Nanohybrid materials from the intercalation of imidazolium ionic liquids in kaolinite

Article abstract of DOI:10.1039/b616922h

A series of novel organic-inorganic nanohybrid materials were obtained by the intercalation in the interlamellar spaces of the clay mineral kaolinite, of ionic liquids based on imidazolium derivatives. The intercalation procedure was successfully accomplished via a melt reaction strategy using the dimethylsulfoxide-kaolinite intercalate (DMSO-K) as a precursor. ¹³C MAS NMR as well as XRD, TGA/DTA and FTIR studies confirmed the complete displacement of DMSO molecules by the imidazolium salts during the intercalation process. Increase of the basal spacing from 1.1 nm in DMSO-K to 1.3-1.7 nm in the nanohybrid materials was observed, indicating that imidazolium derivatives are oriented in a way such that the imidazole ring is parallel, or slightly tilted by an angle of 10-25°, with respect to the kaolinite internal surfaces. The number of moles of organic material loaded in the nanohybrids was obtained from several independent measurements. The intercalation of the imidazolium salts increases the thermal stability of the resulting material by more than 150 °C with respect to DMSO-K. After heating under air at 300 °C for two hours, XRD showed that the structure of the intercalates was kept with only a slight decrease of the intercalation ratio. The original kaolinite structure was recovered after heating the intercalate at 350 °C for an additional two hours. This observed high thermal stability is promising for the use of these nanohybrid materials as precursor for the synthesis of new nanocomposites by incorporation of polymer in kaolinite at high temperature. The Royal Society of Chemistry.

Full text of DOI:10.1039/b616922h

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
Nanohybrid materials from the intercalation of imidazolium ionic liquids in
kaolinite
Sadok Letaief and Christian Detellier*
Received 20th November 2006, Accepted 20th December 2006
First published as an Advance Article on the web 18th January 2007
DOI: 10.1039/b616922h
A series of novel organic–inorganic nanohybrid materials were obtained by the intercalation in
the interlamellar spaces of the clay mineral kaolinite, of ionic liquids based on imidazolium
derivatives. The intercalation procedure was successfully accomplished via a melt reaction strategy
using the dimethylsulfoxide–kaolinite intercalate (DMSO-K) as a precursor. 13C MAS NMR as
well as XRD, TGA/DTA and FTIR studies confirmed the complete displacement of DMSO
molecules by the imidazolium salts during the intercalation process. Increase of the basal spacing
from 1.1 nm in DMSO-K to 1.3–1.7 nm in the nanohybrid materials was observed, indicating that
imidazolium derivatives are oriented in a way such that the imidazole ring is parallel, or slightly
tilted by an angle of 10–25u, with respect to the kaolinite internal surfaces. The number of moles
of organic material loaded in the nanohybrids was obtained from several independent
measurements. The intercalation of the imidazolium salts increases the thermal stability of the
resulting material by more than 150 uC with respect to DMSO-K. After heating under air at 300 uC
for two hours, XRD showed that the structure of the intercalates was kept with only a slight
decrease of the intercalation ratio. The original kaolinite structure was recovered after heating the
intercalate at 350 uC for an additional two hours. This observed high thermal stability is
promising for the use of these nanohybrid materials as precursor for the synthesis of new
nanocomposites by incorporation of polymer in kaolinite at high temperature.
DMSO, N-methylformamide (NMF) or urea has been used
Introduction
successfully as a precursor for further intercalations of other
organic molecules,8 or of polymers.10–14 The intercalation can
Nanohybrid materials, combining organic and inorganic units
at the molecular level, open the way to a large range of new
interlamellar aluminol surfaces.15–22 Recently, the delamina-
functional materials. Among the choices offered for the
lead also to the chemical grafting of organic units on the
inorganic units, clay minerals play a particularly important
tion of kaolinite was considerably promoted from such
grafted derivatives.23
role.1 The layered structuration of most of the phyllosilicates
can be used as a host for the intercalation of organic guests,
organised in a confined two-dimensional phase.2 This inter-
calation is the first step towards the complete delamination
of the layers in an organic polymeric matrix, leading to the
formation of nanocomposite materials.3–6 Due to their swelling
properties, smectites constitute, by far, the most used class of
phyllosilicates for the preparation of nanocomposites.3–6
Ionic liquids based on organic cations generally have a
melting point below 100 uC, and some of them are liquid even
at ambient temperatures. They are generally composed of a
relatively bulky organic cation, such as imidazolium or pyridi-
nium, associated with a halide anion. Their physico-chemical
properties depend on the structure of the ions and can be
adjusted by changing the counter-ion. They are attractive
solvents as they are non-volatile, non-flammable, have a high
thermal stability and are relatively inexpensive to manufacture.
They are currently receiving considerable attention as a new
alternative reaction medium for chemical and biocatalytic
reactions, and qualify as ‘‘green solvents’’.24–26
Kaolinite is an abundant clay mineral, and is widely avail-
able. It is a 1 : 1 phyllosilicate, characterized by a dioctahedral
structure, with the chemical composition Al2Si2O5(OH)4.7 The
structural asymmetry of kaolinite, due to the superposition of
the tetrahedral and octahedral sheets in the 1 : 1 layer, creates
large superposed dipoles, which, in conjunction with H-bonds
between the siloxane macrorings on one side and the aluminol
surface on the other side, result in a large cohesive energy of
the mineral.7 Consequently, the intercalation of molecular
guests in the interlamellar spaces of kaolinite is much less
developed than in the case of the smectites.8,9 However,
kaolinite pre-intercalated with dipolar molecules, such as
The intercalation of alkylimidazolium salts in smectites,
with improved thermal stability compared to the corres-
ponding alkylammonium intercalates, was used to prepare
exfoliated polyamide nanocomposites.27 However, there is no
report on the intercalation of alkylimidazolium salts in
kaolinite. Recently, the intercalation of the ionic liquid ethyl-
pyridinium chloride inside the interlamellar spaces of kaolinite
was reported, using the melt intercalation process.28,29
The intercalation of imidazolium ionic liquids in the
interlamellar spaces of kaolinite was successful in the cases
of the derivatives Im-n (n = 1–7) (see Fig. 1). The resulting
Centre for Catalysis Research and Innovation and Department of
Chemistry, University of Ottawa, 10 Marie Curie, Ottawa, Ontario,
Canada K1N 6N5. E-mail: dete@uottawa.ca
1476 | J. Mater. Chem., 2007, 17, 1476–1484
This journal is ß The Royal Society of Chemistry 2007

View Article Online
Fig. 1 Chemical structures of the ionic liquids used in the preparation of the nanohybrid materials derived from kaolinite.
alkylimidazolium intercalates are characterized by an
increased thermal stability compared to most other interca-
lates, typically above 300 uC. These nanostructured materials
constitute a promising new class of kaolinite derivatives for
advanced functional applications of this abundant mineral.
overnight at 70 uC under N2. The excess of dichloroethane was
removed by evaporation. 1-Methylimidazole ( 2.5 g, 30 mmol)
was added to 5 g of the product obtained (Im-2) (28 mmol).
The mixture was stirred for 20 minutes at room temperature
then at 70 uC under N2 for 72 hours. The excess of 1-methyl-
imidazole was removed by washing with ether. 1H NMR: d
3.87 (s, 6 H, CH3-N); 4.82 (s, 4 H, -CH2-N); 7.78 and 7.84 (s,
4 H, imidazole ring C-CH-N); 9.49 (s, 2 H, imidazole ring
N-CH-N). 13C NMR: d 36.7 (CH3-N); 49.1 (-CH2-N); 122.7 and
124.5 (imidazole ring C-C-N); 137.2 (imidazole ring N-C-N).
Elemental analysis (CHN) calculated (found) for Im-5: C:
45.62% (42.17%); H: 6.08% (6.85%); N: 21.30% (19.57%).
Experimental
Materials
Well-crystallized kaolinite (KGa-1b; Georgia) was obtained
from the Source Clays Repository of the Clay Minerals
Society, Purdue University, West Lafayette, IN, USA. The
purification of KGa-1b and the preparation of the dimethyl-
sulfoxide–kaolinite intercalate (DMSO-K) were done accord-
ing to previously published procedures.28,29 All the ionic
liquids were synthesized using variations of published proce-
dures,30–32 with the exception of Im-8 which was purchased
from Acros Organics. Im-1, Im-2, Im-8, Im-9 are liquids at
room temperature, whereas Im-3, Im-4, Im-5, Im-6 and Im-7
are solids. Their melting points are 33, 36, 74, 38 and 49 uC
respectively. Imidazole, 1-methylimidazole, 1-butylimidazole,
benzyl chloride and naphthyl chloride were purchased from
Aldrich Chemicals and were used as received.
Im-3. Benzyl chloride (15.4 mL, 133 mmol) was added
dropwise to 10 g of 1-methylimidazole (122 mmol). The
mixture was first stirred at room temperature for 20 minutes
then at 80 uC for 72 hours under N2. The excess of benzyl
chloride was removed by washing with diethyl ether. 1H NMR:
d 4.02 (s, 3 H, CH3-N); 5.59 (s, 2 H, Phenyl-CH2-N); 7.3–7.5
(m, 5 H, Phenyl-CH2-N), 7.60 (t, 1 H, imidazole HC-CH-N);
7.74 (t, 1 H, imidazole ring C-CH-N); 10.46 (s, H, imidazole
ring N-CH-N). 13C NMR: d 36.4 (CH3-N); 52.8 (Phenyl-CH2-
N); 122.0 and 123.9 (imidazole ring C-C-N); 128.8, 129.2
and 133.5 (Phenyl-CH2-N); 137.1 (imidazole ring N-C-N).
Elemental analysis (CHN) calculated (found) for Im-3: C:
63.3% (67.7%); H: 6.25% (6.55%); N: 13.4% (9.31%).
Synthesis and characterization of the ionic liquids
Im-1. 1-Bromopropane (12.2 mL, 133 mmol) was added
dropwise to 10 g of 1-methylimidazole (122 mmol). The
mixture was first stirred at room temperature for 20 minutes
then at 80 uC for 48 hours under N2. The excess of
1-bromopropane was removed by washing with diethyl ether.
1H NMR: d 0.21 (t, 3 H, CH3CH2-); 1.26 (m, 2 H, -CH2CH3);
3.39 (s, 3 H, CH3-N), 3.61 (t, 2 H, -CH2-N); 7.10–7.12 (m, 2 H,
imidazole ring); 9.39 ppm (s, H, imidazole ring N-CH-N). 13C
NMR: d 10.1 (CH3CH2-); 23.0 (-CH2CH3); 36.0 (CH3-N), 50.6
(-CH2-N); 122.0 and 123.2 (imidazole ring C-C-N); 136.0
(imidazole ring N-C-N). Elemental analysis (CHN) calculated
(found) for Im-1: C: 41.00% (40.59%); H: 6.35% (7.47%); N:
13.70% (13.75%).
Im-4. 2-Chloromethyl-naphthalene (23.6 g, 133 mmol) was
added dropwise, after gentle heating, to 10 g of 1-methyl-
imidazole (122 mmol). The mixture was first stirred at room
temperature for 20 minutes then at 80 uC for 72 hours under
N2. The excess of 2-chloromethyl-naphthalene was removed by
1
washing with diethyl ether. H NMR: d 3.98 (s, 3 H, CH3-N);
6.00 (s, 2 H, Naphthyl-CH2-N); 7.16 (s, 1 H, imidazole ring
C-CH-N); 7.34–7.88 (m, 7 H, Naphthyl-CH2-N), 8.01 (d, 1 H,
imidazole ring HC-CH-N); 10.26 (s, H, imidazole ring
N-CH-N). 13C NMR d 36.7 (CH3-N); 51.1 (Naphthyl-CH2-
N); 122.8 and 123.5 (imidazole ring C-C-N); 121.7,125.5,
126.6, 127.8, 128.4, 129.0, 130.6, 130.8 and 133.9 (Naphthyl-
CH2-N); 137.7 (imidazole ring N-C-N). Elemental analysis
(CHN) calculated (found) for Im-4: C: 69.60% (68.17%); H:
5.80% (6.49%); N: 10.83% (10.19%).
Im-2 and Im-5. Dichloroethane (50 mL) was added to 10 g
of 1-methylimidazole (122 mmol). The mixture was refluxed
This journal is ß The Royal Society of Chemistry 2007
J. Mater. Chem., 2007, 17, 1476–1484 | 1477

View Article Online
Im-6 and Im-7. 1-Benzylimidazole was first prepared as a
precursor. Imidazole (20 g, 294 mmol) was added to a mixture
of DMSO–KOH (100 mL of DMSO, and 320 mmol of KOH).
The system was stirred for 30 minutes at 80 uC. Benzyl
bromide (55 g, 322 mmol) was then added and the mixture
was stirred for two hours at 80 uC. 1-Benzylimidazole was
extracted with dichloromethane, filtered (celite) and the
volatile solvent was removed using a rotavapor.
of DMSO-K (Im : DMSO-K = 4 : 1 w/w) at room temperature.
The mixture was heated under a flow of nitrogen gas. The
temperature was ramped up to 180 uC. During the heating, all
the solid imidazolium salts became liquid and the modified clay
(DMSO-K) was in suspension in the melt. The suspension was
then magnetically stirred at 180 uC during two hours under a
flow of nitrogen. The molten salt in excess was removed after
four series of washing–centrifugation, using isopropanol. The
recovered solid sample was dried at 60 uC overnight.
Im-6. Benzyl chloride (3.8 mL, 33 mmol) was added to 4.75 g
of 1-benzylimidazole (30 mmol). The mixture was stirred for
20 minutes at room temperature then at 80 uC for 72 hours
under N2. The excess of benzyl chloride was removed by
Characterization
X-Ray diffraction patterns (XRD) were obtained with a
Philips PW 3710 instrument equipped with Ni-filtered and
Cu-Ka radiation (l = 0.15418 nm) operating at 45 kV and
40 mA. X-Ray fluorescence (XRF) experiments were performed
on a Philips PW 2400 fluorimeter. Differential thermal analyses
(DTA) and thermal gravimetric analyses (TGA) were recorded
on a SDT 2960 Simultaneous DSC-TGA instrument under
1
washing with diethyl ether. H NMR d 5.54 (s, 4 H, Phenyl-
CH2-N); 7.33–7.48 (m, 10 H, Phenyl-CH2-N and 2 H,
imidazole ring C-CH-N); 11.01 (s, 1 H, imidazole ring
N-CH-N). 13C NMR d 53.2, (Phenyl-CH2-N); 122.3 and
129.9 (imidazole ring C-C-N); 129.4, 129.8 and 133.5 (Phenyl-
CH2-N); 137.9 (imidazole ring N-C-N). Elemental analysis
(CHN) calculated (found) for Im-6: C: 71.70% (69.65%); H:
5.98% (6.82%); N: 9.84% (9.83%).
N2 flow (100 mL min21) with a heating rate of 10 uC min21
.
Infrared spectra were acquired on a Thermo Nicolet Nexus 670
FT-IR E.S.P. spectrometer under dry air using 128 scans with a
resolution of 4 cm21. The spectra were recorded using KBr
pellets. Before the recording of a spectrum, a flow of hot air was
circulated in the room containing the pellet for 30 minutes to
remove the water molecules adsorbed physically on the sample.
Im-7. 2-Chloromethyl-naphthalene (7.9 g, 45 mmol) was
added to 6.33 g of 1-benzylimidazole (40 mmol). The mixture
was stirred for 20 minutes at room temperature then at
80 uC for 72 hours under N2. The excess of 2-chloromethyl-
1
The H and 13C NMR spectra in solution were recorded on a
1
naphthalene was removed by washing with diethyl ether. H
Bruker 300 MHz spectrometer. The chemical shifts are
NMR d 5.52 (s, 2 H, Phenyl-CH2-N); 6.03 (s, 2 H, Naphthyl-
CH2-N); 7.00 (s, H, imidazole ring N-CH-C) 7.11 (s, H,
imidazole ring N-CH-C); 7.33–8.1 (12 H, Phenyl-CH2-N and
referenced to TMS using CDCl3 as a solvent.
Solid-state 13C NMR CP/MAS spectra were collected on
a Bruker AVANCE 500 NMR spectrometer operating at
125.77 MHz. Approximately 50 mg of sample were packed in
4 mm O.D. zirconia rotors which were spun at the magic
angle at speeds between 10 and 14 kHz. A cross polarization
sequence was used employing a 75%–100% linear ramped pulse
for the 2 ms contact pulse in the 13C channel. Proton 90u pulses
of 3.25 ms were used. Two pulse phase modulation (TPPM)
proton decoupling was used during the 25 ms acquisition with
5.6 ms pulses and an 84 kHz decoupling field. The recycle delay
was 2 seconds. The spectral width was set at 80 kHz and 4 K
data points were collected for each free induction decay. Data
collection times ranged from 1 to 2.5 hours. The 13C chemical
shifts were referenced to TMS at 0 ppm using the high
frequency signal of adamantane at 38.4 ppm as a secondary
standard. 29Si CP/MAS spectra were collected on a Bruker
AVANCE 500 NMR spectrometer operating at 99.35 MHz.
Approximately 50 mg of sample were packed in 4 mm O.D.
zirconia rotors which were spun at the magic angle at 6 kHz.
A conventional cross polarization sequence was used with a
4.35 ms proton 90u pulse, a 10 ms contact time and a 57 kHz
CW proton decoupling field during a 40 ms acquisition time.
The recycle delay was 2 seconds. The spectral width was set
at 25641 Hz and 2 K data points were collected for each
free induction decay. Data collection times ranged from 1 to
1.5 hours. The 29Si chemical shifts were referenced to TMS at
0 ppm using the high frequency signal of tetrakis(trimethyl-
silyl)silane at 29.9 ppm as a secondary standard.
Naphthyl-CH2-N); 11.33 (s, 1 H, imidazole ring N-CH-N). 13
C
NMR d 51.3 (Phenyl-CH2-N); 53.5 (Naphthyl-CH2-N); 121.3,
121.5, 122.7, 125.5, 126.6, 127.9, 128.1, 128.9, 129.1, 129.2,
129.4, 129.5, 130.8, 130.9, 132.8, 133.9, 138.1 (Naphthyl-CH2-
N, Phenyl-CH2-N and imidazole ring -C-C-N-C-N). Elemental
analysis (CHN) calculated (found) for Im-7: C: 75.3% (72.2%);
H: 5.68% (6.60%); N: 8.37% (7.93%).
Im-9. 1-Bromopropane (12.2 mL, 133 mmol) was added
dropwise to 15.1 g of 1-butylimidazole (122 mmol). The
mixture was first stirred at room temperature for 20 minutes
then at 80 uC for 48 hours under N2. The excess of
1-bromopropane was removed by washing with diethyl ether.
1H NMR: d 0.78 (m, 6 H, CH3CH2-); 1.20 (m, 2 H, -CH2-
CH2CH3); 1.79 (m, 4 H, -CH2-CH2-N), 4.09 (m, 4 H, -CH2-N);
7.4 (m, 2 H, imidazole ring); 8.73 (s, H, imidazole ring
N-CH-N). 13C NMR:
d 10.1 (N-CH2-CH2-CH3); 12.6
(N-CH2-CH2-CH2-CH3); 18.6 (N-CH2-CH2-CH2-CH3); 22.9
(N-CH2-CH2-CH3); 31.1 (N-CH2-CH2-CH2-CH3); 49.3
(N-CH2-CH2-CH3); 51.2 (N-CH2-CH2-CH2-CH3; 122.5
(imidazole ring C-C-N); 135.4 (imidazole ring N-C-N).
Elemental analysis (CHN) calculated (found) for Im-9: C:
48.6% (47.01%); H: 7.69% (7.21%); N: 11.34% (11.80%).
Intercalation reaction
The imidazolium salts were intercalated in kaolinite by a melt
intercalation method using DMSO-K as starting material.
Typically, 1.6 g of the imidazolium salt were added to 400 mg
The ChemSketch program was provided by Advancement
Chemistry Development Laboratories (ACD Labs).
1478 | J. Mater. Chem., 2007, 17, 1476–1484
This journal is ß The Royal Society of Chemistry 2007

View Article Online
Fig. 2 Oriented XRD patterns of kaolinite intercalated with DMSO (DMSO-K) and kaolinite intercalated with imidazolium salts (Im-n-K,
n = 1–7).
the intercalation process. The intercalation of imidazolium
Results and discussion
salts into the interlamellar spaces of kaolinite is clearly
A series of novel organic–inorganic nanohybrid materials
observed by a shift of the 001 reflection to lower angles upon
(Im-n-K; n = 1–7) were prepared from the intercalation of a
displacement of DMSO (d001 of DMSO-K is 1.10 nm). The
series of imidazolium salts (Fig. 1) in kaolinite pre-intercalated
values of d001, of Dd001 (obtained by subtracting the kaolinite
with dimethylsulfoxide (DMSO-K). While in the case of
c-spacing, 0.71 nm, from the observed d001), of the intercala-
smectites cationic species can be easily intercalated by ion
tion ratio R (obtained as an approximation from the relative
exchange with the interlamellar cation (usually sodium), the
intensities of the d001 peaks of the intercalate and of the non-
incorporation of ionic liquids into the kaolinite interlamellar
intercalated kaolinite), and of three independent determina-
spaces is more challenging due to the neutral nature of the
tions of the organic content of Im-n-K, nTGA, nEA, and nthm
,
kaolinite layers, and to the strong dipole–dipole and H-bond
interactions between the octahedral and the tetrahedral sheets
of adjacent layers. The intercalation occurred when DMSO-K
was heated in the ionic liquid at temperatures in the range
170–180 uC, not below. In this range of temperature, DMSO
was gradually removed from the kaolinite interlamellar spaces
and could be replaced by the imidazolium salts. The melting
point of the salt is critical in the intercalation process. If the
melting point is higher than 180 uC, the intercalation can not
be successful since the DMSO molecules are released before
the intercalation can take place, resulting in a collapse of the
structure and the recovery of the original kaolinite structure.
For example, in another study in our laboratory, we have
attempted to intercalate some ionic liquid derivatives from
obtained respectively from the TGA analysis, the elemental
analysis and a structural model, are indicated in Table 1. The
d001 expansions (Dd001) values are between 0.6 nm (Im-2; Im-5)
and 1.0 nm (Im-7). They are in good agreement with those
calculated using the ChemSketch program, taking into account
the interatomic distances and the atomic van der Waals
diameters (see below). The intercalation ratio gives an indica-
tion of the percentage of kaolinite lamellae which were
submitted to intercalation. It varies in the range 78 to 98%,
values expected on the basis of previous intercalation studies in
kaolinite.15–23,33,34 The XRD peak characteristic of DMSO-K
is completely removed in all the cases, indicating complete
replacement by Im-n (n = 1–7) of the DMSO intercalated
molecules.
trimethylphosphine which have
a melting point higher
The TGA–DTG traces (Fig. 3) as well as the 13C CP/MAS
NMR spectra (Fig. 5; see below) indicate also a complete
elimination of the DMSO molecules from the kaolinite
than 225 uC. The intercalation process was not successful by
this approach.
The intercalation of alkyl substituted imidazolium salts was
successful when the number of carbon atoms in the alkyl chain
was less than 4. For longer alkyl chains, such as in the case of
Im-8 and Im-9, the clay layers collapsed without insertion of
the imidazolium derivatives into the interlamellar space and
the original kaolinite was recovered. However, imidazolium
salts substituted by aromatic groups such as benzyl and even
naphthyl were successfully intercalated.
Table 1 Basal spacing (d001), interlamellar expansion (Dd001), inter-
calation ratio (R) and organic material content of the nanohybrid
materials (number of moles of the imidazolium derivative per kaolinite
structural unit Al2Si2O4(OH)4) calculated from TGA analysis,
elemental analysis and a theoretical model
a
b
c
d001/nm
Dd001/nm
R (%)
nTGA
nEA
nthm
The oriented XRD patterns of the starting DMSO-K pre-
intercalate as well as of the nanohybrid materials exhibit in
general well developed 00l reflections (001, 002, 003, and 004)
as shown in Fig. 2, indicating that the kaolinite layer structure
is preserved after the intercalation process. In all the cases, the
Im-1-K
Im-2-K
Im-3-K
Im-4-K
Im-5-K
Im-6-K
Im-7-K
a
1.40
1.34
1.53
1.48
1.34
1.56
1.68
0.69
0.63
0.82
0.78
0.63
0.85
0.97
,90
98
89
78
96
85
84
b
0.46
0.48
0.48
0.29
0.27
0.22
0.22
0.46
0.51
0.58
0.29
0.25
0.18
0.20
0.51
0.46
0.40
0.28
0.31
0.27
0.22
˚
d060 reflection of kaolinite (not shown) remains at 1.49 A,
Calculated from TGA analysis.
Calculated from elemental
characteristic of a dioctahedral clay mineral, showing also that
the structure of the layers of kaolinite is largely unaffected by
c
analysis. Calculated from a theoretical model.
This journal is ß The Royal Society of Chemistry 2007
J. Mater. Chem., 2007, 17, 1476–1484 | 1479

View Article Online
Fig. 3 TGA and DTG patterns of kaolinite (K), kaolinite intercalated with DMSO (DMSO-K), and kaolinite intercalated with imidazolium salts
(Im-n-K; n = 1–7).
interlamellar spaces during the intercalation process. The
TGA–DTG traces show that the imidazolium salts are
removed from the kaolinite interlamellar space at maximum
temperatures higher than 270 uC. Typically, a first loss is
observed below 100 uC, corresponding to the removal of
externally adsorbed water. Then, a second and a third losses
are observed in the range of 270 uC to 350 uC and in the range
of 400 uC to 515 uC. The second loss is attributed to the partial
thermal removal of the organic material, while the third one
is attributed to the decomposition of the remaining organic
units concurrently with the kaolinite dehydroxylation. In
the case of the ethylpyridinium chloride intercalate,28,29 the
removal of the organic material was done in one step, with a
maximum at 295 uC. This observed relatively high thermal
stability plausibly results from the strong interactions between
the imidazolium salts and the surface Al–OH (aluminol)
groups through H-p interactions and between the strong ionic
interactions expected with the permanent kaolinite dipoles. In
all the cases (with the exception of Im-3-K), the dehydroxyla-
tion temperature is lower than in kaolinite, as it is usually
observed for intercalated kaolinite derivatives. In all the cases
also, the solid-solid exothermic transformation occurred at
temperatures near 1000 uC, as in the starting kaolinite.
The number of moles of organic material loaded in the
nanohybrids was obtained from several experimental measure-
ments and was compared to calculated values on the basis of
the estimated volume of the salts occupying the kaolinite inter-
lamellar spaces. Table 1 gives the number of moles of imidazo-
lium salt per structural unit of kaolinite, Al2Si2O4(OH)4,
calculated from the TGA analysis (nTGA) and from elemental
analysis (nEA). In some cases the amount of chlorine in the
nanohybrids was determined by semi-quantitative XRF
analysis. Given the large errors associated with the technique,
they were in reasonable agreement with the data obtained by
1480 | J. Mater. Chem., 2007, 17, 1476–1484
This journal is ß The Royal Society of Chemistry 2007

View Article Online
TGA and elemental CHN analysis. The number of organic
molecules loaded in the interlamellar spaces is smaller for the
salts having the larger size, an observation in agreement with
the space occupied by the organic cation along the a,b axes.
The molecular dimensions of the seven intercalated imida-
zolium salts were estimated from a space-filling structural
model using the ChemSketch software from ACD labs. The
volume occupied by the organic cation was approximated to a
parallelepiped shape. In all the cases the molecular thickness
was 0.496 nm, and the sides of the rectangular basis varied
from 0.722 and 1.090 nm in the smallest case (Im-1) to 1.218
and 1.698 nm in the largest case (Im-7), corresponding to
molecular parallelepiped volumes of 0.390 nm3 (Im-1) and
1.026 nm3 (Im-7). The other values fall within this range. Since
the molecular thickness is close to 0.5 nm, while the d001
expansions values are larger, being between 0.6 nm (Im-2;
Im-5) and 1.0 nm (Im-7), the angle by which the molecule
would have to be tilted to account for the observed expansion
was calculated, on the basis of simple geometrical considera-
tions. In all the cases, the values of the angle were found to be
between 10 and 25 degrees. Alternatively, the observed value
could result from an additional space due to the H–p
interactions between the aluminol groups and the imidazolium
cations lying parallel to the kaolinite layers. In any case, given
the small value of the eventual calculated tilt angle, the model
chosen for the calculation was based on an arrangement of the
imidazolium cations parallel to the kaolinite interlamellar
surfaces. The surface occupied by the salt along the a,b axes
was then calculated by adding to the rectangular shape of
the cation (for example, 0.787 nm2 in the case of Im-1 and
2.068 nm2 in the case of Im-7) a square shape for the anion
based on its van der Waals diameter (0.144 nm2 for Cl2). Since
the surface of an unit cell of kaolinite is accepted to be
typically a 6 b = 0.5156 nm 6 0.8945 nm = 0.461 nm2,35
the ratio of the unit cell surface over the surface of the
imidazolium halide should give an estimate of the organic
loading in Im-n-K. The values are also given in Table 1, as
nthm. The agreement of that value with the loading obtained
Fig. 4 IR spectra of kaolinite (K), kaolinite intercalated with
DMSO (DMSO-K), and kaolinite intercalated with imidazolium
salts (Im-n-K; n = 1,2).
attribution of this band to the stretching of the interlamellar
hydroxyl groups pointing perpendicularly to the c-direction.
As such, they are not directly involved in the H-bonding with
organic intercalates. In strong contrast, the other bands are
strongly modified by the intercalation of DMSO and of the
imidazolium salts. The pattern of the IR spectra of Im-1-K and
of Im-2-K in the hydroxyl stretching region, with the
appearance of a new band around 3600 cm21 is similar to
the patterns observed previously in the case of ethylpyridinium
chloride28,29 or polyols18,19 intercalates. The broad band below
3600 cm21 is tentatively attributed to the stretching of the
network of interlamellar hydroxyl groups H-bonded to the
imidazolium cations.
Fig. 5 gives the 13C CP/MAS NMR spectra for DMSO-K
and the seven Im-n-K intercalates. In DMSO-K, the two peaks
characteristic of the unsymmetrical structuration of DMSO
inside the interlamellar spaces of kaolinite are clearly observed
at 43.9 and 42.8 ppm.37 These peaks are totally absent from the
Im-n-K spectra, confirming the complete removal of DMSO
upon intercalation of the imidazolium salts. The 13C NMR
spectra clearly show that the imidazolium moiety remains
intact after the intercalation process. The solid-state spectra
are characterized by chemical shifts close to those of the
starting molecules in solution (see Experimental section). For
example, the C-2 signal is systematically observed in the range
132–139 ppm, either as a well resolved peak or as a shoulder
on the aromatic signals, in good agreement also with chemical
shifts reported for similar compounds,38 and the methyl group
in position 1 is observed in the range 34–37 ppm for Im-n-K
(n = 1–5). All the other signals are also in agreement with
intercalated imidazolium structures corresponding to Fig. 1.
from TGA analysis (nTGA) and from elemental analysis (nEA
)
is good, in support of the proposed structural model of
alternating cationic and anionic species, closely packed in one
layer in the kaolinite interlamellar spaces.
Fig. 4 gives the IR spectra (2600–4000 cm21 region) for
kaolinite (K), the starting material (DMSO-K) and two of the
intercalated materials (Im-1-K and Im-2-K). The spectra of the
other nanohybrids are similar to the ones shown in Fig. 4.
The spectra of kaolinite and of DMSO-K have been described
previously.28,36 In the comparison of the spectra of Im-1-K
and Im-2-K with those of kaolinite and of DMSO-K, one
should particularly note two observations: the relatively sharp
band observed at 3621 cm21 is still observed when the
imidazolium halides have replaced DMSO. This band is
attributed to the stretching vibrations of the internal hydroxyl
groups. Its invariance confirms that the intercalation of
imidazolium halides does not modify the 1:1 layer structure
of kaolinite. The bands attributed to the stretching of the
interlamellar hydroxyl groups are modified. The band at
3695 cm21 in kaolinite and DMSO-K remains observed at a
close position (3697 cm21), which is in agreement with the
This journal is ß The Royal Society of Chemistry 2007
J. Mater. Chem., 2007, 17, 1476–1484 | 1481

View Article Online
Fig. 5 Solid-state 13C CP/MAS NMR spectra for the starting material (DMSO-K) and the nanohybrid materials (Im-n-K, n = 1–7).
The 29Si CP/MAS NMR spectra of kaolinite, DMSO-K and
Im-n-K (n = 1–7) at 99.35 MHz are shown in Fig. 6. At that
frequency of observation the two signals of equal intensities
characterizing kaolinite are well separated at 291.5 and
290.9 ppm.39 In DMSO-K, the signal is shifted to lower
frequencies, at 292.7 ppm, accounting plausibly for the keying
of one of the DMSO methyl groups in the siloxane rings of
the tetrahedral sheets. In contrast, the 29Si NMR spectra of
Im-n-K display a relatively broad envelope around 291.5 ppm,
with, in some cases, finer structures. A similar spectrum was
observed in the case of the ethylpyridinium chloride inter-
calate.28,29 This is an indication that the silicate layer is less
affected by the presence of the imidazolium moiety than it is by
the strong interactions with DMSO. This is in good agreement
with a planar arrangement of the imidazolium cations in the
interlamellar spaces, with no keying of the side chains in the
siloxane macrorings of the tetrahedral sheets. The observation
in some cases of several peaks indicates heterogeneities on the
silicate surface, which is expected on the basis of the dissym-
metric structures of the imidazolium derivatives. Obviously,
the close interactions of phenyl, naphthyl or imidazolium rings
lying parallel to the silicate surfaces lead to a distribution of
dipole-induced dipole interactions, resulting in a distribution
of 29Si chemical shifts.
value of 1.40 nm (Table 1). The TGA–DTG traces indicate a
first loss of organic material at a maximum of 279 uC, followed
by a second loss at 340 uC and the dehydroxylation of the
kaolinite layers at 492 uC (Fig. 3). After heating at 300 uC for
two hours, the XRD pattern shows that the structure of the
intercalate is kept, with a d001 value of 1.38 nm, and an
intercalation ratio of approximately 85%. This indicates that
most of the imidazolium bromide units remain in the inter-
lamellar spaces after heating at 300 uC, while a fraction of
them are thermally removed. In strong contrast, after heating
at 350 uC for an additional two hours, the peak at 1.38 nm
disappears, and the characteristic d001 peak of kaolinite is
obtained at 0.71 nm, showing the recovery of the original
kaolinite structure. This is in agreement with the TGA–DTA
analysis, giving an observed dehydroxylation temperature
close to the one typically observed for kaolinite in its
transformation to metakaolinite, and also with the structural
reorganization of metakaolinite to mullite observed near
1000 uC40 as would be observed for kaolinite. The thermal
stability of Im-1-K is quite remarkable, particularly if com-
pared to what was observed previously in the case of the ethyl-
pyridinium chloride intercalate, where the organic material was
completely removed at 250 uC under the same conditions.28
This new class of organic–inorganic nanohybrid materials,
prepared simply and easily from the dispersion of the DMSO–
kaolinite intercalate in the ionic liquid at 180 uC, are charac-
terized by well-ordered structures and show thermal stability
when heated at temperatures up to 300 uC. The intercalation
process can not be predicted based upon simple factors such
as the molecular size of the salt or the melting point. For
example, the intercalation of Im-7 was successful while that of
In order to evaluate further the thermal stability of these
nanohybrid materials, XRD patterns were recorded after
heating under air for two hours a sample of Im-1-K (Fig. 7).
The intercalation ratio obtained after preparation of the
sample is evaluated at 90% (the overlap of the d002 peak of
Im-1-K with the d001 peak of kaolinite makes difficult a more
precise determination of the intercalation ratio), with a d001
1482 | J. Mater. Chem., 2007, 17, 1476–1484
This journal is ß The Royal Society of Chemistry 2007

View Article Online
Fig. 6 Solid-state 29Si spectra for kaolinite (K), the starting material (DMSO-K) and the nanohybrid materials (Im-n-K, n = 1–7).
Im-9 was not. The former is bulkier, and solid at room
temperature. The second is smaller and liquid at room
temperature. A series of factors contribute to the successful
intercalation, one of them being the ability of the incoming
guest to replace in a concerted way the departing DMSO
molecules to avoid a collapse of the structure and the recovery
of kaolinite. The process occurs in a concerted manner,
concurrently, not in sequence.
In summary, the reactions of imidazolium derivatives with
kaolinite pre-intercalated with DMSO under melt conditions
constitute an efficient route for the preparation of novel
nanostructured materials which are attractive as precursors
for the preparation of other nanohybrid and nanocomposite
materials. The nature and the length of the alkyl chain
substituting the imidazolium cation, as well as the nature of
the counter-anion, are critical for the success of the intercala-
tion process. The high thermal stability of these intercalates
coupled with the electric properties of the imidazolium salts
could provide improved materials with electronic conductivity
behaviour. The relationship between the nanoscale organiza-
tion of the organic salts in the interlamellar spaces of the clay
minerals and the electric conductivity are currently under
investigation in our laboratory.
Acknowledgements
This work was financially supported by a Discovery Grant of
the Natural Sciences and Engineering Research Council of
Canada (NSERC). Dr Raed Abu Rezk is thanked for fruitful
Fig. 7 XRD patterns of an oriented sample of Im-1-K at room
temperature and heated successively during two hours at 300 uC and
350 uC under static air.
This journal is ß The Royal Society of Chemistry 2007
J. Mater. Chem., 2007, 17, 1476–1484 | 1483

View Article Online
18 K. B. Brandt, T. A. Elbokl and C. Detellier, J. Mater. Chem., 2003,
13, 2566–2572.
19 T. A. Elbokl and C. Detellier, Clay Sci., 2005, 12, 38–46.
20 T. Itagaki and K. Kuroda, J. Mater. Chem., 2003, 13, 1064–1068.
21 J. Murakami, T. Itagaki and K. Kuroda, Solid State Ionics, 2004,
279–282.
22 J. E. F. C. Gardolinski and G. Lagaly, Clay Miner., 2005, 40,
537–546.
23 J. E. F. C. Gardolinski and G. Lagaly, Clay Miner., 2005, 40,
547–556.
discussions. Dr Glenn A. Facey is thanked for recording
the NMR spectra and Dr Ron Hartree for the XRF
measurements.
References
1 Handbook of Clay Science, in Developments in Clay Science, 1, ed.
F. Bergaya, B. K. G. Theng and G. Lagaly, Elsevier, Amsterdam,
2006.
2 E. Ruiz-Hitzky, Organic–inorganic materials: from intercalations to
devices, in Functional Hybrid Materials, ed. P. Go´mez-Romero and
C. Sa´nchez, Wiley-VCH, Weinheim, 2004, ch. 2, pp. 15–49.
3 M. Alexandre and P. Dubois, Mater. Sci. Eng., R., 2000, 28, 1–283.
4 S. Sinha Ray and M. Okamoto, Prog. Polym. Sci., 2003, 28,
1539–1641.
5 E. Ruiz-Hitzky and A. Van Meerbeek, Clay Mineral– and
Organoclay–polymer Nanocomposite, in Developments in Clay
Science, 1, Handbook of Clay Science, ed. F. Bergaya, B. K. G.
Theng and G. Lagaly, Elsevier, Amsterdam, 2006, ch. 10.3,
pp. 583–621.
6 Polymer–clay nanocomposites, ed. T. J. Pinnavaia and G. Beall,
Wiley Series in Polymer Science, Wiley, New York, 2000.
7 Hydrous Phyllosilicates, ed. S. W. Bailey, American Mineralogical
Society, Washington DC, 1988.
24 M. J. Earle and K. R. Seddon, Pure Appl. Chem., 2000, 72,
1391–1398.
25 Z. Yang and W. Pan, Enzyme Microb. Technnol., 2005, 37, 19–28.
26 N. Winterton, J. Mater. Chem., 2006, 16, 4281–4293.
27 J. W. Gilman, W. H. Awad, R. D. Davis, J. Shields, R. H. Harris,
C. Davis, A. B. Morgan, T. E. Sutto, J. Callahan, P. C. Trulove
and H. C. DeLong, Chem. Mater., 2002, 14, 3776–3785.
28 S. Letaief and C. Detellier, J. Mater. Chem., 2005, 15, 4734–4740.
29 S. Letaief, T. A. Elbokl and C. Detellier, J. Colloid Interface Sci.,
2006, 302, 254–258.
30 E. A. Turner, C. C. Pye and R. D. Singer, J. Phys. Chem. A, 2003,
107, 2277–2288.
31 P. B. Webb, M. F. Sellin, T. E. Kunene, S. Williamson,
M. Z. S. Alexandra and D. J. Cole-Hamilton, J. Am. Chem.
Soc., 2003, 125, 15577–15588.
32 O. V. Starikova, G. V. Dolgushin, L. I. Larina, P. E. Ushakov,
T. N. Komarova and V. A. Lopyrev, Russ. J. Org. Chem., 2003, 39,
1467–1470.
33 J. J. Tunney and C. Detellier, Can. J. Chem., 1997, 75, 1766–1772.
34 Y. Deng, G. N. White and J. B. Dixon, J. Colloid Interface Sci.,
2002, 250, 379–393.
8 G. Lagaly, M. Ogawa and I. De´ka´ny, Clay Mineral Organic
Interactions, in Developments in Clay Science, 1, Handbook of Clay
Science, ed. F. Bergaya, B. K. G. Theng and G. Lagaly, Elsevier,
Amsterdam, 2006, ch. 7.3, pp. 309–377.
9 R. L. Frost and J. Kristof, Interface Sci. Technol., 2004, 1,
184–215.
35 R. F. Giese, Jr., Kaolin Minerals: Structures and Stabilities, in
Hydrous Phyllosilicates, ed. S. W. Bailey, American Mineralogical
Society, Washington DC, 1988, ch. 3, pp. 29–66.
36 W. N. Martens, R. L. Frost, J. Kristof and E. Horvath, J. Phys.
Chem. B, 2002, 106, 4162–4171.
10 T. A. Elbokl and C. Detellier, J. Phys. Chem. Solids, 2006, 67,
950–955.
11 J. J. Tunney and C. Detellier, Chem. Mater., 1996, 8, 927–935.
12 Y. Komori, Y. Sugahara and K. Kuroda, Chem. Mater., 1999, 11,
3–6.
37 S. Hayashi, Clays Clay Miner., 1997, 45, 724–732.
38 J. Alca´zar, A. de la Hoz and M. Begtrup, Magn. Reson. Chem.,
1998, 36, 296–299.
39 J. Sanz and J. M. Serratosa, J. Am. Chem. Soc., 1984, 106,
4790–4793.
40 L. A. Pe´rez-Maqueda, J. L. Pe´rez-Rodriguez, G. W. Scheiffele,
A. Justo and P. J. Sa´nchez-Soto, J. Therm. Anal., 1993, 39,
1055–1067.
13 T. Itagaki, Y. Komori, Y. Sugahara and K. Kuroda, J. Mater.
Chem., 2001, 11, 3291–3295.
14 J. E. Gardolinski, L. C. M. Carrera, M. P. Canta˜o and F. Wypych,
J. Mater. Sci., 2000, 35, 3113–3119.
15 J. J. Tunney and C. Detellier, Chem. Mater., 1993, 5, 747–748.
16 J. J. Tunney and C. Detellier, J. Mater. Chem., 1996, 6, 1679–1685.
17 J. J. Tunney and C. Detellier, Clays Clay Miner., 1994, 42,
552–560.
1484 | J. Mater. Chem., 2007, 17, 1476–1484
This journal is ß The Royal Society of Chemistry 2007