1994, 36, 179 –
270; b) M. V. Twigg, J. T. Richardson, Ind. Eng. Chem. Res. 2007,
46, 4166 – 4177.
[4] a) F. Svek, J. M. J. Frꢁchet, Science 1996, 273, 205 – 211; b) G. Jas,
A. Kirchning, Chem. Eur. J. 2003, 9, 5708 – 5723; c) M. I.
Burguete, A. Cornejo, E. Garcia-Verdugo, M. J. Gil, S. V. Luis,
J. A. Mayoral, V. Martinez-Merino, M. Sokolova, J. Org. Chem.
2007, 72, 4344 – 4350.
[5] a) N. T. S. Phan, D. H. Brown, P. Styring, Green Chem. 2004, 6,
526 – 532; b) S. J. Haswell, B. O. Sullivan, P. Styring, Lab Chip
2001, 1, 164 – 166.
[8] See the Supporting Information.
[9] a) C. Wiles, P. Watts, S. J. Haswell, Lab Chip 2007, 7, 322 – 330;
b) F. Costantini, W. P. Bula, R. Salvio, J. Huskens, H. J. G. E.
Gardeniers, D. N. Reinhoudt, W. Verboom, J. Am. Chem. Soc.
2009, 131, 1650 – 1651; c) A. R. Bogdan, B. P. Mason, K. T.
Sylvester, D. T. McQuade, Angew. Chem. 2007, 119, 1728 – 1731;
Angew. Chem. Int. Ed. 2007, 46, 1698 – 1701.
[6] a) K. Nakanishi, N. Soga, J. Non-Cryst. Solids 1992, 139, 1 – 24;
b) H. Minakuchi, K. Nakanishi, N. Soga, N. Ishizuka, N. Tanaka,
Anal. Chem. 1996, 68, 3498 – 3501; c) A. Galarneau, J. Iapichella,
D. Brunel, F. Fajula, Z. Bayram-Hahn, K. Unger, G. Puy, C.
Demesmay, J.-L. Rocca, J. Sep. Sci. 2006, 29, 844 – 855; d) C. D.
Liang, S. Dai, G. Guiochon, Chem. Commun. 2002, 2680 – 2682;
e) J. Babin, J. Iapichella, B. Lefevre, C. Biolley, J.-P. Bellat, F.
Fajula, A. Galarneau, New J. Chem. 2007, 31, 1907 – 1917; f) H.
Maekawa, J. Esquena, S. Bishop, C. Solans, B. F. Chmelka, Adv.
Mater. 2003, 15, 591 – 596; g) Z.-G. Shi, L.-Y. Xu, Y.-Q. Feng, J.
[10] T. Okuhara, Chem. Rev. 2002, 102, 3641 – 3666.
[11] a) A. Corma, H. Garcia, Chem. Rev. 2003, 103, 4307 – 4365; b) A.
Corma, J. Catal. 2003, 216, 298 – 312; c) E. De Vos, M. Dam, B. F.
Sels, P. A. Jacobs, Chem. Rev. 2002, 102, 3615 – 3640; d) B. F. Sels,
E. De Vos, M. Buntinx, P. A. Jacobs, J. Catal. 2003, 216, 288 – 297;
Y. Ono, J. Catal. 2003, 216, 406 – 415.
4972 www.angewandte.org
ꢀ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2009, 48, 4969 –4972