K-10 montmorillonite: An efficient and reusable catalyst for the aerobic CC bond cleavage of α-substituted ketones

Article abstract of DOI:10.1016/j.molcata.2012.07.022

A commercially available acid-activated montmorillonite clay catalyst, K10 montmorillonite, was tested for the catalytic oxidation of cyclic ketones in the presence of molecular oxygen under mild conditions (343 K and atmospheric pressure). K10 montmorillonite catalyzed the oxidative cleavage of CC bonds in 2-methylcyclohexanone, 2-phenylcyclohexanone, 2-hydroxylcyclohexanone and 1,2-cyclohexanedione with good activity and excellent selectivity toward the formation of the corresponding ketoacids and diacids. The effects of acidity, amount of catalyst, temperature and solvent on the catalytic activity were investigated. Furthermore, this catalyst was reusable without any appreciable loss in activity and selectivity.

Full text of DOI:10.1016/j.molcata.2012.07.022

Journal of Molecular Catalysis A: Chemical 363–364 (2012) 437–445
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
K-10 montmorillonite: An efficient and reusable catalyst for the aerobic
C C bond cleavage of ␣-substituted ketones
Iman El Younssia, Tarik Rhadfia,c, Ahmed Atlamsania,∗, Jean-Paul Quisefitb,
Frédéric Herbstc, Khalid Draouia
a Université Abdelmalek ESSAADI, Faculté des Sciences Tétouan, Laboratoire des Matériaux et des Systèmes Interfaciaux, P.B. 2121, 93000 Tétouan, Morocco
b LISA, Unité Mixte de Recherche Université Paris Diderot, Université Paris Est Créteil Val de Marne et CNRS UMR 7583, Faculté des Sciences et Technologie,
61 avenue du Général de Gaulle, 94010 Créteil Cedex, France
c ITODYS, Université Paris Diderot, CNRS UMR 7086, 15 rue Jean-Antoine de Baïf, 75205 Paris Cedex 13, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 9 May 2012
Received in revised form 16 July 2012
Accepted 18 July 2012
Available online 8 August 2012
A commercially available acid-activated montmorillonite clay catalyst, K10 montmorillonite, was tested
for the catalytic oxidation of cyclic ketones in the presence of molecular oxygen under mild conditions
(343 K and atmospheric pressure). K10 montmorillonite catalyzed the oxidative cleavage of C C bonds in
2-methylcyclohexanone, 2-phenylcyclohexanone, 2-hydroxylcyclohexanone and 1,2-cyclohexanedione
with good activity and excellent selectivity toward the formation of the corresponding ketoacids and
diacids. The effects of acidity, amount of catalyst, temperature and solvent on the catalytic activity
were investigated. Furthermore, this catalyst was reusable without any appreciable loss in activity and
selectivity.
Keywords:
K-10 montmorillonite
Dioxygen
␣-Substituted ketones
Oxidative cleavage
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
been made to develop homogeneous catalysts [6] and eco-friendly
heterogeneous catalysts for the synthesis of industrially important
The oxidative cleavage of cyclic ketones and their substituted
derivatives is the main pathway for the synthesis of diacids and
ketoacids on both the laboratory and industrial scales [1]. Due to
ever-growing environmental concerns, there is a strong need for
the establishment of promising catalytic protocols using molecular
oxygen as an oxidizing non-toxic agent [2]. For that reason, many
transition metal-catalyzed aerobic processes, both homogeneous
and heterogeneous, have been developed.
The conversion of cyclohexanone or of cyclohexanol–
cyclohexanone mixtures (KA oil) to adipic acid, one of the starting
materials of Nylon® 6-6, is the most important application of this
reaction [3]. Nitric acid is used as the oxidant in current processes,
thereby leading to pollution with the formation of side-products
such as NOx and N2O [4]. Similarly, laboratory-scale experiments
based on stoichiometric oxidants like Pb(OAc)4, NaIO4, etc. [1,5]
result in the formation of large amounts of solid toxic wastes which
are not easily recyclable. The use of such oxidants have also some
drawbacks that include corrosion, loss of catalyst, environmental
issues and formation of unwanted side products along with the
desired one reducing the yield of the latter. Efforts have therefore
ketoacids. In this context, we have demonstrated that Nafion®
supported vanadium oxo species [7] catalyze efficiently the aerobic
cleavage of C C bonds in some ␣-ketols with good yields and can be
recycled.
The clay catalysts and montmorillonite in particular, have
received considerable attention for different organic syntheses
because of their environmental harmlessness, low cost, high selec-
tivity, reusability and operational simplicity [8]. The reactions
catalyzed by montmorillonite are generally carried out under mild
conditions, the separation of the spent catalyst is achieved by fil-
tration, and the product is recovered by mere evaporation of the
solvent. Furthermore, the montmorillonite catalysts can be regen-
erated easily and reused [9].
Montmorillonite is classified as a 2:1 clay, which means that
one octahedral sheet is sandwiched between two silica tetrahedral
sheets, and belongs to a highly disordered group of smectites [10].
Water molecules are readily absorbed by montmorillonite, forming
hydration shells around the interlayer cations rather than contin-
uous sheets [11]. The cation exchange capacity (CEC) is defined as
the maximum amount of any one cation that can be taken up by a
given clay [10]. Montmorillonite, as smectites in general, has a high
cation exchange capacity, which ranges from 70 to 130 mequiv.
per 100 g [12]. Most of the exchange capacity (80%) is due to
∗
Corresponding author. Tel.: +212 539 97 24 23; fax: +212 539 99 45 00.
E-mail addresses: atlamsani@uae.ma, atlamsani@hotmail.com (A. Atlamsani).
1381-1169/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcata.2012.07.022

438
I. El Younssi et al. / Journal of Molecular Catalysis A: Chemical 363–364 (2012) 437–445
substitution within the structure, but a smaller amount (20%) is
due to the charges at the edges of the sheets [13].
2.3. Characterization techniques
Acid-activated clays are efficient and relatively inexpensive
solid acid catalysts for liquid phase processes [14]. There is, there-
fore, a great variety in the type and activity of acid treated clays.
Despite the discovery of many new inorganic mesoporous molecu-
lar sieves on which solid acids can be based, activated clays remain
one of the most important and widely used classes of mesoporous
solid acid catalyst available in industry [15].
One commercial example is K-10 montmorillonite, which is
obtained from the natural montmorillonite by treatment with min-
eral acids at 96–98 ◦C. The natural montmorillonite structure is
progressively destroyed, resulting in a loss of crystallinity but a sig-
nificant increase in surface area and acidity in comparison to the
natural clay [16]. K-10 montmorillonite clay has been widely used
as catalyst in a large variety of organic reactions [17–21] and it has
also received considerable attention in different areas of organic
synthesis [22–27].
X-ray diffraction patterns were obtained with a Panalytical
X’Pert Pro diffractometer equipped with a X’Celerator detector
using Fe-filtered Co-K␣ radiation. The data were collected at room
temperature with a 0.017◦ step size in 2ꢀ (scan step time = 25 s),
from 2ꢀ = 5 to 110◦. The crystalline phase was identified by com-
parison with ICSD reference files.
X-ray fluorescence analyses were carried out using a Panalyti-
cal Minipal 4 spectrometer equipped with a rhodium anode X-ray
tube. The classical preparation method, with approximately 5 g of
the solid, was not used. An original method based on the analysis
of thin layers [29] and adapted to the geometry of the spectrom-
eter (the analytical area of the Minipal 4 is exactly centered on
the sample holder and corresponds to a diameter of 8 mm) was
developed. Briefly, the intensity was first calibrated by depositing
15 ␮L of a known solution (1 g/L) of iron or calcium salt on polycar-
bonate thin membranes (uncertainties less than 1%). For the other
elements, a deposit of a very small quantity of a geological powder
material (CMR GH-CRPG) was deposited on a second polycarbonate
film from a suspension in cyclohexane. For this standard material,
the concentrations of all the elements under study are very well
known and allow, comparatively to the response for pure iron or
calcium deposits, to determine the sensitivity of other elements
such as potassium, silicon, aluminum or magnesium. The sensitivity
for all elements of interest is thus obtained. A few hundred of micro-
grams of the montmorillonite sample was deposited exactly in the
same manner as for the geological material. Owing to the principle
of thin layer analysis, the intensity is directly proportional to the
mass of the analyzed element in the analytical area and allows to
quantitatively analyze all the deposited powder. We consequently
obtain the elemental mass of all the elements of interest in the
montmorillonite sample.
SEM-FEG images were obtained using a Zeiss SUPRA 40 FESEM
that is completely controlled from a computer workstation. The
electron source is a thermal field emission gun (Schottky type).
Images are created using the SMARTSEM software. Prior to anal-
ysis, the samples were coated with a 5 nm carbon layer using a
Cressington 208 carbon high vacuum carbon coater.
TEM experiments were performed using a Jeol JEM 100CXII
transmission electron microscope operating at 100 kV. One drop of
an ethanolic suspension of the as-produced particles was deposited
on the carbon membrane of the microscope grid for the observa-
tions.
In the present study, we report the aerobic cleavage of C
C
bonds in ␣-substituted ketones catalyzed by K-10 montmorillonite
(referred to as K-10MT). Most experimental parameters such as the
addition of organic and inorganic acids, amount of catalyst, temper-
ature and solvent nature were systematically studied. The results
are compared to those obtained with a synthetic transition-metal
free montmorillonite.
2. Experimental
2.1. Chemicals
Aerosil silica (Degussa), magnesium acetate (Prolabo, 98%),
boehmite alumina (Sassol, Pural SB-1), sodium acetate (Pro-
analysi, ≥99%), acetic acid (Riedel de Haën, ≥99%), acetonitrile
(Riedel de Haën, ≥99.5%), methanol (Riedel de Haën, ≥99.8%),
sulfuric acid (Solvachim, 98%), fluorhydric acid (VWR, 40%), hep-
tanoic acid (Aldrich, ≥97%), 2-methylcyclohexanone (Aldrich, 99%),
2-phenylcyclohexanone (Aldrich, 99%), 2-hydroxycyclohexanone
(Aldrich, 99%), 1,2-cyclohexanedione (Acros, 98%) were used as
received without any further purification.
The K-10 montmorillonite (K-10MT) used in this work was pur-
chased from Aldrich with a surface area (SBET) of 269 m2 g−1. The
composition (wt.%) of K-10MT determined by X-ray fluorescence
(XRF) was: 82.67% SiO2, 13.93% Al2O3, 1.61% Fe2O3, 0.32% TiO2,
0.10% CaO and 1.36% K2O. The calculated cation exchange capacity
was 0.8 mequiv./g clay.
Adsorption and desorption nitrogen isotherms were obtained
at 77 K using a Micromeritics ASAP 2020 apparatus. The samples
were outgassed at 423 K and 0.1 Pa for 12 h before measure-
ments. Specific surface area (SBET) values were obtained using the
Brunauer–Emmett–Teller equation with relative pressures in the
range 0.05–0.20.
2.2. Preparation of synthetic montmorillonite
In order to evaluate the possible activity of a transition-metal
free montmorillonite, a Mg containing synthetic clay, denoted S-
MMT, was prepared according to a procedure previously reported
by Reinholdt et al. [28]. Briefly, a transition-metal free sample was
synthesized in an acidic fluoride medium (caution should be taken
when using HF which can cause severe burns to tissue and is lethal)
using a Teflon-lined stainless steel autoclave under hydrothermal
conditions with the following molar gel composition:
NH3 chemisorption experiments were realized with a Belsorp
apparatus. Samples were first outgassed at 423 K and 10 Pa for
12 h before measurements. The total number of acid sites (Lewis
and Brönsted) was determined using the following method. A first
isotherm corresponding to physically and chemically adsorbed
ammonia was obtained at 373 K. Then outgassing at 423 K for 3 h
was performed and a second ammonia adsorption experiment per-
formed at 373 K gave a second isotherm. This second isotherm
corresponds to physically only adsorbed ammonia. The difference
between the first and the second isotherm is the total chemisorbed
amount (the total acidity).
1SiO2:0.2Al2O3:0.1MgO:0.05Na2O:0.05HF:96H2O
After 72 h reaction time at 493 K, the autoclave was allowed to
cool and the white precipitate was filtered. The pH of the filtrate
was 4.2. The solid was thoroughly washed using Milli-Q® water and
finally dried at 333 K for 12 h.
2.4. Catalytic tests
All catalytic tests were carried out using Schlenk techniques
(20 mL) which was attached to a vacuum line with a manometer

I. El Younssi et al. / Journal of Molecular Catalysis A: Chemical 363–364 (2012) 437–445
439
˚
montmorillonite are 10.3 and 14.5 A, respectively. The d0 0 1 inter-
reticular distance of a synthetic material natural montmorillonite
under a relative humidity of 80% has been reported by Reinholdt
1.490 Å
(c)
(a)
˚
et al. to be about 15 A [28]. Because of the swelling properties
of smectites, the position of the (0 0 1) reflexion depends on the
water content of the material which in turn depends on the rela-
tive humidity of the atmosphere in which the materials have been
stored. The different (0 0 1) positions observed for K-10MT and S-
MMT could thus be explained by different water amounts in the
interlamellar space of the clay mineral. The basal spacing d0 0 1 of
montmorillonite K-10MT “with K as predominant exchangeable
1.502 Å
*
*
(001)
(001)
72 73 74 75
*
*
˚
*
*
cations” is 10.3 A in agreement with zero-layer hydration state.
*
*
*
*
*
*
(a)
The synthetic montmorillonite under high humidity conditions
and in presence of Na+ exchangeable cation shows the increase in
the basal spacing corresponding to the two layers hydration state
˚
(d0 0 1 = 14.5 A). Moreover, the (0 0 1) peak is clearly broader for S-
MMT than for K-10MT which can be explained considering: (i) a
higher crystallization state for K-10MT than for S-MMT and/or (ii)
smaller crystallite sizes for S-MMT (this latter point is in agree-
ment with the SEM-FEG observations, vide infra). The two samples
exhibit a peak at about 73◦ attributed to the (0 6 0) reflection of
dioctahedral phyllosilicates [28]. It is well known that for diocta-
hedral smectites, i.e. containing vacancies in the octahedral layer,
*
*
(001)
(b)
(c)
(110),(020)
(130),(200)
(060)
(210)
10
20
30
40
50
60
70
80
2θ (KαCo), °
˚
the (0 6 0) reflection is expected at 1.49 A, whereas for trioctahe-
˚
dral materials, this reflection is at 1.52 A [28,31]. For S-MMT d0 6 0 is
Fig. 1. X-ray diffraction patterns for: (a) uncalcined K-10MT, (b) K-10MT after catal-
ysis test (corresponding to Table 2, entry 3) and (c) synthetic montmorillonite
S-MMT: ꢁ: quartz (JCPDS no. 01-078-1252). Inset: enlargement of the (0 6 0) reflec-
tion.
˚
˚
1.49 A whereas for K-10MT, it is 1.50 A. Both values are indicative of
dioctahedral smectites. Note that for K-10MT, peaks corresponding
to quartz (JCPDS no. 01-078-1252) are also unambiguously identi-
fied; this has already been observed by other authors [31,32].
The SEM-FEG micrographs of samples K-10MT and S-MMT are
presented in Fig. 2a and b, respectively. They show clearly the
lamellar nature of the materials. The S-MMT clay exhibits a more
divided state than K-10MT and is made of crumpled platelets as
reported previously by Reinholdt et al. [28].
and a gas inlet. Prior to each catalytic test, the clay was dried at
373 K for 2 h. In a typical experiment, the Schlenk was charged with
the solvent (5.0 mL; solvent: CH3CN, H2O, CH3OH, CH3COOH or a
mixture of CH3COOH and H2O) and K-10MT (0.20 g). In some tests
H2SO4 (0–0.37 mmol) was also added. The substrate (5.0 mmol)
was then added and the vessel was immersed in an oil bath pre-
heated at 70 ◦C. O2 was introduced at atmospheric pressure and the
mixture was stirred magnetically for 24 h. Three parallel catalytic
experiments were carried out for each test.
The TEM image displayed in Fig. 2c shows also the lamellar
nature of the K-10MT montmorillonite. The d0 0 1 value, determined
˚
with the numerical electron diffraction pattern, is found to be 10.1 A
and is in excellent agreement with the value determined by X-ray
diffraction.
Dioxygen uptake was determined using a thermostated (70 ◦C)
gas burette system connected to the reactor. The addition of
heptanoic acid as an internal standard at the beginning of the
experiment permitted to quantitatively analyze the products using
a Shimadzu GC-17A gas chromatograph equipped with a TRA-5
column (30 m × 0.25 mm internal diameter) and a flame ioniza-
tion detector. GC conditions: initial temperature, 80 ◦C for 10 min;
ramp rate, 10 ◦C min−1; final temperature, 220 ◦C; injection tem-
perature, 220 ◦C; detector temperature, 250 ◦C; carrier gas, He at
25 mL min−1. Absolute errors were estimated to be about 5%.
After catalysis test, the catalyst was recovered following the
recommendations proposed by Sheldon [30]: the solid was fil-
tered at the reaction temperature, i.e. 70 ◦C, washed three times
with distilled water and dried at 373 K for 12 h. The powders were
characterized by X-ray diffraction (XRD) and transmission elec-
tron microscopy (TEM). The recovered catalyst was also used for
recycling experiments.
The clays were also characterized using physisorption tech-
niques. Both materials exhibit type II isotherms according to
the IUPAC classification [33] with a H3 hysteresis type, usually
obtained for aggregates of platy particles or adsorbents containing
slit-shaped pores (Fig. S1, Supplementary materials). The C BET con-
stants found for K-10MT and S-MMT are 111 and 264, respectively.
The significant higher C BET value for S-MMT than for K-10MT can
indicate the existence of microporosity for the former. The spe-
cific surface areas, determined using the BET method, are 269 and
132 m2 g−1 for K-10MT and S-MMT, respectively. The important
difference in the SBET values can be explained taking into account
that the commercial K-10MT sample has been activated by mineral
acids which leads to a loss of crystallinity but a significant increase
in surface area compared to the natural clay [16]. Conversely, even
if S-MMT has been prepared in a highly acidic fluoride medium,
it has not undergone a post-treatment which can alter its textural
properties.
3. Results and discussion
3.1.2. Ammonia chemisorption
3.1. Characterization of the materials
Ammonia chemisorption has been carried out at 373 K in order
to evaluate the acidic properties of the two samples. The obtained
isotherms are presented in Fig. 3 and the number of total (Lewis
and Brønsted) acid sites is given for the two samples in Table 1.
The values (expressed in mmol g−1) obtained here are close to that
reported by Flessner et al. for similar solids [34]. However, when
normalized to the specific surface area value, S-MMT displays a
3.1.1. Structural and morphological studies
The X-ray diffraction patterns of the K-10MT and synthetic
montmorillonites are presented in Fig. 1a and c, respectively. They
exhibit characteristic reflections usually observed for smectites
[28,31]. The basal spacing (d0 0 1) for K-10MT and the synthetic

440
I. El Younssi et al. / Journal of Molecular Catalysis A: Chemical 363–364 (2012) 437–445
Fig. 2. SEM-FEG images of (a) dried K-10MT and (b) S-MMT and TEM images of (c) dried K-10MT (inset: numerical electron diffraction pattern) and (d) K-10MT after catalysis
test (corresponding to Table 2, entry 3).
higher number of acid sites per unit surface area than K-10MT. This
result could be explained considering that: (i) S-MMT has been
synthesized in a highly acidic fluoride medium and (ii) protons can
readily exchange with other cations present in the inter-lamellar
space.
45
(a)
40
35
30
25
20
15
10
5
3.2. Reactivity
3.2.1. Oxidation of 2-methylcyclohexanone
We have first studied the oxidative cleavage of carbon carbon
bonds in ␣-substituted ketones using commercial K-10MT and
synthetic montmorillonite S-MMT as heterogeneous catalysts and
dioxygen as a clean oxidant. While K-10MT contains iron and tita-
nium cationic species, S-MMT is transition-metal free (see Sections
2.1 and 2.2). Initially, the oxidation of 2-methylcyclohexanone, 1,
to 6-oxoheptanoic acid, 2, was chosen in order to investigate the
catalytic activity, selectivity, and stability of the clays (Scheme 1).
Note that test experiments realized under the same conditions, but
in the absence of a catalyst, showed that the ketoacid cannot be
obtained.
0
0
10
20
30
40
50
60
70
80
P(NH3), kPa
(b) 45.0
40.0
The catalytic results obtained with the two sets of catalysts, K-
10MT and S-MMT are presented in Table 2. In methanol, the data
clearly show that K-10MT is an efficient catalyst contrary to S-MMT,
since no conversion is observed with the latter. Nevertheless, a
moderate yield is obtained with K-10MT if no acid is added in the
35.0
30.0
25.0
20.0
15.0
10.0
5.0
Table 1
Specific surface areas and number of total acid sites determined by ammonia
chemisorption at 373 K.
0.0
0
10
20
30
40
50
60
70
80
Sample
SBET, m2 g−1
Number of acid
sites, mmol g−1
Number of acid
sites, ␮mol m−2
P(NH3), kPa
K-10MT
S-MMT
269
132
0.23
0.34
0.85
2.57
Fig. 3. NH3 chemisorption isotherms at 373 K for (a) K-10MT and (b) S-MMT: ♦, first
isotherm; ꢀ, second isotherm.

I. El Younssi et al. / Journal of Molecular Catalysis A: Chemical 363–364 (2012) 437–445
441
O
1
80
70
60
50
(a)
K-10MT
O
CH3
or S-MMT
CH3
HO
Solvent, O2, 1 atm
O
2
40
30
Scheme 1. Heterogeneous oxidation of 2-methylcyclohexanone, with dioxygen and
catalyzed by commercial K-10MT and synthetic montmorillonite S-MMT.
20
10
0
reaction medium indicating that the intrinsic acidity of the clay
is not sufficient to induce appreciable conversion of the substrate.
The improvement of the catalytic activity with the addition of acids
is well known [6,7,35,36]. Several reasons have been proposed: (i)
the oxidation reaction proceeds first via the formation of the enol
tautomer from the cyclic ketone, which is enhanced in the pres-
ence of protons [6,7,35,36], (ii) peracetic acid could be generated
in situ from acetic acid, the former acting as an oxidant [36] and
(iii) the protonation of the cyclohexanone may facilitate its disso-
lution as far as water is used as the solvent [36] (the solubility of
2-methylcyclohexanone is about 2 g per 100 mL at 20 ◦C [37]). For
instance, previous results obtained with the homogeneous system
Cu(NO3)2·3H2O/O2/AcOH/H2O at 60 ◦C have shown that the high-
est 2-methylcyclohexanone conversion and highest ketoacid yield
were obtained for the AcOH/H2O = 4.5/0.5 mL/mL composition [6].
Recently, Cavani et al. have also shown the dramatic influence of
acetic acid used as a co-solvent for the oxidation of cyclohexanone
into adipic acid in the presence of H3+n[PMo12−nVnO40]·aq (n = 1 and
2) polyoxometalates: the highest conversion was observed for an
equivolumetric mixture of water and acetic acid [36]. Moreover it
was shown that, in presence of acetic acid, besides the redox mech-
anism, a radical-chain auto-oxidation mechanism prevailed when
a very low catalyst/cyclohexanone was used. For K-10MT, the use
of a water/acetic acid mixed solvent also allowed to increase con-
siderably the conversion and the yield toward the ketoacid (see
Table 2, entry 2). The catalytic activity could be further increased
adding a small amount of sulfuric acid (see Table 2, entry 3). The
results displayed in Table 2 also show that the transition metal-free
S-MMT is inactive toward the oxidation of 2-methylcyclohexanone,
indicating that if acidity is needed, transition metal species are
mandatory.
0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8
Amount of K-10MT, g
80
70
60
50
40
30
(b)
0
1
2
3
4
5
6
7
8
9
10
2-methylcyclohexanone, mmol
Fig. 4. (a) Effect of the catalyst mass on the conversion of 2-methylcyclohexanone.
Conditions: substrate (5 mmol), AcOH/H2O (4.5 mL/0.5 mL), H2SO4 (0.37 mmol),
t = 24 h, T = 70 ◦C, P(O2) = 0.1 MPa and (b) effect of the 2-methylcyclohexanone
amount with 0.2 g of catalyst at the same reaction conditions.
overlapping reaction mechanisms can account for the product dis-
tributions: a redox mechanism (predominant in absence of acetic
acid) and a radical auto-oxidation one (favored in the presence of
acetic acid and low catalyst amounts) [36]. In our work, the decrease
in ketone conversion could be explained by chain-termination
effects as reported previously for Co-containing molecular sieves
for the liquid phase oxidation of cyclohexane to adipic acid [38].
The effect of the initial amount of 2-methylcyclohexanone was
investigated on the conversion in the range of 1.25–8.75 mmol, at
70 ◦C, by keeping the same amount of catalyst (0.2 g). As shown in
Fig. 4b, the conversion increased with increase in quantity of 2-
methylcyclohexanone reaching a maximum of 70% for 5 mmol. In
the following, all the catalysis tests were thus performed using 0.2 g
of solid montmorillonite and 5 mmol of substrate.
3.2.1.1. Effect of catalyst amount and effect of substrate amount.
Fig. 4a shows the effect of the amount of K-10MT catalyst on the
conversion of 2-methylcyclohexanone, using a mixed water/acetic
acid (4.5/0.5 mL/mL) solvent in the presence of a small amount of
H2SO4. The amount of K-10MT catalyst was increased from 0 to
0.7 g. In the absence of catalyst, the ketoacid could not be obtained.
The conversion of the ketone increased with the amount of catalyst,
reaching a maximum of 70% for 0.2 g. This point can be explained by
the increase in the number of acid and redox sites available for oxi-
dation reaction. Further increase in catalyst amount, beyond 0.2 g,
resulted in a decrease of the conversion of the substrate. Cavani
et al. have recently reported that in the presence of acetic acid, two
3.2.1.2. Effect of solvent. Preliminary studies with other catalytic
systems have emphasized the important role of the solvent
[6,39,40]. In this study, the effect of the solvent is also clear: after
Table 2
Oxidation of 2-methylcyclohexanone with K-10MT and S-MMT.
Entry
Catalyst
K-10MT
Solvent (mL)
mmol O2
Conversion (%)
6-Oxoheptanoic acid yield (%)
1a
2
CH3OH (5)
0.53
2.65
3.50
0
0
0
10
53
70
–
–
–
8
51
68
–
–
–
AcOH/H2O (4.5/0.5)
AcOH/H2O (4.5/0.5)
CH3OH (5)
AcOH/H2O (4.5/0.5)
AcOH/H2O (4.5/0.5)
3b
4a
5
S-MMT
6b
Conditions: substrate (5 mmol, 0.6 mL), catalyst (0.2 g), t = 24 h, 70 ◦C, p (O2) = 0.1 MPa.
a
Reaction carried out at 60 ◦C.
Test carried out with 0.37 mmol H2SO4.
b

442
I. El Younssi et al. / Journal of Molecular Catalysis A: Chemical 363–364 (2012) 437–445
Table 3
Table 4
Effect of solvent on the oxidation reaction of 2-methylcyclohexanone.
Effect of temperature and acidity in the oxidation reaction of 2-
methylcyclohexanone.
Entry
Solvent (mL)
mmol O2
Conversion (%)
6-Oxoheptanoic
acid yield (%)
Entry Temperature
(◦C)
mmol
sulfuric acid
mmol
O2
Conversion 6-Oxoheptanoic
(%)
acid yield (%)
1
CH3CN (5)
CH3OH (5)
H2O (5)
AcOH/H2O (3/2)
AcOH/H2O (4/1)
AcOH (5)
0.41
2.50
1.10
1.60
2.97
4.00
2.00
8
50
22
32
59
80
40
7
48
19
30
57
77
38
2a
3
1
2
3
4
5
6
7
8
9c
r.t.
40
70
70
70
70
70
60
70
0.37
0.37
–
0.52
1.35
2.65
3.00
3.50
3.50
1.65
0.60
–
10
27
53
60
70
70
33
12
12
8
26
51
58
68
68
30
10
10
4
5
0.18
0.37
0.55
0a
6b
7
–
Conditions: substrate (5 mmol, 0.6 mL), K-10MT (0.2 g), H2SO4 (0.37 mmol), t = 24 h,
70 ◦C, p (O2) = 0.1 MPa.
0a,b
0a,b
a
Reaction carried out at 60 ◦C.
Reaction time 30 h.
b
Conditions: substrate (5 mmol, 0.6 mL), AcOH/H2O (4.5 mL/0.5 mL), K-10MT (0.2 g),
t = 24 h, p (O2) = 0.1 MPa.
a
Reaction carried out with a K-10MT sample (1.0 g) first treated with H2SO4
(1.85 mmol) in water (25 mL) for 5 h at 96 ◦C, recovered by filtration and dried at
100 ◦C for 12 h.
24 h, the conversion reaches only 8% in acetonitrile whereas it is 22%
in aqueous media (Table 3, entries 1 and 3). With pure methanol,
higher values (50%) were obtained (Table 3, entry 2). It is difficult to
explain the effect of solvent on the basis of their dipole moments or
their dielectric constants, but, whatever the solvent used, excellent
selectivities toward 6-oxoheptanoic acid were observed, between
86 and 97%. We believe that the solvent effect could be explained
by the fact that the solubility of dioxygen differs from one solvent
to another. Probably, when the mole fraction of dioxygen increase
the catalytic activity increases.
Moreover, X-ray fluorescence analyses have shown that iron and
titanium are present in K-10MT whereas the synthetic montmoril-
lonite is transition metal free (vide supra). Thus, the active species
could correspond to Fe(II) and/or Fe(III) species and/or Ti(IV). We
think that the mixed solvent AcOH/H2O (4.5 mL/0.5 mL) facilitates
the reduction rate of Mn+ (M = Fe or Ti) and increases the solubil-
ity of dioxygen. By-products such as cyclopentanone are identified,
but they are always produced in very low yield.
Note that without solvent, a moderate activity (38% yield) is
observed but higher than with water (Table 3, entries 3 and 7). It is
a very interesting result regarding environmental concerns because
the exclusion of pollutants from an unselective preparation method
is always preferable to subsequent treatment. In this context, the
goal of modern organic synthesis is to develop efficient catalytic
methods that can produce compounds with atom economy and
environmental advantages.
With pure acetic acid (Table 3, entry 6) the conversion of 2-
methylcyclohexanone can reach up to 80% but a longer reaction
time is necessary (30 h). This is explained by an observed induction
time (about 6 h) before reaction starts (see Fig. 5).
b
Reaction carried out in methanol.
In methanol under reflux conditions.
c
Based on previous studies [39,40] realized with homogeneous
catalytic systems, mixtures of water and acetic acid were also eval-
uated. The best result (Table 2, entry 3) has been obtained for
the AcOH/H2O = 4.5/0.5 v/v composition with a 68% yield for 6-
oxoheptanoic acid after 24 h. Indeed, even if the initial rate is higher
for the AcOH/H2O = 4/1 (v/v) composition than for the latter (see
Fig. 5), the final conversion after 24 h is substantially lower.
In the following, the tests have been carried out with the
AcOH/H2O = 4.5/0.5 (v/v) mixed solvent.
3.2.1.3. Effect of temperature and acidity. We have further studied
the role of the temperature and the acidity on the conversion of
compounds 1 to 2. In this way, we realized the oxidation reaction
at room temperature (Table 4, entry 1), at 40 ◦C (Table 4, entry 2)
and at 70 ◦C (Table 4, entry 5). The yield in ketoacid was gradually
increased from 8 to 68% after 24 h.
Moreover, previous work has shown the dramatic influence
of the Brønsted acidity on the catalytic activity. For instance, the
remarkable efficiency of HPA-n, i.e. H3+n[PMo12−nVnO40]·aq, was
explained by the combination of their intrinsic Brønsted acidity
together with their redox properties [39,40]. In order to obtain
higher conversions, the acidity of the catalytic system was fur-
ther increased via the addition of H2SO4 into the reaction mixture.
It seems clear that its presence enhances the catalytic activity in
the reaction of oxidative cleavage of 2-methylcyclohexanone (see
entries 3–5 in Table 4). Note that a further addition in sulfuric acid
(entry 6, Table 4) did not allow to improve the product yield.
It is noteworthy that for a K-10 MT sample first treated ex situ
with sulfuric acid, a decreased conversion was observed compared
to what is obtained for a test performed with K-10MT and the
joint addition of sulfuric acid (see Table 4, entry 7 compared to
Table 4, entries 3 and 5 and see Table 4, entries 8 and 9 compared to
Table 3, entry 2). This could be explained considering that the pre-
treatment of K-10MT performed in acidic aqueous medium has led
to a partial depletion in some cations within the octahedral sheet
(please see XRF results below). Consequently, the number of pro-
tons in the interlamellar space was reduced because they balance
the net negative charge on the layer which arises from isomorphous
substitution in the octahedral sheet [41].
Using clay catalysts, the question of structural stability always
arises. Structural evolution of recovered catalyst at the end of the
reaction (carried in the presence of H2SO4 at 70 ◦C) was moni-
tored with X-ray diffraction (Fig. 1b) and transmission electron
microscopy (Fig. 2d).
Fig. 5. O2 consumption at 70 ◦C for different solvents using K10-MT for the catalytic
oxidation of 2-methylcyclohexanone in (ꢀ) water, (ꢁ) AcOH and different AcOH/H2O
mixtures: (ꢁ) 3 mL/2 mL; (ꢂ) 4 mL/1 mL; (᭹) 4.5 mL/0.5 mL.
As shown in Fig. 1b, the addition of sulfuric acid at 70 ◦C to the
reaction mixture did not modify the basic structural characteristics

I. El Younssi et al. / Journal of Molecular Catalysis A: Chemical 363–364 (2012) 437–445
443
Conversion
Yield
1.05
1.00
0.95
0.90
0.85
0.80
0.75
0.70
80
60
40
20
0
0
1
2
3
4
5
Number of run
1
2
3
4
5
Run
Fig. 7. Variations of the relative metal contents (M/M0) determined by XRF for K-
10MT powder vs the number of runs: ꢁ, Si; ꢃ, Ca; ᭹, Al; ꢄ, K; ꢅ, Ti; ꢀ, Fe.
Fig. 6. Recycling of K-10MT catalyst in the oxidation of 2-methylcyclohexanone by
dioxygen. Conditions: substrate (5 mmol, 0.6 mL), AcOH/H2O (4.5 mL/0.5 mL), H2SO4
(0.37 mmol), K-10MT (0.2 g), t = 24 h, p (O2) = 0.1 MPa.
K-10MT decrease in the order of Ca2+, Al3+, Fe3+, Si4+, K+ and Ti4+
with the increase in number of run. There is a marked decrease in
the Ca element and, to a lesser extent, in the Fe and Al elements,
together with a relative increase of the Si, Ti and K elements. Similar
observations have been made by Breen et al. for various smectites
treated with different combinations of acid concentrations, tem-
peratures and reaction times [41]. These results can be explained
considering that the K+ cations between 2:1 layers of montmoril-
lonite and the Si4+, Ti4+ cations at tetrahedral sites of smectite are
not extracted from the framework by the addition of sulfuric acid
in the reaction mixture. By opposition there is a partial leaching
of iron and aluminum under the reacting conditions for the first
four tests. After four tests, their amounts in the solid material are
nearly constants. It is well known that iron(III) salts can catalyze the
oxidative cleavage of cyclic ketones [43]. We thus can suspect the
contribution of solvated iron species to the global catalytic activity
in the first four runs. However, more importantly, the performances
of the catalyst after 5 runs are the same than that of the fresh cata-
lyst whereas the metal leaching in solution, if any, is negligible (see
Figs. 6 and 7). This indicates that K-10MT can be considered as a
true heterogeneous catalyst.
of K-10MT, since the original characteristic peaks of the clay are still
present. TEM image (Fig. 2d) reveals that the recovered catalyst
is still laminated like the fresh clay (Fig. 2c and d). This observa-
tion is in good agreement with the X-ray diffraction results. The
results given above by XRD and TEM indicate that K-10MT was not
degraded under the chosen reaction conditions.
3.2.2. Catalyst regeneration
The clay obtained after the catalytic test corresponding to
Table 2, entry 3 was separated at the reaction temperature from the
reaction mixture by filtration and washed four times with distilled
water. The resulting clay was dried at 373 K for 10 h, grounded into
a fine powder and used again to study its activity for the oxidation
reaction.
Fig. 6 shows that the clay can be recycled several times without
any appreciable loss in activity and selectivity (the small observed
variations are within the experimental error).
3.2.3. Change in the chemical composition after recycling
As it is well known [41,42], acid-activation often results in sub-
stantial changes of the montmorillonite clay such as the removal
of octahedral cations or the modification of specific surface areas
and pore volumes. The XRF chemical analyses (wt.%) of the fresh
clay and recovered catalysts are given in Table 5. The variations of
the relative content of metals (M/M0) in the K-10MT with num-
ber of runs are given in Fig. 7 in which M0 and M are the wt.% of
the oxide elements in the fresh clay and recovered catalysts sam-
ples. Different behaviors are observed depending on the nature of
the element. The relative contents of the leached elements in the
3.2.4. Oxidation of other ˛-substituted ketones
Encouraged by the remarkable results obtained with the
above reaction conditions, and in order to show the scope and
possible generalization of this new protocol, we used this cat-
alytic system, K-10MT/AcOH-H2O/H2SO4/O2 for the oxidation of
other ␣-substituted ketones (Table 6). 2-Phenylcyclohexanone
produces mainly 5-benzoyl valeric acid (entry 1). Treatment of
2-hydroxylcyclohexanone gives adipic acid in good yield with
excellent selectivity (Table 6, entry 2). Dimethyl adipate was actu-
ally obtained as a result of the in situ esterification of the crude
mixture with methanol. Adipic acid can also been obtained from
1,2-cyclohexanedione but with a lower yield (Table 6, entry 3).
In both cases, very low yields of dimetyl glutarate and dimethyl
succinate derivatives were obtained.
Table 5
X-ray fluorescence analyses of the fresh and recovered catalysts.
Run
% (w/w)
SiO2
4. Discussion
Al2O3
Fe2O3
TiO2
CaO
K2O
0
1
2
3
4
5
82.67
83.37
83.69
83.99
84.15
84.19
13.93
13.30
12.99
12.70
12.55
12.51
1.61
1.54
1.53
1.52
1.50
1.50
0.32
0.32
0.32
0.32
0.33
0.33
0.10
0.09
0.09
0.08
0.08
0.08
1.36
1.38
1.38
1.39
1.39
1.39
This paper describes the aerobic oxidative C C bond cleavage
of ␣-substituted ketones using montmorillonites as bifunctional
catalysts, i.e. both solid and redox catalysts. Two materials were
considered: a commercial montmorillonite, K-10MT, and a syn-
thetic product, denoted S-MMT. Both solids exhibit comparable
acidic properties as evidenced by NH3 chemisorption at 373 K. Nev-
ertheless, no conversion of the ␣-substituted ketone was observed
Conditions: substrate (5 mmol, 0.6 mL), AcOH/H2O (4.5 mL/0.5 mL), H2SO4
(0.37 mmol), K-10MT (0.2 g), t = 24 h, p (O2) = 0.1 MPa.

444
I. El Younssi et al. / Journal of Molecular Catalysis A: Chemical 363–364 (2012) 437–445
Table 6
Catalytic oxidation of ␣-substituted ketones by the catalytic system K-10MT/AcOH-H2O/H2SO4/O2.
Entry
␣-Substituted ketones
mmol O2
Conversion (%)
Product and yield (%)
OH
O
O
O
1
1.85
37
36
OMe
O
O
OH
O
2a
2.50
1.30
50
26
OMe
48
OMe
O
O
O
O
3a
OMe
24
Conditions: substrate (5 mmol), K-10MT (0.2 g), AcOH/H2O (4.5 mL/0.5 mL), H2SO4 (0.37 mmol), t = 24 h, 70 ◦C, p (O2) = 0.1 MPa.
a
Conversions and yields were determined by GC analysis after the in situ esterification of the crude mixture with methanol using methyl heptanoate as internal standard.
with S-MMT showing the fundamental role of transition metal
species in the reaction pathway.
5. Conclusion
Montmorillonite is
a
2:1 layered aluminosilicate with an
In summary, we have successfully developed a simple and effi-
cient method for the synthesis of ketoacids or diacids derivatives
from various ␣-substituted ketones using cheap and readily avail-
able K-10MT as catalyst using molecular oxygen as oxygen source.
The main features of this procedure are the appreciable yields of
products, the clean reaction profiles (i.e. the good selectivities) and
the availability of the reagents at low cost and enhanced rates which
makes it a useful and attractive process. The simple experimental
and product isolation procedures as well as the easy recovery and
reuse of the catalyst are expected to play an important role in the
development of this new method. Moreover, these results open the
way for the use of this inexpensive material in other applications.
Even if acidity is a necessary condition to allow the formation
of the enol and the subsequent C C bond oxidative cleavage, the
presence of transition metals species is also a prerequisite. Indeed,
with a transition metal-free synthetic montmorillonite prepared
by hydrothermal synthesis, no conversion was observed. Future
efforts will be directed to the preparation of transition metal (such
as copper, iron, titanium or vanadium) containing synthetic mont-
morillonites via cation exchange procedures and to the evaluation
of their catalytic properties. Further research will also develop
toward elucidating the catalytic oxidation mechanism.
octahedral sheet (predominantly aluminate) sandwiched by two
tetrahedral layers (predominantly silicate). Due to the isomor-
phous substitution of Al3+ within the octahedral layer by divalent
cations such as Mg2+ or Fe2+ and of Si4+ by Al3+ in the tetrahedral
layer, the resulting charge deficit is balanced by the incorporation
of exchangeable cations in the interlamellar space. X-ray fluores-
cence analyses have shown that iron and titanium are present in
K-10MT whereas the synthetic montmorillonite is transition metal
free.
Iron(III) has been shown to be active toward the aerobic
oxidative C C bond cleavage of ␣-substituted ketones [43]. The
dimension of the hexagonal cavity in the tetrahedral layer is about
˚
5 A [44]. To the best of our knowledge, no data has been reported
concerning the kinetic diameter of 2-methylcyclohexanone but for
cyclohexane and trimethylbenzene, these diameters are 6.0 and
˚
7.3 A, respectively [45]. Thus, it appears that the physical access
of the substrate to the octahedral layer is restricted. However,
even if Milone et al. [32] have shown by X-ray fluorescence that
Fe cations were mainly located in the octahedral sheet for K-10
montmorillonite, the ion-exchange reaction of a native sample (Fe
amount: 2.0 wt.%) with a NaCl aqueous solution yielded a mate-
rial with a reduction of the iron amount (Fe amount: 1.7 wt.%)
as seen by X-ray fluorescence. Although it is a small diminution
(0.3 wt.%), it indicates that some Fe cations could be located inside
the interlamellar space of the clay and could thus be available for
catalysis.
To date, titanium(IV) has not been reported as an efficient cat-
alyst for such a reaction, but its participation in the catalytic cycle
cannot be discarded. TiO2 solids are well known to activate dioxy-
gen under photocatalytic conditions [46]. Thus, in this work, the
active species could correspond to Fe(II) and/or Fe(III) species inter-
calated between the clay layers and/or Ti(IV) species which are
located in the tetrahedral sheet.
Acknowledgements
I. El Younssi thanks the Ministère de l’éducation nationale,
de l’enseignement supérieur, de la formation des cadres et de la
recherche scientifique of Morocco for the allocation grant. T. Rhadfi
gratefully acknowledges the Agence Universitaire de la Franco-
phonie and the CNRST for financial support. Thanks are due to Drs.
J.-Y. Piquemal and L. Sicard for constructive discussions, for cor-
recting the manuscript and for physicochemical analysis facilities,
and to reviewers for valuable comments.

I. El Younssi et al. / Journal of Molecular Catalysis A: Chemical 363–364 (2012) 437–445
445
Appendix A. Supplementary data
[21] F. Bigi, L. Chesini, R. Maggi, G. Sartori, J. Org. Chem. 64 (1999) 1033–1035.
[22] E.C. Taylor, C.-S. ChiangeTaylor, Synthesis (1977) 467.
[23] S. Hunig, E. Benzing, E. Lucke, Chem. Ber. 90 (1957) 2833–2840.
[24] J.-F. Roudier, A. Foucaud, Tetrahedron Lett. 25 (1984) 4375–4378.
[25] B. Labiad, D. Villamin, Synthesis (1989) 143–144.
[26] D. Fishman, J.T. Klug, A. Shani, Synthesis (1981) 137–138.
[27] M. Onaka, T. Shinoda, Y. Izumi, R. Nolen, Chem. Lett. (1993) 117–120.
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/
j.molcata.2012.07.022.
[28] M. Reinholdt, J. Miehé-Brendlé, L. Delmotte, M.-H. Tuilier, R. le Dred, R. Cortès,
A.-M. Flank, Eur. J. Inorg. Chem. (2001) 2831–2841.
[29] P. de Chateaubourg, J.-P. Quisefit, S. Garivait, E. Steiner, C. Goyon, Analusis 21
(1993) 293–298.
[30] R.A. Sheldon, M. Wallau, I.W.C.E. Arends, U. Schuchardt, Acc. Chem. Res. 31
(1998) 493–495.
[31] A. Bakandritsos, A. Simopoulos, D. Petridis, Nanotechnology 17 (2006)
1112–1117.
[32] C. Milone, M. Dhanagopal, S. Santangelo, M. Lanza, S. Galvagno, G. Messina, Ind.
Eng. Chem. Res. 49 (2010) 3242–3249.
[33] K.S.W. Sing, D.H. Everett, R.A.W. Haul, L. Moscou, R.A. Pierotti, J. Rouquerol, T.
Siemieniewska, Pure Appl. Chem. 57 (1985) 603–619.
[34] U. Flessner, D.J. Jones, J. Rozière, J. Zajac, L. Storaro, M. Lenarda, M. Pavan, A.
Jiménez-López, E. Rodríguez-Castellón, M. Trombetta, G. Busca, J. Mol. Catal. A:
Chem. 168 (2001) 247–256.
[35] M. Vennat, J.-M. Brégeault, Appl. Catal. A: Gen. 386 (2010) 9–15.
[36] F. Cavani, L. Ferroni, A. Frattini, C. Lucarelli, A. Mazzini, K. Raabova, S. Alini, P.
Accorinti, P. Babini, Appl. Catal. A: Gen. 391 (2011) 118–124.
[37] R.M. Stephenson, J. Chem. Eng. Data 37 (1992) 80–95.
[38] I. Belkhir, A. Germain, F. Fajula, E. Fache, J. Chem. Soc. Faraday Trans. 94 (1998)
1761–1764.
[39] A. Atlamsani, J.-M. Brégeault, M. Ziyad, J. Org. Chem. 58 (1993) 5663–5665.
[40] L. El Aakel, F. Launay, A. Atlamsani, J.-M. Brégeault, Chem. Commun. (2001)
2218–2219.
[41] C. Breen, F. Deeba Zahor, J. Madejová, P. Komadel, J. Phys. Chem. B 101 (1997)
5324–5331.
[42] P. Komadel, J. Madejová, Acid activation of clay minerals, in: F. Bergaya, B.K.G.
Theng, G. Lagaly (Eds.), Handbook of Clay Science. Development in Clay Science,
vol. 1, Elsevier Ltd., 2006, pp. 263–288.
[43] S. Ito, M. Matsumoto, J. Org. Chem. 48 (1983) 1133–1135.
[44] C. Swearingen, J. Wu, J. Stucki, A. Fitch, Environ. Sci. Technol. 38 (2004)
5598–5603.
References
[1] R.C. Larock, Comprehensive Organic Transformations, second ed., Wiley, New
York, 1999, pp. 1664–1669.
[2] K. Kervinen, H. Korpi, M. Leskelä, T. Repo, J. Mol. Catal. A: Chem. 203 (2003)
9–19.
[3] A.K. Suresh, M.M. Sharma, T. Sridhar, Ind. Eng. Chem. Res. 39 (2000) 3958–3997.
[4] A. Shimizu, K. Tanaka, M. Fujimori, Chemosphere: Global Change Sci. 2 (2000)
425–434.
[5] R. Floresca, M. Kurihara, D.S. Watt, A. Demir, J. Org. Chem. 58 (1993) 2196–2200.
[6] A. Atlamsani, J.-M. Brégeault, Synthesis (1993) 79–81.
[7] L. El Aakel, F. Launay, J.-M. Brégeault, A. Atlamsani, J. Mol. Catal. A: Chem. 212
(2004) 171–182.
[8] A. Cornelius, P. Laszlo, Synlett (1994) 155–161.
[9] T.-K. Huang, R. Wang, L. Shi, X.-X. Lu, Catal. Commun. 9 (2008) 1143–1147.
[10] W.E. Worrall, Clays and Ceramic Raw Materials, second ed., Elsevier Scientific,
London, 1986.
[11] W.A. Deer, R.A. Howie, J. Zussman, An Introduction to the Rock-Forming Min-
erals, second ed., Addison-Wesley-Longman Ltd., Harlow, 1996.
[12] J.T. Kloprogge, E. Mahmutagic, R.L. Frost, J. Colloids Interface Sci. 296 (2006)
640–646.
[13] C.E. Weaver, L.D. Pollard, The Chemistry of Clay Minerals, Elsevier Scientific,
Amsterdam, 1973.
[14] M.P. Hart, D.R. Brown, J. Mol. Catal. A: Chem. 212 (2004) 315–321.
[15] A. Vaccari, Catal. Today 41 (1998) 53–71.
[16] T. Cseri, S. Békassy, F. Figueras, E. Cseke, L.C. de Menorval, R. Dutartre, Appl.
Catal. A: Gen. 132 (1995) 141–155.
[17] C. Cativiela, F. Figueras, J.M. Fraile, J.I. Garcia, J.A. Mayoral, Tetrahedron: Asymm.
4 (1993) 223–228.
[18] B.M. Choudary, N.S. Chowdari, M.L. Kantam, R. Kannan, Tetrahedron Lett. 40
(1999) 2859–2862.
[19] F.P.R. Crisóstomo, R. Carrillo, T. Martín, V.S. Martín, Tetrahedron Lett. 46 (2005)
2829–2832.
[45] A. Jakob, V. Valtchev, M. Soulard, D. Faye, Langmuir 25 (2009) 3549–3555.
[46] M. Dan-Hardi, C. Serre, T. Frot, L. Rozes, G. Maurin, C. Sanchez, G. Férey, J. Am.
Chem. Soc. 131 (2009) 10857–10859.
[20] T. Kawabata, T. Mizugaki, K. Ebitani, K. Kaneda, Tetrahedron Lett. 44 (2003)
9205–9208.