Nitroaldol condensation catalyzed by topologically modulable cooperative acid-base chitosan-TiO2 hybrid materials

Article abstract of DOI:10.1039/c4ra04590d

Chitosan-TiO₂ shaped as macroporous aerogels, lamellar cryogels or electrospun films act synergistically as acid-base bifunctional catalysts. Depending on the topology of the material, a marked difference in the selectivity for nitroaldol condensation is observed. the Partner Organisations 2014.

Full text of DOI:10.1039/c4ra04590d

RSC Advances
View Article Online
View Journal | View Issue
COMMUNICATION
Nitroaldol condensation catalyzed by topologically
modulable cooperative acid–base chitosan–TiO2
hybrid materials†
Cite this: RSC Adv., 2014, 4, 33160
a
b
Abdelhfid Aqil, Abdelkrim El Kadib, Mohamed Aqil,ac Mosto Bousmina,bd
Received 15th May 2014
Accepted 7th July 2014
*
*
Abderrahman Elidrissi,c Christophe Detrembleura and Christine Jeromea
´ ˆ
DOI: 10.1039/c4ra04590d
www.rsc.org/advances
Chitosan–TiO2 shaped as macroporous aerogels, lamellar cryogels or of amino groups in the chitosan backbone in close proximity
electrospun films act synergistically as acid–base bifunctional cata- to the nanosized Lewis acidic titanium oxide clusters
lysts. Depending on the topology of the material, a marked difference “NH2/Ti–O–Ti” confers an unusual acid–base cooperative
in the selectivity for nitroaldol condensation is observed.
character to these materials, which has been advantageously
exploited for ne chemical synthesis (Fig. 1).10–13
Among the exciting advantages of chitosan are its ability to
afford open framework hydrogels by pH inversion (coagulation
of acidic soluble polymer in base bath),11 and its versatility for
providing different macroscopic topologies (microspheres,
monoliths, thinner or thicker lms and ultra-ne particles).
Moreover, trimodal micro-, meso- and macro-porosity of the
resultant polymeric network can also be tuned.11,14 Chitosan
and its blended polymeric versions can also be shaped in the
form of nano- and micro-brils by electro-spinning and other
processes for membrane applications.15 Furthermore, the
replication of inorganic oxide occurring within the entangled
porous light-weight aerogels (at the brillar level) can be
reproduced within the electrospun chitosan-based nanobers,
for which the material is shaped as a membrane (at surface)
rather than porous microspheres (convex surface).
Considering the paucity of literature reports regarding the
inuence of the macroscopic topology and related textural
properties on the cooperative catalytic activity, these materials
seem to be an ideal prototypical quest. Herein, we focus on the
preparation of three textured chitosan–titania catalysts,
featuring similar chemical composition “NH2/Ti–O–Ti” but
with different topologies, i.e. convex CS–TiO2–Aero or at
Inspired by the ability of enzymes to tune cooperative interac-
tions in conned environments, sustained efforts have been
devoted to design acid–base bifunctional building blocks for
applications such as ligand–receptor binding,1 non-metallic
activation,2 adsorption3 and catalysis.4 In contrast to the
homogeneous acid–base motifs, for which an abundant litera-
ture exists, acid–base synergy in heterogeneous materials is
relatively less addressed and there are plenty of opportunities
for such materials to be discovered and explored. Very recently,
cooperative solid materials, including aluminophosphates,5
lamellar clay,6 mesoporous organosilica7 and metal organic
framework,8 were reported. Attaching acidic and basic species
in close vicinity through chemical synthesis is thus possible, but
generally it requires complex and time-consuming multi-step
synthesis routes. In contrast, conferring acid–base bifunc-
tional character through the entrapment of specic selected
nanoparticles in a given basic or acidic chemical medium is an
attractive strategy that can be applied for a wide spectrum of
cost-effective applications.9 In this respect, we have recently
shown that chitosan, a natural amino-carbohydrate driven from
biomass resources, can self-assemble to control the nucleation
and growth of titanium-based sol–gel species at “the brillar
level” affording—aer CO2 supercritical drying—macroporous
chitosan–titania (CS–TiO2) hybrid materials.10,11 The presence
a
`
Center for Education and Research on Macromolecules (CERM), University of Liege,
`
B6a B-4000 Liege, Belgium. E-mail: a.aqil@ulg.ac.be
b
`
`
Euro-Mediterranean University of Fez (UEMF), Fes-Shore, Route de Sidi harazem, Fes,
Morocco. E-mail: a.elkadib@ueuromed.org
cUniversity Mohammed Ier, Oujda, BV Mohammed VI B.P. 524, Oujda 60000, Morocco
dHassan II Academy of Science and Technology, Morocco
† Electronic supplementary information (ESI) available: Synthesis and
characterizations of precursors and materials. See DOI: 10.1039/c4ra04590d
Fig. 1 Illustration of the acid–base cohabitation within the framework
of chitosan–titania hybrid materials.
33160 | RSC Adv., 2014, 4, 33160–33163
This journal is © The Royal Society of Chemistry 2014

View Article Online
Communication
RSC Advances
surfaces CS–TiO2–F and outer-sphere surfaces such as open CS– assignable to the methyl and carbonyl groups of the acetamide
TiO2–Aero or lamellar CS–TiO2–Lyo, and examine the similari- moieties. Four peaks attributable to the pyranose ring skeleton
ties and the differences in their behaviour during synergistic have also been shown at 55, 75, 82 and 105 ppm, indicating that
acid–base heterogeneous catalysis.
no transformation occurred on the polymeric chains (neither
The starting microspheres were prepared by solubilizing fragmentation nor ring opening or disproportionation) as a
chitosan in acidic solution (to generate a soluble ammonium consequence of titanium alkoxyde mineralisation. Two addi-
polymer), and its subsequent introduction as droplets in 0.1 N tional peaks are observed at 118 and 195 ppm, unambiguously
NaOH base bath via a needle. This induces deprotonation of the assigned to residual acetylacetonate groups bridged to the
ammoniums, which causes the insoluble amines to immedi- metal center.16 This is consistent with the fast hydrolysis–
ately form an entangled brillar hydrogel-shaped as self- condensation of the alkoxyde groups (compared to the acac
standing microspheres (2 wt% of the polymer dispersed in 98 fragments) in Ti(OiPr)2(acac)2, therby forming oxo-clusters in
wt% water).11 Aer extensive washing until pH ꢀ7, the beads which acetylacetonates are ligated on the surface of the growing
can be exchanged to alcohol, and then to CO2 for supercritical hybrid materials.18 Owing to the presence of titanium clusters,
drying (CS–Aero) or can be directly lyophilized (CS–Lyo).16 DRUV analysis of CS–TiO2 shows a broad oxygen / metal
Although the open framework of the initial hydrogel could be charge transfer (CT) absorption band with a maximum at 312
stabilized by supercritical drying (CS–Aero), partial collapse of nm, indicative of the incipient oligomerisation of Ti(IV)
the lyophilized microspheres (CS–Lyo) is observed, causing a species.19
signicant alteration of the brillar network to generate a
The presence of titania has been conclusively corroborated
layered lamellar structure. The third material is prepared by by EDX and XPS analyses. EDX was used for Ti mapping, and it
electro-spinning chitosan to generate polysaccharide thin lms revealed the homogeneous distribution of titanium species
built from nanobers (CS–F).15 The common thread in these within the polymeric network, thereby excluding the formation
materials is their brillar network (Fig. 2: middle), amenable to of core–shell-type hybrid materials.16 XPS was performed in
replicate titanium alkoxide precursors leading to three CS–TiO2 order to elucidate the nature of the titania species in these
materials.16 Although Ti(OiPr)4 coating is found to be hardly composite materials. Indeed, an expected signal has been
controllable, switching to the less reactive Ti(OiPr)2(acac)2 observed at 458.3 eV, typical of titanium oligomeric clusters.20
affords an aggregation-free, nely decorated polysaccharide Insight on the chemical interplay between the glucosamine
brillar network. SEM analyses of the three CS–TiO2 have evi- units of the chitosan backbone and titanium dioxide centers
denced that the petrication occurs at the brillar surface with has been gained by means of XPS analysis. The observation of
no phase separation or uncontrolled growth outside of the two peaks at 399 and 401 eV indicates the presence of two
bers, and the resultant materials maintain textural properties nitrogen species. The former at 399 eV is characteristic of free
similar to those of the starting polymer CS (Fig. 2: right). To amine (similar to those observed in native chitosan), while the
accurately establish the chemical structure of these materials latter at 401 eV is typical of quaternized, lower metal coordi-
aer sol–gel coating, a MAS 13C NMR has been recorded and nated NH2 / Ti.16 The amount of the loaded titanium oxide
compared with literature data of nonmodied chitosan.17 clusters estimated by thermogravimetric analysis (TGA) ranges
CS–TiO2 materials exhibit two peaks around 24 and 174 ppm, from 15 to 23 wt%. Some intrinsic characteristics of these three
materials are gathered in Table 1.
To verify the synergistic acid–base cooperation of CS–TiO2
hybrid materials, the nitroaldol condensation, commonly
referred to as the Henry reaction (Scheme 1) has been selected
as a prototype reaction (extensively studied reaction). Indeed,
Katz et al. previously correlated the type of products (nitro-
styrene I or nitroaldol II) to the dielectric-outer spheres of the
amine catalyst tethered on the silica surface.21 Asefa nicely
tuned the selectivity in the nitroaldol condensation by sup-
porting primary, secondary or tertiary amines on the silica
Table 1 Textural properties of CS–TiO2
a
b
Material
SBET
% TiO2
SBET
Fiber diameter
CS–TiO2–Aero
CS–TiO2–Lyo
CS–TiO2–F
120
71
—
23
15
20
480
120
—
35 nm
1.05 mm
230 nm
Fig. 2 Left. Digital photos of the prepared materials. (a) CS–Aero. (b)
CS–Lyo (inset: SEM of the croygel bead) and (c) CS–F. Middle: SEM
photos showing the fibrillar network of the polymer. Right: SEM photos
of the respective materials after sol–gel coating (CS–TiO2). The fibrillar
network is preserved.
a
SBET (m2 gꢁ1
) from nitrogen sorption analysis before titania
mineralization. SBET (m2 gꢁ1) from nitrogen sorption analysis aer
b
titania mineralization.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 33160–33163 | 33161

View Article Online
RSC Advances
Communication
signicantly boosts the catalytic activity, which enables quan-
titative conversion of the reactants (Fig. 3b).
When comparing the three catalysts, the same trend (as for
native chitosan) is observed with CS–TiO2–Aero being the most
active one (98% aer 3 h), followed by CS–TiO2–Lyo (99% in 24
h), while the attened surface CS–TiO2–F needed 48 hours to
accomplish quantitative conversion. This unambiguously
demonstrates that, independent of the porosity type, CS–TiO2
materials preserve their acid–base synergistic cooperation in
catalysis. However, the state of the framework (opened, lamellar
or attened) plays a key role in accelerating the conversion
kinetics, as expected.
Scheme 1 Henry reaction (aldol condensation) yielding nitroalkene I
or nitroalcohol II depending on the outer-sphere polarity of the
surrounding catalyst.
surface.22 Mechanistically speaking, this reaction is shown to
proceed via ion-pair mechanism (formation of nitroaldol II) or
imine intermediates (formation of nitrostyrene I).
To ascertain rst whether the three chitosan forms (CS–Aero,
CS–Lyo and CS–F) are valuable for catalysis, these amines
crowded in various environments were subjected to nitro-
methane for a para-methoxybenzaldehyde condensation and
compared to homogeneous NEt3 and EtNH2. Comparison of the
catalytic activity of CS–Aero, CS–Lyo and CS–F reveals that
CS–Aero is the most active catalyst, performing a satisfactory
conversion of 68% in 24 hours, whereas CS–Lyo and CS–F
provide, aer the same period of time, only 23% and 27%
conversions, respectively (Fig. 3a). This behaviour is due to the
accessibility issue as CS–Aero features the highest surface area
and the largest macroporous network making more than 75% of
its amine groups accessible for catalysis.11,23 The layered
framework in the case of CS–Lyo and the densely distributed
polymeric bers in CS–F cause a great number of amine groups
to be sequestered within the packed network and consequently
devoided from any catalytic activity. Homogeneous EtNH2
accomplishes an outstanding 90% conversion aer only 5
hours, illustrating thus the higher reactivity of homogeneous
species even over the highly porous CS–Aero (Fig. 3a).
Taking a closer look at the forming products (aldol versus
alkene products), a dramatic switch in the reaction selectivity is
observed (Fig. 4). Although CS–TiO2–Aero affords almost selec-
tively the nitrostyrene product (ratio of alkene : alcohol ¼
95 : 5), it is totally inverted in the case of CS–TiO2–Lyo (ratio of
alkene : alcohol ¼ 9 : 91) and for CS–TiO2–F (ratio of alke-
ne : alcohol ¼ 7 : 93). The predominance of nitrostyrene over
nitroalcohol in the case of CS–TiO2–Aero indicates that the
condensation occurs herein by an imine formation mechanism
reminiscent of that occurring with the hydrophilic silica teth-
ered aminopropyl catalyst.21 In this latter case, the surface
silanols act in synergy with tethered amine groups as Si–OH are
involved in the formation of imine and iminium (both of them
are intermediates in nitrostyrene formation).21 Hydrophobic
silica in which Si–OH are end-capped to provide Si–OSiMe3 does
not allow this synergy and under these conditions, the reaction
proceeds via ion pair mechanism to preferentially produce the
nitroaldol product.21 Similarly herein, CS–TiO2–Aero is distin-
guished from its lyophilized CS–TiO2–Lyo or attened-surface
CS–TiO2–F version by the abundance of Bronsted OH groups
and accessible Lewis acid titanium centers, both of them can
catalyze, similar to silanol, the imine and iminium formation.
This explanation seems to be reasonable as the brillar network
of the polymeric chains appears more accessible in the aerogels
as a result of the three-dimensional network, drawn one-each-
other in the layered, partially collapsed lyophilized cryogels
and well-packed in the case of the attened lms. In the two
latter cases (layered or packed), this probably leads to: (i) the
connement of the replicated species in a constrained space,
making the simultaneous accessibility of the two reactants to
the collaborative active sites environment more difficult (kinetic
effect), and (ii) the consumption of the hydroxylic groups
belonging to the organic polymer and those resulting from the
Having validated that the openness of the framework affects
the reaction kinetics, the question was whether or not these
textured materials could preserve their synergistic acid–base
character or if the evolution of the gelling microstructure to
lamellar structure in CS–Lyo and the atness of the surface in
CS–F would suppress these properties. To this end, the behav-
iour of CS–TiO2–Aero, CS–TiO2–Lyo and CS–TiO2–F was inves-
tigated in the Henry condensation.
The comparison of hybrids CS–TiO2 with the starting native
chitosan CS reveals that the presence of titanium clusters
Fig. 3 (a) Left: conv. (%) obtained using different aminated catalysts.
(b) Right: comparison of the conv. (%) using mono-(CS) versus
bifunctional (CS–TiO2) catalysts.
Fig. 4 Aldol-to-alkene ratio obtained with the three CS–TiO2.
33162 | RSC Adv., 2014, 4, 33160–33163
This journal is © The Royal Society of Chemistry 2014

View Article Online
Communication
RSC Advances
incomplete condensation of titania species (Ti–OH), thereby
affecting the outsphere polarity (selectivity effect).
7 For review, see: (a) E. L. Margelefsky, R. K. Zeidan and
M. E. Davis, Chem. Soc. Rev., 2008, 37, 1118–1126; (b)
U. Diaz, D. Brunel and A. Corma, Chem. Soc. Rev., 2013, 42,
4083–4097; (c) S. Shylesh and W. R. Thiel, ChemCatChem,
2011, 3, 278–287; (d) M. Tada, K. Motokura and
Y. Iwasawa, Top. Catal., 2008, 48, 32–40.
The recyclability of CS–TiO2 materials has been assessed and
regardless of the topology of the material, the three mineralized
catalysts can be recovered and reused several times (up to four
times for CS–TiO2–Aero) without any decrease of their catalytic
activity. In contrast, a fast decay of the native chitosan is
observed within the rst two runs, probably because of its
limited thermal and chemical stability.
To sum up, the outersphere environment has been hitherto
tailored by means of chemical modication to provide tethered
catalyst grown in hydrophilic or hydrophobic medium. Quite
recently, supramolecular chemistry based on foldamers24 and
8 F. Vermoortele, R. Ameloot, A. Vimont, C. Serrec and D. De
Vos, Chem. Commun., 2011, 47, 1521–1523.
9 See for instance: (a) B. Voit, Angew. Chem., Int. Ed., 2006, 45,
4238–4240; (b) B. Helms, S. J. Guillaudeu, Y. Xie,
M. McMurdo, C. J. Hawker and J. M. Frechet, Angew.
Chem., Int. Ed., 2005, 44, 6384–6387; (c) F. Gelman, J. Blum
and D. Avnir, Angew. Chem., Int. Ed., 2001, 40, 3647–3649.
micellar systems25 have been proven effective for shielding 10 A. El Kadib, K. Molvinger, C. Guimon, F. Quignard and
active sites in a conned medium. Herein, the versatility in D. Brunel, Chem. Mater., 2008, 20, 2198.
processing the starting polymer and the control over drying the 11 A. El Kadib and M. Bousmina, Chem.–Eur. J., 2012, 18, 8264–
sol–gel modied chitosan hydrogels enable access to various 8277.
multi-functional enzyme-like catalysts. This textural engi- 12 A. El Kadib, K. Molvinger, M. Bousmina and D. Brunel, Org.
neering has been the driving force to invert the classical selec-
tivity of primary amines for nitroaldol condensation.
Lett., 2010, 12, 948.
13 A. El Kadib, K. Molvinger, M. Bousmina and D. Brunel, J.
Catal., 2010, 273, 147.
14 A. El Kadib, M. Bousmina and D. Brunel, J. Nanosci.
Nanotechnol., 2014, 14, 308–331.
Acknowledgements
´ ´
The authors are grateful to the ‘Politique Scientique Federale’
´
15 K. Ziani, C. Henrist, C. Jerome, A. Aqil, J. I. Mate and
for nancial support in the framework of the Interuniversity
Attraction Poles Programme (IAP VII-05): Functional Supramo-
lecular Systems (FS2), to the Region wallonne (DG06) for
support in the frame of GoCell project (grant nos 516025 and
616262) and to the Euro-Mediterranean University for nancial
support. The authors are thankful to Professor Albert Demon-
R. Cloots, Carbohydr. Polym., 2011, 83, 470–476.
16 See ESI† for details.
17 M. Rinaud, Prog. Polym. Sci., 2006, 31, 603–632.
18 C. Sanchez, J. Livage, M. Henry and F. Babonneau, J. Non-
Cryst. Solids, 1988, 100, 65–76.
19 L. Marchese, E. Gianotti, V. Dellarocca, T. Maschmeyer,
F. Rey, S. Colluccia and J. M. Thomas, Phys. Chem. Chem.
Phys., 1999, 1, 585–592.
`
ceau (University of Liege) for helpful assistance in GC analysis.
20 C. D. Wagner, W. M. Riggs, L. E. Davis, J. F. Moulder and
G. E. Muilenberg, Handbook of X-ray Photoelectron
Spectroscopy, Perkin-Elmer Corp., Eden Prairie, MN, 1979,
p. 68.
21 D. Bass, A. Solovyov, A. J. Pascall and A. Katz, J. Am. Chem.
Soc., 2006, 128, 3737.
Notes and references
1 M. W. Pantoliano, R. A. Horlick, B. A. Springer, D. E. Van
Dyk, T. Tobery, D. R. Wetmore, J. D. Lear,
A. T. Nahapetian, J. D. Bradley and W. P. Sisk, Biochemistry,
1994, 33, 10229–10248.
2 D. W. Stephan and G. Erker, Angew. Chem., Int. Ed., 2010, 49,
46–76.
22 A. Anan, R. Vathyam, K. K. Sharma and T. Asefa, Catal. Lett.,
2008, 126, 142.
3 N. A. Brunelli, S. A. Didas, K. Venkatasubbaiah and
C. W. Jones, J. Am. Chem. Soc., 2012, 134, 13950–13953.
4 J. M. Notenstein and A. Katz, Chem.–Eur. J., 2006, 12, 3954.
23 R. Valentin, K. Molvinger, F. Quignard and D. Brunel, New J.
Chem., 2003, 27, 1690.
24 E. Huerta, P. J. M. Stals, E. W. Meijer and A. R. A. Palmans,
Angew. Chem., Int. Ed., 2013, 52, 2906.
´
5 M. J. Climent, A. Corma, V. Fornes, R. Guil-Lopez and
S. Iborra, Adv. Synth. Catal., 2002, 344, 1090–1096.
6 K. Motokura, N. Fujita, K. Mori, T. Mizugaki, K. Ebitani and
K. Kaneda, J. Am. Chem. Soc., 2005, 127, 9674–9675.
25 P. Cotanda, A. Lu, J. P. Patterson, N. Petzetakis and
R. K. O'Reilly, Macromolecules, 2012, 45, 2377–2384.
This journal is © The Royal Society of Chemistry 2014
RSC Adv., 2014, 4, 33160–33163 | 33163