Co-Polymeric Nanosponges from Cellulose Biomass as Heterogeneous Catalysts for amine-catalyzed Organic Reactions

Article abstract of DOI:10.1002/cctc.202001157

Heterogeneous catalysts prepared from biomass waste sources are attracting increasing interest. The reasons rely on the possibility of combining the virtuous approach of circular economy with the consolidated advantages of heterogeneous catalysis, namely the recycling of the system and the possibility to drive selectivity towards desired products. Herein we report a highly porous cellulose-based nanosponge (CNS) and its use as a recoverable catalyst for Henry and Knoevenagel reactions, two classical amino-catalyzed transformations. The material is obtained by cross-linking between TEMPO-oxidized cellulose nanofibers (TOCNF) and branched polyethyleneimine 25 kDa (bPEI) in the presence of citric acid. CNS have been developed as sorbent materials for water remediation but their use as heterogeneous catalysts was never investigated. The fully characterized micro- and nano-porous system guarantees a complete penetration of CNS, allowing reagents to diffuse within. Indeed, by modulating reaction conditions (catalyst loading, temperature, solvent, microwave versus conventional heating, relative ratio of reagents) it was possible to drive selectivity towards the desired products, while maintaining high efficiency in terms of conversion. The catalyst could be re-used several times without losing in catalytic efficiency. In most cases the products’ distribution is quite different from homogeneous conditions, this much more emphasizing the importance of this heterogeneous solution.

Full text of DOI:10.1002/cctc.202001157

The European Society Journal for Catalysis
Supported by
Accepted Article
Title: Co-Polymeric Nanosponges from Cellulose Biomass as
Heterogeneous Catalysts for amine-catalyzed Organic Reactions
Authors: Alessandro Sacchetti, Laura Riva, and Carlo Punta
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: ChemCatChem 10.1002/cctc.202001157
Link to VoR: https://doi.org/10.1002/cctc.202001157
01/2020

10.1002/cctc.202001157
ChemCatChem
FULL PAPER
Co-Polymeric Nanosponges from Cellulose Biomass as
Heterogeneous Catalysts for amine-catalyzed Organic Reactions
Laura Riva,[a] Carlo Punta[a],[b] and Alessandro Sacchetti*[a]
[a] L. Riva, Prof. C. Punta and Prof. A. Sacchetti
Department of Chemistry, Materials and Chemical Engineering, “Giulio Natta”
Politecnico di Milano
Via Luigi Mancinelli 7, 20131, Milano, Italia.
alessandro.sacchetti@polimi.it
[b] Prof. C. Punta
C. N. R. Istituto di Chimica del Riconoscimento Molecolare (ICRM)
Milan, Italy
Supporting information for this article is given via a link at the end of the document.
Abstract: Heterogeneous catalysts prepared from biomass waste
sources are attracting increasing interest. The reasons rely on the
possibility of combining the virtuous approach of circular economy
with the consolidated advantages of heterogeneous catalysis, namely
the recycling of the system and the possibility to drive selectivity
towards desired products. Herein we report a highly porous cellulose-
based nanosponge (CNS) and its use as a recoverable catalyst for
Henry and Knoevenagel reactions, two classical amino-catalyzed
transformations. The material is obtained by cross-linking between
TEMPO-oxidized cellulose nanofibers (TOCNF) and branched
polyethyleneimine 25 kDa (bPEI) in the presence of citric acid. CNS
have been developed as sorbent materials for water remediation but
their use as heterogeneous catalysts was never investigated. The fully
characterized micro- and nano-porous system guarantees a complete
penetration of CNS, allowing reagents to diffuse within. Indeed, by
modulating reaction conditions (catalyst loading, temperature, solvent,
microwave versus conventional heating, relative ratio of reagents) it
was possible to drive selectivity towards the desired products, while
maintaining high efficiency in terms of conversion. The catalyst could
be re-used several times without losing in catalytic efficiency. In most
cases the products’ distribution is quite different from homogeneous
conditions, this much more emphasizing the importance of this
heterogeneous solution.
building blocks for the design of heterogeneous catalysts. Despite
the improvements in the field of biopolymer-based catalysis, the
development of catalytic systems bearing several active sites and
capable to drive the selectivity of the processes remain rather rare.
Cellulose represents a source of major interest among those
derived from biomass. It is the most ubiquitous structural
amphiphilic renewable polymer resource in the biosphere,[5] with
an estimated annual production of 7.5 × 1010 tons. This nearly
unlimited supply of sustainable polysaccharide possesses
remarkable physical and chemical properties, suggesting its use
in a wide range of materials and products.[6] In recent years,
heterogeneous cellulose-based catalytic systems have been
developed for promoting different chemical reactions,[7,8]
obtaining excellent yields and very short reaction times. These
systems are often reusable, and this fact represents a key point
from an industrial application point of view.[9]
Introduction
The advantages of heterogeneous catalysis over homogeneous
one are numerous. Among the most evident are certainly the
possibility of easily recovering and reusing the catalyst from the
reaction medium, and consequently isolating the final products in
a simpler way. Other advantages are the duration of the catalyst
and a considerable reduction in costs.
Figure 1. Various applications of nanocellulose obtained from
recycled sources.
Microporous nanomaterials have been widely proposed as
heterogeneous catalysts,[1] due to the remarkable catalytic
efficiency of these solutions, often designed to drive selectivity
towards desired products. However, in most cases the high
energy demand for their synthesis and the end-of-life disposal of
these systems, with consequent production of waste, negatively
impact on the environment.
Given the raising trend towards the discovery of alternative
materials that may be less impacting, greener, more sustainable
and renewable, the widespread use of materials derived from
natural biomass is increasing also in the field of catalysis. In
particular, in the recent literature, biopolymers such as wool,[2]
chitosan,[3] and lignin[4] are increasingly taking root as ideal
Recently our research group has developed a new class of
cellulose-based nanostructured aerogels, obtained by valorizing
different biomass waste sources of cellulose, such as beached
Posidonia Oceanica and recycled paper pulp. These cellulose
nanosponges (CNS) are produced by following a two-step
protocol. First, cellulose fibers are chemically treated through a
well-consolidated TEMPO (2,2,6,6-tetramethylpiperidine-1-oxyl
radical)-mediated NaClO/NaBr regioselective oxidation.[10] After
ultra-sonication in an alkaline medium, we obtain TEMPO-
oxidized cellulose nanofibers (TOCNF), characterized by
micrometric length and nanometric diameter, and provided with a
high content of carboxylic groups..
1
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001157
ChemCatChem
FULL PAPER
In the second step TOCNFs are thermally cross-linked with 25
kDa branched polyethyleneimine (bPEI), promoting the formation
of amidic bonds and providing micro- and nano-porous aerogels.
These systems were first proposed as efficient sorbent materials
for water decontamination from heavy metal ions and organic
contaminants.[11,12,13] In this context, they have also been made
eco-safe through an in-depth eco-design study, consisting in a life
cycle analysis (LCA)[14] and in an ecotoxicological evaluation,[15,16]
obtaining an optimized synthetic strategy of the material, which
saw the introduction of citric acid as a third component, and
ensuring the absence of risks for aquatic biota.
Moreover, their high versatile properties and the possibility to
easily functionalize both the polymeric components suggested an
extension of their potential application, so that they resulted to be
successful drug-delivery systems,[17] useful templates for
organic[18] and inorganic coatings,[19] and suitable heterogeneous
sensors for specific anions, when functionalized with adequate
fluorophores.[20,21]
By continuing our investigation to completely exploit the
potentialities of this material, we envisioned that the combination
of the chemical composition and the specific morphology, made
of micro and nano-porosity, could make CNS ideal candidates as
sustainable heterogeneous solid catalysts. In particular, the
presence of bPEI as a structural component makes them suitable
for the promotion of amine-catalyzed reactions. Indeed, this
polyamine has many primary, secondary, and tertiary amino groups
that can act as catalytic sites in reactions usually promoted by
conventional bases such as triethylamine or pyridine. Herein we
report the role of CNS in promoting two different organic
transformations, Henry and Knoevenagel[22-24] reactions, which
were chosen as models of study to understand the efficiency and
the behavior of our eco-compatible catalyst. We focused on these
aldolic condensations also because they both represent key
transformations for the synthesis of bioactive molecules in the
pharmaceutical and fragrance industries, where, the use of
heterogeneous catalysts for base-catalyzed reactions has allowed
to overcome some problems related to homogeneous catalysis, first
of all the poor recovery of the organo-catalyst from the reaction
medium.[25-27]
content of carboxylic units. The same result cannot be obtained
by pushing the conversion during oxidative process, as further
oxidation leads to the depolymerization of TOCNF.[28] This
formulation provides CNS with excellent mechanical properties.[17]
Scheme 2 summarizes the synthetic procedure. Each step is
essential to attribute specific properties to the material. The
addition of polyethyleneimine (which has an estimated amount of
primary amines of 7.43 mmol/g)[21] is of fundamental importance
to ensure the presence of a high number of primary, secondary
and tertiary amines on the surface of the final material, traceable
through elemental analysis[17] and through titration (see
paragraph Experimental Section). With this latter we were able
to determine the presence of about 4.5 mmol of amines per gram
of nanosponge. The freeze-drying process is crucial for
determining the morphology and the porosity of the nanosponge.
The final CNS exhibits a classical 2D sheet-like structure, with a
porosity close to 70% of bulk material, as determined by
microcomputed tomography,[14] and which includes both micro-
porosity, evidenced by scanning electron microscopy (SEM)
technique (Scheme 2), and nano-porosity, previously determined
through a detailed small angle neutron scattering (SANS) analysis,
which afforded a short-range correlation length in a range
between 25 and 35 Å,[29] and more recently by a combined
investigation of the FTIR-ATR spectra of CNS hydrated with H2O
and D2O.[30]
The material was grinded in a mortar before use to guarantee a
higher contact surface area. Particle sizes resulted to be in a
range between 50 and 400 μm, with a maximum distribution at
130 μm.
CNS as catalysts for Henry reaction
The Henry reaction,[31] is a very versatile C–C bond formation
reaction leading to nitro-derivatives, resulting from the coupling of
a nucleophilic nitroalkane with an electrophilic aldehyde or ketone.
The main product is a β-nitro alcohol (Scheme 1), which can be
further transformed into amino-alcohols, useful for the synthesis
of many drugs and natural products. The reaction occurs under
basic conditions in the presence of bases, such as carbonates,
alkali metal hydroxides, alkoxides, or organic nitrogen bases. In
the view of a more sustainable approach, environmentally friendly
processes using heterogeneous catalysis,[32] green and non-toxic
solvents,[33] microwave irradiation[34] have been proposed.
To investigate the catalytic activity of the CNS system for Henry
transformation, the reaction of nitromethane with p-
fluorobenzaldehyde was selected as the model. Methanol was
chosen as a typical solvent for this reaction. Several different
parameters as temperature, reaction time, nitromethane
equivalents, and catalyst loading were considered. In all the
reactions, the products were easily recovered in quantitative
amounts by filtering the nanosponge powder and washing with
methanol. The crude residue was then submitted to analysis by
NMR spectroscopy.
Results and Discussion
Synthesis and Characterization of Heterogeneous catalyst
TOCNF were obtained from virgin and renewable sources,
including recycled paper pulp and beached Posidonia Oceanica, by
selective oxidation of the primary alcoholic moieties in C6 position
of the glucopyranose units to the corresponding carboxylic acids.
At basic pH and under ultrasonication, the electrostatic repulsion
between the negatively charged cellulosic chains, due to
deprotonation of carboxylic groups, promotes the formation of
nanofibers by defibrillation. The sponge-like catalytic material
described in this work has been synthesized starting from TOCNF
and bPEI 25 kDa. These two building-blocks underwent cross-
linking through a thermal process at about 100 °C, via formation of
amidic bonds between the carboxylic moieties on the cellulosic
nanofiber backbone and the primary amines of bPEI polymer. The
optimized formulation herein investigated includes the use of citric
acid (CA) as co-reticulant. The addition of CA guarantees a higher
crosslinking between the two polymers by increasing the overall
The outcome of the reaction was evaluated in terms of percentage
of conversion of the starting aldehyde and in selectivity toward the
final product. It is known that three possible products can form in
this reaction, namely the expected nitroaldol 1, the elimination
product β-nitrostyrene 2 and the bis-adduct 3 (Scheme 1). Under
standard conditions, 1 is the main product but in some cases 2 or
3 can also be obtained.
2
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001157
ChemCatChem
FULL PAPER
Scheme 1. Scheme of the synthesis of the nanosponge, with a SEM image at 300x magnification showing the micro-pores of the
catalytic system
NO2
O
OH
Table 1. Results from reaction between nitromethane and p-
fluorobenzaldehyde in methanol as solvent.
NO2
NO2
NO2
H
+
+
CH3NO2
F
F
F
F
1
2
3
T
(h)
T
(°C)
CH3NO2
equiv.
Conv.
(%)a
4
(%)b
2
(%)b
3
(%)b
entry
Scheme 2. Possible products resulting from the Henry reaction
between nitromethane and p-fluorobenzaldehyde.
1
2
3
4
5
48
24
24
2
rt
40
2
2
2
2
5
8
12
0
79
97
21
98
29
10
3
In a first study, the catalyst loading was fixed to a 20% molar ratio
referred to the free amino groups present on the catalyst, which
were determined to be 4,5 mmol/g (see experimental for details).
Results are reported in Table 1.At room temperature, after 48 h
the main product resulted to be the β-nitrostyrene 2 but with a
poor 8% conversion (entry 1). An increase of the temperature
from rt to 40 °C allowed to obtain complete selectivity toward 2
after 24h in 37 % conversion (entry 2). By heating the reaction
mixture at reflux for 24h, the conversion increased to a good 98%
(entry 3), but the selectivity was decreased. The amount of 1,4
adducts 3 increased and notably, the main product of the reaction
was found to be the derivative 4 (Scheme 3) which was obtained
in 54% selectivity
37
98
97
93
reflux
40c
54
1
25
1
2
80c
53
18
Reaction conditions: methanol 2.5 mL, aldehyde 0.4 mmol, cat. 20% mol.
Conversion was measured with 1H-NMR on the crude with respect to
a
unreacted aldehyde.
b Percentage of compound in the mixture of products as obtained by 1H-NMR
on the crude.
c Reaction was heated under microwave (MW) irradiation.
O
O
We then investigated the effect of catalyst loading. Different m/m
percentages of catalysts were employed. Results are reported in
Table 2. The mol% of catalyst calculated on the amount of
p-fluorobenzaldehyde is also showed.
NO2
20%
catalyst
H
+
+
3
2
CH3NO2
MeOH, MW
F
F
4
Scheme 3. Products from the Henry reaction between
nitromethane and p-fluorobenzaldehyde in methanol as solvent.
Different experiments were run at different times to find out the best
balance between yield, selectivity, reaction time and amount of
catalyst. The first acceptable 73% yield was achieved with 5% m/m
of catalyst (2.8 mol%) after 4h, with a complete selectivity toward 2
(entry 4). After the same time and using 20% m/m (11 mol%) of
catalyst, the yield was 91% (entry 8). Finally, when 35% m/m (20
mol%) of catalyst was employed, the yield was 97% (100%
selectivity) after only 2h (entry 9). To better evaluate the effectiveness
of the CNS catalyst, the two single components of the catalyst, i.e.
TOCNF and PEI, were tested separately in the same conditions. With
the use of TOCNF, no reaction occurred (entry 10). As expected,
when pure PEI was used as catalyst at 20% mol, 2 was obtained with
good selectivity but in poor conversion (entry 11). At higher
This product can be explained as the result of a Michael addition
of methanol on β-nitrostyrene 2; similarly, the formation of 3 is the
result of reaction between 2 and nitromethane (vide infra). In order
to optimize these results, we performed the reaction by operating
under microwave irradiation instead of conventional heating. We
could obtain a complete selectivity toward the formation of 2 with
a 97% conversion after 2h at 40°C (entry 4). Finally, a further
increase of the temperature up to 80 °C produced a good
conversion (93%) but a lower selectivity (entry 5).
3
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001157
ChemCatChem
FULL PAPER
concentration of PEI (50% mol), the conversion increased but the
selectivity lowered. These experiments confirm that PEI is the active
component of the catalyst, however the TOCNF not only supply the
structural properties of the material but also provides the correct
environment to drive the reaction toward high conversion and
selectivity.
nitrostyrene 2, although no trace of the alkene could be detected in
the mixture. Even in this case, at higher temperatures or with longer
reaction times the selectivity decreased and mixtures of products 1,
2 and 3 were obtained (entry 3,4).
Table 3. Results from reaction between nitromethane and p-
fluorobenzaldehyde in water as solvent.
Table 2. Results from screening of the CNS catalyst loading in the
reaction between nitromethane and p-fluorobenzaldehyde.
T
(h)
T
(°C)
CH3NO2
equiv.
Conv.
(%)a
1
(%)
2
(%)b
3
(%)b
entry
catalyst
(% mol)
catalyst
(% m/m)
time
(h)
selectivity
(%)b
entry
Conv. (%)a
1
2
3
4
2
2
8
2
40
40
2
5
5
5
28
99
99
65
100
89
0
0
0
11
58
45
1c
2 c
1.1
1.1
2.8
2.8
5.5
5.5
11
2
2
2
4
2
4
2
4
2
4
2
4
2
2
0
n.d
n.d
80
42
0
0
105
26
29
3 c
5
32
73
54
74
60
91
97
0
100
100
100
100
100
100
100
n.d.
99
Reaction conditions: water 2.5 mL, aldehyde 0.4 mmol, cat. 20% mol, MW
4 c
5
heating
a Conversion was measured with 1H-NMR on the crude with respect unreacted
5 c
10
10
20
20
35
20
n.d.
n.d.
aldehyde.
b Percentage of compound in the mixture of products as obtained by 1H-NMR
on the crude.
6 c
7 c
Since we could observe the presence of the unanticipated bis
adduct 3, with the aim to switch the selectivity towards this product,
the same reaction was performed using neat nitromethane as
solvent, at 80 °C under MW irradiation (Table 4). As expected, 3
was obtained as the major product (70%) with a good 91%
conversion yield (entry 1). Notably, under conventional heating,
the major product resulted to be the β-nitrostyrene b (entry 2).
8 c
11
9 c
20
10d
11e
12e
n.d.
20
36
56
50
71
Table 4. Results from reaction between nitromethane and p-
fluorobenzaldehyde in neat nitromethane as solvent.
Reaction conditions: methanol 2.5 mL, aldehyde 0.4 mmol, MW heating at 40°C.
a
Conversion was measured with 1H-NMR on the crude with respect to
unreacted aldehyde.
b Percentage of compound 2 in the mixture of products as obtained by 1H-NMR
on the crude.
T
(h)
T
(°C)
entry
Conv. (%)a
1 (%)
2 (%)b
3 (%)b
c CNS was used as catalyst
d TOCNF was used as catalyst
e PEI was used as catalyst
1
2
8
2
80
91
68
0
9
30
71
70
20
80c
These results showed that the catalyst is very effective and good
conversions can be achieved with only 2.8% mol loading (5% m/m
in the case of p-fluorobenzaldehyde). Nevertheless, our main focus
was to obtain the highest yield/selectivity in the mildest conditions
and in the shortest time possible and this result could be reached
with 20%mol of catalyst. The catalyst herein proposed can be easily
prepared in huge amounts from cheap starting materials and with
little efforts. Moreover, our catalyst can be easily recovered and
recycled without significative loss of activity (vide infra). Following
these remarks, we decided to use the 20% mol loading as the
optimized condition for our further studies.
Reaction conditions: nitromethane 2.5 mL, aldehyde 0.4 mmol, cat. 20% mol,
MW heating
a Conversion was measured with 1H-NMR on the crude with respect unreacted
aldehyde.
b Percentage of compound in the mixture of products as obtained by 1H-NMR
on the crude.
c Reaction was heated with conventional heating.
After these tests, we were able to define three different sets of
conditions which allowed to switch the selectivity among the three
expected products of the Henry reaction of nitromethane with p-
fluorobenzaldehyde. In all cases, the amount of catalyst was 20 %
mol with respect the aldehyde and the reaction mixtures were
heated under microwave irradiation for 2h at the indicated
temperature. The only changes were the solvent and the amount of
nitromethane used. The final set of conditions are described in
Table 5. To resume, the nitroaldol product 1 can be obtained in
water at 40°C with 5 equiv. of nitromethane (Table 5, cond. B)
whereas the β-nitrostyrene 2 is achieved in methanol at the same
temperature with 5 equiv of nitromethane (Table 5, cond. A). Lastly,
in neat nitromethane at 80°C the 1,3-dinitropropane derivative 3 is
the main product (Table 5, cond. C).
Due to the highly hydrophilic nature of the material, we then decided
to investigate the effect of using water as the solvent (Table 3). By
heating the reaction mixture at 40 °C (MW irradiation) for 2 hours
with 2 equivalents of nitromethane, despite a lower conversion of
28 %, it was possible to push the reaction to a complete selectivity
for the nitroaldol product 1 (entry 1). After some trials, quantitative
conversion was achieved with 5 equivalents of nitromethane even
if with a lower selectivity (89% of product 1, entry 2) for the presence
of some amounts of product 3. The formation of this last compound
can only result from the addition of nitromethane on the β-
4
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001157
ChemCatChem
FULL PAPER
Table 5. Resume of the optimized reaction conditions between
nitromethane and p-fluorobenzaldehyde
In water the nitroaldol 5 was recovered in good yields whereas in
methanol the β-nitrostyrene 6 was the main product. In neat
nitropropane the principal product was again 6, together with
moderate amounts of the unprecedented bis-adduct 7 which
could be isolated as a mixture of diastereoisomers (see SI).
T
CH3NO2
equiv.
Cond.
Solv.
Selectivity
Product
[°C]
A
B
C
CH3OH
Water
40
40
80
2
5
-
98%
89%
70%
2
1
3
Table 6. Results from reaction between nitropropane and p-
fluorobenzaldehyde. Only the main product is reported.
CH3NO2
entry
Solv.
Water
T [°C]
40
yield
84%
84%
88%
Product
Reaction conditions: solvent 2.5 mL, aldehyde 0.4 mmol, cat. 20% mol, MW
heating for 2 h.
1
2
5
6
6
CH3OH
80
In a last experiment, to verify the hypothesis on the formation of
4, we performed an experiment in which the β-nitrostyrene 2 was
dissolved in methanol and heated to 80°C in the presence of 20%
mol of the catalyst. After few hours we could detect the formation
of the expected product 4 in 37% yield. Notably, in the same way,
when the β-nitrostyrene 2 was treated in water as solvent at 100°,
after 2h it was converted in 54% of the nitroaldol 1. Finally, as
expected, the treatment of 2 with neat nitromethane at 80°C
produced the bis-adduct 3 with a selectivity of 66%. These
experiments suggested that the first product of the reaction is
probably always the β-nitrostyrene 2, then a further addition to the
double bond occurs from a nucleophile, i.e. the solvent, to
produce the main product (Scheme 4). It is worth noting that the
addition of oxygen nucleophiles to β-nitrostyrenes is not common
and it is usually accomplished by the use of metal catalysis[35] or
with strong alkoxide nuclephiles.[36]
3a
80
CH3CH2NO2
Reaction conditions: solvent 2.5 mL, nitropropane 2 mmol, aldehyde 0.4 mmol,
cat. 20% mol, MW heating for 4 h.
a60 equivalents of nitropropane were used
We then explored the scope of this protocol and other aldehydes
were employed (Scheme 6). Depending on the reaction
conditions used, a main product was obtained, according to the
results reported in Table 7.
NO2
OR
O
20% catalyst
solvent, MW
CH3NO2
NO2
NO2
NO2
+
+
Ar
Ar
Ar
Ar
H
9
10
8
Scheme 6. Products from the Henry reaction between
nitromethane and various aldehydes.
OMe
MeOH
80°C
NO2
NO2
37%
54%
4
Table 7. Results from Henry reaction between nitromethane and
different aromatic aldehydes under optimized conditions A, B and
C, as described in Table 5.
F
F
OH
NO2
mol
H2O
20%
catalyst
1
F
100°C
NO2
NO2
Conditions
2
CH3NO2
80°C
entry
Ar.
A (% of 8) a
B (% of 9) a
C (% of 10) a
66%
F
3
1
2
3
4
5
6
7
8
p-NO2Ph
o-NO2Ph
p-BrPh
12b
tracesb
74
98
96
76
71
91
88
92
92
87c
56c
81
76
86
61
58
56
Scheme 4. Reactions of β-nitrostyrene
2 with different
nucleophiles.
The reaction of p-fluorobenzaldehyde with nitropropane was also
investigated (Scheme 5). According to the optimized reaction
conditions, product 5, 6, and 7 were obtained (Table 6).
m-BrPh
98
o-BrPh
81
O
p-MePh
p-MeOPh
m-MeOPh
60
20% catalyst
H
NO2
79
solvent, MW
F
84
NO2
NO2
OH
NO2
a Conversion was measured with 1H-NMR on the crude with respect unreacted
NO2
+
+
aldehyde.
Percentage of compound in the mixture of products as obtained by 1H-NMR on
the crude.
F
F
F
5
6
7
b A complex mixture of inseparable products was obtained
cThe main product was 9.
Scheme 5. Products from the Henry reaction between
nitropropane and p-fluorobenzaldehyde.
5
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001157
ChemCatChem
FULL PAPER
O
Results from Table 7 are generally in agreement with the previous
achievements. Good yields can be obtained with both electron
donating (EDG) and electron withdrawing (EWG) substituents on
the aromatic ring, in particular for the formation of compounds 8
and 9. The formation of the 1,3-dinitropropane derivatives was
more difficult. In particular, in the case of the strongly donating
methoxy group (entry 7,8), and the methyl group (entry 6), the
double adduct 10 was obtained in 56-61% yields, together with
the olefin 9 (35 to 40% yield). The presence of the EDG made the
Michael addition to the double bond less effective. Surprisingly,
when the highly reactive nitrobenzaldehyde was employed, the
only product which we could obtain was the nitroaldol 8 (entries 1
and 2) in all the conditions tested. In particular, with methanol as
the solvent, the yields were very low for the formation of a
complex mixture of products. All the attempts to obtain a different
selectivity on these substrates failed.
At the end of the reactions, the catalyst was usually recovered by
easy filtration and washing with organic solvent. Recyclability of
solid heterogeneous catalysts is of great interest for practical
applications. We then decided to test the feasibility to recycle the
catalyst. We conducted the test using once again
p-fluorobenzaldehyde as model substrate to afford product 2, by
operating according to B procedure. We could observe that the
catalyst could be reused up to five times in the same conditions
without significant changes in both selectivity and conversions
(Figure 2).
O
O
O
O
H
RO
RO
OR
F
20% catalyst
11a
11b
R=Me
R=Et
OR
solvent, heating
F
Scheme 7. The Knoevenagel reaction between DMM and DEM
and p-fluorobenzaldehyde.
According to the previous results, reactions were performed both
in methanol and in water using p-fluorobenzaldehyde as the
reference substrate. In both solvents, the expected condensation
product 11a,b were obtained as a single product. Again, the use
of microwaves as source of heating proved to be more beneficial
with respect to classical heating. High conversions could be
obtained in water by using 20% mol of catalyst and after 4h (see
Table 8).
Table 8. Results from Knoevenagel reaction between malonate
diester and p-fluorobenzaldehyde.
entry
Producta
Yield (%)b
Yield (%)c
1
2
11a
11b
73
72
56
93
Reaction conditions: solvent 2.5 mL, aldehyde 0.4 mmol, malonate diester 0.48
mmol, cat. 20% mol.
a Conversion was measured with 1H-NMR on the crude with respect unreacted
aldehyde.
b Reaction was performed in methanol at 80°C under microwave irradiation for
4h.
c Reaction was performed in water at 100°C under microwave irradiation for 4h.
Table 9. Results from Knoevenagel reaction between
diethylmalonate (DEM) and different aromatic aldehydes under
optimized conditions.
O
O
O
catalyst 20%
Figure 2. Results from recycle of the catalyst. Reaction
conditions: methanol 2.5 mL, aldehyde 0.4 mmol, catalyst 20%
mol, MW heating at 40°C for 2 h.
DEM
O
O
Ar
H
H
2O,
100°C
Ar
12
It is also important to highlight how the same results in terms of
conversion and selectivity were obtained independently of the
cellulosic source of choice. In fact, nanosponges produced with
TOCNF derived from waste sources of cellulose (i.e. recycled
paper pulp and beached Posidonia Oceanica) provided the same
catalytic performances obtained by using cotton linters as starting
biomass material. Moreover, it is worth noticing that cotton linters
itself represent the discharge stock in textile industry.[37]
entry
Ar
Yield (%)a
1
2
3
4
5
6
7
p-BrPh
96
79
97
97
77
92
89
m-BrPh
o-BrPh
p-MePh
o-MeOPh
m-MeOPh
p-MeOPh
CNS as catalyst for Knoevenagel reaction
As a further application, the nanosponges were also tested as
basic catalyst in the Knoevenagel reaction. The product of this
reaction is an olefin resulting from the condensation of an
activated methylene group on a generic aromatic aldehyde,
usually with the use of a secondary amine as catalyst. As the
methylene component, different malonate ester were used,
namely dimethyl (DMM) and diethyl (DEM) (Scheme 8).
Reaction conditions: water 2.5 mL, aldehyde 0.4 mmol, malonate diester 0.48
mmol, cat. 20% mol, MW heating (T = 100°C) for 4 h.
a Conversion was measured with 1H-NMR on the crude with respect unreacted
aldehyde.
6
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001157
ChemCatChem
FULL PAPER
Following these results, the reaction was then repeated with
diethylmalonate (DEM) on different aromatic aldehydes using
water as solvent and heating at 100°C under microwave
irradiation for 4h.The Knoevenagel product was obtained in all the
cases with moderate to very good yields. The product could be
easily isolated without further purification after filtering of the
catalyst and solvent extraction.
Stereoscan 360) at 100/120 Pa with a detector VPSE. The operating
voltage was 20 kV with an electron beam current intensity of 150 pA. The
focal distance was 8 mm. The specimens were used without any treatment
Cellulose Oxidation with TEMPO/NaClO/KBr system
In a 1 L beaker, 100 g of cotton linter paper were minced with gradual
addition of deionized water. Simultaneously, on a 8 L keg placed on a
magnetic stirrer, 2.15 g of tetramethyl-piperidine-N-oxide (TEMPO) and
15.42 g of KBr were dissolved in 2 L of deionized water. Once the paper
was homogeneously blended with water, the solution was transferred in
the keg and water was added in order to obtain a total volume of 5.7 L.
While keeping the solution stirred, a pH meter and two dropping funnels
were installed above the keg, one of the funnels containing 437 mL of
aqueous solution of NaClO and the other 250 mL of NaOH 4 M. NaClO
was then slowly dripped into the solution and pH was monitored to kept it
above 10.5-11 by dripping NaOH 4 M. Once consumed the NaClO, some
of the residual NaOH 4 M was dripped again to maintain basic pH until
stability above 10.5-11, then the solution was left stirring overnight. After
12-16 hours the oxidized cellulose was acidified with concentrated HCl for
aggregation of the cellulose fibers and their easy separation from water.
The oxidized cellulose was then filtered on a Buchner funnel and washed
with deionized water (2 L, 5÷6 times) until neutrality.
Conclusion
In this work, we reported the use of
a nanostructured
heterogeneous catalyst easy to prepare starting from waste
sources containing cellulose, by combining TOCNFs and bPEI.
These nanosponges (CNS) were tested in two typical amine-
catalyzed reactions, the Henry nitroaldol and the Knoevenagel
condensation. In the first case an unexpected but remarkable
selectivity toward different products could be observed depending
on the reaction conditions. In the second case, high yields could
be obtained with different substrates. In both the reactions, the
use of microwave irradiation was crucial to speed up the reaction
still maintaining the catalyst preserved and active for up to five
cycles. Notably, this material can be obtained in high amount and
short time, making it a good candidate for the scale-up of these
reactions. Further investigations on this promising material will be
done in order to explore the scale-up feasibility and the application
to other reactions.
TEMPO-Oxidized Cellulose Nanofibers (TOCNFs) Titration
Estimation of the obtained concentration of carboxylic acids on the
cellulose structure was performed with NaOH using phenolphthalein as
colorimetric indicator. The change of color happens around pH 9.5, when
all the carboxylate groups have gone through base hydrolysis.
NaOH Titration
NaOH solution was first titrated by means of potassium hydrogen phthalate.
218.6 mg were dissolved in water, a drop of phenolphthalein solution 10
mM in acetonitrile was added and then the NaOH solution was dripped in
the beaker under continuous stirring. After neutralization of the phthalate
with 11.35 mL of NaOH, the molar concentration of the solution was
calculated to be 0.0943 M.
Experimental Section
The reagents were purchased from Sigma-Aldrich. The starting material
(cotton linter) was obtained from Bartoli paper factory (Lucca, Italy).
Deionized water was produced within the laboratories with a Millipore Elix®
Deionizer with Progard® S2 ion exchange resins. All 1H-NMR spectra were
recorded on a 400 MHz Brüker NMR spectrometer. Microwave reactions
were conducted in a Biotage® Initiator+. Other equipment used in the
procedures include a Branson SFX250 Sonicator and a SP Scientific
BenchTop Pro Lyophilizer. Scanning electron microscopy (SEM) was
TOCNFs Titration
This solution was then used to titrate the oxidized cellulose obtained with
the TEMPO/NaClO/KBr system: 616.0 mg of TOCNF were dispersed in
.
performed using
a variable-pressure instrument (SEM Cambridge
Scheme 8. Scheme of the equations used for the titration of the NS.
7
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001157
ChemCatChem
FULL PAPER
water and sonicated to obtain a homogeneous water/cellulose solution. A
drop of the same phenolphthalein solution was added and then the NaOH
solution previously titrated was dripped in the beaker under continuous
stirring: 9,75 mL were required to neutralize the carboxylic acids of the
oxidized cellulose. The concentration of the carboxylic groups was
calculated by means of the equation 1 (see Scheme 9).
NS Titration
27 mg of NS were added to 10 mL HCl (0.0962 M). The mixture was
shaken for 16 hours in a sealed container and then filtered to remove the
solid from the solution. Immediately after filtration, to the solution was
added a drop of phenolphthalein solution 10 mM in acetonitrile and then
HCl was titrated with NaOH (previously titrated, 0.1114 M), using a seal of
parafilm in order to avoid probable HCl evaporation and so change in acid
concentration. The consumed HCl by NS was accounted to be equivalent
with the amount of amine functions on their surface. Consequently, the
difference between initial moles of HCl and moles of HCl consumed by
NaOH titration represents the amount of amine functions on the NS. So,
this amount was calculated following equation 3 (see Scheme 9), where
Ca, Va and Cb and Vb were the concentration and volume of the initial acid
as well base solution and the sample weight represents the dry weight of
NS subjected to the titration.
Synthesis of the Nanosponge (NS)
In a 1 L becker, 3.5 g of oxidized cellulose were suspended in 140 mL
deionized water, in order to obtain a 2.5% w/v solution. While mixing it,
granular NaOH in a stoichiometric quantity was added to the solution to
obtain basic pH (evaluated with litmus paper), obtaining a slightly yellow
viscous solution. The beaker was placed in an ice bath and the solution
was sonicated to further promote the separation of the nanofibers,
obtaining a transparent and more homogeneous solution. This latter was
then acidified with concentrated HCl and filtered on a Buchner funnel under
vacuum, then washed and re-suspended. This procedure was repeated
with deionized water until a neutral pH was reached. Wet cellulose was
removed from the filter paper and weighted; 3.5 g were assumed to
correspond to the starting cellulose, while the remaining mass represented
the residual water content. The remaining water necessary to obtain a 2.8-
3% w/v TOCNFs solution for the next phase was split up in three quotas,
in order to re-suspend the cellulose and dissolve the cross-linking polymer
(branched polyethylenimine, bPEI 25kDa) and the co-reticulant agent
(citric acid monohydrate), in three separate batches. The amount of the
co-reticulant was calculated with the equation 2 (see Scheme 9).
Once dissolved in water, the cross-linker and the co-reticulant agents were
slowly added to the cellulose solution, while continuously stirring until
obtaining a white and homogeneous hydrogel, which was placed in 24-
wells well-plates (about 15.6 mm diameter, half-filled with about 2 mL
each) and quickly frozen at -35°C. After a few hours, the well-plates were
moved to the freeze-dryer, which allowed water removal by sublimation,
thus retaining the internal structure of the freezed hydrogels. At the end of
the process, white cylindrical-shaped spongy aerogels were obtained.
They were removed from the wells and placed in the laboratory oven, at
the initial temperature of 55°C. The temperature was then slowly raised up
to 102°C in 4 hours, in order to avoid sudden overheating and prevent the
risk of burning the material. Above 100°C the water formed due to the
reticulation reactions between carboxylic moieties and ammines
(formation of amidic bonds) was quickly evaporated, favoring the formation
of amidic bonds. The sponges were removed from the oven after 16 hours.
Their aspect was yellow and their texture slightly soft but hardened while
cooling down. If not kept in an anhydrous atmosphere, they tended to
adsorb water and become softer. At the end of the process, nanosponges
were washed 6 times with deionized water and the last time with ethanol
to eliminate the excess of bPEI.
Catalytic Activity
Henry reaction: 50 mg of the substituted aldehyde, nitromethane (2 or 5
equivalents with methanol or water as solvent), 2.5 mL of the solvent
(methanol, water or nitromethane) and the desired amount of powdered
nanosponge (CNS) catalyst were added in a 5 mL-microwave vial
equipped with a magnetic stirrer. The reaction was heated at the desired
temperature (from 40°C to 105°C) for the required reaction time.
Alternatively, when conventional heating was used, the reaction was
performed in a two-neck round-bottom flask equipped with a magnetic
stirrer, oil bath, and water condenser. Similar amount of reagents, catalyst
(CNS) and solvent were used. The reaction mixture was stirred at different
temperature for different reaction times. After the reaction was complete,
the catalyst was filtered and washed three times with 5 mL of methanol.
The solvent was then removed under reduced pressure and the crude
were analyzed by NMR on a Brüker NMR spectrometer, acquiring 1H-
spectra in CDCl3 (see SI for further information). NMR analytical data for
all the known compounds were in agreement with the literature.[38] A full
NMR characterization of compound 7 is reported in the SI.
Knoevenagel reaction: 50 mg of the substituted aldehyde, 1.2 equivalent
of diethylmalonate (DEM), 2.5 mL of the solvent (water) and 0.2 molar
equivalents of the powdered nanosponge (CNS) catalyst were added in a
5-mL-microwave vial equipped with a magnetic stirrer. The reaction was
heated at 100°C for 4h. After the reaction was complete, the catalyst was
filtered and washed three times with 5 mL of methanol. The solvent was
then removed under reduced pressure and the crude were analyzed by
NMR on a Brüker NMR spectrometer, acquiring 1H-spectra in CDCl3. NMR
analytical data for all the known compounds were in agreement with the
literature.[39]
Acknowledgements
Nanosponge Titration
Laura Riva thanks Innovhub – Stazioni Sperimentali per l’Industria
– Area Carta, Cartoni e Paste per Carta for financial support.
Estimation of the obtained concentration of amino groups on the surface
of NS was performed with HCl and NaOH using phenolphthalein as
colorimetric indicator.
Keywords: • nanocellulose • aerogels • polyethylenimine •
reusable catalyst • Henry reaction • Knoevenagel reaction •
heterogeneous catalysis
NaOH Titration
NaOH solution was first titrated by means of potassium hydrogen phthalate.
179.7 mg were dissolved in water, a drop of phenolphthalein solution 10
mM in acetonitrile was added and then the NaOH solution was dripped in
the beaker under continuous stirring. After neutralization of the phthalate
with 7.9 mL of NaOH, the molar concentration of the solution was
calculated to be 0.1114 M.
References
[1]
[2]
A. P. Wight, M. E. Davis, Chem. Rev. 2002, 102(10), 3589–3614.
H. Chang, Q. Li, X. Cui, H. Wang, Z. Bu, C. Qiao, T. Lin, J. CO2 Util.
2018, 24, 174-179.
HCl Titration
[3]
[4]
J. Tharun, D. W. Kim, R. Roshan, Y. Hwang, D. W. Park, Catal. Commun.
2013, 31, 62-65.
To 15.0 mL of HCl solution, a drop of phenolphthalein solution 10 mM in
acetonitrile was added and then the NaOH solution previously titrated was
dripped in the beaker under continuous stirring. After neutralization of the
phthalate with 7.9 mL of NaOH, the molar concentration of the solution
was calculated to be 0.0962 M.
L. Guo, R. Dou, Y. Wu, R. Zhang, L. Wang, Y. Wang, Z. Gong, J. Chen,
X. Wu, ACS Sustain. Chem. Eng, 2019, 7(19), 16585–16594.
G. Biliuta, S. Coseri, Coord. Chem. Rev. 2019, 383, 155-173.
B. Thomas, M. C. Raj, B. K. Athira, H. M. Rubiyah, J. Joy, A. Moores, G.
L. Drisko, C. Sanchez, Chem. Rev. 2018, 118(24), 11575–11625.
A. Maleki, E. Akhlaghi, R. Paydar, Appl. Organomet. Chem. 2016, 30(6),
[5]
[6]
[7]
8
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001157
ChemCatChem
FULL PAPER
382-386.
[8]
J. Liu, A. Plog, P. Groszewicz, L. Zhao, Y. Xu, H. Breitzke, A. Stark, R.
Hoffmann, T. Gutmann, K. Zhang, et al., Chem. - A Eur. J. 2015, 21(35),
12414-12420.
[9]
F. Osanlou, F. Nemati, S. Sabaqian, Res. Chem. Intermed. 2017, 43,
2159–2174.
[10] G. Pierre, C. Punta, C. Delattre, L. Melone, P. Dubessay, A. Fiorati, N.
Pastori, Y. M. Galante, P. Michaud, Carbohydr. Polym. 2017, 165, 71-85.
[11] L. Melone, B. Rossi, N. Pastori, W. Panzeri, A. Mele, C. Punta,
Chempluschem 2015, 80(9), 1408-1415.
[12] I. Corsi, A. Fiorati, G. Grassi, I. Bartolozzi, T. Daddi, L. Melone, C. Punta,
Materials (Basel). 2018, 11(7), 1228.
[13] L. Riva, N. Pastori, A. Panozzo, M. Antonelli, C. Punta, Nanomaterials
2020, 10(8), 1570.
[14] I. Bartolozzi, T. Daddi, C. Punta, A. Fiorati, F. Iraldo, J. Ind. Ecol. 2020,
24(1), 101-115.
[15] A. Fiorati, G. Grassi, A. Graziano, G. Liberatori, N. Pastori, L. Melone, L.
Bonciani, L. Pontorno, C. Punta, I. Corsi, J. Clean. Prod. 2020, 246.
[16] G. Liberatori, G. Grassi, P. Guidi, M. Bernardeschi, A. Fiorati, V. Scarcelli,
M. Genovese, C. Faleri, G. Protano, G. Frenzilli, et al., 2020, 10(7), 1283.
[17] A. Fiorati, G. Turco, A. Travan, E. Caneva, N. Pastori, M. Cametti, C.
Punta, L. Melone, Chempluschem 2017, 82(6), 848-858.
[18] L. Panzella, L. Melone, A. Pezzella, B. Rossi, N. Pastori, M. Perfetti, G.
D’Errico, C. Punta, M. D’Ischia, Biomacromolecules 2016, 17(2), 564–
571.
[19] L. Melone, L. Altomare, I. Alfieri, A. Lorenzi, L. De Nardo, C. Punta, J.
Photochem. Photobiol. A Chem. 2013, 261, 53-60.
[20] L. Riva, A. Fiorati, A. Sganappa, L. Melone, C. Punta, M. Cametti,
Chempluschem 2019, 84(10), 1512-1518.
[21] L. Melone, S. Bonafede, D. Tushi, C. Punta, M. Cametti, RSC Adv. 2015,
5, 83197-83205.
[22] J. Han, Y. Xu, Y. Su, X. She, X. Pan, Catal. Commun. 2008, 9(10), 2077-
2079.
[23] C. Paun, J. Barklie, P. Goodrich, H. Q. N. Gunaratne, A. McKeown, V. I.
Pârvulescu, C. Hardacre, J. Mol. Catal. A Chem. 2007, 269(1-2), 64-71.
[24] A. Karmakar, G. M. D. M. Rúbio, M. F. C. Guedes da Silva, A. J. L.
Pombeiro, ChemistryOpen 2018, 7(11), 865-877.
[25] S. S. Ganesan, A. Ganesan, J. Kothandapani, Synlett 2014, 25(13):
1847-1850.
[26] W. Jin, X. Li, B. Wan, J. Org. Chem. 2011, 76, 484–491.
[27] J. D. White, S. Shaw, Org. Lett. 2012, 14(24), 6270–6273.
[28] A. Isogai, T. Saito, H. Fukuzumi, Nanoscale 2011, 3, 71–85.
[29] G. Paladini, V. Venuti, L. Almásy, L. Melone, V. Crupi, D. Majolino, N.
Pastori, A. Fiorati, C. Punta, Cellulose 2019, 26, 9005–9019.
[30] G. Paladini, V. Venuti, V. Crupi, A. Fiorati, C. Punta, Cellulose 2020,
https://doi.org/10.1007/s10570-020-03380-7.
[31] L. Henry, Bull. Acad. R. Belg. 1896, 32, 33.
[32] M. Mokhtar, B. F. A. Alhashedi, H. A. Kashmery, N. S. Ahmed, T. S.
Saleh and K. Narasimharao, ACS Omega, 2020, 5, 6532–6544.
[33] a) T. Jiang, H. Gao, B. Han, G. Zhao, Y. Chang, W. Wu, L. Gao and G.
Yang, Tetrahedron Lett., 2004, 45, 2699–2701. b)
R. Ballini, L.
Barboni, F. Fringuelli, A. Palmieri, F. Pizzo and L. Vaccaro, Green Chem.,
2007, 9, 823–838
[34] R. S. Vanna, R. Dahiya and S. Kumar, Tetrahedron Lett., 1997, 38,
5131–5134
[35] S. Payra, A. Saha and S. Banerjee, RSC Adv., 2016, 6, 95951–95956.
[36] M. A. Torres, M. C. Rezende and B. Cassels, Synth. Commun., 1995, 25,
1239–1247.
[37] A. Sczostak, Macromol. Symp. 2009, 280(1), 45-53.
[38] a) S. Roshandel, L. Gurung, T. Mathew and G. K. S. Prakash,
Tetrahedron Lett., 2017, 58, 2842–2845. b) S. Jyoti Singha Roy and S.
Mukherjee, Chem. Commun., 2014, 50, 121–123. c) P. Bora, P. P. Bora,
B. Wahlang and G. Bez, Can. J. Chem., 2017, 95, 1261–1266. d) K.
Venkatanna, S. Yeswanth Kumar, M. Karthick, R. Padmanaban and C.
Ramaraj Ramanathan, Org. Biomol. Chem., 2019, 17, 4077–4086. e) D.
Xu, Q. Sun, Z. Quan, W. Sun and X. Wang, Tetrahedron Asymmetry,
2017, 28, 954–963. f) E. Gao, M. Li, L. Duan, L. Li and Y. M. Li,
Tetrahedron, 2019, 75, 130492.
[39] a) S. Ghosh and C. K. Jana, Org. Biomol. Chem., 2019, 17, 10153–
10157. b) J. Wang, Y. Zhou, L. Zhang, Z. Li, X. Chen and H. Liu, Org.
Lett., 2013, 15, 1508–1511.
9
This article is protected by copyright. All rights reserved.

10.1002/cctc.202001157
ChemCatChem
FULL PAPER
Entry for the Table of Contents
Insert graphic for Table of Contents here. (max 11.5 x 2.5 cm)
Insert text for Table of Contents here. (max 450 characters including spaces)
Heterogeneous catalysis has many advantages and the use of derivatives from natural biomass is increasing also in this field, with
several examples of heterogeneous cellulose-based catalysts. Here we present a class of nanostructured aerogels obtained from
TEMPO-oxidized cellulose nanofibers derived from waste sources and branched polyethylenimine, proposing them as innovative eco-
safe heterogeneous catalysts for Henry and Knoevenagel reactions.
10
This article is protected by copyright. All rights reserved.