An effective route to montmorillonite intercalation with imidazole complexes: Experiment and theory

Article abstract of DOI:10.1016/j.molstruc.2009.02.005

Intercalation of 1H-imidazole (IM) and his derivative 2-phenylimidazole (2-PhIM) into the interlayer spaces of both natural Ca(II)-montmorillonite (Ca-MMT) and Cu(II)-montmorillonite (Cu-MMT) was carried out by two ways, solid-state intercalation and inte

Full text of DOI:10.1016/j.molstruc.2009.02.005

Journal of Molecular Structure 923 (2009) 85–89
Contents lists available at ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
An effective route to montmorillonite intercalation with imidazole complexes:
Experiment and theory
b
a
a
a
Sylva Holešová a, , Lenka Kulhánková , Grazyna Simha Martynková , Jana Kukutschová , Pavla Capková
*
ˇ
ˇ
a Nanotechnology Centre, VŠB-Technical University of Ostrava, 17.listopadu 15/2172, 708 33 Ostrava-Poruba, Czech Republic
b Faculty of Metallurgy and Materials Engineering, VŠB-Technical University of Ostrava, 17.listopadu 15/2172, 708 33 Ostrava-Poruba, Czech Republic
a r t i c l e i n f o
a b s t r a c t
Article history:
Intercalation of 1H-imidazole (IM) and his derivative 2-phenylimidazole (2-PhIM) into the interlayer
spaces of both natural Ca(II)-montmorillonite (Ca-MMT) and Cu(II)-montmorillonite (Cu-MMT) was car-
ried out by two ways, solid-state intercalation and intercalation from solution. Only the presented solid-
state intercalation gave required results. In situ formation of Cu complexes in the interlayer spaces of
Cu(II)-MMT is not a feasible way to prepare required clay–coordination–intercalation compounds. The
Cu complexes of IM and 2-PhIM have been synthesized separately and they were anchored on surface
and edges of silicate layers, consequently they were not intercalated into the interlayer spaces of Ca-
MMT. X-ray powder diffraction (XRD) and Fourier Transform Infrared (FTIR) spectroscopy were applied
to characterize nanocomposites of mentioned above heterocyclic compounds and their Cu complexes
with Ca-MMT and/or Cu-MMT. Infrared spectra of intercalation products indicating, that these com-
pounds have been anchored only by non-bonding interactions into interlayer spaces. The experimental
data are supported by calculated structure models using molecular dynamics simulations.
Ó 2009 Elsevier B.V. All rights reserved.
Received 12 December 2008
Received in revised form 5 February 2009
Accepted 9 February 2009
Available online 14 February 2009
Keywords:
Imidazole
Montmorillonite
Intercalation
Molecular modelling
1. Introduction
One of the most important challenges in homogeneous catalysis is
to transform a successful homogeneous catalyst into a heterogeneous
Clay minerals have continuously attracted attention also for the
possibility of modifying their layered structure by intercalation,
thus creating new materials, such as adsorbents [1], catalysts [2],
nanocomposite precursors [3], materials with attractive electric
or optical properties [4]. Among possible layered solids, the smec-
tite group of layered clay minerals (especially montmorillonites)
provides attractive features, such as large surface areas, swelling
behaviour and ion exchange properties. The clay mineral montmo-
rillonite, a member of dioctahedral smectite group, is the most
important smectite used in catalytic applications. Smectites have
a layered structure formed by two tetrahedral sheets linked with
an octahedral sheet. The octahedral sites of montmorillonite are
occupied mainly by Al3+ cations which are to some extent replaced
by Fe3+ and/or Mg2+ ions. The tetraedra contain as the central
atoms mostly Si4+, partially substituted by Al3+. Non-equivalent
substitution of the central atoms in the octahedra and/or tetrahe-
dra generates a negative charge on the layers which is balanced
by hydrated exchangeable cations in the interlayers, most fre-
quently Ca2+, Mg2+ and Na+ in natural samples.
catalytic system, due to the intrinsic advantages of catalyst easy sep-
aration and recycling and product separation. Heterocyclic com-
pounds act as ligands for creation complexes with metals, which
are used as catalysts of organic syntheses. One of these ligands may
be imidazole. Imidazole is a part of biological significant compound
(amino-acid) histidine. Molecules of imidazole are involved as a
nucleophile or proton-transfer processes in biological systems [9].
Imidazole issignificant inenzyme reactions playing an important role
in the binding of metallic ions to proteins [10]. Polar organic mole-
cules (e.g. imidazole) can replace water molecules, which occupy
the interlayer spaces in the montmorillonite. Introducing transition
metal ions and adding suitable ligands, coordination–intercalation
compound may be formed in the interlayers of montmorillonite [11].
Owing to the excellent intercalation properties, we expect to
use MMT and modified MMT as hosts to achieve nanocomposites
with intercalated structures. In this paper, we report a study of
intercalation of the 1H-imidazole and his derivative 2-phenylimi-
dazole into the interlayer space of montmorillonite (a member of
2:1 smectite group) and a possibility of in situ formation Cu com-
plexes of these compounds in the interlayer space of Cu(II)-MMT.
X-ray powder diffraction (XRD) and Fourier Transform Infrared
(FTIR) spectroscopy were employed to characterize structure, as
well as surface. Interlayer structure and arrangement of guests in
host matrix has been studied using a combination of XRD empirical
data and molecular modelling (molecular mechanics and molecu-
The development of new heterogeneous catalysts to promote
organic reactions is a field of growing interest due to the inherent
advantages of heterogeneous over homogeneous catalysts [5–8].
* Corresponding author. Tel.: +420 597 321 575; fax: +420 597 321 640.
E-mail address: sylva.holesova@vsb.cz (S. Holešová).
0022-2860/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2009.02.005

86
S. Holešová et al. / Journal of Molecular Structure 923 (2009) 85–89
lar dynamics), which is illustrative tool both for image presumable
arrangement of guest molecules in host matrix and for finding pre-
sumable number of intercalated guest molecules.
were recorded in the 4000–400 cmꢁ1 spectral range with a resolu-
tion of 4 cmꢁ1 at room temperature using the KBr pressed disc
technique (0.8 mg of sample and 290 mg of KBr).
2.6. Strategy of molecular modelling
2. Experimental details
The initial model of the host montmorillonite layer was built in
Materials Studio (Crystal builder module) using structure data pub-
lished by Tsipursky and Drits (1984) [14]: space group, C2/m. The
unit cell parameters according to Méring (1967) [15] have been used
to define the planar unit cell dimensions: a = 5.208 Å and
b = 9.020 Å. The composition of montmorillonite layer according to
2.1. Materials
ˇ
The natural Ca-MMT from a deposit Ivancice in the Czech
Republic, selected as a starting material for the experiments, was
fractionated to less than <45 lm. The structural formula of this
montmorillonite as calculated from the chemical analysis is ðAl2:52
the chemical analysis was ðAl2:52Fe3þ Mg0:90Ti0:04ÞðSi7:96Al0:04
Þ
3þ
Fe Mg0:90Ti0:04ÞðSi7:96Al0:04ÞO20ðOHÞ4ðCa0:24K0:06Na0:09 Mg0:10Þ. Its
0:54
0:54
O20ðOHÞ4. To create the supercell of reasonable size for calculations
cation exchange capacity (CEC) is 105 cmol (+)/kg.
the structure formula was approximated by: ðAl23Mg8Fe3þÞ
The other materials are purchased as analytical grade chemi-
cals, including 1H-imidazole, 2-phenylimidazole, CuCl2ꢀ2H2O, eth-
anol, diethyl ether and acetone as solvents.
5
ðSi72ÞO180ðOHÞꢁ for the supercell 3a ꢃ 3b ꢃ 1c (ninefold single cell)
36
with the total negative layer charge (ꢁ8). In this supercell series of
models with variable interlayer structure have been created accord-
ing to the experimental sample preparation:
2.2. Cation-exchange procedure
1. Ca-MMT intercalated with IM (MI-1).
2. Ca-MMT intercalated with 2-PhIM (MI-2).
3. Cu-MMT intercalated with IM (MI-3).
4. Cu-MMT intercalated with 2-PhIM (MI-4).
The Cu-MMT was prepared from Ca-MMT by addition of a 1 M
CuCl2ꢀ2H2O solution and the mixture was stirred and heated at
70 °C for 2 h. After centrifugation, CuCl2 solution was added again
to the solid phase, stirred and heated as previously. This procedure
was repeated three times. The solid product was then washed by
water in order to remove the Clꢁ anions and finally dried at 50 °C.
In all the initial models the concentration, orientation and posi-
tion of guest molecules was systematically varied.
The models for the analysis of surface structure have been built
using the basic 3D model described above using the surface builder
modulus in Materials Studio modelling environment. The size of 2D
supercell was 3a ꢃ 3b. These surface models consisted of one sili-
cate layer and charge compensating Ca2+-cations on the lower
and upper side of silicate layer. The starting surface guest structure
was created by the same way that means systematic changes in
concentration, orientation and position of surface guest molecules.
Molecular mechanics calculations and classical dynamic simu-
lations have been performed in Materials Studio modelling environ-
ment in Forcite module, where the Universal force field [16] has
been used to describe the potential energy of initial models. Elec-
trostatic and van der Waals energy was calculated using Ewald
summation method [17]. After the first energy minimization the
molecular dynamic simulations have been started.
2.3. Intercalation of Ca-MMT and Cu-MMT with IM and 2-PhIM
The mixture of 0.5 g Ca-MMT or 0.5 g Cu-MMT with 0.5 g IM or
0.5 g 2-PhIM was heated at 70 °C for 5 h.
Prepared samples are: Ca-MMT+IM (MI-1), Ca-MMT+2-PhIM
(MI-2), Cu-MMT+IM (MI-3) and Cu-MMT+2-PhIM (MI-4).
2.4. Preparation of Cu complexes and their intercalation into the
interlayers of Ca-MMT
CuL4X2 (L = 1H-imidazole, X = Cl). Dark blue-purple crystals were
separated on mixing 1H-imidazole and CuCl2ꢀ2H2O in a 4.4:1 mole
ratio in ethanol (50 ml). The mixture was stirred at room temper-
ature for 24 h. The product was washed with acetone and diethyl
ether and then was dried at 40 °C [12].
CuL2X2ꢂ0.5H2O (L = 2-phenylimidazole, X = Cl). CuCl2.2H2O
(0.68 g, 4 mmol) was added to a diethyl ether (120 ml) solution
of 2-phenylimidazole (2.3 g, 16 mmol). The reaction mixture was
stirred at room temperature for 24 h. A green precipitate was
formed, which was filtered off and washed with diethyl ether.
The product was dried at 40 °C [13].
The mixture of 0.2 g Ca-MMT with 0.2 g Cu complex of 1H-imid-
azole (CuL4X2) or 2-phenylimidazole (CuL2X2ꢂ0.5H2O) was heated
at 70 °C for 5 h.
3. Results and discussion
3.1. Diffraction pattern analysis
Fig. 1a shows XRD patterns of original Ca-MMT (d001 = 1.56 nm)
and its intercalated forms. The changes of basal spacing reflect the
intercalation stages of IM (d001 = 1.80 nm) and2-PhIM (d001
=
1.84 nm) into interlayer spaces of Ca-MMT (MI-1, MI-2). The XRD
patterns (Fig. 1b) of the Cu-MMT shows decrease of basal spacing
(d001 = 1.36 nm) in comparison with starting material Ca-MMT.
Exchanging interlayer cations of Ca-MMT with Cu cations the
interlayer space decrease involving the fact that it is causes by a
difference in the size of hydrated form between Cu2+ and Ca2+ ions
and ionic radius of Cu2+ ion is smaller than Ca2+ ion.
The Cu complexes of IM and 2-PhIM in system with original Ca-
MMT exposed different type of interaction, than in case of simple
compounds. Ca-MMT after treatment with CuL4X2 complex (MI-
5) is decreasing the original value of basal diffraction by 0.2 nm
(Fig. 1a). Similar values of d001 are known from dehydration–rehy-
dration behaviour of Ca-MMT at 70 °C [18]. The intercalation of
huge molecule of the complex is beyond possibility, therefore we
presume that changes are caused due to loss of water molecules
in crystal structure of MMT, namely hydrate packing of calcium
Prepared samples are: Ca-MMT+CuL4X2 (MI-5) and Ca-MMT+
CuL2X2ꢂ0.5H2O (MI-6).
2.5. Analytical methods and equipment
The XRD patterns were recorded using the X-ray diffractometer
INEL equipped with a curved position-sensitive detector CPSD 120
(reflection mode, Ge-monochromatized CuK radiation). Diffrac-
a
1
tion patterns of the samples were taken in ambient atmosphere
under constant conditions (2000 s, 35 kV, 20 mA). The samples
were located in a flat rotation holder; the measurement was re-
peated three times. The XRD measurement was performed at nor-
mal laboratory conditions (25 °C, 43% of humidity).
The mid-infrared spectra were obtained on a Perkin Elmer 2000
Fourier transform infrared spectrometer. For each sample, 32 scans

S. Holešová et al. / Journal of Molecular Structure 923 (2009) 85–89
87
a
b
1.93nm
1.59nm
1.36nm
1.80nm
1.84nm
1.80nm
1.56nm
1.36nm
Ca - original
Cu-original
2
1
2
3
4
5
6
7
8
1
3
4
5
6
7
8
°2Θ Cu Kα
°2Θ Cu Kα
Fig. 1. XRD patterns of (a) Ca-MMT and the modified forms MI-1 (1.80 nm), MI-2 (1.84 nm), MI-5 (1.36 nm) and MI-6 (1.59 nm) and (b) Cu-MMT and the modified forms MI-3
(1.80 nm) and MI-4 (1.93 nm).
cations. This phenomenon can be caused by the presence of organ-
ic molecules of the complex located on surface or edges of silicate
layers. These molecules are attracting the water molecules, creat-
ing partly hydrated complexes, for that reason the water is not
truly leaving the structure.
In second case of Ca-MMT with CuL2X2ꢂ0.5H2O complex (MI-6)
the basal spacing slightly increase (d001 = 1.59 nm) in comparison
with Ca-MMT. In both cases the Cu complexes have not interca-
lated Ca-MMT.
of structural O–H groups (3631 cmꢁ1
) and water molecules
(3423 cmꢁ1). Shift of O–H groups stretching band to lower wave-
numbers in the spectrum of MI-1 (Fig. 2b) was observed which
may be due to intercalation of IM molecules into interlayer space
of the Ca-MMT. On the contrary, intercalation of CuL4X2 complex
(Fig. 2c) into the Ca-MMT (MI-5) interlayer space was not proved
according to the absence of the O–H band shift. Bending vibration
of O–H groups in water molecules at 1638 cmꢁ1 are absent in the
MI-1 spectrum (Fig. 2b). The presence of water molecules in MI-
5 (Fig. 2c) was observed but intensity of the band was significantly
weaker [19].
Similarly for MI-3 is d001 value 1.80 nm and for MI-4 is d001 va-
lue 1.93 nm, indicating that these compounds have been interca-
lated into Cu-MMT interlayer spaces (Fig. 1b).
The IR spectra in Fig. 3 in agreement with the XRD data demon-
strate intercalation of 2-PhIM (Fig. 3b) into the interlayer space of
the Ca-MMT proved by the slight shift of O–H band to 3629 cmꢁ1
3.2. FTIR spectroscopy
The mid-IR spectra for MI-1 and MI-5 are shown in Fig. 2. Broad
band in the Ca-MMT spectrum is attributed to stretching vibrations
Fig. 2. IR spectra of (a) Ca-MMT, (b) MI-1, (c) MI-5.
Fig. 3. IR spectra of (a) Ca-MMT, (b) MI-2, (c) MI-6.

88
S. Holešová et al. / Journal of Molecular Structure 923 (2009) 85–89
caused by length increase of the O–H bond. Intercalation process of
IM and 2-PhIM and sorption of their Cu complexes are likely to be
similar which may be seen in the spectra Figs. 2b, 3b and 2c, 3c,
respectively.
Nevertheless, the IR spectrum does not reveal any shift in the OH
stretching band region because of N–H group(at 3659 cmꢁ1
belonging to the 2-PhIM.
)
Compared to the Ca-MMT, the Cu-MMT (Fig. 4a) has additional
band appearing at 3365 cmꢁ1 which can be attributed to the O–H
stretching vibration in dimerized Cu2+ ions bridging two hydroxyls
[20]. Shift of O–H groups stretching band from 3631 to 3624 cmꢁ1
in the spectrum of MI-3 (Fig. 4b) was found which may be caused
by intercalation of IM molecules into interlayer space of the Cu-
MMT. According to the XRD data 2-PhIM intercalation occurred.
3.3. Molecular modelling
For our purposes in accord to experimental data ones of the
possible models was created. The prepared models have only illus-
trative character. Calculated structure models were sorted using
two criteria: (1) the total crystal energy and (2) agreement of cal-
culated and experimental basal spacing. Created models are:
Ca-MMT intercalated with IM (MI-1). The optimized structure of
MI-1 is in the Fig. 5a. This supercell contains 41 1H-imidazole mol-
ecules, which are strongly disordered in the interlayer space. The
strong dependence of calculated basal spacing on the guest con-
centration allowed to estimate the imidazole content in the inter-
layer space quite precisely, i.e. 41 molecules per 3a ꢃ 3b ꢃ 1 c
supercell. The calculated basal spacing 1.81 nm is in perfect agree-
ment with the XRD measured value.
Ca-MMT intercalated with 2-PhIM (MI-2). In the Fig. 5b one can
see the optimized model of MI-2, where the guests exhibit ten-
dency to certain type of ordering. The guest molecules are almost
parallel with each other and perpendicular (with small angle devi-
ation) to the silicate layer. The cations of Ca2+ are adjacent to the
silicate layers. The guest concentration estimated from modelling
is 19–20 guest molecules with calculated d001 = 1.80–1.87 nm.
Measured value for MI-2 is 1.84 nm.
Cu-MMT intercalated with IM (MI-3). In this case the interlayer
structure is very similar to the MI-1, this means the disordered
guest and Cu2+ cations adjacent to the silicate layer. However
Cu2+ cations are immersed deeper in the pseudo-hexagonal cavi-
ties of the silicate layer (see Fig. 6a). The calculated basal spacing
1.80 nm is in perfect agreement with the XRD measured value, too.
Cu-MMT intercalated with 2-PhIM (MI-4). The structure of MI-4
(see Fig. 6b) differs from that of MI-2 in type of guests ordering.
While in case of MI-2 guest molecules are ordered with their
long-axis almost perpendicularly to the silicate layers, for MI-4
guests are more disordered and consequently the basal spacing is
slightly higher, than for MI-2. Another difference is in the position
of interlayer cations Ca2+ and Cu2+
.
Surface structure of Cu complexes of IM or 2-PhIM anchored on the
Ca-MMT surface. The optimized structures of MI-5 and MI-6 display
the position and orientation of complexes with respect to the sili-
Fig. 4. IR spectra of (a) Cu-MMT, (b) MI-3, (c) MI-4.
Fig. 5. Ca-MMT intercalated with (a) 41 molecules of IM (MI-1) and (b) 19 molecules of 2-PhIM (MI-2). Ca2+ cations are shown as balls.
Fig. 6. Cu-MMT intercalated with (a) 42 molecules of IM (MI-3) and (b) 20 molecules of 2-PhIM (MI-4). Cu2+ cations are shown as balls.

S. Holešová et al. / Journal of Molecular Structure 923 (2009) 85–89
89
cate layer. In both cases there is no regular set of complexes on the
References
silicate layer observed. Comparing the non-bond energy for both
models we can conclude, that CuL2X2ꢂ0.5H2O exhibits slightly
higher adhesion energy to the silicate layer than the CuL4X2.
[1] P.X. Wu, Z.W. Liao, H.F. Zhang, J.G. Guo, Environ. Int. 26 (2001) 401.
[2] J.-Q. Ji, L. Black, P.G. Weidler, M. Janek, Langmuir 20 (2004) 9796.
[3] P. Maiti, K. Yamada, M. Okamoto, K. Ueda, K. Okamoto, Chem. Mater. 14 (2002)
4654.
[4] I. Chashechnikova, L. Dolgov, T. Gavrilko, G. Puchkovska, Ye. Shaydyuk, N.
Lebovka, V. Moraru, J. Baran, H. Ratajczak, J. Mol. Struct. 563 (2005) 744.
[5] H.-U. Blaser, B. Pugin, Scope and Limitations of the Application of
Heterogeneous Enantioselective Catalysis, Chiral Reactions in Heterogeneous
Catalysis, Plenum Press, New York, 1995. p. 33.
[6] J.M. Fraile, J.I. García, J.A. Mayoral, T. Tarnai, Tetrahedron: Asymmetry 8 (1997)
2089.
[7] J.M. Fraile, J.I. García, J.A. Mayoral, T. Tarnai, Tetrahedron: Asymmetry 9 (1998)
3997.
4. Conclusion
Two types of montmorillonites have been researched for inter-
calation ability of 1H-imidazole and/or 2-phenylimidazole and
their Cu complexes. The intercalation of both IM and 2-PhIM have
been proven using XDR observing the increase of interlayer spac-
ing. IR spectroscopy indicates that these compounds have non-
bonding interactions to the silicate layers.
[8] AI. Fernández, J.M. Fraile, J.I. García, C.I. Herrerías, J.A. Mayoral, L. Salvatella,
Catal. Commun. 2 (2001) 165.
The modification of MMT with complexes exposed different
type of interactions. The complexes were not intercalated into
MMT interlayer, but were anchored on surface and edges of silicate
layers. This modification caused that interlayer water of natural
MMT escapes during the thermal preparations of composite, and
the rehydration is complicated with presence of organic molecules.
IR spectra of these samples show decrease of interlayer water band
intensity, which reflex that amount of water molecules is reduced.
Based on experimental data the most optimal models were
established using molecular modelling.
[9] R.J. Sundberg, R.B. Martin, Chem. Rev. 74 (1974) 471.
[10] D. Garfinkel, J.T. Edsall, J. Am. Chem. Soc. 80 (1958) 3807.
[11] E. Jóna, E. Rudinská, M. Kubranová, M. Sapietová, M. Pajtášová, V. Jorík, Chem.
Pap. 59 (2005) 248.
[12] D.M.L. Goodgame, M. Goodgame, P.J. Hayward, G.W. Rayner-Canham, Inorg.
Chem. 7 (1968) 2447.
[13] M. Pellei, C. Pettinari, A. Cingolani, A. Lacchè, Main Group Metal Chem. 23
(2000) 673.
[14] J. Tsipursky, V.A. Drits, Clay Miner. 19 (1984) 177.
[15] J. Méring, A. Oberlin, Clays Clay Miner. 27 (1967) 3.
ˇ
[16] G.S. Martynková, L. Kulhánková, P. Maly´, P. Capková, J. Nanosci. Nanotechnol. 8
(2008) 2069.
[17] K. Rappé, C.J. Casewit, K.S. Colwell, W.A. Goddard III, W.M. Skiff, J. Am. Chem.
Soc. 114 (1992) 10024.
[18] Z. Weiss, M. Chmielová, A. Buchal, L. Bílková, M. Rieder, Geol. Carpath. 6 (1997)
Acknowledgments
91.
[19] H. He, D. Yang, P. Yuan, W. Shen, R.L. Frost, J. Colloid Interface Sci. 297 (2006)
This work was supported by the Czech Grant Agency (project
GACR No. 205/08/0869), and by the Ministry of education, youth
and sports of the Czech Republic (MSM 6198910016).
235.
ˇ
[20] M. Tanaka, A. Itadani, T. Abe, H. Taguchi, M. Nagao, J. Colloid Interface Sci. 308
(2007).