A novel nickel metal-organic framework with fluorite-like structure: Gas adsorption properties and catalytic activity in Knoevenagel condensation

Article abstract of DOI:10.1039/c3dt52698d

A new non-interpenetrating 3D metal-organic framework {[Ni ₄(μ₆-MTB)₂(μ₂-H ₂O)₄(H₂O)₄] ·10DMF·11H₂O}_n (DMF = N,N′- dimethylformamide) built from

Full text of DOI:10.1039/c3dt52698d

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
A novel nickel metal–organic framework with
fluorite-like structure: gas adsorption properties
and catalytic activity in Knoevenagel
condensation†
Cite this: Dalton Trans., 2014, 43,
3730
Miroslav Almáši,a Vladimír Zeleňák,*a Maksym Opanasenkob and Jíří Čejka*b
A new non-interpenetrating 3D metal–organic framework {[Ni4(μ6-MTB)2(μ2-H2O)4(H2O)4]·10DMF·11-
H2O}n (DMF = N,N’-dimethylformamide) built from nickel(II) ions as connectors and methanetetrabenzo-
ate ligands (MTB4−) as linkers has been synthesized and characterized. The single crystal X-ray diffraction
showed that complex exhibits CaF2-like fluorite structure topology and four types of 3D channels with
sizes about 12.6 × 9.4 Å2, 9.4 × 8.0 Å2, 12.6 × 11.7 Å2 and 14.9 × 14.9 Å2, which are filled with guest mole-
cules. Conditions of the activation of the compound have been studied and optimized by powder X-ray
diffraction during in situ heating, thermogravimetric analysis and infrared spectroscopy. Nitrogen and
carbon dioxide adsorption showed that the activated sample exhibits a BET specific surface area of
700 m2 g−1 and a carbon dioxide uptake of 12.36 wt% at 0 °C, which are the highest values reported for
the compounds of the MTB4− series. The complex was tested in Knoevenagel condensation of aldehydes
and active methylene compounds. Straightforward dependence of the substrate conversion on the size of
used aldehyde was established. A possible mechanism of Knoevenagel condensation over a MTB4− con-
taining a metal–organic framework was proposed.
Received 27th September 2013,
Accepted 17th December 2013
DOI: 10.1039/c3dt52698d
www.rsc.org/dalton
Particularly, the use of MOFs in different catalytic processes
is a very intensively studied area in recent years, in particular,
Introduction
Porous metal–organic frameworks (MOFs) represent a class of the issues of MOF design for heterogeneous catalysis8 and
crystalline micro-/mesoporous materials constructed by assem- application of MOFs in asymmetric catalysis.9 Acidic and basic
bly of metal cations or their clusters as connectors, and MOFs can exhibit high catalytic activity in condensation reac-
usually anionic organic ligands as linkers.1 The final structure tions like the Pechmann,10 Knoevenagel,11 and Prins reac-
and dimensionality of MOFs depends on several factors, such tions.12 The performance of MOFs frequently surpasses those
as molar ratio of reactants, composition and ratio of the of conventional porous solid catalysts such as clays and zeo-
mixture of solvents, choice of organic linker and inorganic lites.13 On the other hand, the lower stability and possibility of
salt, pH value, reaction temperature etc.2 One of the most deactivation of MOF catalysts during the course of the reaction
important properties of porous MOFs is their high surface should be taken into account when considering MOFs in the
area, which may lead to applications in gas storage and separ- catalysis.14
ation,3 achiral and chiral catalysis,4 drug delivery,5 sensor tech-
nology,6 and polymerization reactions in the channels.7
The other emerging and intensively investigated area is gas
adsorption, which is targeted at the use of MOFs in gas separ-
ation and storage. One of the most technologically important
gases is carbon dioxide. The selective capture of CO2 in porous
materials has potential for purification of fuels or flue gases or
for CO2 sequestration. It was reported that the adsorption
capacities of carbon dioxide at ambient temperature in some
porous MOFs can be significantly higher than in currently
used commercial materials.15
Synthesis of MOFs can be considered as a construction
game, because the resulting architecture, design and robust-
ness of compounds depend on appropriately selected building
components and their compatibility. In our work, we have
aDepartment of Inorganic Chemistry, Faculty of Science, P. J. Šafárik University,
Moyzesova 11, 041 54 Košice, Slovak Republic. E-mail: vladimir.zelenak@upjs.sk;
Tel: +00421 55 2342343
bDepartment 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,
128 23 Prague, Czech Republic. E-mail: jiri.cejka@jh-inst.cas.cz;
Fax: +00420 2865 823 07; Tel: +00420 26605 3795
†Electronic supplementary information (ESI) available. CCDC 963547. For ESI
and crystallographic data in CIF or other electronic format see DOI: 10.1039/
c3dt52698d
3730 | Dalton Trans., 2014, 43, 3730–3738
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
used the methanetetrabenzoate (MTB4−) ligand, which has an Ni4C88H138N10O44 (2274.86 g mol−1); calculated: C, 46.10%; H,
ideal stereochemical arrangement for creating novel MOFs 6.15%; N, 6.11%; found: C, 46.40%; H, 6.24%; N, 6.24%. FT-IR
with 3D porous structure. To date, only five complexes contain- (KBr technique, cm−1): ν(OH) 3400(m, br); ν(CH)aliph. 2958(w),
ing the MTB4− ligand have been reported. These compounds 2931(w); ν(CvO) 1665(s); ν(COO−)asym. 1597(s); ν(CvC)ar. 1542(s),
contain Zn(II),16 Cu(II),17 Ni(II),18 Co(II)19 and Cd(II)20 metal 1495(m); ν(COO−)sym. 1393(s); δ(CCH)ar. 1186(w), 1146(w),
centres and have four different structures and topologies. The 1102(m); δ(COO−) 783(s) (s – strong, m – medium, w – weak,
porous frameworks of zinc(II) and copper(II) forms are isostruc- aliph. – aliphatic, sym – symmetric, asym – antisymmetric,
tural having composition {[Zn2(μ8-MTB)(H2O)2]·3DMF·3H2O}n ar – aromatic, br – broad).
and {[Cu2(μ8-MTB)(H2O)2]·6DEF·2H2O}n (DEF = N,N′-diethyl-
formamide), respectively. Both of them contain paddle-wheel
secondary building units (SBUs) leading to non-interpenetrat-
ing PtS-like structure. The Ni(II) form has composition {[Ni2-
(μ4-MTB)(κ4-CYC)2]·4DMF·8H2O}n and its structure is formed
by the fourfold interpenetrating diamond-like network. The
cadmium(II) complex has formula {[Cd4(μ8-MTB)2(DMF)4]·
4DMF·4H2O}n with structure created by cyclotetranuclear clus-
ters spatially arranged in a CaF2-like framework.
Characterization
The elemental analysis was performed using
a CHNOS
Elemental Analyzer vario MICRO from Elementar Analysensys-
teme GmbH. The infrared spectra of the sample at room temp-
erature and after heating to 100, 200, 300, and 400 °C were
recorded using an Avatar FTIR 6700 spectrometer in the range
4000–400 cm−1. The sample was prepared in KBr pellets with a
complex/KBr weight ratio of 1/100. Before IR measurements,
KBr was dried at 500 °C for 3 h in an oven and cooled in a
desiccator. Thermal behaviour was studied by thermogravi-
metry (TGA) with simultaneous differential thermal analysis
(DTA). The sample with a weight of 20 mg, placed in an Al2O3
crucible, was heated in the region of 25–1000 °C in the
dynamic atmosphere of air (80 cm3 min−1), using STA Netzsch
In this study, a new compound of the MTB4− series, namely
{[Ni4(μ6-MTB)2(μ2-H2O)4(H2O)4]·10DMF·11H2O}n (1), was syn-
thesized by the solvothermal method. After activation, com-
pound 1 represents a porous framework with high adsorption
capacity to CO2. The investigation of the catalytic properties of
1 in Knoevenagel condensation of different bulky aldehydes
(heptanal, cyclohexane carbaldehyde, benzaldehyde, 4-tert-
butylbenzaldehyde and 3,5-ditertbutylbenzaldehyde) with active
methylene compounds (malononitrile, methyl cyanoacetate
and ethyl acetoacetate) showed that compound 1 exhibits high
activity in acid-catalysed Knoevenagel condensation, reaching
up to 100% yields without a significant loss in activity.
409-PC apparatus and a heating rate of 5 °C min−1
.
Powder (PXRD) and high-energy powder X-ray diffraction
(HE-PXRD) experiments
HE-PXRD measurements were carried out at the BW5 wiggler
beamline of the DORIS positron storage ring in DESY
(Hamburg, Germany). The patterns were collected during
in situ heating with a heating rate of 10 °C min−1 and a beam
Experimental section
Materials
All chemicals used in the synthesis of the complex and
reagents used in catalytic reactions were obtained from Sigma-
Aldrich Company and used without further purification.
Methanetetracarboxylic acid (H4MTB) was prepared according
to a literature procedure.16 Heptanal (95%), cyclohexane car-
baldehyde (97%), benzaldehyde (≥99%), 4-tertbutylbenzalde-
hyde (97%), 3,5-ditertbutylbenzaldehyde (97%), malononitrile
(≥99%), methyl cyanoacetate (99%) and ethyl acetoacetate
(≥99%) were received as commercial samples, n-dodecane
(≥99%) was used as an internal standard and p-xylene (≥99%)
as the solvent in catalytic experiments.
energy of 100 keV, corresponding to
a wavelength of
0.123984 Å. The gradient Ge/Si (111) single crystal was used to
monochromatize the polychromatic beam. Two pairs of slits
were used to set the beam width and height to 1 × 1 mm2. The
diffracted photons were collected using an image plate detec-
tor from Perkin-Elmer (2300 × 2300 pixels, each pixel having a
size of 150 × 150 μm2) carefully aligned orthogonal to the X-ray
beam. The sample-to-detector distance and the detector tilt
were determined from diffraction patterns obtained from LaB6
(SRM 660a) standard reference material. The two-dimensional
diffraction patterns were integrated using the QXRD software.
The observed X-ray scattering intensities were corrected for
background, absorption, polarization, incoherent and multiple
Synthesis of {[Ni4(μ6-MTB)2(μ2-H2O)4(H2O)4]·10DMF·11H2O}n
The mixture of Ni(NO3)2·6H2O (0.2 mmol, 58.6 mg) and scattering effects.
H4MTB (0.04 mmol, 20 mg) in 3 ml DMF, 3 ml ethanol and
The crystallinity of the sample after catalytic runs was deter-
2.4 ml H2O was placed in a 20 ml glass Ace® tube. The reaction mined using a Bruker AXS D8 ADVANCE diffractometer in the
mixture was heated to 80 °C with a heating rate of 1 °C min−1 Bragg-Brentano geometry using CuKα (1.54056 Å) radiation
and kept at 80 °C at this temperature for 48 h. Then, the with a NaI dynamic scintillation detector (Vantec-1) in the
mixture was cooled down to room temperature with a rate of range 2θ = 1–30°. To limit the effect of the preferential orien-
0.1 °C min−1. The green crystals of {[Ni4(μ6-MTB)2(μ2-H2O)4- tation of individual crystals, gentle grinding of the sample to
(H2O)4]·10DMF·11H2O}n were filtered, slightly washed with decrease their size and careful packing into the holder were
DMF and dried. Elemental analysis of the compound performed.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 3730–3738 | 3731

View Article Online
Paper
Dalton Transactions
Structure determination and refinement
experiment workstation StarFish. Before catalytic experiments,
50 mg of the catalyst was activated at 250 °C for 90 min with a
temperature heating rate of 10 °C min−1 in a stream of air.
Typically, 4 mmol of aldehyde, 0.4 g of dodecane (internal
standard), 50 mg of catalyst and 10 ml of p-xylene (solvent)
were added to the three-necked vessel, equipped with a con-
denser and a thermometer, stirred and heated. Then 6 mmol
of a methylene compound was added into the reaction vessel
through a syringe when the desired reaction temperature was
reached.
0.2 ml of the reaction mixture was sampled using a syringe
with a needle after 10, 30, 60, 120, 240, 360 and 1440 min. The
reaction products were analysed by gas chromatography (GC)
using an Agilent 6850 with a 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 (Varian Unity 300
spectrometer). 1H chemical shifts in CDCl3 were given with
tetramethylsilane as the internal standard. The observed
spectra corresponded to the previously published data.26
To evaluate the potential influence of leaching of the active
species from the heterogeneous catalysts, a part of the reaction
mixture was taken, immediately centrifuged and the obtained
liquid phase was again investigated in the Knoevenagel con-
densation under the same reaction conditions.
The compound {[Ni4(μ6-MTB)2(μ2-H2O)4(H2O)4]·11H2O·10DMF}n
crystallizes in the form of green needles. The intensity data
were collected at −100 °C with a graphite monochromator
using MoKα radiation (λ = 0.71073 Å) on a four-circle κ-axis
Xcalibur2 diffractometer equipped with
a CCD detector
Sapphire2 (Oxford Diffraction). The CrysAlis software
package21 was used for data collection and reduction. The
structure was solved by direct methods using SIR9222 incorpor-
ated in the WINGX program package23 and refined by full-
matrix squares on F2 using the SHELXT-97 program package.24
All non-hydrogen atoms were refined anisotropically and all
hydrogen atoms were included in the calculated positions. The
isotropic thermal parameters were fixed at 1.2Ueq of the parent
atom for all H atoms. The structure figures were drawn using
DIAMOND software.25 Crystal data for {[Ni4(μ6-MTB)2(μ2-H2O)4-
(H2O)4]·11H2O·10DMF}n: green crystals 0.55 × 0.14 × 0.12 mm3,
orthorhombic, space group Pnnm, a = 14.993(2) Å, b = 17.232(2)
Å, c = 24.798(3) Å, V = 6406.7(12) Å3, Z = 4, Dclcd = 0.792 g cm−3
,
μ = 0.621 mm−1. Of the 6782 total reflections collected
(−18 ≤ h ≤ 10, −21 ≤ k ≤ 20, −19 ≤ l ≤ 31), 3242 were
unique (Rint = 0.0837) and were used to solve the structure.
Based on these data and 212 refined parameters, the final
R indices [I > 2σ(I)] R1 = 0.0863, wR2 = 0.2356; R indices
(all data), R1 = 0.1365, wR2 = 0.2556 and the goodness-of-fit on
F2 is 0.899.
For the 2nd and 3rd catalytic runs, catalysts from the reac-
tion mixtures were separated by centrifugation and washed
with ethanol several times. Then the samples were activated at
250 °C.
Nitrogen and carbon dioxide adsorption measurements
Nitrogen adsorption isotherms were taken using a Quanta-
chrome Nova 1200e automatic adsorption analyser at −196 °C.
Before the nitrogen adsorption, the sample was activated by
evacuation at different temperatures (150, 250, and 300 °C) for
24 h to remove the guest molecules from the channel
system. Different evacuation temperatures were used to find
Results and discussion
the optimal conditions for the activation of the compound (see Description of the structure
ESI, Fig. S2†). The adsorption isotherms of the desolvated
The crystal structure of 1 is shown in Fig. 1, and the principal
sample were collected in a relative pressure range from p/p0 =
0.005 to 1. Based on the nitrogen adsorption measurements,
the BET specific surface area (SBET) of the sample was evalu-
ated using adsorption data in a p/p0 range from 0.005 to 0.20.
Carbon dioxide adsorption isotherms were recorded at
0 and 22 °C using the Iso-Therm thermostat (e-Lab Services,
Czech Republic) maintaining the temperature of the sample
with an accuracy of 0.01 °C. The adsorption isotherms of
carbon dioxide were measured on the same batch of the
sample as nitrogen adsorption. Since the batch of the sample
used for CO2 adsorption was already desolvated for nitrogen
adsorption experiments, for CO2 sorption the outgassing pro-
cedure was shorter and typical outgassing was performed at
250 °C for 8 h.
bond lengths and angles are summarized in Table 1. Single
X-ray diffraction analysis showed that the compound crystal-
lized in the orthorhombic space group Pnnm with four
formula units in the cell. Compound 1 is isostructural with
the Co(II) analogue.19 Its crystal structure is formed by linear
tetranuclear nickel(II) clusters of composition [Ni4(μ2-COO)4-
(COO)4(μ2-H2O)4(H2O)4] (COO = carboxylate), creating eight-
connected elongated tetragonally prismatic SBUs (Fig. 1a,
dashed line). The cluster Ni1–Ni2–Ni2A–Ni1A is built from two
crystallographically independent cations (Ni1 and Ni2), which
are connected by four water molecules (O1, O2, O1A, O2B) and
four syn–syn carboxylate groups of four different MTB4−
anions. The Ni⋯Ni distances in the clusters are 3.4856(14) Å
for Ni1–Ni2 and 3.2552(12) Å for Ni2–Ni2A, respectively. Each
nickel atom is coordinated by six oxygen atoms forming an
elongated tetragonal bipyramid (see Fig. 1a). Four equatorial
Catalysis and reaction product analysis
Knoevenagel condensation of aldehydes and active methylene positions of an elongated tetragonal bipyramid around Ni1
compounds was performed in liquid phase under atmospheric atoms are occupied by two oxygen atoms (O5, O5C) of mono-
pressure and temperatures of 100–130 °C in
a
multi- dentate MTB4− anions and by two oxygen atoms (O4, O4C) of
3732 | Dalton Trans., 2014, 43, 3730–3738
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Fig. 2 View showing the CaF2-like network of 1.
between Ni1 and Ni2 atoms. Equatorial positions of Ni2 are
occupied by four oxygen atoms (O2, O2B, O3, O3C).
This connectivity gives rise to the 3D structure with struc-
ture channels extending along the a, b crystallographic axes
and along the [−4,5,0] crystallographic direction (Fig. 1b–d).
Sizes of the channel apertures are 12.6 × 9.4, 9.4 × 8.0, 12.6 ×
11.7 and 14.9 × 14.9 Å2, respectively. The channels are filled
with DMF and water molecules. Due to the high positional dis-
order of guest molecules these could not be refined unambigu-
ously by X-ray diffraction data. The presence of guest
molecules was determined by the combination of elemental
analysis, IR spectroscopy and TGA analysis (see below). The
void space in the total crystal volume after removal of the guest
molecules, determined using the PLATON program, is 4350 Å3
(cell volume 6407 Å3) and corresponds to 68% of the crystal
volume. Compound 1 exhibits rare fluorite (CaF2) topology
(see Fig. 2). The central C atom of the MTB4− ligand can be
considered the F vertex and the Ni4 cluster decorates the Ca2+
positions in the CaF2 structure. The formed structure is stable
in air and in water. No hydrolysis and structure change were
observed after the immersion of the compound into water for
several days.
Fig. 1 (a) Coordination environment of Ni4 clusters. Hydrogen atoms
have been omitted for clarity. Symmetry operations: A: 1 − x, −y, z; B:
1 − x, −y, −z; C: x, y, −z. View of the crystal structure along the a axis (b),
b axis (c), and along the [−4,5,0] crystallographic direction (d).
Table 1 Selected bond distances (Å) and angles (°) in {[Ni4(μ6-MTB)2-
(μ2-H2O)4(H2O)4]·11H2O·10DMF}n
Bond distances
Ni1–O4A
Ni1–O4
Ni1–O5
Ni1–O5A
Ni1–O7
Ni1–O1
Ni2–O3A
Ni2–O3
Ni2–O1
2.015(4)
2.015(4)
2.060(4)
2.060(4)
2.108(6)
2.110(5)
1.997(4)
1.997(4)
2.042(5)
Ni2–O8
Ni2–O2
Ni2–O2B
O5–C9
O3–C1
O4–C1
2.088(6)
2.143(4)
2.143(4)
1.211(7)
1.246(6)
1.225(6)
2.143(4)
1.290(7)
Spectral, thermal and HE-PXRD studies
O2–Ni2B
O6–C9
The IR spectra of the {[Ni4(μ6-MTB)2(μ2-H2O)4(H2O)4]·
11H2O·10DMF}n are shown in Fig. 3 and wavenumbers of the
principal IR spectral bands are listed in the Experimental
section. The spectra measured at room temperature exhibit a
Bond angles
O4A–Ni1–O4
O4A–Ni1–O5
O4–Ni1–O5
O4A–Ni1–O5A
O4–Ni1–O5A
O5–Ni1–O5A
O4A–Ni1–O7
O4–Ni1–O7
O5–Ni1–O7
O5A–Ni1–O7
O4A–Ni1–O1
O4–Ni1–O1
O5–Ni1–O1
O5A–Ni1–O1
O7–Ni1–O1
OC–Ni2–O3
O3A–Ni2–O1
O3–Ni2–O1
O3A–Ni2–O8
O3–Ni2–O8
92.6(3)
O1–Ni2–O8
O3A–Ni2–O2
O3–Ni2–O2
O1–Ni2–O2
O8–Ni2–O2
O3A–Ni2–O2A
O3–Ni2–O2B
O1–Ni2–O2B
O8–Ni2–O2B
O2–Ni2–O2B
C9–O5–Ni1
Ni2–O1–Ni1
C1–O3–Ni2
C1–O4–Ni1
Ni2–O2–Ni2B
O4–C1–O3
177.7(2)
169.27(15)
89.54(17)
93.04(11)
85.22(13)
89.54(17)
169.27(15)
93.04(11)
85.22(13)
81.2(2)
127.2(4)
114.2(2)
135.1(4)
135.5(4)
98.8(2)
174.41(16)
88.91(18)
88.91(18)
174.41(16)
89.1(2)
87.60(16)
87.60(16)
87.09(15)
87.09(15)
93.02(15)
93.02(15)
92.26(14)
92.26(14)
179.1(2)
99.2(2)
126.4(5)
115.4(5)
118.1(5)
125.2(6)
119.2(5)
92.82(15)
92.82(15)
88.67(16)
88.67(16)
O4–C1–C2
O3–C1–C2
O5–C9–O6
O5–C9–C10
μ2-bridging carboxylate MTB4− anions. The axial positions of
the elongated tetragonal bipyramids are occupied by two water
molecules (O1 and O7). The water molecule O1 makes a bridge
Fig. 3 FT-IR spectra of complex 1 measured at different temperatures.
Dalton Trans., 2014, 43, 3730–3738 | 3733
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Dalton Transactions
broad absorption band centred at 3400 cm−1 corresponding to
stretching vibrations, ν(OH), of coordinated and clathrated
water molecules. The presence of DMF molecules in the
sample is evident from absorption bands at 2958 cm−1 and
2931 cm−1 corresponding to aliphatic stretching vibrations of
C–H groups of DMF and a very intensive band of the carbonyl
(CvO) group centred at 1665 cm−1. The MTB ligand shows
strong IR bands at 1597 cm−1 and 1393 cm−1, which can be
assigned to antisymmetric and symmetric stretching vibrations
of carboxylate groups, and the band at 783 cm−1, corres-
ponding to δ(COO−) vibrations.
To investigate solvent removal from {[Ni4(μ6-MTB)2(μ2-H2O)4-
(H2O)4]·11H2O·10DMF}n, IR spectra of the sample 1 after
heating to 100, 200, 300 and 400 °C were measured (see
Fig. 3). The spectra demonstrate a gradual release of water and
DMF molecules from the framework. After heating to 100 °C, a
decrease in the intensity of the band at 3400 cm−1 corres-
ponding to removal of water molecules was observed. The
spectrum measured at 200 °C showed diminishment of ν(OH)
vibration due to the release of water molecules. However, the
presence of DMF in the channels was still confirmed by
stretching vibrations of aliphatic C–H groups around
3000 cm−1 and a strong absorption band of ν(CvO) vibration
at 1665 cm−1. Further heating to 300 °C led to complete
removal of DMF molecules. The spectrum measured after
heating the compound to 400 °C showed the decomposition of
the framework.
Fig. 4 (a) Comparison of HE-PXRD patterns of as synthesized sample 1
(red line) and the pattern of this compound calculated from single X-ray
diffraction data (black line). (b) The patterns measured during in situ
heating of 1 in the temperature range 25–600 °C. (c) The comparison of
the pattern of the final decomposition product of compound
1
measured at 600 °C (green line) and the patterns of NiO (black line) and
Ni (red line).
molecules and is stable up to 300 °C. In the range 300–440 °C,
the intensities of strong diffraction peaks, centred at 0.53 Å−1
and 0.83 Å−1, gradually decreased with increasing temperature,
indicating decomposition of the framework. Above 440 °C the
mixture of Ni and NiO was detected as a final decomposition
product (Fig. 4c).
Thermal stability and solvent removal were further studied
by TGA analysis. A thermogravimetric curve of 1 (see ESI,
Fig. S1†) exhibits four mass loss steps. The first three steps
take place in the range of 25–300 °C with a mass loss of 45.2%
Gas sorption properties
and correspond to removal of water and DMF molecules Porosity of the sample was studied using nitrogen adsorption
(calculated weight 46.8%). The stepwise elimination of solvent at −196 °C and carbon dioxide adsorption at 0 and 22 °C.
molecules (water and DMF) was accompanied by endothermic From the results of IR spectroscopy, TGA, and in situ HE-PXRD
peaks on the DTA curve at 84, 152 and 240 °C. The last step in measurements it was shown that a gradual release of the sol-
the TGA curve, in the range 300–500 °C, is due to decompo- vents takes place in the range 150–300 °C. To optimize the con-
sition of the MTB ligand with an observed mass loss of 43.5% ditions of the sample activation the nitrogen adsorption
(calculated 42.40%). The process of decomposition of MTB isotherms were measured after activation at 150, 250, 300 °C
was accompanied on the DTA curve by two exothermic peaks (see ESI, Fig. S2†). It is obvious that the best textural character-
at 392 °C and 460 °C. The final decomposition product, a istics were observed after activation at 250 °C. Activation at
mixture of Ni and NiO, was identified by high energy powder this temperature led to a change of crystal colour from green
X-ray diffraction (HE-PXRD).
to brown. This change is reversible and after cooling of the
The stability of the framework and its collapse at tempera- compound and its immersion into the water–DMF solution,
tures above 330 °C were further evidenced by experimental the green colour of the crystals was recovered (see ESI,
HE-PXRD patterns measured during in situ heating. Fig. 4a Fig. S3†). The reversibility of the colour change, together with
shows comparison of experimental HE-PXRD patterns of 1 and the results of HE-PXRD presented in Fig. 4, may indicate
the patterns calculated from single crystal X-ray data. Very single-crystal to single-crystal transformation (SCSC)27 in com-
good agreement between the measured and the calculated pound 1.
data was observed, indicating the presence of a pure phase of
Nitrogen adsorption isotherm of compound 1 can be classi-
{[Ni4(μ6-MTB)2(μ2-H2O)4(H2O)4]·11H2O·10DMF}n. Differences fied as type I (Fig. 5a), which is characteristic of microporous
observed in diffraction line intensities can be attributed to the materials. Evaluation of the adsorption data using the BET
variation of the preferred orientation of crystallites in the pow- equation gave the value of specific surface area (SBET) of
dered sample. Fig. 4b shows HE-PXRD patterns of compound 700 m2 g−1. The micropore volume (Vp) and the pore size dia-
1 measured at room temperature and during in situ heating to meter of the sample were determined as 0.291 cm3 g−1 and
600 °C. The patterns demonstrate that the framework of the 16.31 Å, respectively. The determined pore volume is approxi-
compound does not collapse after removal of the solvent mately 53% of the crystallographic pore volume of 1. It
3734 | Dalton Trans., 2014, 43, 3730–3738
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
as zinc(II) or isostructural cobalt(II) complexes.16,19 Larger CO2
adsorption is probably related to a larger surface area and a
larger pore volume of compound 1 (SBET = 700 m2 g−1, Vp
=
0.291 cm3 g−1) in comparison with Co(II) and Zn(II) compounds
(SDR = 356 m2 g−1 and Vp = 0.165 cm3 g−1 for Co(II) compound
and SBET = 248 m2 g−1 and Vp = 0.0891 cm3 g−1 for Zn(II)
compound.
Knoevenagel condensation
The Knoevenagel condensation (Scheme 1) of aldehydes with
active methylene compounds (containing two electron with-
drawing groups) is one of the most useful C–C bond forming
reactions having wide applications.28 The condensation is
usually catalysed by bases or acids and it requires high reac-
tion temperatures or microwave irradiation.29 In addition to
the main product, the reaction mixture may also contain some
byproducts from the consecutive reactions as self-conden-
sation and oligomerization of the primary reaction product.
Liquid phase Knoevenagel condensation was performed
using aldehydes and active methylene compounds different in
size and reactivity (Scheme 1). In general, the highest conver-
sions after 360 min of reaction time were achieved using
malononitrile as the substrate, except for condensation with
4-tertbutylbenzaldehyde (Table 2). However, even in the last
case after prolongation of the catalytic process (24 h), activity
of methylene compounds changed in the order: malononitrile
(96% conversion of 4-tertbutylaldehyde) > ethyl acetoacetate
(75%) > methyl cyanoacetate (52%). Such a sequence of
substrate reactivity is in good agreement with our previous
reports.13,30 The selectivity to the target product in most cases
was higher for condensation of aldehydes with malononitrile
and methyl cyanoacetate (Table 2).
Fig. 5 (a) Nitrogen adsorption isotherm of compound 1 activated at
250 °C. (b) Carbon dioxide adsorption isotherms measured at 0 °C (red
line) and 22 °C (black line).
suggests that compound 1 partially collapsed or shrunk by the
removal of the guest moles and coordinated aqua ligands.
A comparison of the BET surface area of 1 with the litera-
ture data reported for other MOFs containing the MTB4−
ligand shows that compound 1 has a larger surface area in
comparison with the corresponding Zn(II)16 and Cu(II)17 com-
plexes, for which surface areas of 248 and 526 m2 g−1 were
observed. However, when comparing these values, it should be
noted that different pretreatment conditions were used for the
gas adsorption experiments on the respective compounds. For
the Co(II) complex,19 which adsorbs a very limited amount of
nitrogen (21.6 cm3 g−1 at STP), the surface area was estimated
from carbon dioxide adsorption isotherm. The determined DR
(Dubinin–Radushkevich) surface area of the Co(II) complex was
356 m2 g−1. It should be mentioned that compound 1 and its
isostructural Co(II) analogue show different adsorption behav-
iour with respect to nitrogen. The differences can be caused by
smaller effective aperture sizes of the channels in the Co(II)
compound. In this analogue the channels extending along the
a, b, and c axes have the sizes 6.0 × 8.0 Å2, 5.8 × 6.3 Å2 and
13.8 × 11.5 Å2, respectively, while in compound 1 the sizes of
the channel apertures extending along the crystallographic
axes and along the [−4,5,0] crystallographic direction are
larger with the values 12.6 × 9.4 Å2, 9.4 × 8.0 Å2, 12.6 × 11.7 Å2
and 14.9 × 14.9 Å2, respectively.
During the course of the condensation of heptanal, cyclo-
hexane carbaldehyde and benzaldehyde with malononitrile,
compound 1 showed steady selectivity in contrast to the reac-
tion of the mentioned aldehydes with methyl cyanoacetate and
ethyl acetoacetate. In the last cases, formation of bulky side
products (i.e. ethyl-2-(cyclohexylmethylene)-3-oxobutanoate for
cyclohexane carbaldehyde condensation with ethyl aceto-
acetate) was detected by GC-MS (Scheme 2).
Since the conversions of aldehydes with similar accessibility
and polarity of the carbonyl group (benzaldehyde, 4-tertbutyl-
benzaldehyde,
3,5-ditertbutylbenzaldehyde)
significantly
Carbon dioxide adsorption isotherms of 1 measured at
0 and 22 °C are shown in Fig. 5b. The complex adsorbs
12.36 wt% (71.8 cm3 g−1, 3.21 mmol g−1) of CO2 at 0 °C and
101 kPa and 6.53 wt% (38.4 cm3 g−1, 1.59 mmol g−1) at 22 °C
and 101 kPa. According to the literature data, the CO2 adsorp-
tion capacities were published only for two MTB-containing
16
compounds, namely {[Zn2(μ8-MTB)(H2O)2]·3DMF·3H2O}n
and {[Co4(μ6-MTB)2(μ2-H2O)4(H2O)4]·13DMF·11H2O}n.19 Their
CO2 adsorption capacities at 101 kPa and 0 °C were 5.67 and
7.02 wt%, respectively. It indicates that 1 exhibits significantly
Scheme 1 Knoevenagel condensation of aldehydes with active methyl-
larger carbon dioxide uptake (12.36 wt% at 0 °C and 101 kPa) ene compounds.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 3730–3738 | 3735

View Article Online
Paper
Dalton Transactions
Table 2 Conversion and selectivity in Knoevenagel condensation over
catalyst 1 at 130 °C
Aldehyde
(approximate
kinetic
Conversion
after
360 min
Methylene
compound
Selectivity at
50% conversion
diameter (Å))
84
81
45
87
1st run
2
3
100
100
93
84
78
79
Fig. 6 Proposed mechanism of Knoevenagel condensation of benz-
aldehyde with ethyl acetoacetate over 1.
nd run
rd run
82
58
decreasing the reaction temperature from 130 to 100 °C. While
the selectivity to the target product was practically unchanged
(almost 100% in both cases), conversion of aldehyde (45 and
78% at 100 and 130 °C, respectively) was much higher at
130 °C (Table 2).
89
49
94
64
46
100
We observed that 1 was highly active also in the 2nd and 3rd
catalytic runs (Table 2). Despite a lower initial rate of the reac-
tion over the reactivated catalyst, yield of the product of hepta-
nal condensation with malononitrile was comparable with the
results on the fresh catalyst. Reactivated MOF under investi-
gation exhibited 100% yield of the target product after
480 min of the reaction time.
58
n.d.
130 °C
100 °C
78
45
55
100
100
56
42
93
Since the Knoevenagel condensation over 1 was chosen as
the model reaction for the investigation of the influence of the
size and nature of substrates on the activity of 1, the perfect
catalytic performance of the used solid was not our goal.
Indeed, conversions achieved over 1 at relatively high tempera-
ture (130 °C) were comparable with those reached at 25–40 °C
over other MOF catalysts (especially containing basic sites)
tested in Knoevenagel condensation of aldehydes and methyl-
ene compounds.31
34
19
97
59a
22
24
100a
94
a At maximal conversion.
It was of interest to study whether the framework of 1 is
stable during the catalytic runs. As it follows from powder
X-ray patterns (Fig. 7), small changes in the diffraction
Scheme 2 Formation of the adduct of the Knoevenagel condensation
product with ethyl acetoacetate.
depend on the size of the substrate (78, 34, 24% for not substi-
tuted, monomethyl and dimethyl benzaldehydes, respectively),
we assume that the Knoevenagel reaction proceeds inside the
pore system of 1 under the conditions applied. Taking into
account the mentioned sequence of aldehyde activity and only
slight dependence of the substrate conversion on the nature of
the active methylene compound, the possible mechanism of
Knoevenagel condensation catalyzed by 1 can be proposed
(Fig. 6).
It was found that the activity of 1 in condensation of benz-
Fig. 7 PXRD patterns of 1 before and after the Knoevenagel conden-
aldehyde with malononitrile significantly decreased after sation (reaction with malononitrile at 130 °C).
3736 | Dalton Trans., 2014, 43, 3730–3738
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
patterns of the activated sample (the sample before the immer-
sion into the solvent) and the patterns measured on the
sample after liquid-phase Knoevenagel condensation were
observed. The observed changes of the intensity ratios of diffr-
action peaks and small shifts of the corresponding diffraction
lines are probably related to the interaction of the framework
of 1 with reaction products/substrates.
V. R. dos, S. Malta, J. D. L. Dutra, N. B. da Costa Jr.,
R. O. Freire and S. A. Júnior, J. Solid State Chem., 2013, 197,
7.
3 S. S. Han, J. L. Mendoza-Cortés and W. A. Goddard III,
Chem. Soc. Rev., 2009, 38, 1460; W. Wang, X. Dong, J. Nan,
W. Jin, Z. Hu, Y. Chen and J. Jiang, Chem. Commun., 2012,
48, 7022.
4 L. Ma, C. Abney and W. Lin, Chem. Soc. Rev., 2009, 38,
1248; J.-M. Gu, W.-S. Kim and S. Huh, Dalton Trans., 2011,
40, 10826.
Conclusions
5 Z. Ma and B. Moulton, Coord. Chem. Rev., 2011, 255, 1623;
P. Horcajada, T. Chalati, C. Serre, B. Gillet, C. Sebrie,
T. Baati, J. F. Eubank, D. Heurtaux, P. Clayette, C. Kreuz,
J. S. Chang, Y. K. Hwang, V. Marsaud, P. N. Bories,
L. Cynober, S. Gil, G. Ferey, P. Couvreur and R. Gref, Nat.
Mater., 2010, 9, 172.
6 B. Cen, Y. Yang, F. Zapata, G. Lin and G. Qian, Adv. Mater.,
2007, 19, 1693; M. D. Allendorf, C. A. Bauer, R. K. Bhakta
and R. J. T. Houk, Chem. Soc. Rev., 2009, 38, 1330.
7 T. Uemura, S. Horike and S. Kitagawa, Chem.–Asian J., 2006,
1, 36; T. Uemura, N. Yanai and S. Kitagawa, Chem. Soc.
Rev., 2009, 38, 1228.
8 A. Corma, H. Garcia and F. Xamena, Chem. Rev., 2010, 110,
4606; Z. Wang, G. Chen and K. Ding, Chem. Rev., 2009, 109,
322.
9 M. Yoon, R. Srirambalaji and K. Kim, Chem. Rev., 2012,
112, 1196.
We have prepared and characterized a new non-interpenetrat-
ing metal–organic framework constructed by MTB4− ligands
and Ni4 clusters that imitates the fluorite structure. The IR
spectroscopy, thermal analysis and in situ HE-PXRD measure-
ments revealed that removal of guest molecules from the chan-
nels leads to the opened porous framework. The comparison
of sorption properties of complex 1 with other reported com-
plexes containing MTB4− ligand shows that complex 1 exhibits
the highest BET surface area (700 m2
g
−1) and the highest
carbon dioxide uptake (12.36 wt% at 0 °C and 101 kPa) among
the MTB compounds. The prepared metal–organic framework
provides high activity in Knoevenagel condensation, reaching
up to 100% yields without significant loss in activity after
several catalytic runs. It was demonstrated that the reaction
proceeds mainly inside the pore system of compound 1 and
the framework of the catalyst has reasonable stability under
the used conditions.
10 M. Opanasenko, M. Shamzhy and J. Čejka, ChemCatChem,
2013, 5, 1024.
11 M. Hartmann and M. Fischer, Microporous Mesoporous
Mater., 2012, 164, 38; F. Xamena, F. G. Cirujano and
A. Corma, Microporous Mesoporous Mater., 2012, 157, 112;
L. T. L. Nguyen, K. K. A. Le, H. X. Truong and N. T. S. Phan,
Catal. Sci. Technol., 2012, 2, 521; A. Dhakshinamoorthy,
M. Opanasenko, J. Čejka and H. Garcia, Adv. Synth. Catal.,
2013, 355, 247.
12 M. Opanasenko, A. Dhakshinamoorthy, J. S. Chang,
H. Garcia and J. Čejka, ChemSusChem, 2013, 6, 865.
13 A. Dhakshinamoorthy, M. Opanasenko, J. Čejka and
H. Garcia, Catal. Sci. Technol., 2013, 3, 2509.
Acknowledgements
This work was supported by VEGA project, Ministry of Edu-
cation of Slovak Republic under contract no. 1/0583/11, and
the Slovak Research and Development Agency under contract
no. APVV 0132-11. V.Z. thanks DESY/HASYLAB project no.
I-20110282 EC for the support during the synchrotron related
measurements. J.Č. acknowledges the Czech Science Foun-
dation for the support of this research (P106/12/G015). The
authors would like to thank Dr J. Kuchár for help with the
single crystal XRD measurements and refinement.
14 M. Opanasenko, A. Dhakshinamoorthy, J. Čejka and
H. Garcia, ChemCatChem, 2013, 5, 1553.
15 A. R. Millward and O. M. Yaghi, J. Am. Chem. Soc., 2005,
217, 17998; P. L. Llewellyn, S. Bourrelly, C. Serre,
A. Vimont, M. Daturi, L. Hamon, G. De Weireld,
J.-S. Chang, D.-Y. Hong, Y. K. Hwang, S. H. Jhung and
G. Férez, Langmuir, 2008, 24, 7245; H. Furukawa, N. Ko,
Y. B. Go, N. Aratani, S. B. Choi, E. Choi, A. Ö. Yazaydin,
R. Q. Snurr, M. O’Keeffe, J. Kim and O. M. Yaghi, Science,
2010, 329, 424.
Notes and references
1 A. Y. Robin and K. M. Fromm, Coord. Chem. Rev., 2006,
250, 2127; J. L. C. Rowsell and O. M. Yaghi, Microporous
Mesoporous Mater., 2004, 73, 3.
2 E. Tynan, P. Jensen, P. E. Kruger and A. C. Lees, Chem.
Commun., 2004, 776; J. Lincke, D. Lässig, J. Moellmer,
C. Reichenbach, A. Puls, A. Moeller, R. Gläser, G. Kalies,
R. Staudt and H. Krautscheid, Microporous Mesoporous 16 J. Kim, B. Chen, T. M. Reineke, H. Li, M. Eddaoudi,
Mater., 2011, 142, 62; Y. Li, J. P. Zou, W. Q. Zou,
F. K. Zheng, G. C. Guo, C. Z. Lu and C. J. Huang, Inogr.
Chem. Commun., 2007, 10, 1026; Y. B. Go, X. Wang,
D. B. Moler, M. O’Keeffe and O. M. Yaghi, J. Am. Chem.
Soc., 2001, 123, 8239; M. Almáši, V. Zeleňák, R. Gyepes,
A. Zukal and J. Čejka, Colloids Surf., A, 2013, 437, 101.
E. V. Anokhina and A. J. Jacobson, Inorg. Chem., 2005, 44, 17 L. Ma, A. Jin, Z. Xie and W. Lin, Angew. Chem., Int. Ed.,
8265; C. A. F. de Oliveira, F. D. da Silva, I. Malvestiti,
2009, 48, 9905.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 3730–3738 | 3737

View Article Online
Paper
Dalton Transactions
18 Y. E. Cheon and M. P. Suh, Chem.–Eur. J., 2008, 14, 3961.
19 Y. E. Cheon and M. P. Suh, Chem. Commun., 2009, 2296.
20 H. Chun, D. Kim, D. N. Dybtsev and K. Kim, Angew. Chem.,
Int. Ed., 2004, 43, 971.
28 L. F. Tietze and U. Beifuss, Comprehensive Organic Synthesis,
ed. B. M. Trost, I. Fleming, Pergamon Press, Oxford, 1991,
vol. 2, p. 341; F. Freeman, Chem. Rev., 1981, 80, 329;
L. F. Tietze, Chem. Rev., 1996, 96, 115.
21 Oxford Diffraction, CrysAlis RED and CrysAlis CCD software 29 S. A.-E. Ayoubi, F. Texier-Boullet and J. Hamelin, Synthesis,
(ver. 1.171.1), Oxford Diffraction Ltd, Abingdon, Oxford-
shire, England, 2003.
22 A. Altomare, G. Cascarano, C. Giacovazzo and
A. Guagliardi, J. Appl. Crystallogr., 1993, 26, 343.
23 L. J. J. Farrugia, Appl. Crystallogr., 1999, 32, 837.
24 G. M. Sheldrick, SHELXL-97, Program for the Refinement of
1994, 258; D. Prajapati, K. C. Lekhok, J. S. Sandhu and
A. C. Ghosh, J. Chem. Soc., Perkin Trans. 1, 1996, 959;
S. Saravanamurugan, M. Palanichamy, M. Hartmann and
V. Murugesan, Appl. Catal., A, 2006, 298, 8; J. S. Yadav,
B. V. Subba Reddy, A. K. Basak, B. Visali, A. V. Narsaiah
and K. Nagaiah, Eur. J. Org. Chem., 2004, 546.
Crystal Structures, University of Göttingen, Göttingen, 1997. 30 M. Opanasenko, A. Dhakshinamoorthy, M. Shamzhy,
25 K. Brandenburg, DIAMOND, Version 2.1e. Crystal Impact
P. Nachtigall, M. Horáček, H. Garcia and J. Čejka, Catal.
Sci. Technol., 2013, 3, 500.
GbR, Bonn, 2000.
26 Spectral Database for Organic Compounds, National Insti- 31 S. Hasegawa, S. Horike, R. Matsuda, S. Furukawa,
tute of Advanced Industrial Science and Technology, http://
riodb01.ibase.aist.go.jp/sdbs/
27 M. Fujita, Metal-Organic Frameworks: Design and Appli-
cation, ed. L. R. MacGillivray, John Willey and Sons, Inc.,
2010, pp. 1–32.
K. Mochizuki, Y. Kinoshita and S. Kitagawa, J. Am. Chem.
Soc., 2007, 129, 2607; P. Serra-Crespo, E. V. Ramos-
Fernandez, J. Gascon and F. Kapteijn, Chem. Mater., 2011,
23, 2565; J. Gascon, U. Aktay, M. D. Hernandez-Alonso,
G. P. M. Van Klink and F. Kapteijn, J. Catal., 2009, 261, 75.
3738 | Dalton Trans., 2014, 43, 3730–3738
This journal is © The Royal Society of Chemistry 2014