Efficient and Reusable Pb(II) Metal–Organic Framework for Knoevenagel Condensation

Article abstract of DOI:10.1007/s10562-018-2471-8

A microporous lead–organic framework {[Pb₄(μ₈-MTB)₂(H₂O)₄]·5DMF·H₂O}_n (MTB = methanetetrabenzoate, DMF = N,N′-dimethylformamide) was synthesized and studied as a catalyst in Knoevenagel condensation reactions. The framework is built from tetranuclear [Pb₄(μ₃-COO)(μ₂-COO)₆(COO)(H₂O)₄] clusters and exhibits a 3D structure, with repeated 1D jar-like cavities with sizes about 14.98 × 7.88 and 14.98 × 13.17??² and BET specific surface area of 980?m²?g^?1. To obtain open framework with unsaturated Pb(II) sites needed for catalysis, the thermal activation of the solvent exchanged sample was performed (DMF was exchanged by EtOH). The activated compound was tested in Knoevenagel condensation of bulky aldehydes and active methylene compounds at different temperatures. Excellent catalytic conversion and selectivity in condensation of small-sized aldehydes with malononitrile was observed, which indicates that the opened Pb(II) sites play a significant role in the heterogeneous catalytic process. Leaching test confirmed the stability of the catalyst in catalytic reactions. Moreover, the compound displayed good recyclability after several reuses without significant decrease in the original catalytic activity. Graphical Abstract: Novel Pb(II) metal–organic framework was tested in Knoevenagel condensation. The catalyst showed excellent catalytic conversion, selectivity and recyclability. Aldehydes with lower kinetic diameter demonstrated high conversions and yields. Catalyst is less efficient for condensation of larger aromatic aldehydes. [Figure not available: see fulltext.].

Full text of DOI:10.1007/s10562-018-2471-8

Catalysis Letters
https://doi.org/10.1007/s10562-018-2471-8
Efficient and Reusable Pb(II) Metal–Organic Framework
for Knoevenagel Condensation
Miroslav Almáši1 · Vladimír Zeleňák1 · Maksym V. Opanasenko2 · Jiří Čejka2
Received: 1 February 2018 / Accepted: 24 June 2018
© Springer Science+Business Media, LLC, part of Springer Nature 2018
Abstract
A microporous lead–organic framework {[Pb4(µ8-MTB)2(H2O)4]·5DMF·H2O}n (MTB=methanetetrabenzoate, DMF=N,N′-
dimethylformamide) was synthesized and studied as a catalyst in Knoevenagel condensation reactions. The framework is
built from tetranuclear [Pb4(µ3-COO)(µ2-COO)6(COO)(H2O)4] clusters and exhibits a 3D structure, with repeated 1D jar-like
cavities with sizes about 14.98×7.88 and 14.98×13.17 Å2 and BET specific surface area of 980 m2 g−1. To obtain open
framework with unsaturated Pb(II) sites needed for catalysis, the thermal activation of the solvent exchanged sample was per-
formed (DMF was exchanged by EtOH). The activated compound was tested in Knoevenagel condensation of bulky aldehydes
and active methylene compounds at different temperatures. Excellent catalytic conversion and selectivity in condensation of
small-sized aldehydes with malononitrile was observed, which indicates that the opened Pb(II) sites play a significant role
in the heterogeneous catalytic process. Leaching test confirmed the stability of the catalyst in catalytic reactions. Moreover,
the compound displayed good recyclability after several reuses without significant decrease in the original catalytic activity.
Graphical Abstract
Novel Pb(II) metal–organic framework was tested in Knoevenagel condensation. The catalyst showed excellent catalytic
conversion, selectivity and recyclability. Aldehydes with lower kinetic diameter demonstrated high conversions and yields.
Catalyst is less efficient for condensation of larger aromatic aldehydes.
Electronic supplementary material The online version of this
article (https://doi.org/10.1007/s10562-018-2471-8) contains
supplementary material, which is available to authorized users.
Extended author information available on the last page of the article
Vol.:(0123456789)
1 3

M. Almáši et al.
Keywords Lead(II) · Methanetetrabenzoate · Metal–organic framework · Sorption properties · Knoevenagel condensation
sites on metal ions [41–45]. To date different MOFs such as
UMCM-1 and IRMOF-3 [35, 43], MIL-101 [36, 43], Ni-
MTB [38], UIO-66 [39], HKUST-1 [2], PCN-100 [46] and
PCN-101 [46] were investigated as catalysts in Knoevenagel
condensation. In the case of PCN-100 and PCN-101 a size-
selective catalytic activity was studied by varying the size
and shape of the substrates [46].
1 Introduction
The Knoevenagel condensation is one of the most important
C–C forming condensation reactions in organic chemistry
[1]. Products of this reaction are α,β-unsaturated com-
pounds, which have been widely used as final substances or
intermediates in the synthesis of fine chemicals [2], pharma-
ceuticals [3], natural products [4], perfumes [5], agrochemi-
cals [6] and functional polymers [7]. Knoevenagel conden-
sation can be performed under homogeneous conditions in
the presence of catalysts such as organic amines [8], alkali
salts [9], ionic liquids [10, 11], self-assembled solids [12],
polyoxometalates [13], and others. In recent years, Knoev-
enagel condensations promoted by microwave irradiation
have been developed [14]. Unfortunately, reactions under
these conditions often need high power microwaves and are
not applicable for large-scale synthesis.
In our research we are interested in the preparation of
the new MOFs using methanetetrabenzoate linker (MTB).
Based on our knowledge and experience this linker prefers
coordination with central atoms in oxidation state +2. To
date we have prepared and studied MOF compounds con-
taining Zn(II) [25, 49] and Ni(II) [38] ions as metal nodes.
As a continuation of these works, in the subsequent study,
we selected lead atom, which also prefers oxidation state
+2 and moreover the stereochemistry of lead(II) ions is due
to presence of lone electron pair in the valence sphere very
interesting for the chemists working in the field of coordina-
tion chemistry. Using lead(II) ion and MTB linker we have
prepared and investigated the compound {[Pb4(µ8-MTB)2(
H2O)4]·5DMF·H2O}n (1). The properties of this compound,
like crystal structure, low and high-pressure gas adsorption
and preliminary catalytic properties were described in detail
in our recent work [40]. The catalytic properties of 1 seemed
us so interesting, that we studied them more in detail. The
obtained results we describe in the present study dealing
with the Knoevenagel condensation reaction over MOF with
composition {[Pb4(µ8-MTB)2(H2O)4]·5DMF·H2O}n. The
tested catalytic reaction was selected based on our previous
results when we explored and compared the effect of zeo-
lites versus MOF activity in Knoevenagel condensation [2].
Moreover, the Knoevenagel condensation was selected also
based on the literary review showing, that the Knoevenagel
reactions are one of the most frequent reactions used for
tests of new prepared catalysts today. It is to note, based on
searches in different databases, there are no studies in the
literature dealing with the application of lead-based MOFs
in the Knoevenagel condensation reaction.
Nowadays there is an urgent need to develop green
chemistry processes, where the generation of waste can be
avoided, but considering both environmental and economic
points of view. Therefore, the current research is focused
on the development of new heterogeneous catalysts, which
can help to overcome these tasks. To catalyse Knoevenagel
condensation, amines [15], metal nanoparticles [16] or metal
oxides [17] immobilized on mesoporous silica [18], alumina
[19], polyacrylamide [20] and polyacrylonitrile fiber [21],
mesoporous carbon nitride [22] or zeolites were tested [23,
24].
Among potential catalysts, metal–organic frameworks
(MOFs) have attracted tremendous interest in the last decade
[24]. MOFs represent crystalline porous solids resembling
zeolites, having metal ions/clusters and organic ligands to
build a final framework with large surface area, predictable,
controllable and post-synthetic modifiable pores and cavi-
ties. Due to their fascinating diversity, MOFs are intensively
studied in the different practical areas including gas storage
[25, 26] and separation [27], drug delivery [28], sensing
[29], magnetism [30] and heterogeneous catalysis [31–40].
As for catalysis, MOFs have been investigated as solid
catalysts or catalysts support for organic transformations.
The coordinatively unsaturated metal sites within MOF
framework can serve as catalytic centres acting as Lewis
acid catalysts for various organic reactions, such as Beck-
mann rearrangement [31], Henry [32] and Prins [33] reac-
tions, Friedel–Crafts alkylation [34] and of course Kno-
evenagel condensation [35–40]. In the case of different
porous coordination polymers, Knoevenagel condensation
is usually catalysed by basic sites (e.g. amine group located
on linker in the polymeric framework, imidazole in ZIFs,
M–O pairs) or acid sites such as coordinatively unsaturated
In the present study an interesting size-selectivity screen
has been observed revealing the relationship between the
size of the substrates and the catalytic activity of 1 in the
Knoevenagel condensation of different bulky aldehydes
(n-heptanal, cyclohexane carbaldehyde, benzaldehyde,
4-tert-butylbenzaldehyde, 3,5-ditert-butylbenzaldehyde,
2-naphthaldehyde) with active methylene compounds (malo-
nonitrile, methyl cyanoacetate and ethyl acetoacetate) at dif-
ferent temperatures (70–130 °C).
1 3

Efficient and Reusable Pb(II) Metal–Organic Framework for Knoevenagel Condensation
solvent was necessary for dissolution of substrates. Methyl
2 Experimental
groups on phenyl ring in p-xylene are in positions 1 and
4 and therefore less steric limitation for diffusion of
these molecules into channel system compared to o- and
m-xylene. Finally, in comparison to benzene (b.p. 80.1 °C)
and toluene (b.p. 110.6 °C), p-xylene (b.p. 138.4 °C) has
higher boiling point and we can reach reaction temperature
130 °C, used in our study.
2.1 Materials
All chemicals used in the synthesis of the complex and
reagents for catalytic reactions were obtained from Sigma-
Aldrich or Acros Organics companies and used without
further purification. The catalyst was prepared according to
the literature procedure [40]. n-heptanal (95%), cyclohexane
carbaldehyde (97%), benzaldehyde (≥99%), 4-tert-butylb-
enzaldehyde (97%), 3,5-ditert-butylbenzaldehyde (97%),
2-naphthaldehyde (98%), malononitrile (≥ 99%), methyl
cyanoacetate (99%), ethyl acetoacetate (≥99%), n-dodecane
(≥ 99%, internal standard) and p-xylene (≥ 99%, solvent)
used in the catalytic experiments were obtained from Sigma-
Aldrich and Across Organics.
Typically, 4 mmol of aldehyde, 0.4 g of dodecane (inter-
nal standard), 50 mg of catalyst and 10 cm3 of p-xylene
(solvent) were added to the 25 cm3 three-necked vessel,
equipped with condenser and thermometer, stirred and
heated. Then 6 mmol of methylene compound was added
into the reaction vessel through syringe with needle when
the desired reaction temperature was reached.
0.1 cm3 of the reaction mixture was sampled after 10,
30, 60, 120, 240, 360 and 1440 min. After the sampling, the
reaction suspension was immediately centrifuged for 5 min
at 4000 rpm. After this procedure, the reaction products
were analysed by gas chromatography (GC) using an Agi-
lent 6850 with FID detector equipped with a nonpolar HP1
column (diameter 0.25 mm, thickness 0.2 µm and length
30 m). The reaction products were identified using GC–MS
analysis (ThermoFinnigan, FOCUS DSQ II Single Quad-
rupole GC/MS). The identity of the reaction products was
also confirmed using 1H NMR spectroscopy measurements
carried out on a Varian Unity 400 spectrometer. 1H chemical
shifts in CDCl3 were given with tetramethylsilane (TMS) as
the internal standard. The observed spectra corresponded to
the previously published data [47].
2.2 Characterization
The catalyst was analysed by an Avatar FT-IR 6700 spec-
trometer using KBr and ATR techniques to identify the
organic ligands of the catalyst, to monitor the desolvation
process and immobilization of the substrates/products after
catalysis. The crystallinity of the samples was monitored by
powder X-ray diffraction (Bruker AXS D8 Advance). Ther-
mogravimetric analysis of 1, solvent exchanged sample and
reused catalyst was performed in the temperature range of
30–900 °C under dynamic atmosphere of air with heating
rate 9 °C min−1 (STA Netzsch 409-PC). The surface area
and pore volume of the samples after different activation
procedures were analysed and determined from nitrogen
physisorption at −197 °C (Quantachrome Nova 1200e). The
details of analysis procedures are described in [40].
To evaluate the potential influence of leaching of the
active species from the heterogeneous catalysts, a part of the
reaction mixture was taken away, immediately centrifuged
and the obtained liquid phase was again investigated in the
Knoevenagel condensation under the same reaction condi-
tions. The reaction mixture was sampled after 5, 10, 20,
30, 40 and 60 min. Leaching test was also performed using
ICP-MS measurement on ICP-MS 7700 instrument (Agilent
Technologies). Before measurement, reaction suspension
composed from the catalyst, malononitrile and n-heptanal
was heated at 130 °C for 24 h. After this time the catalyst
was separated by filtration and liquid phase evacuated in
vacuum. Final residue was mineralized with aqua regia and
solution was analysed by ICP-MS.
2.3 Knoevenagel Condensation Reactions
The Knoevenagel condensation of aldehydes and active
methylene compounds was performed in a liquid phase
under atmospheric pressure and temperature of 130, 100
and 70 °C on a multi-experiment work station StarFish.
Before catalytic experiments the single crystals of the
catalyst 1, after solvent exchange with ethanol, were gen-
tly grinded with aim to decrease the size of grains and
increase their active sites for catalytic experiments. After
this procedure, the catalyst was activated at 200 °C for
90 min with a temperature heating rate 10 °C min−1 in
a stream of air (50 cm3 min−1). The p-xylene was used
as a solvent for Knoevenagel condensation. We have per-
formed condensation reaction in different solvents such
as benzene, toluene and p-xylene, but best results were
obtained for p-xylene. Moreover, the p-xylene as nonpolar
Recyclability of the catalyst was tested in the reactions
of n-heptanal/cyclohexane carbaldehyde with malononi-
trile in a three repetition cycles. After each test, the cata-
lyst was separated by centrifugation, stirred in solvents
with different polarity (ethanol, acetone, n-hexane), fil-
tered off, reactivated at 200 °C for 90 min and then reused
in the next run. Moreover, the catalyst recycled after 3rd
run in the reaction of n-heptanal and malononitrile was
analysed by a combination of infrared spectroscopy and
1 3

M. Almáši et al.
TG/DTG–DTA in order to demonstrate the immobilization
of reactants/products in the pores of the material.
3 Results and Discussion
3.1 Description of the Crystal Structure
The crystal structure of 1 is formed by tetrahedral meth-
anetetrabenzoate linkers and lead(II) ions (see Fig. 1a).
Pb(II) cations are bridged by carboxylate groups under
formation of asymmetrical tetranuclear clusters with com-
position [Pb4(µ3-COO)(µ2-COO)6(COO)(H2O)4] in which
the µ3-carboxylato functional group is coordinated in che-
late-anti-anti fashion and connected to three lead(II) ions
Pb1, Pb3 and Pb4 (Fig. 1a). The other six carboxylates are
bonded to Pb(II) ions in µ2-coordination mode and connect
lead ions into the final cluster. Cluster is strongly distorted
because of the presence of the stereochemically active 6s2
lone-pair electrons on the Pb(II) central atoms. An inter-
esting feature of the cluster is presence of four coordi-
nated water molecules, which could be easily removed by
thermal treatment (see Sect. 3.2 below) and this way four
unsaturated centers as Lewis acid metal sites for catalysis
could be obtained.
As depicted Fig. 1b, lead–organic framework contains
open channels propagating along crystallographic axis a,
which are formed by repeated jar-like cavities with nar-
row entrances (14.98 × 7.88 Å2) and wider inside pockets
(14.98 × 13.17 Å2). Each channel is constructed by the
methanebenzoate subunit of MTB4− with [Pb4(µ3-COO)
(µ2-COO)6(COO)(H2O)4] and is further connected with
five neighboring MTB4− molecules, forming a 3D open
framework, with 50.3% solvent accessible voids. The
detailed description of crystal structure can be found in
[40].
3.2 Study of Activation Process
and Characterization of the Catalyst
The activation process of the compound was studied by
the combination of thermal analysis (TG/DTG–DTA) and
infrared spectroscopy measured at different temperatures
(see Sect. 2.2). The framework stability was monitored by
powder X-ray diffraction (PXRD) and nitrogen adsorption/
desorption measurements.
As it is evident from TG/DTG–DTA curves (Fig. S1 in
Supplementary information), compound 1 is thermally sta-
ble after heating up to 110 °C. The desolvation processes
take place in the temperature range 100–240 °C in one
decomposition step. The observed weight loss of 19.8%
in this region corresponds to release of five water and five
clathrated DMF molecules (calcd. 20.1%). The desolva-
tion process is characterized by two endothermic effects
on DTA at 105 and 210 °C, corresponding to the release
Fig. 1 a Building blocks used in the synthesis of 1 and its tetranu-
clear lead(II) cluster [Pb4(µ3-COO)(µ2-COO)6(COO)(H2O)4] with
marked catalytic active sites on Pb2 and Pb4 ions (blue ellipsoids). b
View of the final 3D polymeric framework with 1D channels propa-
gating along a axis
1 3

Efficient and Reusable Pb(II) Metal–Organic Framework for Knoevenagel Condensation
of water and DMF, respectively. The desolvated form of
1, {[Pb4(µ8-MTB)2]}n (denoted as 1′ in the further text),
is thermally stable in the temperature range 240–340 °C,
as evidenced by a plateau on the TG curve. In the range
340–750 °C, decomposition of the polymeric framework
takes place with the observed mass change of 38.5%
(calcd. 37.8%) on TG curve. Decomposition process was
accompanied by three exothermic effects on DTA at 375,
510 and 620 °C. The final product of the decomposition
was PbO2 (obs. 41.7%, calcd. 42.2%).
aliphatic stretching vibrations of C–H bonds, (ν(CH)aliph), at
2975 and 2928 cm−1.
To study the liberation of guest molecules from {[Pb4(µ8-
MTB)2(H2O)4]·5DMF·H2O}n, the infrared spectra at dif-
ferent temperatures 100, 200, 300 and 400 °C were meas-
ured (see Fig. 2). IR spectra at the respective temperatures
illustrate the decrease of the intensity of the bands corre-
sponding to water and DMF molecules and thus indicate
release of these solvents from the framework of 1. In the
concrete, after heating to 100 °C a decrease of the inten-
sity of the broad band around 3400 cm−1, corresponding
to the ν(OH) stretching vibration in water molecules, was
observed. The spectrum measured at 200 °C showed dimin-
ishment of the intensity of this band due to complete release
of water molecules. On the other hand, stretching vibration
of aliphatic C–H bonds in the region under 3000 cm−1 and
strong adsorption band of the carbonyl ν(C=O) vibration
at 1659 cm−1 of DMF were still observed in the spectrum.
Further heating to 300 °C leads to the complete removal of
N,N′-dimethylformamide molecules. After the desolvation
(removal of water and DMF molecules), the framework is
stable and keeps its structure integrity as confirmed by pow-
der X-ray diffraction. The details about PXRD can be found
in [40]. Finally, heating to 400 °C led to the decomposi-
tion of the framework. It should be noted, that the presented
conclusions observed from IR spectra measured at different
temperatures are in good agreement with the results obtained
from TG/DTG–DTA analysis.
In addition to the as synthesised sample, the sample after
solvent exchange process (DMF was exchanged by ethanol),
was also studied by TGA (see Fig. S2 in the Supplemen-
tary information). As it can be seen from TG curve, in the
ethanol exchanged sample, molecules of solvents (EtOH and
H2O) are evolved in temperature range 50–150 °C and after
this process framework of the compound is stable in region
150–300 °C. At higher temperatures, the decomposition of
the polymeric framework is similar as for as-synthesized
sample described in the paragraph above.
IR spectra of compound 1 (see Fig. 2, black curve)
show characteristic bands of the antisymmetric vibra-
tions, (ν(COO)as), of methanetetrabenzoate ligands at
1588, 1532 cm−1, symmetric vibrations, (ν(COO)s), at
1389 cm−1 and absorption band at 777 cm−1 correspond-
ing to δ(COO−) vibration. The absence of characteristic
absorption bands of protonated carboxylic groups in the
range 1730–1690 cm−1 indicates the complete deprotona-
tion of H4MTB. The broad band at 3440 cm−1 confirms the
presence of coordinated water molecules. Very strong and
sharp band at 1659 cm−1 corresponds to stretching vibra-
tion ν(C=O), which could be attributed to DMF molecules.
The presence of DMF molecules is also evident from weak
Relatively large accessible volume in 1 (50.3% of the
cell volume 17 704 Å3) led us to investigate the textural
properties of the compound using nitrogen adsorption
Fig. 2 IR spectra of compound 1 measured at room temperature
(black line), and after heating to 100 °C (red line), 200 °C (green
line), 300 °C (wine line) and 400 °C (orange line)
Fig. 3 Nitrogen adsorption isotherms of as-synthesized/solvent
exchanged samples subsequently activated at different temperatures
(100, 150 and 200 °C)
1 3

M. Almáši et al.
measurements under different activation conditions. The
solvent molecules in the cavities of as-synthesized sample
1 were removed using different thermal treatment proce-
dures or solvent exchange processes with the aim to find the
activation conditions leading to the best textural properties.
Freshly prepared sample was first activated at 100, 150 and
200 °C for 24 h under vacuum, but according to nitrogen
adsorption measurements the compound adsorbed only a
limited amount of nitrogen (see Fig. 3). Therefore, we have
tried to activate the compound by solvent exchange with
ethanol (5 days, while ethanol was changed every day), fol-
lowed by outgassing at 100, 150 and 200 °C under vacuum
for 24 h. Measured adsorption isotherms are illustrated in
Fig. 3.
could be eliminated by thermal treatment to provide CUSs
as Lewis acid sites for catalysis.
On the other side, the nature of basic sites of MOFs is
unclear. However, we can speculate, that their basic sites
arise from the oxygen atoms of carboxylate groups of
organic linkers connected to metal ions of SBUs (M–O
bond) from benzene rings or from basic functional groups
in linkers [41]. So, in the sense of our suggestion, the pos-
sible basic sites in 1′ could came from carboxylate oxygen
atoms bonded to lead(II) ions (Pb–O), or from inert electron
pairs on the valence sphere of central atoms.
In this work, Knoevenagel condensation reactions of
different aromatic and non-aromatic aldehydes with active
methylene compounds at different temperatures were chosen
as catalytic test reactions. A size-selectivity screening has
been studied to reveal the relationship between the size of
the aldehyde substrate and the catalytic activity of 1′. Kno-
evenagel condensation proceeds according to the reaction
shown in Fig. 4, the summary of conversion and selectivity
to the targeted products are presented in Table 2 and the size
of the substrates used in catalytic reactions are shown in Fig.
S3 in the Supporting information.
The highest nitrogen uptake was observed for etha-
nol exchanged sample after evacuation at 200 °C for 24 h
(denoted as 1′). Nitrogen adsorption measurement of 1′ at
−196 °C revealed a type I isotherm classified by IUPAC typ-
ical for microporous materials (see Fig. 3, the isotherm with
orange colour). At the relative pressure (p/p0 =0.95), the N2
uptake capacity reaches 278 cm3 g−1, which corresponds to
a total microporous volume of 0.666 cm3 g−1 calculated by
density functional theory (DTF) method. Evaluation of the
nitrogen adsorption isotherm in the p/p0 = 0.05–0.2 range
using the Brunauer–Emmett–Teller (BET) equation gave the
specific surface area (SBET) of 980 m2 g−1. It is to note, that
the observed SBET surface area is the highest, when com-
pared to the other reported MOFs constructed by the meth-
anetetrabenzoate linker (see Table 1).
3.3 Knoevenagel Condensation Reactions
According to Lewis acid/base theory, Lewis acids and bases
are atoms, molecules or ions that tend to act as an elec-
tron acceptor or donor, respectively. Lewis acid sites, i.e.,
coordinatively unsaturated sites (CUSs) can be formed by
different ways, such as defect engineering in MOFs, the
removal of labile water molecules or coordinating solvent
molecules from the metal centre. In compound 1, there are
four terminal water molecules per each [Pb4(µ3-COO)(µ2-
COO)6(COO)(H2O)4] cluster in the asymmetric unit, which
Fig. 4 Scheme of Knoevenagel condensation with experimental con-
ditions and used substrates
Table 1 Textural properties
Compound
Adsorbate Surface
(m2 g−1
Pore volume References
of up to date prepared MOFs
)
(cm3 g−1
)
containing MTB4− linker
{[Ni2(µ4-MTB)(κ4-CYC)2]·4DMF·8H2O}n
{[Zn2(µ8-MTB)(H2O)2]·3DMF·3H2O}n
{[Co4(µ6-MTB)2(µ2-H2O)4(H2O)4]·13DMF·11H2O}n CO2
CO2
N2
141
248
356
526
644
700
980
0.055
0.089
0.165
–
0.246
0.291
0.666
[48]
[49]
[50]
[51]
{[Cu2(µ8-MTB)(H2O)2]·6DEF·2H2O}n
N2
Ar
N2
N2
{[Zn2(µ4-MTB)(κ4-CYC)2]·2DMF·7H2O}n
{[Ni4(µ6-MTB)2(µ2-H2O)4(H2O)4]·10DMF·11H2O}n
{[Pb4(µ8-MTB)2(H2O)4]·5DMF·H2O}n
[25]
[38]
This work
1 3

Efficient and Reusable Pb(II) Metal–Organic Framework for Knoevenagel Condensation
Table 2 Results of Knoevenagel
Aldehyde
Methylene comp.
MNT
Repetition
130 °C
Conv.b
100 °C
Conv.b
70 °C
Conv.b
condensation performed at
different temperatures
Sel.a
Sel.a
Sel.a
n-HCHO
1st
2nd
3rd
100
100
100
100
97
100
100
100
61
100
96
97
96
42
100
97
98
100
55
100
100
87
98
61
48
98
65
100
99
69
55
100
–
–
95
88
100
–
79
–
–
28
26
85
–
83
–
–
57
–
–
MCA
EAA
MNT
–
91
–
–
–
–
35
–
–
24
–
–
30
–
–
38
–
–
87
–
–
–
–
66
–
–
45
–
–
84
–
–
n
–
–
ChCHO
1st
2nd
3rd
–
–
MCA
EAA
MNT
MCA
EAA
MNT
MCA
EAA
MNT
MCA
EAA
MNT
MCA
EAA
95
68
72
62
63
38
30
12
26
19
18
56
38
26
92
25
93
92
55
85
28
6
93
23
n
76
18
3
57
PhCHO
100
100
68
56
45
39
45
26
21
4T-PhCHO
3,5T-PhCHO
2-NaphCHO
83
40
31
–
The yield is calculated by GC using n-dodecane as the internal standard
n not detected, – not performed, n-HCHO n-heptanal, ChCHO cyclohexane carbaldehyde, PhCHO ben-
zaldehyde, 4T-PhCHO 4-tert-butylbenzaldehyde, 3,5T-PhCHO 3,5-ditert-butylbenzaldehyde, 2-NaphCHO
2-naphthaldehyde, MNT malononitrile, MCA methyl cyanoacetate, EAA ethyl acetoacetate
aSelectivity after 360 min of reaction
bConversion after 360 min of reaction
In addition to the main product, the reaction mixture may
contain some by-products from the consecutive reactions
of primary product. According to the results of GS–MS,
carboxylic acids from the corresponding aldehydes, formed
via an oxidation, were found as by-products. Other identified
side products were cyanoacetic acid formed during dees-
terification reaction of methyl cyanoacetate, or compounds
created by hydrolysis of cyano group of malononitrile or
methyl cyanoacetate (or its deesterification form).
The test reaction, Knoevenagel condensation of malonon-
itrile with aldehydes at 130 °C, resulted in the condensation
product with excellent selectivity about 100% (n-HCHO),
100% (ChCHO), 100% (PhCHO), 98% (4T-PhCHO), 98%
(3,5T-PhCHO) and 99% (2-NaphCHO) yielding the desired
product after 360 min of reaction. Our obtained results for
reaction PhCHO with malononitrile over 1′ (see Table 2)
are comparable to conversion and selectivity for compounds
CAU-1–NH2 (conv. 98%, sel. 100%, 40 °C, 420 min),
Fe–MIL-101–NH2 (conv. 90%, sel. 100%, 80 °C, 180 min)
or Zn–MOF–NH2 (conv. 98%, sel. 100%, 80 °C, 270 min),
Fig. 5 Conversion curves of different bulky aldehydes in the reaction
with malononitrile (130 °C)
1 3

M. Almáši et al.
which in contrast to our sample, contain basic Lewis sites
in their frameworks [52]. As it can be seen from the Table 2
and Fig. 5, the conversions for n-heptanal and benzalde-
hyde are similar and are faster in comparison with the other
aldehydes, giving the condensation product with 100% yield
in 240 min, in contrast to 480 min in the case of 2-naph-
thaldehyde. On the other hand, the catalytic activity for
cyclohexane carbaldehyde lies between those of n-heptanal/
benzaldehyde and 2-naphthaldehyde as it takes 360 min to
reach 100% conversion. Although we extended the time of
reaction to 24 h, bulkier aldehydes such as 4T-PhCHO and
3,5T-PhCHO did not reach 100% conversion. For the bulkier
aldehydes, their conversions (selectivities) under the same
conditions were 45% (100%), 56% (95%), and 83% (95%)
for 3,5-ditert-butylbenzaldehyde, 4-tert-butylbenzaldehyde,
and 2-naphthaldehyde, respectively. Since the conversions of
aldehydes with similar accessibility and polarity of the car-
bonyl group significantly depend on the size of the substrate,
we assume that the Knoevenagel condensation proceeds
inside the pore system of 1′ under the applied conditions.
The 100% conversions in the reactions of methyl cyanoac-
etate with selected aldehydes after 360 min was observed
only for n-heptanal and benzaldehyde. The size effect was
also observed in the case of bulkier aldehydes (see Fig. S3 in
the Supporting information) and the conversions decreases
with increasing size of substrate as follows, NaphCHO
(83%), ChCHO (61%), 4T-PhCHO (56%), and 3,5T-PhCHO
(45%). The observed selectivities were lower compared with
malononitrile and the values are in the range 60–100%.
In addition, ethyl acetoacetate has been proven as not
efficient active methylene compound for Knoevenagel con-
densation, because of low conversions and selectivities were
observed in the catalytic tests (see Table 2). Similar low
conversion for reaction PhCHO with ethyl acetoacetate were
also observed for compound IRMOF-3, where different sol-
vents and reaction temperatures were applied: 65% (DMSO,
80 °C), 54% (DMF, 80 °C), 54% (MeOH, 60 °C) or 39%
(DMF, 60 °C) [35].
the reactivity difference of the methylene compounds fol-
lows with their acidity scale, which increases in the order
EAA (pKa =13.1)<MCA (pKa =12.8)<MNT (pKa =11.1)
[53–55].
An explanation of the catalytic results in terms of used
aldehyde is that n-heptanal, benzaldehyde and cyclohex-
ane carbaldehyde have sizes of 2.1×8.1 Å, 4.0×6.0 Å and
2.6 × 6.3 Å (see Fig. S3 in the Supporting information),
which allows them to enter and exit from 1′ freely. As it
was described above, in the framework of 1, there are 1D
jar-like cavities with effective pore size about 14.98×7.88 Å
and 14.98×13.17 Å. This is the main reason why catalyst
demonstrated excellent yields towards the Knoevenagel
condensation reaction of these aldehydes. While slightly
larger aldehyde substrates, such as 4-tert-butylbenzaldehyde,
3,5-ditert-butylbenzaldehyde and 2-naphthaldehyde, showed
significantly decreased conversions and yields. The sizes of
4T-PhCHO, 3,5T-PhCHO and 2-NaphCHO are 4.1×8.3 Å,
8.6 × 8.4 Å and 4.0 × 8.7 Å, respectively which are quite
close to the pore size of 1. Moreover, these results also show
on the size effect, the probability of bulkier aldehydes with
methyl cyanoacetate or ethyl acetoacetate to form transition
state complexes is significantly reduced due to the limited
space in the pore system of the catalyst.
It was found that the activity and selectivity of 1′ in con-
densation reactions significantly decreased after decreas-
ing of the reaction temperature from 130 to 70 °C. As an
example, the reaction of benzaldehyde with malononitrile
is presented. When the reaction temperature decreases from
130 to 100 °C and finally to 70 °C, conversion (selectivity)
decreases in the order 100% (100%), 72% (93%) and 35%
(66%), respectively. The results could reflect high energy
barrier of transition state, which could not be achieved at
lower reaction temperature.
Although conversion and selectivity over catalyst 1′ gave
satisfactory results, they were achieved at relatively high
temperature. For comparison, MOF catalysts giving the best
results in Knoevenagel condensation of aldehydes and active
methylene compounds (especially containing basic sites)
reached the comparable conversion and selectivity values at
lower temperatures (25–40 °C) [2, 41, 42].
From the results of catalytic measurements, it can be con-
cluded, that the selectivity (more than 98% for all reactions)
and conversion to the target product were in most cases
higher for condensation of aldehydes with malononitrile
compared to methyl cyanoacetate and ethyl acetoacetate (see
Table 2). The sequence of substrate reactivity is in good
agreement with our previous reports [2, 38].
The key advantage of the use of heterogeneous catalysts
is their easy recovery from the reaction mixtures and their
reuse. We examined the recyclability of 1′ in Knoevenagel
condensation reactions of n-heptanal and cyclohexane car-
baldehyde with malononitrile by reusing recovered catalysts
for the 2nd and 3rd catalytic runs under the same condi-
tions as the 1st run. In each test, the catalyst was separated
by centrifugation, stirred in solvents with different polar-
ity (ethanol, acetone, n-hexane), filtered off, reactivated at
200 °C and then reused in the next run. As shown in Fig. 6,
the conversion curves of reused catalyst in the reaction of
malononitrile with n-heptanal displayed steady acceleration,
An explanation of the catalytic results in terms of used
active methylene compounds is that the reactivity trend
depends on the strong accelerating influence of electron
withdrawing groups bonded to active methylene cen-
trum (–CN > –OMe > –OEt). The stronger electron with-
drawing ability of the substituent, the easier activation of
active methylene compound for electrophilic attack at the
R–CH2–R group and hence higher conversion. Furthermore,
1 3

Efficient and Reusable Pb(II) Metal–Organic Framework for Knoevenagel Condensation
Pb(II) ions from the framework of 1′, performed under the
Knoevenagel condensation of n-heptanal with malononitrile
at 130 °C, provided that leaching of the active sites is not
responsible for the high catalytic activity of 1′. After 10 min
reaction, the solid catalyst was filtered off and the same reac-
tion mixture was continued to react for further 60 min with
aliquots being taken at regular time intervals and analysed by
GC. As can be seen from Fig. 7, GC analysis proved stop of
reaction at ~60% conversion, indicating absence of leaching
of the active species from 1′. Negative leaching test of Pb(II)
ions form 1′ was also confirmed by ICP-MS measurement.
The possible incorporation or immobilization of sub-
strates in the catalyst was investigated in the reaction of
n-heptanal with malononitrile by the combination of ther-
mogravimetry and infrared spectroscopy. Figure S4 in SI
shows TG/DTG–DTA analysis of the used, refreshed and
activated catalyst in the temperature range 30–900 °C. After
heating of the sample to 300 °C no significant mass change
on TG was observed indicating the absence of solvents in the
pore system of 1′. The thermal decomposition takes place in
the temperature range 300–600 °C, which is in good agree-
ment with pyrolysis of the framework (see Figs. S1 and S2
in the Supporting information). The observed mass change
Fig. 6 Conversion and selectivity (inset) curves using reused catalyst
after 1st (black squares), 2nd (red ball) and 3rd (green triangle) cata-
lytic run in the condensation reaction of n-heptanal and malononitrile
at 130 °C
with slight decreasing of selectivity during three repetitions.
A similar trend was observed in the repetition reactions with
cyclohexane carbaldehyde and malononitrile (see Table 2).
Results indicate, that 1′ possesses good recyclability without
significant degradation in the original catalytic activity.
When a solid catalyst is employed for a liquid-phase
organic transformation, an important problem, that should
be seriously considered, is the possibility that some of the
active species could release into the reaction solution. In
some cases, these leached species could contribute signifi-
cantly to the catalytic activity during the reaction. Therefore,
we have performed a leaching test. The leaching test for
Fig. 7 Leaching test of Pb(II) ions from the framework of 1′ under
heterogeneous and homogeneous (after centrifugation procedure)
conditions in the Knoevenagel condensation of n-heptanal and malo-
nonitrile at 130 °C
Fig. 8 Infrared spectra of a malononitrile, b n-heptanal and c 1′ after
reaction of heptanal and malononitrile at 130 °C
1 3

M. Almáši et al.
products, as confirmed by the thermal analysis and IR meas-
urements described above (Fig. S4 and Fig. 8).
4 Conclusions
Lead–organic framework with composition {[Pb4(µ8-MTB)2
(H2O)4]·5DMF·H2O}n was prepared and tested as catalyst in
Knoevenagel condensation. Compound properties and pro-
cess of activation for catalysis was studied by different phys-
icochemical techniques such as thermogravimetry, infrared
spectroscopy measured at different temperatures, PXRD and
nitrogen adsorption measurements. To obtain the best poros-
ity and surface area characteristics, conditions of activation of
compound at different temperatures and time intervals were
investigated. The best textural characteristics with BET sur-
face area of 980 m2 g−1 and micropore volume 0.666 cm3 g−1
were obtained after ethanol exchange process and evacuation
at 200 °C under dynamic vacuum. Activation procedure led
to the formation of open framework and generation of unsatu-
rated coordination sites on lead(II) ions as active centres for
catalysis. The catalytic properties of activated compound (1′)
were tested in condensation reactions of different aldehydes
with active methylene compounds at different temperatures
(130–70 °C). Based on the results of the catalytic experiments
it was shown, that catalyst is less efficient for Knoevenagel
condensation of larger aromatic aldehydes such as 4-tert-butyl-
benzaldehyde, 3,5-ditert-butylbenzaldehyde and 2-naphthalde-
hyde with selected active methylene compounds and the con-
version decreased when the aldehyde substrate became bulkier.
On the other hand, aldehydes with lower kinetic diameter like,
n-heptanal, cyclohexane carbaldehyde and benzaldehyde dem-
onstrated high conversions and yields to desired products. It
was demonstrated that the reaction proceeds mainly inside the
pore system of the catalyst and the framework has reasonable
stability under the used conditions. Moreover, 1′ possesses
good recyclability without significant decrease in the original
catalytic activity.
Fig. 9 PXRD patterns of 1 before and after the Knoevenagel conden-
sation (reactions with malononitrile at 130 °C)
in reused catalyst was 70.0% and it is higher in comparison
to that observed in the thermal decomposition of 1 (37.8%).
The difference between mass changes corresponds to immo-
bilization of four molecules of 2-heptylidenemalononitrile
as condensation product per formula unit of 1′.
Furthermore, immobilization of substrates/products was
also studied by infrared spectroscopy. IR spectra of malon-
onitrile, n-heptanal and 1′ after the third catalytic run, are
illustrated in Fig. 8. Malononitrile is characterized in the
spectra by two absorption bands at 2966 and 2933 cm−1
corresponding to ν(CH)aliph of active methylene centrum
(–CH2–) and one absorption band at 2272 cm−1 of nitrile
group (ν(CN)), Fig. 8a. In the IR spectrum of n-heptanal,
ν(CH)aliph and ν(C=O) vibrations in region 2980–2860 cm−1
and at 1727 cm−1, respectively are characteristic (Fig. 8b).
IR spectrum of reused catalyst (Fig. 8c) exhibits some differ-
ences compared to that of freshly prepared sample (Fig. 2).
As it can be seen from the Fig. 8c, new characteristic vibra-
tions for aldehyde in the range 2980–2860 cm−1 (ν(CH)aliph
)
Acknowledgements This work was supported by the Scientific Grant
Agency of the Slovak Republic (VEGA) Project No. 1/0745/17, by
the Slovak Research and Development Agency under the contracts
APVV-0073-14 and APVV-15-520 and the project from P. J. Šafárik
University No. VVGS-2016-249. V.Z. and M.A. thank the Ministry of
Education, Science, Research and Sport of the Slovak Republic for the
financial support of the TRIANGEL team in the frame of the scheme
“Top Research Teams in Slovakia”. J.Č. acknowledges the Czech Sci-
ence Foundation for the support of the project P106/12/G015.
and nitrile bending vibration (ν(CN)) at 2272 cm−1 for
malononitrile were observed. The absence of ν(CH)aliph
originated from malononitrile and ν(C=O) of aldehyde in
the spectrum of reused 1′ confirm the immobilisation of
condensation product, 2-heptylidenemalononitrile in the
framework.
The stability of the catalyst after 24 h reaction of aldehydes
with malononitrile at 130 °C was monitored by PXRD. As it
follows from the PXRD patterns (see Fig. 9), small changes in
the diffraction patterns of 1 and the patterns obtained from the
sample after Knoevenagel condensation were observed. The
observed changes of the intensity ratios of diffraction peaks
and small shifts of the corresponding diffraction lines are
related to the interaction of the framework of 1 with reaction
References
1. Jones G (2004) The Knoevenagel condensation. In: Organic reac-
tions, 2nd edn, vol 15. Wiley, New York, pp 204–599
2. Dhakshinamoorthy A, Opanasenko M, Čejka J, Garcia H (2013)
Catal Sci Technol 3:2509–2540
1 3

Efficient and Reusable Pb(II) Metal–Organic Framework for Knoevenagel Condensation
30. Kurmoo A (2009) Chem Soc Rev 38:1353–1379
3. Walker SD, Borths CJ, DiVirgilio E, Huang L, Liu P, Morrison H,
Sugi K, Tanaka M, Woo JCS, Faul MM (2011) Org Process Res
Dev 15:570–580
31. Opanasenko M, Shamzhy M, Lamač M, Čejka J (2013) Catal
Today 204:94–100
32. Zhu L, Liu XQ, Jiang HL, Sun LB (2017) Chem Rev
117:8129–8176
4. Jung JC, Lim E, Lee Y, Min D, Ricci J, Park OS, Jung M (2012)
Molecules 17:2091–2102
33. Hu Z, Zhao D (2017) CrystEngComm 19:4066–4081
34. Zhu NX, Zhao CW, Wang JC, Li YA, Dong YB (2016) Chem
Commun 52:12702–12705
5. Tietze LF, Rackelmann N (2004) Pure Appl Chem 76:1967–1983
6. Gouda MA, Abu-Hashem AA (2012) Green Chem Lett Rev
5:203–209
35. Burgoyne AR, Meijboom R (2013) Catal Lett 143:563–571
36. Toyao T, Fujiwaki M, Horiuchi Y, Matsuoka M (2013) RSC Adv
3:21582–21587
7. Liang F, Pu YJ, Kurata T, Kido J, Nishide H (2005) Polymer
46:3767–3775
8. Bhat AR, Selokar RS, Meshram JS, Dongre RS (2014) J Mater
Environ Sci 5:1653–1657
37. Canivet J, Vandichela M, Farrusseng D (2016) Dalton Trans
43:4090–4099
9. Pasha MA, Manjula K (2011) J Saudi Chem Soc 15:283–286
10. Ma M, Li H, Yang W, Wu Q, Shi D, Zhao Y, Feng C, Jiao Q
(2018) Catal Lett 148:134–143
38. Almáši M, Zeleňák V, Opanasenko M, Čejka J (2014) Dalton
Trans 43:3730–3738
39. Luan Y, Qi Y, Gao H, Andriamitantsoa RS, Zheng N, Wang G
(2015) J Mater Chem A 3:17320–17331
11. Priede E, Brica S, Bakis E, Udris N, Zicmanis A (2015) New J
Chem 39:9132–9142
40. Almáši M, Zeleňák V, Gyepes R, Bourrelly S, Opanasenko M,
Llewellyn PL, Čejka J (2018) Inorg Chem 57:1774–1786
41. Panchenko VN, Timofeeva MN, Jhung SH (2016) Catal Rev
58:209–307
12. Zhu F, Sun X, Lou F. An L, Zhao P (2015) Catal Lett
145:1072–1107
13. Wang H, Wang C, Yang Y, Zhao M, Wang Y (2017) Catal Sci
Technol 7:405–417
42. Valvekens P, Vandichel M, Waroquier M, Van Speybroeck V, De
Vos D (2014) J Catal 317:1–10
14. Maloo P, Roy TK, Sawant DM, Pardasani RT, Salunkhe MM
(2016) RSC Adv 6:41897–41906
43. Serra-Crespo P, Ramos-Fernandez EV, Gascon J, Kapteijn F
(2011) Chem Mater 23:2565–2572
15. Isobe K, Hoshi T, Suzuki T, Hagiwara H (2005) Mol Divers
9:317–320
44. Tran UPN, Le KKA, Phan NTS (2011) ACS Catal 1:120–127
45. Amarante SF, Freire MA, Mendes DTSL, Freitas LS, Ramos ALD
(2017) Appl Catal A 548:47–51
16. Khurana JM, Vij K (2010) Catal Lett 138:104–110
17. Li JPH, Stockenhuber M (2015) Catal Today 245:108–115
18. Ikeue K, Miyoshi N, Tanaka T, Machida M (2011) Catal Lett
141:877–881
46. Fang QR, Yuan DQ, Sculley J, Li JR, Han ZB, Zhou HC (2010)
Inorg Chem 49:11637–11642
19. Motokura K, Viswanadham N, Dhar GM, Iwasawa Y (2009) Catal
Today 141:19–24
47. Spectral database for organic compounds, National Institute of
Advanced Industrial Science and Technology. http://riodb01.ibase
.aist.go.jp/sdbs/
20. Tamami B, Fadavi A (2005) Catal Commun 6:747–751
21. Li G, Xiao J, Zhang W (2012) Green Chem 14:2234–2242
22. Xu J, Shen K, Xue B, Li YX, Cao Y (2013) Catal Lett
143:600–609
48. Cheon YE, Suh MP (2008) Chem Eur J 14:3961–3967
49. Almáši M, Zeleňák V, Gyepes R, Zukal A, Čejka J (2013) Colloids
Surf A 437:101–107
23. Srasra M, Delsarte S, Gaigneaux EM (2009) Catal Lett
52:1541–1548
50. Cheon YE, Suh MP (2009) Chem Commun 2296–2298. https://
doi.org/10.1039/B900085B
24. Morris RE, Čejka J (2015) Nat Chem 7:381–388
25. Almáši M, Zeleňák V, Zukal A, Kuchár J, Čejka J (2016) Dalton
Trans 45:1233–1242
51. Ma L, Jin A, Xie Z, Li W (2009) Angew Chem Int Ed
48:9905–9908
52. Dhakshinamoorthy A, Heidenreich N, Lenzen D, Stock N (2017)
CrystEngComm 19:4187–4193
26. Qiu S, Xue M, Zhu G (2014) Chem Soc Rev 43:6116–6140
27. Li JR, Kuppler RJ, Zhou HC (2009) Chem Soc Rev 38:1477–1504
28. Ke F, Yuan YP, Qiu LG, Shen YH, Xie AJ, Zhu JF, Tian XY,
Zhang LD (2001) J Mater Chem 21:3843–3848
29. Qu XL, Gui D, Zheng XL, Li R, Han HL, Li X, Li PZ (2016)
Dalton Trans 45:6983–6989
53. Olmstead WN, Margolin Z, Bordwell FG (1980) J Org Chem
45:3295–3299
54. Bordwell FG, Fried HE (1981) J Org Chem 46:4327–4331
55. Ugale B, Nagaraja CM (2016) RSC Adv 6:28854–28864
Affiliations
Miroslav Almáši1 · Vladimír Zeleňák1 · Maksym V. Opanasenko2 · Jiří Čejka2
1
* Vladimír Zeleňák
Department of Inorganic Chemistry, Faculty of Science,
vladimir.zelenak@upjs.sk
P. J. Šafárik University, Moyzesova 11, 041 54 Košice,
Slovak Republic
Miroslav Almáši
2
miroslav.almasi@upjs.sk
Department of Synthesis and Catalysis, J. Heyrovský
Institute of Physical Chemistry of the ASCR, v.v.i., Academy
of Sciences of the Czech Republic, Dolejškova 2155/3,
182 23 Prague 8, Czech Republic
Maksym V. Opanasenko
maksym.opanasenko@jh-inst.cas.cz
Jiří Čejka
jiri.cejka@jh-inst.cas.cz
1 3