Calcium and nitrogen species loaded into SBA-15-a promising catalyst tested in Knoevenagel condensation

Article abstract of DOI:10.1039/d0dt01621g

Mesoporous silica of the SBA-15 type was used as a support for basic active centers generated by the incorporation of calcium species and (3-aminopropylo)trimethoxysilane (APTMS) or imidazole. The samples were characterized by low temperature N2 adsorption/desorption, XRD, XPS, FTIR spectroscopy, CO2-TPD, and elemental and thermal analyses. Calcium containing samples were analysed in 2,5-hexanedione dehydration and cyclization, while the activities of all the samples were examined in Knoevenagel condensation between benzaldehyde and malononitrile. It was demonstrated that the calcium species interacted with a silica support increasing the stabilization of organosilanes on the SBA-15 surface. A very high activity of the catalysts in Knoevenagel condensation indicated a synergistic interaction between calcium and the organic modifiers.

Full text of DOI:10.1039/d0dt01621g

Dalton
Transactions
View Article Online
View Journal
PAPER
Calcium and nitrogen species loaded into
SBA-15 – a promising catalyst tested in
Knoevenagel condensation†
Cite this: DOI: 10.1039/d0dt01621g
Dorota Kryszak, Katarzyna Stawicka * and Maciej Trejda
Mesoporous silica of the SBA-15 type was used as a support for basic active centers generated by the
incorporation of calcium species and (3-aminopropylo)trimethoxysilane (APTMS) or imidazole. The
samples were characterized by low temperature N2 adsorption/desorption, XRD, XPS, FTIR spectroscopy,
CO2-TPD, and elemental and thermal analyses. Calcium containing samples were analysed in 2,5-hexa-
nedione dehydration and cyclization, while the activities of all the samples were examined in Knoevenagel
condensation between benzaldehyde and malononitrile. It was demonstrated that the calcium species
interacted with a silica support increasing the stabilization of organosilanes on the SBA-15 surface. A very
high activity of the catalysts in Knoevenagel condensation indicated a synergistic interaction between
calcium and the organic modifiers.
Received 4th May 2020,
Accepted 23rd June 2020
DOI: 10.1039/d0dt01621g
rsc.li/dalton
densation reaction, including Lewis acids, weak bases, ionic
liquids, exchange resins, modified zeolites or mesoporous
Introduction
Knoevenagel condensation is one of the most important reac- silica and metallosilicates.6–12
tions in organic chemistry. It involves the condensation of an
Thanks to their large surface area allowing the incorpor-
aldehyde or ketone (e.g. benzaldehyde and furfural) with an ation of large amounts of an active phase, mesoporous silicas
active methylene compound (e.g. ethyl cyanoactetate, ethyl are commonly used as supports for the synthesis of catalysts.
acetoacetate, and malononitrile) followed by elimination of One of the methods used to generate basicity in mesoporous
one water molecule to form a C–C bond between substrates.1 silicas is their modification with specific organic compounds
Methylene compounds show different structures such as Z1– containing basic active species. In recent years a lot of basic
CH2–Z2, Z1–CHR–Z2 and Z1–CHR1R2, where Z1 and Z2 are elec- modifiers have been applied for the synthesis of catalysts dedi-
tron-withdrawing groups. These groups promote the dis- cated to the Knoevenagel condensation. Among them are
sociation of a proton from the neighboring methylene group amines,13–15 imidazoles,16 1,2,3-triazol-4-ylmethanamine,17
as a result of the destabilization of a methylene bridge in the melamine,18 urea,19 cytosine20 and polyvinylamine.21 The basi-
active methylene compound.2 The final products of conden- city of the above-mentioned modifiers results from the pres-
sation, α,β-unsaturated compounds, have found numerous ence of a basic nitrogen atom which acts as the Brønsted basic
applications. They can be used in the production of fine site.22 Another group of modifiers applied for the synthesis of
chemicals or intermediates in the pharmaceutical and cos- active basic catalysts based on a silica support is basic oxides,
metic industry.3–5 Knoevenagel condensation can also be used e.g. calcium oxide23–25 and cesium oxide.26 The basicity of
as a test reaction for probing the basicity of solid catalysts. A such modifiers is due to the presence of basic oxygen species.
number of materials have been tested as catalysts in this con- As illustrated in Table 1 there are many catalysts modified with
basic nitrogen species or calcium oxide which were tested in
Knoevenagel condensation. The different substrate concen-
trations, catalyst loading, reaction temperature and reaction
time were investigated in Knoevenagel condensation as well as
Adam Mickiewicz University in Poznan, Faculty of Chemistry, ul. Uniwersytetu
Poznanskiego 8, 61-614 Poznan, Poland. E-mail: kstawick@amu.edu.pl
solvent addition.
†Electronic supplementary information (ESI) available: Binding energies of the
O 1s, Si 2p, Ca 2p and N 1s region of obtained samples, XRD patterns of silica
and metallosilicates, the FTIR spectra of synthesized samples in the range of
4000–400 cm−1, XP spectra of N 1s for SBA-15-AP and SBA-15-Im, the scheme of
possibilities of amine and imidazole anchored in mesoporous silicas and metal-
losilicates, ATR-FTIR spectra of Ca/SBA-15_550-Im before and after its reuse and
wash with acetone. See DOI: 10.1039/D0DT01621G
It has been proven that the presence of additional species
on the support, such as metal oxides, besides the modifiers
with a basic nitrogen atom, can affect the activity of catalyst.
The centers of both types can promote the Knoevenagel con-
densation. Basic centers, which come from basic nitrogen or
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Table 1 Examples of Knoevenagel condensation performed over the supports modified with base-silanes or calcium oxide
Yield of the
main product
Catalyst
Reactants
Reaction conditions
Ref.
14
NH2(50%)-MIL-53
Benzaldehyde (B)
Malononitrile (M)
5 mmol (B) : 5 mmol (M)
10 ml DMF as solvent
5.3 mg ml−1 of catalysts
5 h, 80 °C
5 mmol (B) : 5 mmol (M)
10 ml DMF as solvent
5.3 mg ml−1 of catalysts
5 h, 80 °C
89
71
99
94
NH2(100%)-MIL-53
Benzaldehyde (B)
Malononitrile (M)
14
23
15
FSM-16(amino)
Benzaldehyde (B)
Cyanoacetate (C)
1.25 mmol (B) : 1.30 mmol (C)
5 ml of toluene as catalyst
0.1 g of catalyst
20 h, 80 °C
N-GO-1.00 (GO – graphene oxide)
Benzaldehyde (B)
Malononitrile (M)
10 mmol (B) : 10 mmol (M)
5 ml CH3CN as solvent
0.1 g of catalyst
4 h, 40 °C
30 mmol (B) : 30 mmol (M)
0.2 g of catalyst
6 h, 60 °C
10 mmol (B) : 10 mmol (M)
0.1 g of catalyst
MCM-41-mela
Urea(4.8)/SBA-15
CaO
Benzaldehyde (B)
Malononitrile (M)
91
95
84
18
19
24
Benzaldehyde (B)
Malononitrile (M)
4 h, 40 °C
Benzaldehyde (B)
Malononitrile (M)
10 mmol (B) : 12 mmol (M)
5 ml EtOH as solvent
0.44 g of catalyst
2 min, microwave irradiation
1.25 mmol (B) : 1.30 mmol (C)
5 ml of toluene as catalyst
0.2 g of catalyst
20 h, 80 °C
1.5 mmol (B) : 1 mmol (E)
0.3 g of catalyst
Ca-FSM-16
Benzaldehyde (B)
Cyanoacetate (C)
64
72
23
25
FAC (fly ash supported calcium oxide)
Benzaldehyde (B)
Ethylcyanoacetate (E)
2 h, 90 °C
oxygen, are active in the adsorption of a proton from a methyl- silica. The impact of metals, niobium and cerium, depends on
ene group, which leads to the generation of a carbanion. This the pKa value of the methylene compound used for the
carbanion is able to attack the carbonyl group of an aldehyde Knoevenagel condensation. At lower pKa (9) niobium modified
or ketone. The additional role of metal species, which can act samples showed higher activity, whereas with increasing pKa
as Lewis acid sites (LAS) is to activate the carbonyl group in (from 9 to 13.3) the impact of cerium species dominated. The
the aldehyde or ketone. As a result of this reaction a carbo- metallic modifier affected not only the activity of imidazole
cation is formed.10 The modification with both metal species modified samples but also the selectivity.12 An increase in the
and nitrogen containing compounds was employed in the syn- activity of bifunctional catalysts in the Knoevenagel conden-
thesis of SBA-15 modified with zirconium, molybdenum or sation has also been observed for the samples modified with
niobium and (3-aminopropyl)trimethoxysilane, which led to gold species. The modification of amine-grafted MCF and Nb/
the formation of a very active bifunctional catalyst. The MCF with gold species increased significantly the conversion
samples synthesized were tested in the Knoevenagel conden- of 2–4-dichlorobenzaldehyde with ethyl cyanoacetate. It was
sation between ethyl acetoacetate, diethyl malonate or ethyl concluded that negatively charged gold nanoparticles and gold
cyanoacetate and benzaldehyde. The basic oxygen species cations could interact with 2,4-dichlorobenzaldehyde and this
from the metal oxides allow the abstraction of a proton from increased the effectiveness of the attack of the carbanion
the methylene group in ethyl acetoacetate or diethyl malonate formed from ethyl cyanobenzaldehyde on the carbon atom
but not from ethyl cyanoactetate. The abstraction of a proton from the aldehyde.27 Recently, a positive impact of tantalum
from this substrate is possible only by anchored amine species on the catalytic performance of MCF modified with
groups.11 The metal species loaded into silica may also influ- alkaline metals has been observed in comparison with those
ence the activity of other basic modifiers such as imidazole. In obtained over MCF modified with niobium and alkaline
our recent paper we have demonstrated that the Lewis acidity metals.28
originating from niobium species increased the catalytic
In this work we have applied for the first time calcium
activity of imidazole anchored on mesoporous silica.16 Not modified SBA-15 as a support for (3-aminopropyl)trimethoxysi-
only niobium but also another metal – cerium – play an impor- lane or imidazole species. Calcium was incorporated in two
tant role in establishing the basicity of imidazole modified ways, i.e. by a one-pot synthesis approach and by a post-syn-
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
thesis impregnation method. The use of these two synthesis was filtered, washed with 130 cm3 of toluene, 65 cm3 of dis-
methods allows obtaining the samples with different calcium tilled water and 15 cm3 of acetonitrile, respectively, and dried
species loaded on the silica surface and with different basic at 100 °C. The samples obtained were denoted as SBA-15-AP,
strengths. In our earlier work we have proved that an imidazole CaSBA-15-AP, Ca/SBA-15_550-AP and Ca/SBA-15_700-AP,
containing catalyst exhibited higher activity in the respectively.
Knoevenagel condensation between benzaldehyde and ethyl
Modification with imidazole. First, the sample (1 g) was
cyanoacetate than that between triazole and aminotriazole dried overnight at 100 °C and then immersed in a flask con-
modified samples.12 No direct comparison of the activity of taining 15 cm3 of dried toluene and 1 cm3 of (3-chloropropyl)
(3-aminopropyl)trimethoxysilane and imidazole anchored into trimethoxysilane (ClPTMS). The solution obtained was stirred
calcium modified SBA-15 has been performed yet, according slowly at 110 °C for 24 h. After that the solid product was fil-
to our knowledge.
tered, washed with 15 cm3 of toluene and dried at 100 °C. Next
the sample modified with ClPTMS (1 g) was immersed in a
flask containing 15 cm3 of toluene, 0.75 g of imidazole and
1.42 cm3 of triethylamine (for the neutralization of HCl
obtained during the modification). The mixture obtained was
then heated at 110 °C for 48 h. The product was filtered,
Experimental
Synthesis of catalysts
Synthesis of calcium containing SBA-15. The calcium modi- washed with 30 cm3 of dichloromethane and 10 cm3 of
fied SBA-15 sample was prepared according to ref.29. First, toluene and dried at 100 °C. The samples obtained were
Pluronic P123 (4 g) was dissolved in a HCl solution (8.76 g of denoted as SBA-15-Im, CaSBA-15-Im, Ca/SBA-15_550-Im and
HCl and 110 g of distilled water) at 35–40 °C in a polypropyl- Ca/SBA-15_700-Im, respectively.
ene bottle. Then calcium nitrate tetrahydrate (1.595 g) was dis-
solved in distilled water (31.24 g) and added to the synthesis
Catalyst characterization
mixture. After 10 minutes TEOS (8.527 g) was dropwise added The obtained samples were characterized by XRD, N2 ads./
to the synthesis mixture. The mixture was then stirred for 20 h des., XPS, FTIR spectroscopy, CO2-TPD, elemental analysis,
at the same temperature and heated in an oven at 100 °C for and DTA/TGA and analysed in the cyclization and dehydration
24 h. The mixture obtained was then evaporated using a of 2,5-hexanedione.
vacuum evaporator until the water was removed. The solid
The X-ray diffraction measurements (XRD) were performed
product was dried overnight at RT (room temperature) and at using a Bruker AXS D8 Advance diffractometer using Cu Kα
100 °C for 24 h and finally calcined at 550 °C for 8 h. The radiation (γ = 0.154 nm) with a step of 0.02° and 0.05° at a
sample obtained was denoted as CaSBA-15.
small and wide-angle, respectively.
Two other calcium modified samples were prepared by an
N2 adsorption/desorption measurements were performed
impregnation method. For this purpose the SBA-15 support using a Micromeritics ASAP 2020 apparatus. Before the ana-
was synthesized as follows. Pluronic P123 (16 g) was dissolved lysis the sample (ca. 100 mg) was outgassed under vacuum at
in a 0.7 M HCl solution (600 g) at 35–40 °C in a polypropylene 300 °C for 8 h (supports) or 120 °C for 10 h (organosilane
bottle. After the surfactant dissolution, TEOS (34.108 g) was modified samples). The surface area was calculated using the
added. The mixture was then stirred for 20 h at the same temp- BET method. The pore volume and diameter were determined
erature and then heated in an oven at 100 °C for 24 h. The by the DFT method appropriate for mesoporous silicas with a
solid product was filtered, washed with 1200 cm3 of distilled cylindrical pore system attached to the MicroActive software
water and dried at RT. The template was removed by calcina- (Version 4.0) from Micromeritics.
tion at 500 °C for 8 h under static conditions. Next, 3 g of
X-ray photoelectron measurements (XPS) were performed
dried silica (which was previously evaporated at 80 °C for 1 h) using an Ultra High Vacuum System based on a Phoibos 150
was added into the flask together with calcium nitrate tetrahy- NAP analyzer (Specs). The materials were irradiated with Al Kα
drate (1.624 g) previously dissolved in distilled water (30 cm3). radiation (1486.6 eV). The binding energies were referenced to
The mixture obtained was rotated in a vacuum evaporator at C 1s at 284.6 eV. The spectra obtained were analyzed using
80 °C until the water removal. Then the solid material was CasaXPS. The deconvolution of the spectra was performed
dried for 12 h at 120 °C and calcined at 550 °C for 8 h (temp- using the Gaussian function.
erature ramp 1 °C min−1). The second sample was calcined at
Infrared spectra were acquired with a Vertex 70 spectro-
700 °C for 5 h (temperature ramp 2 °C min−1). The samples meter in the range from 4000 cm−1 to 400 cm−1 (resolution
obtained were denoted as Ca/SBA-15_550 and Ca/SBA-15_700, 4 cm−1 and the number of scans = 64) using an in situ vacuum
respectively.
cell. Firstly, the materials were pressed under low pressure into
Modification with (3-aminopropyl)trimethoxysilane. First, a thin wafer of ca. 10 mg cm−2 and immersed in a cell. The
the sample (1 g) was dried overnight at 100 °C and then samples were activated at 100 °C for 1 h under vacuum and
immersed in a flask containing 100 cm3 of dried toluene. then the FTIR spectrum of the catalyst was recorded. The
Then 0.96 cm3 of (3-aminopropyl)trimethoxysilane (APTMS) FTIR-ATR (attenuated total reflectance) spectra were recorded
was introduced to the solution. The solution obtained was for the powdered samples using a Platinum ATR accessory
stirred slowly at 110 °C for 24 h. After that the solid product (Bruker).
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
The temperature-programmed desorption of CO2 was metallosilicate supports. The modification employed was
carried out using a quartz tube reactor. Before the measure- expected to increase the basicity of the obtained catalysts as a
ment the sample (0.03 g – particles with 0.5–1 mm diameter) result of the interaction between basic oxygen in CaO and
was activated at 250 °C for 2 h under a helium flow (40 cm3 basic nitrogen in organosilanes, relative to that of silica modi-
min−1). Then carbon dioxide was adsorbed for 30 min. To do fied with a single modifier, alkali oxide, APTMS or Im. The
so the sample was purged with 10 vol% of CO2 in He (cm3 samples synthesized are dedicated to the reactions requiring
min−1) at RT. After that the sample was kept under a helium the presence of basic active sites. Thus, in this paper we
flow to remove unbound CO2. Then the desorption of carbon decided to investigate the activity of the obtained catalysts in
dioxide was proceeded up to 800 °C (temp. ramp 10 °C min−1). the Knoevenagel condensation between benzaldehyde and
Carbon dioxide was detected by using a thermal conductivity malononitrile under mild reaction conditions.
detector.
Texture/structural parameters of calcium modified silica
SBA-15
Elemental analyses were carried out using an Elementar
Analyser Vario EL III.
DTA/TGA measurements were performed in an air atmo-
sphere using an SETARAM SETSYS-12 apparatus in the temp-
erature range of 20–1000 °C at a temperature ramp of 5 °C
The synthesis of SBA-15 allowed obtaining the mesoporous
support with an ordered hexagonal structure as confirmed by
the N2 adsorption/desorption isotherm and XRD pattern
(Fig. 1 and 2, respectively). The isotherm obtained was of type
IVa typical of mesoporous solids with a hysteresis loop of type
H1, characteristic of the materials which show a narrow range
of uniform pores.30 The low angle XRD pattern of synthesized
SBA-15 shows diffraction peaks at 2 theta values of ca. 1.0°,
1.7° and 1.9°. The most intense peak at a 2 theta value of ca.
1° was assigned to the presence of regular interspace distances
between the walls of mesoporous channels, whereas the other
two peaks evidenced the long range ordering of the silica
structure.31
The addition of a calcium precursor into SBA-15 by a one-
pot synthesis approach (CaSBA-15) or by post-synthesis
impregnation (Ca/SBA-15_550 and Ca/SBA-15_700) did not
affect significantly the mesoporous structure of the support.
All modified materials show the isotherms of type IVa typical
of mesoporous solids (Fig. 1) and low angle XRD patterns with
diffraction peaks assigned to the well-ordered SBA-15 structure
(Fig. 2). Interestingly, the wide angle XRD patterns of the
calcium modified silicas do not show diffraction peaks typical
of calcium oxide (Fig. 2). For all materials only a wide peak at a
2 theta value of 15–35° was detected. This diffraction peak in
min−1
.
The cyclization and dehydration of 2,5-hexanedione was
performed in a tabular down-flow reactor under atmospheric
pressure using nitrogen as a carrier gas. The catalyst bed
(0.05 g – particles with 0.5–1 mm diameter) was first activated
for 2 h at 400 °C under a nitrogen flow (40 cm3 min−1). Next
the temperature was decreased to 350 °C and 0.5 cm3 of 2,5-
hexanedione was passed continuously through the catalyst bed
for 30 min by using a pump system (KD Scientific). The reac-
tion products were collected in a cold trap (the mixture of
2-propanol and liquid nitrogen) and analyzed by using an SRI
310 C chromatograph equipped with a capillary DB-1 column
(30 m) and TCD detector.
Catalytic activity
The catalytic activity of all samples was evaluated in solvent
free Knoevenagel condensation with a glass reactor using an
EasyMax workstation. Prior to the reaction the material was
activated at 100 °C overnight. 0.017 g of the catalyst was added
to the equimolecular mixture of benzaldehyde (20 mmol,
2.12 g) and malononitrile (20 mmol, 1.32 g). After the tempera-
ture reached 60 °C the reaction was performed for 30 min.
After that the sample was cooled down to RT and 5 cm3 of
acetone was added to dissolve the solid product of the reac-
tion. The reaction mixture was analyzed by using a gas chro-
matograph (Thermo Scientific) equipped with a 60 m VF-5 ms
capillary column and FID detector. The catalyst was separated
by centrifugation, washed with acetone and applied in the next
catalytic cycle.
Results and discussion
The aim of this work was to synthesize the mesoporous silica
of the SBA-15 type modified with calcium species together
with organosilanes containing a basic nitrogen atom such as
(3-aminopropyl)trimethoxysilane (APTMS) and N-(3-trimethox-
ysilylpropyl)imidazole (Im). The latter modifier was syn-
thesized via a reaction between imidazole and (3-chloropropyl)
trimethoxysilane which was previously anchored to silica and Fig. 1 N2 adsorption/desorption isotherms of the obtained samples.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
Fig. 2 Small and wide angle XRD patterns of the supports, and amine and imidazole modified samples.
the literature is commonly interpreted as the evidence of the for all calcium modified silicas a decrease in the unit-cell para-
amorphous character of the obtained samples.32 The lack of meter (a0) and interspace distance (d100) was observed. This
diffraction peaks typical of the crystalline phase of calcium implies the interaction between the silica and calcium species
species33,34 for the CaSBA-15, Ca/SBA-15_550 and Ca/ which resulted in a shrinkage of the hexagonal structure of
SBA-15_700 samples suggests that the crystals of the calcium SBA-15. To confirm, if the decrease of the unit-cell parameter
species on the surface of the materials studied are too small to and interspace distance in the calcium modified materials was
be detected by this technique or they are not formed upon cal- caused by the interaction of the calcium species with silica (i.e.
cination performed below 700 °C, if calcium nitrate is used as not only by the high temperature of metallosilicate calcina-
a modifier of silica, as deduced in ref.35, 36.
tion), the SBA-15 support was heated at 700 °C for 5 h (temp-
Greater differences between SBA-15 and calcium doped erature ramp 2 °C min−1). This material is denoted as
silicas were observed in their texture parameters estimated SBA-15_700. It can be seen that a peak at ca. 1.0° for Ca/
from the nitrogen adsorption and low angle XRD patterns SBA-15_550 and Ca/SBA-15_700 is located at higher value of 2
shown in Table 2. The surface area and pore volume decreased theta than that for SBA-15_700 (Fig. S1†). This clearly demon-
after calcium addition into SBA-15, independently of the tech- strates that the impregnated calcium species into silica inter-
nique used for metal incorporation. A decrease in the pore dia- acted with the support and caused the differences in the inter-
meter was only observed for the samples to which the calcium spaces distance of SBA-15.
precursor was added by the impregnation method. The
The suggested interaction between the calcium precursor
decrease was greater for Ca/SBA-15_700 than that for Ca/ and SBA-15 might be confirmed by using the Si 2p XP spectra
SBA-15_550. This feature should be related to the higher temp- of metallosilicates presented in Fig. 3. The spectrum of meso-
erature of Ca/SBA-15_700 calcination in comparison with that porous silica shows only one band at 103.1 eV, assigned to the
of Ca/SBA-15_550 calcination, which could result in the for- silicon atom bonded to four oxygen atoms. In the spectrum of
mation of bigger agglomerates of the active phase in the pores calcium modified silicas an additional component at
of Ca/SBA-15_700 than those of Ca/SBA-15_500. Interestingly, 101.6–102.2 eV was detected, and it corresponds to the silicon
Table 2 Texture/structural parameters
Sa, m2 g−1
Smicro a, m2 g−1
da, nm
Va, cm3 g−1
dwall b, nm
a0 b, nm
d100
b
Catalyst
SBA-15
SBA-15-AP
SBA-15-Im
CaSBA-15
CaSBA-15-AP
CaSBA-15-Im
Ca/SBA-15_550
Ca/SBA-15_550-AP
Ca/SBA-15_550-Im
Ca/SBA-15_700
Ca/SBA-15_700-AP
Ca/SBA-15_700-Im
888
270
482
446
312
298
457
294
302
415
272
302
283
—
65.9
29.3
—
4.4
65.1
—
9.8
7.9
8.5
9.9
9.2
8.5
9.4
7.8
7.8
8.1
7.3
7.3
0.91
0.31
0.56
0.64
0.46
0.40
0.59
0.39
0.37
0.51
0.34
0.36
1.44
2.95
2.11
1.72
2.37
2.62
1.77
2.66
2.54
1.90
2.76
2.72
10.14
8.91
9.59
8.74
9.29
9.29
8.40
9.29
8.57
8.02
8.82
9.00
8.87
7.71
8.31
7.57
8.04
8.04
7.28
8.04
7.42
6.95
7.64
7.80
—
53.2
3.2
14.4
a Calculated from the adsorption/desorption of nitrogen. b Calculated from XRD.
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fig. 3 XP spectra of O 1s and Si 2p of SBA-15, CaSBA-15, Ca/SBA-15_550 and Ca/SBA-15_700.
Fig. 4 The scheme of defect formation in SBA-15 after calcium species incorporation.
atom with three instead of four oxygen neighbors.37,38 The for-
mation of such an element might be due to the etching of
silicon in the presence of a base as a calcium precursor and
further calcination of the samples at 550 and 700 °C which
resulted in the dehydration of metallosilicates. As a result, the
surface defects in the calcium doped silicas might be formed
as shown in Fig. 4. The presence of such defect centers in pure
silica SBA-15 in the form of the so called E′ centre defined as
Siu• was confirmed by EPR measurements in ref.39.
Moreover, it is seen that the band assigned to the silicon atom
bonded to four oxygen atoms was shifted towards higher BE in
the XP spectra of calcium doped silicas (Fig. 3 and Table S1†),
which furthermore implies the interaction between the silicon
species and the calcium precursor.
The presence of calcium species in the SBA-15 support
caused an increase in the wall thickness as shown in Table 2.
This implies that the modifier is located on the surface of all
metallosilicates. This observation can be confirmed by the
FTIR spectra presented in Fig. 5 (the FTIR spectra in the range
of 4000–400 cm−1 are included in SD – Fig. S2†). It is well
visible that the intensities of the bands typical of the surface
silanol species vibration at 3743 cm−1 and 959 cm−1 (ref. 40)
decreased significantly after the SBA-15 modification with the
calcium precursor, especially for the post-synthetically modi-
fied samples. Thus, it can be concluded that the calcium
species cover or interact with the –SiOH groups on the surface
of the silica support and consequently the wall thickness of
Fig. 5 (A) FT-IR and (B) ATR-FTIR spectra of the supports.
the metallosilicates is increased.
It is well known that the modification of silica with the
calcium precursor leads to the formation of different types of
calcium species on the surface of the silica support, such as typical of the above-mentioned species, additional measure-
CaO, Ca(OH)2 or CaSiO3.29,36,41 As the XRD patterns of the ments using XPS, FTIR spectroscopy and CO2-TPD were per-
obtained materials (Fig. 2) did not show any diffraction peaks formed in order to confirm this issue. The Ca 2p XPS spectra
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
Fig. 6 XP spectra of Ca 2p of CaSBA-15, Ca/SBA-15_550 and Ca/SBA-15_700.
of all metallosilicates presented in Fig. 6 show two well-
resolved components at ca. 351.2 eV and 347.6 eV assigned to
the CaO, Ca(OH)2 or CaCO3 species, dependently on cited
paper.42–44 The presence of CaCO3 on the surface of all
calcium modified silicas was confirmed by the FTIR spectra
shown in Fig. 5 which displayed the bands at 1494 cm−1
,
1422 cm−1 and 867 cm−1 assigned to the vibration mode of the
carbonate groups.29,45 The mentioned bands show lower inten-
sity in the spectrum of Ca/SBA-15_700 in comparison with
those in the spectra of CaSBA-15 and Ca/SBA-15_550. As car-
bonates could not be formed directly from the calcium precur-
sor (Ca(NO3)2) during metallosilicate calcination, the presence
of CaCO3 species on the surface of CaSBA-15, Ca/SBA-15_550
and Ca/SBA-15_700 can be explained by assuming the exist-
ence of the CaO phase on the metallosilicate surface which
reacted with CO2 from air. The presence of calcium oxide on
Fig. 7 CO2-TPD profile of the supports.
the surface of all metallosilicates can be confirmed by the O 1s of defects on the metallosilicate surface, could moreover influ-
XP spectra and CO2-TPD profiles presented in Fig. 3 and 7, ence the formation of calcium silicate as a consequence of the
respectively. The XP spectrum of SBA-15 revealed one com- etching of silicon atoms from the silica surface as concluded
ponent at 532.5 eV assigned to oxygen in the silica support from the Si 2p XPS spectra of metallosilicates shown in Fig. 3.
(–Si–O–Si–),46 while in the XP spectra of metallosilicates an It is worth noting that the peaks typical of CO2 desorption
additional component at ca. 531.1–531.8 eV was detected after from CaO and CaSiO3 are less intense for Ca/SBA-15_700 than
spectral deconvolution. This additional component is typical those for CaSBA-15 and Ca/SBA-15_550. Moreover, for all
of lattice oxygen in metal oxides, as in CaO species.47 It is metallosilicates, differences in the position of the peaks
worth noting that the band assigned to oxygen in silica was characteristic of CO2 desorption from both CaO and CaSiO3
shifted towards higher BE in the XP spectra of all metallosili- species were observed. The mentioned peaks are shifted to
cates, which suggests the interaction between SBA-15 and the higher temperatures as follows: CaSBA-15 > Ca/SBA-15_550 >
calcium precursor (Fig. 3 and Table S1†).
Ca/SBA-15_700. This suggests that the calcium species exhibit
The presence of CaO on the surface of calcium doped higher basicity in the one pot synthesized metallosilicates
silicas might also be confirmed by CO2-TPD analysis (Fig. 7). than in the impregnated silicas.
The CO2 desorption profile of SBA-15 shows only a desorption
The results obtained based on the FTIR spectra (Fig. 5) and
peak at around 60–140 °C which is typical of CO2 desorption CO2-TPD profiles (Fig. 7) indicated that the calcium species in
from weak basic species in the silica support.29,48 The CO2 de- Ca/SBA-15_700 form larger agglomerates in the pores of silica
sorption profiles of all silicas modified with calcium are totally than those in the other calcium doped materials. Moreover, the
different and show two desorption peaks centered in different surface composition investigated by XPS measurements, pre-
regions of the CO2-TPD profile. The more intense peak at sented in Table 3, implies that the calculated Ca/Si molar ratio
around 290–500 °C is characteristic of CO2 desorption from slightly decreases as follows: CaSBA-15 (0.115) > Ca/SBA-15_550
calcium oxide29,49 which thus confirmed the formation of CaO (0.111) > Ca/SBA-15_700 (0.103). These results support the data
species on the surface of metallosilicates as shown in the O 1s obtained from the N2 adsorption/desorption presented in
XP spectra in Fig. 3. The less intense peak in the CO2-TPD pro- Table 2, showing the highest decrease in the pore diameter for
files of calcium doped silicas at ca. 60–250 °C is assigned to Ca/SBA-15_700 as a result of the greater agglomeration of the
different calcium species, in the system studied with CaSiO3.29 calcium precursor in the pores of the SBA-15 support.
Thus, it can be concluded, that the interaction between the
In order to confirm if the calcium species are less agglom-
calcium precursor and SBA-15 support, besides the formation erated and thus more available on the surface of CaSBA-15 and
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Table 3 Composition of the catalyst estimated from elemental analysis
and XPS
angle XRD pattern of SBA-15-Im, three well resolved diffraction
peaks at 2 theta values of 0.9°, 1.6° and 1.8° were observed,
while for SBA-15-AP only one diffraction peak at a 2 theta value
of 1.0° was detected (Fig. 2). Therefore, it can be concluded
that the grafting of amine into SBA-15 leads to better efficiency
of APTMS anchoring than that of imidazole, although two
modification steps should be applied for the latter material. It
is important to note that the amount of amine and imidazole
loaded into SBA-15 differs, depending on the modifier used
for silica anchoring (Table 3). The amount of the amine and
imidazole species in both samples was calculated on the basis
of the detection of nitrogen by elemental analysis and XPS.
Nitrogen was not detected in the calcium modified silicas to
which calcium nitrate was added. Thus, the nitrogen content
in the amine or imidazole modified metallosilicates originates
from both organic modifiers. The imidazole modified samples
could be contaminated with the Et3NH+Cl− salt synthesized in
the reaction between Et3N (added during the silica and metal-
losilicate modification with imidazole species) and HCl
(formed during the modification process). However, the FTIR
spectra of all imidazole modified materials shown in Fig. S2†
do not contain bands typical of the NH+ vibration from tertiary
Elemental analysis
XPS
Nitrogen compound, Nitrogen compound,
Catalyst
mmol g−1
mmol g−1
Ca/Si
SBA-15-AP
SBA-15-Im
CaSBA-15
2.59
0.92
—
2.69
0.86
—
—
—
0.115
0.116
0.105
0.111
0.104
0.091
0.103
0.107
0.108
CaSBA-15-AP
CaSBA-15-Im
Ca/SBA-15_550
Ca/SBA-15_550-AP 1.24
Ca/SBA-15_550-Im 0.86
Ca/SBA-15_700
Ca/SBA-15_700-AP 0.91
Ca/SBA-15_700-Im 0.71
1.18
1.08
—
1.22
0.92
—
1.21
0.91
—
—
0.98
0.82
Table 4 Conversion and selectivity to MCF in the cyclization and de-
hydration of 2,5-hexanedione
Catalyst
Conversion, %
MCP selectivity, %
CaSBA-15
Ca/SBA-15_550
Ca/SBA-15_700
52
40
24
99.8
99.6
99.6
amines in the range of 2770–2380 cm−1 51
Thus, the nitrogen
.
content measured for the imidazole modified samples orig-
inates only from the imidazole species.
Ca/SBA-15_550 than Ca/SBA-15_700, the dehydration and cycli-
According to the results, the grafting of APTMS on SBA-15
zation of 2,5-hexanedione (2,5-HDN) was performed. If the resulted in a greater modifier loading than the modification
tested catalyst has basic active species 3-methyl-2-cyclopente- with imidazole. Thus, a significant decrease in the intensity of
none (MCP) is formed, while in the presence of acid sites 2,5- the XRD diffraction peaks observed for SBA-15-AP could be
dimethylfurane (DMF) is obtained. The results of this reaction due to a high loading of amine species into SBA-15. Obviously,
are presented in Table 4. It is worth mentioning that in the the anchoring of both modifiers did not affect the amorphous
presence of all samples, 3-methyl-2-cyclopentenone only was nature of the obtained samples, as deduced from the wide
obtained, which confirmed the basic character of the obtained angle XRD patterns shown in Fig. 2. However, in the XRD
samples.50 A big difference in the activity of the catalysts was pattern of SBA-15-Im, a new intense diffraction peak at a 2
observed despite a quite similar metal loading of all the theta value of 18° was detected. The presence of this peak
samples. Indeed, the results support the hypothesis that the could be explained by the contamination of the sample with a
calcium species on the surface of Ca/SBA-15_700 are less acces- small amount of Teflon52 from the stirrer during the
sible to 2,5-hexanedione than those in the other samples. As a SBA-15 modification with imidazole species follow by ClPTMS
result, the obtained conversion of 2,5-HDN under Ca/ anchoring.
SBA-15_700 was the lowest among all calcium doped silicas.
The textural parameters of the samples modified with
organic species calculated from the N2 adsorption/desorption
isotherms and small-angle XRD patterns are presented in
Table 2. It can be seen that all parameters decreased after
The impact of APTMS and imidazole on the texture/structure
and surface properties of SBA-15
The anchoring of (3-aminopropyl)trimethoxysilane or N-(3-tri- basic modifier loading into SBA-15 except the wall thickness.
methoxysilylpropyl)imidazole into SBA-15 caused visible The reduction in the surface area, pore volume and diameter
changes in the texture/structural parameters of the catalysts. as well as the unit-cell parameter and interspace distance was
Thus, the shape of the isotherms (Fig. 1) and intensities of greater for SBA-15-AP than that for SBA-15-Im. This feature
diffraction peaks in the small-angle XRD patterns (Fig. 2) of could be due to the higher loading of amine species than that
the samples modified with amine or imidazole species are of imidazole into the silica support, as shown in Table 3. The
different compared to those obtained for unmodified silica. increase in the wall thickness observed for SBA-15 after amine
The volume of adsorbed nitrogen decreased significantly for or imidazole species anchoring implies the formation of a
SBA-15-AP and SBA-15-Im, especially for the amine modified chemical bond between the organic species and silica surface.
silica. However, the N2 adsorption/desorption isotherms of The anchoring of amine and imidazole species on the surface
both samples are still of type IVa with a hysteresis loop of type of SBA-15 was moreover confirmed by the FTIR and N 1s XP
H1, as observed for SBA-15, which means that the mesoporous spectra, shown in Fig. 8 and S3,† respectively. The anchoring
nature of the materials obtained is preserved. In the small- of APTMS was confirmed by the presence of the IR bands at
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
Fig. 8 FT-IR spectra of the amine and imidazole modified samples.
3360 cm−1, 3299 cm−1 and 1594 cm−1 characteristic of the peaks detected for SBA-15-Im could be explained by a different
amine species vibration (Fig. 8),53 and the peak at 399.4 eV in location of the imidazole ring in this sample. The first peak is
the XP spectra assigned to the –NH2 groups (Fig. S3†).54 The related to the decomposition of imidazole connected to the
IR bands typical of the primary amines in SBA-15-AP were propyl chain from ClPTMS, while the second one is assigned
moreover accompanied by the band typical of the –N–H to the decomposition of imidazole anchored directly to the
vibration at 1647 cm−1 and by the bands assigned to the C–N siliceous cation as deduced from our earlier paper (Fig. S4†).16
vibration at 1561 cm−1 and 1410 cm−1 55
The loading of imid-
.
The impact of nitrogen containing organosilanes on the texture/
structure and surface properties of calcium doped silicas
azole into SBA-15 was confirmed by the characteristic IR bands
of the imidazole ring at 1570 cm−1 (C–N) and 1515 cm−1
(CvN)56 (Fig. 8) and by the peak at 399.0 eV assigned to CvN
in the N 1s XP spectrum of this sample16 (Fig. S3†). Moreover,
for SBA-15-Im a band at 1563 cm−1 typical of the C–N vibration
was visible, and it originates from the connection between the
nitrogen atom of imidazole and carbon from the propyl
chain.55 The connection of imidazole via the propyl chain
from the previously anchored ClPTMS was also evidenced by
the N 1s XP spectrum in which the band typical of nitrogen
bonded to the propyl chain was detected at 400.8 eV.16
The stability of the anchored modifiers was studied by
thermogravimetric analysis. The results are shown in Fig. 9.
The DTA/TG curves indicated that the amine species incorpor-
ated into SBA-15 decomposed at around 283 °C, whereas the
imidazole species were more stable and decomposed at higher
temperatures, i.e. 393 °C and 548 °C. The presence of two
The N2 adsorption/desorption isotherms of calcium doped
silicas after amine or imidazole loading are shown in Fig. 1.
The incorporation of both organosilanes results in a decrease
in the volume of adsorbed nitrogen. However, the mesoporous
character of the obtained samples was preserved as concluded
from the isotherm shape of type IVa, according to the IUPAC
classification.30 The modification of calcium doped materials
with amine or imidazole species did not destroy the hexagonal
arrangement of the pores as deduced from the small-angle
XRD patterns, in which the diffraction peaks typical of SBA-15
support are still visible (Fig. 2). It is worth noting that the iso-
therms and small angle XRD patterns of CaSBA-15-AP, Ca/
SBA-15_550-AP and Ca/SBA-15_700-AP are less affected than
those of SBA-15-AP after amine anchoring. This feature could
be related to the decrease in the amount of anchored amine in
Fig. 9 Thermogravimetric analysis of the materials containing amine or imidazole.
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
calcium doped silica in comparison with its content in SBA-15 fied with amine than that for the imidazole species, probably
(Table 3). A similar phenomenon of decreasing the efficiency due to the higher efficiency of APTMS than that of imidazole
of basic modifier anchoring in the metallosilicates was also anchoring (Table 3). The unit-cell parameter (a0), interspace
observed for the samples modified with imidazole species, distance (d100) and especially wall thickness increased after
except for CaSBA-15-Im. According to our results, the calcium organosilane anchoring into the metallosilicates. An opposite
species loaded into the SBA-15 support affect the effectiveness phenomenon was observed for SBA-15 after APTMS and imid-
of anchoring of both organosilanes.
azole loading. Therefore, it seems that the calcium modified
Surprisingly, the wide angle XRD patterns of the calcium samples interacted differently with the organosilanes than
and APTMS modified silicas, presented in Fig. 2, show with the silica support. The impact of the organosilanes on
additional diffraction peaks at 2 theta of 29° and 39°. These the surface properties of the calcium doped silicas is proved by
diffraction peaks did not appear in the XRD patterns of the the O 1s and Si 2p XP spectra presented in Fig. 10. In the O 1s
imidazole modified metallosilicates. The additional diffraction spectra of all materials the area of the peak assigned to lattice
peaks observed for the APTMS containing metallosilicates oxygen from CaO was observed to decrease (529.9–531.4 eV).
could be due to the presence of Ca(OH)2 species.42 It was con- As the Ca/Si molar ratio of the organosilane modified metallo-
firmed that the calcium doped silicas have surface calcium silicates was not changed significantly in comparison with
oxide species, which could react with water used as a solvent that of the appropriate supports (Table 3), the decrease could
during the washing process after the APTMS modification. As be due to the covering of calcium oxide by the anchored orga-
a result of this reaction the Ca(OH)2 species could be formed. nosilanes. As a result CaO was not detected by XPS. Moreover,
The calcium and imidazole modified silicas were washed only the anchored organosilanes caused a decrease in the area of
with water-free solvents, i.e. dichloromethane and methanol. the peak assigned to trigonal silicon (100.8–102.6 eV), except
Thus, on the surface of these materials no calcium hydroxide for CaSBA-15-Im. This observation could be associated with a
could be formed.
change in the coordination number of silicon from trigonal to
The textural parameters of the samples obtained, presented tetrahedral as a result of the interaction between metallosili-
in Table 2, show a decrease in the surface area, pore volume cates and organosilanes. The decrease in the area of the peak
and diameter after the APTMS and imidazole anchoring into from trigonal silicon was higher for the amine than that for
metallosilicates. The decrease is higher for the materials modi- the imidazole modified samples, which could be associated
Fig. 10 XP spectra of O 1s and Si 2p of the amine and imidazole modified samples.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
Table 5 Yield of benzylidenemalononitrile after 30 min of the reaction
with the higher efficiency of APTMS loading (Table 3). The
anchoring of amine and imidazole species into calcium doped
silicas resulted moreover in a shift of the BE of the peaks in
the region of O 1s, Si 2p, Ca 2p and N 1s as shown in
Table S1.† These shifts were more significant for the amine
than for the imidazole modified materials which could
suggest a greater interaction between the metallosilicates and
anchored APTMS than Im.
Catalyst
Product yield, %
Blank test
SBA-15
14
1
A stronger interaction between amine and metallosilicates
could be confirmed by the FTIR spectra and DTA/TG curves,
presented in Fig. 8 and 9, respectively. The IR bands assigned
SBA-15-AP
SBA-15-Im
CaSBA-15
49
55
19
87
87
21
87
93
22
75
80
to the –NH2 vibration (3360 cm−1, 3299 cm−1 and 1594 cm−1
)
CaSBA-15-AP
CaSBA-15-Im
Ca/SBA-15_550
Ca/SBA-15_550-AP
Ca/SBA-15_550-Im
Ca/SBA-15_700
Ca/SBA-15_700-AP
Ca/SBA-15_700-Im
were shifted to higher wavenumbers in the spectra of
CaSBA-15-AP, Ca/SBA-15_550-AP and Ca/SBA-15_700-AP in
comparison with their positions in the spectrum of SBA-15-AP.
This shift was not observed for the materials modified with
imidazole species. The greater interaction observed between
the amine species and calcium doped materials could increase
the stability of the anchored organosilane species. Indeed, the
DTA/TG curves (Fig. 9) confirm that the exothermic effect,
typical of amine species decomposition, was observed at a
higher temperature, ca. 25 °C. In the imidazole modified
materials this shift was only observed for Ca/SBA-15_700-Im
(ca. 10 °C). Interestingly, in the DTA/TG curves of the calcium
and amine and imidazole modified silicas, the second exother-
mic effect was detected (between 408–548 °C). This effect
could be assigned to the decomposition of imidazole or amine
species bonded directly to the silicon in the organosilane
molecules as observed for the imidazole modified materials in
ref.16 The presence of calcium in the metallosilicates caused a
decrease in the stability of such bonded imidazole and amine
species. This can be due to a closer distance of the imidazole
and –NH2 to calcium species (Fig. S4†), in contrast to their
further distance to the calcium species if they are located at
the end of the propyl chain of the organosilanes.
Catalytic activity in the Knoevenagel condensation between
benzaldehyde and malononitrile
Liquid phase Knoevenagel condensation between benz-
aldehyde and malononitrile was performed for the characteriz-
ation of the catalytic properties of the samples prepared. In all
cases, the yield of benzylidenemalononitrile increased with
the reaction time. As can be seen in Table 5, pure SBA-15 was
not active in the tested reaction. The obtained yield of benzyli-
denemalononitrile was even lower than that obtained in the
blank test. This can be due to the strong adsorption of
malononitrile on the surface of SBA-15.
Fig. 11 Yield of benzylidenemalononitrile in the reaction with malono-
nitrile of amine and imidazole modified materials. Reaction conditions:
benzaldehyde to malononitrile molar ratio 20 mmol : 20 mmol, 0.017 g
of catalyst, reaction temperature: 60 °C.
The calcium modified supports after 30 min of the reaction SBA-15_550-AP). However, after the reaction time was
revealed a little bit higher yield of benzylidenemalononitrile increased up to 30 min the imidazole containing samples
than that obtained in the case of the blank test (∼20% and exhibited higher activity. This suggests that for the APTMS
14% respectively) independently of the way of incorporation of modified catalysts the reaction rate is higher at the beginning
the calcium species. The anchoring of organosilanes improved of the process.
significantly the catalytic properties of all supports (Fig. 11).
The modification of calcium containing materials with
After 15 min of the reaction the amino modified samples APTMS enhances the catalytic activity of the materials
showed a higher yield of benzylidenemalononitrile than the obtained (Fig. 11), although the amount of active nitrogen
appropriate imidazole modified materials (except for Ca/ species is more than two times higher in SBA-15-AP. The
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
loading of amine species on the surface of the calcium con-
taining material caused a significant shift in the BE of oxygen
typical of CaO than that of nitrogen assigned to the amine
species (Table S1†). Thus, it seems that the modification of
metallosilicates with amine species enhances the basicity of
oxygen in CaO and does not influence so much the basicity of
nitrogen in the amine groups. This conclusion is proved by the
fact that the yield of benzylidenemalononitrile of the modified
metallosilicates increased with the amount of APTMS
anchored, because of the increasing number of the amine
species, not the increase in their basicity.
Fig. 12 The proposed mechanism of Knoevenagel condensation
between benzaldehyde and malononitrile in the presence of bifunc-
tional catalysts. R – imidazole or amine species.
The catalytic activity of the imidazole modified samples also
increased after the incorporation of calcium species into silica
SBA-15 (Fig. 11). However, the yield of benzylidenemalononitrile
did not depend on the amount of the anchored organosilane.
The highest activity was achieved for Ca/SBA-15_550-Im. This
sample was characterized by the medium efficiency of imidazole
loading and the lowest interaction between the imidazole and
metallosilicate support (the lowest shift of the binding energy
of in calcium, oxygen, silicon and nitrogen in the XP spectrum,
(Fig. 10 and Table S1†). For CaSBA-15-Im, in which the nitrogen
amount was the highest, a decrease in the basicity of imidazole
species was observed. This feature might be related to the great-
est impact of the metallosilicate support on the anchored imid-
azole. For CaSBA-15-Im the highest shift of the BE typical of
both nitrogen species C–N and CvN was observed, which could
result in a decrease in the basicity of the imidazole species. The
important finding is that for all samples modified with organic
species and calcium the activity obtained is much higher than
the sum of activities of the corresponding calcium containing
SBA-15 and SBA-15 modified only with organic species. This
clearly indicates the synergism of both modifiers.
The results of the Knoevenagel condensation suggest that
the synergistic interaction between the anchored APTMS or
imidazole species with the calcium species loaded into the
silica support enhances the basicity of CaO. This is proved by a
significant shift in the BE of oxygen typical of calcium oxide.
The BE of oxygen in CaO decreased, while the BE of nitrogen
in APTMS and imidazole species did not change significantly
or slightly increased (in some cases for the latter one)
(Table S1†). Thus one can assume that the activity of the N
species does not change much in the presence of CaO,
however the activity of the latter species increases.
The activity of the calcium modified silicas together with
APTMS or imidazole species tested in the Knoevenagel conden-
sation between benzaldehyde and malononitrile is much
higher in comparison with the current literature data. A much
smaller amount of bifunctional materials (0.017 g) added to
the reaction shows a similar high yield of benzylidenemalono-
nitrile (more than 80%) in comparison with monofunctional
catalysts described in the literature (catalysts loading between
0.1–0.44 g) in a shorter reaction time (30 min in comparison
with 2–20 h).
The stability of the most active sample (Ca/SBA-15_550-Im)
was tested in the next catalytic cycles. After the reaction the
catalyst was separated from the reaction mixture by centrifu-
gation and then washed with 5 cm3 of acetone under stirring
for 30 min. After such a treatment the catalyst was reused in
the Knoevenagel condensation. A significant decrease in the
yield of benzylidenemalononitrile was observed in the second
run as shown in Fig. 13. Interestingly, only a small decrease
in the catalytic activity was observed in the next two cycles.
For this reason the washing procedure was changed: the cata-
lyst was mixed with acetone for 2 hours or 3 times for 30 min.
The same dependence was observed for both new methods of
reuse. In order to examine if leaching of the active phase
from Ca/SBA-15_550-Im occurred the hot filtration test was
performed. After 1 min of the reaction the catalyst was separ-
Based on the catalytic results and XPS measurements
(Table S1†) the proposed mechanism and role of CaO species
in the Knoevenagel condensation are shown in Fig. 12. First,
the migration of electrons from basic nitrogen species to CaO
occurs. Then an electron from CaO migrates to the hydroxyl
group located on the oxide surface. The hydroxyl species in
CaO are able to abstract a proton from malononitrile, which
leads to the generation of a carbanion and water as a bypro-
duct. The formed carbanion is then able to attack the carbonyl
group of benzaldehyde. The generated enol is then dehydrated
which gives an unsaturated final product of the Knoevenagel
condensation. The produced water in the Knoevenagel conden-
sation regenerates the hydroxyl species on the CaO surface.
Fig. 13 Yield of benzylidenemalononitrile after 30 min of the reaction
with fresh and reused (after regeneration by mixing with acetone for
30 min, 2 h or 3 × 30 min) Ca/SBA-15_550-Im. Benzaldehyde to
malononitrile molar ratio 20 mmol : 20 mmol, 0.017 g of catalyst, reac-
tion temperature: 60 °C.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
ated from the reaction mixture. Then the reaction was contin-
ued without the catalyst for 15 minutes. No change in the
Acknowledgements
yield of benzylidenemalononitrile was observed, which indi- The National Science Centre in Poland (Project No. UMO-2018/
cated no leaching of the active phase. The ATR-FTIR spectra 29/B/ST5/00137) is gratefully acknowledged for its financial
of the catalyst after the reuse and washing with acetone for support. The work was also supported by grant no.
30 min showed new bands in the range of 1700–1530 cm−1 POWR.03.02.00-00-I023/17 co-financed by the European Union
which did not appear for the fresh sample. These bands are through the European Social Fund under the Operational
characteristic of the reagents, which suggests the presence of Program Knowledge Education Development.
reactants and/or products in the pores of the catalyst
(Fig. S5†). It could be concluded that after the reaction, a part
of the active centers was still blocked after the catalyst regen-
eration, and the washing treatment applied was insufficient
Notes and references
for the full recovery of the catalytic activity.
1 G. Jones, Org. React., 1967, 15, 204–599.
2 W. Li, R. Li, X. Yu, X. Xu, Z. Guo, T. Tan and S. N. Fedosov,
Biochem. Eng. J., 2015, 101, 99–107.
3 J. L. Scott and C. L. Raston, Green Chem., 2000, 2, 245–247.
4 A. Corma, R. M. Martin-Aranda and F. Sanchez, J. Catal.,
1990, 126, 192–198.
5 F. Bigi, L. Chesini, R. Maggi and G. Sartori, J. Org. Chem.,
1999, 64, 1033–1035.
6 A. V. Narsaiah and K. Nagaiah, Synth. Commun., 2003, 33,
3825–3832.
7 J. Guan, B. Liu, X. Yang, J. Hu, C. Wang and Q. Kan, ACS
Sustainable Chem. Eng., 2014, 2, 925–933.
8 L. C. Player, B. Chan, P. Turner, A. F. Masters and
T. Maschmeyer, Appl. Catal., B, 2018, 223, 228–233.
9 H. Ishitani, Y. Saito, Y. Nakamura, W. J. Yoo and
S. Kobayashi, Asian J. Org. Chem., 2018, 7, 2061–2064.
Conclusions
The catalysts having basic active sites and supported on the
mesoporous structure of SBA-15 were prepared. Two sources
of basicity (separately or together) were incorporated: the
oxygen species in the form of a Ca precursor, and the nitro-
gen species as (3-aminopropyl)trimethoxysilane (APTMS) or
imidazole. The one pot and post synthesis modification of
silica with the calcium precursor allows us to obtain the CaO
and CaSiO3 species on the silica support. The CaSiO3 species
are formed as a result of the interaction between the silica
and calcium precursor. A difference in the calcium species
agglomeration was observed depending on the calcination
temperature. The higher value of this parameter, i.e. 700 °C, 10 A. Wojtaszek-Gurdak, V. Calvino-Casilda, A. Grzesinska,
caused higher agglomeration of the calcium species in the
mesoporous channels of SBA-15 making them less accessible
R. Martin-Aranda and M. Ziolek, Microporous Mesoporous
Mater., 2019, 280, 288–296.
for the reactants as demonstrated by the results of 2,5-hexa- 11 V. Calvino-Casilda, M. Olejniczak, R. M. Martin-Aranda and
nedione dehydration and cyclization. A higher efficiency of M. Ziolek, Microporous Mesoporous Mater., 2016, 224, 201–207.
organic modifier incorporation was observed for the APTMS 12 D. Kryszak, K. Stawicka, M. Trejda, V. Calvino-Casilda,
species, however imidazole was more thermally stable on the
surface of the SBA-15 support. The process of metallosilicate
R. Martin-Aranda and M. Ziolek, Catal. Sci. Technol., 2017,
7, 5236–5248.
modification with APTMS, in which water was used as a 13 J. Mondal, A. Modak and A. Bhaumik, J. Mol. Catal. A:
solvent for sample washing, caused the formation of Ca(OH)2 Chem., 2011, 335, 236–241.
species on the catalysts surface. The interaction between the 14 F. Martínez, G. Orcajo, D. Briones, P. Leo and G. Calleja,
organic modifiers and calcium was analysed by the XPS and Microporous Mesoporous Mater., 2017, 246, 43–50.
FTIR measurements, which resulted in an increase of the 15 B. Xue, J. Zhu, N. Liu and Y. Li, Catal. Commun., 2015, 64,
modifier stability on the surface of the support. The catalytic 105–109.
activities of the materials obtained were examined in the 16 D. Kryszak, K. Stawicka, V. Calvino-Casilda, R. Martin-
Knoevenagel condensation between benzaldehyde and Aranda and M. Ziolek, Appl. Catal., A, 2017, 531, 139–150.
malononitrile. A very high activity was observed especially for 17 K. Stawicka, K. Drazkiewicz and M. Ziolek, Microporous
the materials containing both modifiers, pointing to the Mesoporous Mater., 2018, 258, 41–54.
synergetic interaction between them. The decrease in the 18 J. N. Appaturi, R. J. Ramalingam and H. A. Al-Lohedan,
activity in the reaction after the regeneration of the catalyst J. Porous Mater., 2018, 25, 629–641.
was explained by the blockage of the active centers by the 19 J. N. Appaturi, M. R. Johan, R. J. Ramalingam and H. A. Al-
reactants which were not removed during the regeneration
process.
Lohedan, Microporous Mesoporous Mater., 2018, 256, 67–74.
20 F. Rajabi, F. Fayyaz and R. Luque, Microporous Mesoporous
Mater., 2017, 253, 64–70.
21 A. Wach, M. Drozdek, B. Dudek, E. Szneler and
P. Kustrowski, Catal. Commun., 2015, 64, 52–57.
22 K. Stawicka, D. Prukała, T. Siodła, M. Sikorski and
M. Ziolek, Appl. Catal., A, 2019, 570, 339–347.
Conflicts of interest
There are no conflicts to declare.
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
23 K. Ikeue, N. Miyoshi, T. Tanaka and M. Machida, Catal. 41 D. Kryszak, M. Trejda, N. Benedyczak and M. Ziolek, Catal.
Lett., 2011, 141, 877–881.
Today, 2019, 325, 11–17.
24 Y. Peng and G. Song, Indian J. Chem., 2003, 42B, 924–926.
25 D. Jain, C. Khatri and A. Rani, Fuel Process. Technol., 2010,
91, 1015–1021.
26 K. R. Kloetstra and H. van Bekkum, Stud. Surf. Sci. Catal.,
1997, 105, 431–438.
42 W. Gu, D. W. Bousfield and C. P. Tripp, J. Mater. Chem.,
2006, 16, 3312–3317.
43 N. Zhang, H. Xue and R. Hu, RSC Adv., 2018, 8, 32922–
32929.
44 D. H. Chu, M. Vinoba, M. Bhagiyalakshmi, I. Hyun Baek,
S. C. Nam, Y. Yoon, S. H. Kim and S. K. Jeong, RSC Adv.,
2013, 3, 21722–21729.
27 I. Sobczak, V. Calvino-Casilda, L. Wolski, T. Siodla, R. Martin-
Aranda and M. Ziolek, Catal. Today, 2019, 325, 81–88.
28 A. Smuszkiewicz, J. López-Sanz, I. Sobczak, R. M. Martin- 45 Y. Tang, L. Li, S. Wang, Q. Cheng and J. Zhang, Environ.
Aranda, M. Ziolek and E. Pérez-Mayoral, Catal. Today, 2019,
325, 47–52.
Prog. Sustainable Energy, 2016, 35, 257–262.
46 F. Márquez, C. Morant, V. López, F. Zamora, T. Campo and
E. Elizalde, Nanoscale Res. Lett., 2011, 6, 495–502.
47 D. K. Dumbre, V. R. Choudhary, N. S. Patil, B. S. Uphade
and S. K. Bhargava, J. Colloid Interface Sci., 2013, 415, 111–
116.
48 R. Wang, H. He, L. C. Liu, H. X. Dai and Z. Zhao, Catal. Sci.
Technol., 2012, 2, 575–580.
49 L. B. Sun, J. H. Kou, Y. Chun, J. Yang, F. N. Gu, Y. Wang,
J. H. Zhu and Z. Z. Gang, Inorg. Chem., 2008, 47, 4199–
4208.
29 H. Sun, J. Han, Y. Ding, W. Li, J. Duan, P. Chen, H. Lou
and X. Zheng, Appl. Catal., A, 2010, 390, 26–34.
30 M. Thommes, K. Kaneko, A. V. Neimark, J. P. Olivier,
F. Rodriguez-Reinoso, J. Rouquerol and K. S. W. Sing, Pure
Appl. Chem., 2015, 87, 1051–1069.
31 D. Zhao, Q. Huo, J. Feng, B. F. Chmelka and G. D. Stucky,
J. Am. Chem. Soc., 1998, 120, 6024–6036.
32 G. Wang, L. Zhang, J. Deng, H. Dai, H. He and C. T. Au,
Appl. Catal., A, 2009, 355, 192–201.
33 N. Kaur and A. Ali, RSC Adv., 2015, 5, 13285–13295.
34 C. Zhao and G. Wang, Miner. Process. Extr. Metall. Rev.,
2015, 36, 385–390.
35 Q. Meng, J. Lin, L. Fu, H. Zhang, S. Wang and Y. Zhou,
J. Mater. Chem., 2001, 11, 3382–3386.
36 M. Trejda, D. Kryszak and M. Ziolek, Mater. Res. Bull., 2018,
97, 530–536.
37 A. Hemeryck, N. Richard, A. Estève and M. Djafari
Rouhani, Surf. Sci., 2007, 601, 2082–2088.
50 A. Gervasini, J. Fenyvesi and A. Auroux, Catal. Lett., 1997,
43, 219–228.
51 W. E. Thompson, R. J. Warren, I. B. Eisdorfer and
J. E. Zarembo, J. Pharm. Sci., 1965, 54, 1819–1821.
52 R. A. P. O. d’Amorim, M. I. Teixeira, L. V. E. Caldas and
S. O. Souza, J. Lumin., 2013, 136, 186–190.
53 S. Y. Li, W. H. Ma, Y. Zhuo, X. H. Chen, M. Y. Ma, Y. H. Xu,
Z. Ding and X. H. Wu, Int. J. Electrochem. Sci., 2013, 8,
1802–1812.
38 W. L. Zhang, S. Zhang, M. Yang and T. P. Chen, J. Phys. 54 J. Ederer, P. Janoš, P. Ecorchard, J. Tolasz, V. Štengl,
Chem. C, 2010, 114, 2414–2420.
H. Beneš, M. Perchacz and O. Pop-Georgievski, RSC Adv.,
2017, 7, 12464–12473.
55 C. C. Chuang, W. C. Wu, M. X. Lee and J. L. Lin, Phys.
Chem. Chem. Phys., 2000, 2, 3877–3882.
39 M. Trejda, M. Ziolek, P. Decyk and D. Duczmal,
Microporous Mesoporous Mater., 2009, 120, 214–220.
40 G. Dingemans, C. A. A. van Helvoirt, D. Pierreux,
W. Keuning and W. M. M. Kessels, J. Electrochem. Soc., 56 B. Stuart, Infrared Spectroscopy: Fundamentals and
2012, 159, H277–H285. Applications, John Wiley & Sons, 2004.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020