Generation of singlet molecular oxygen from H2O2 with molybdate-exchanged layered double hydroxides: Effects of catalyst composition and reaction conditions

Article abstract of DOI:10.1006/jcat.2000.3070

(Mg, Al)-layered double hydroxides (LDHs) with different Mg/Al ratios in the octahedral layer were prepared via the coprecipitation method and were exchanged with varying amounts of molybdate. The composition of the LDH supports and the state of the molyb

Full text of DOI:10.1006/jcat.2000.3070

Journal of Catalysis 197, 139–150 (2001)
doi:10.1006/jcat.2000.3070, available online at http://www.idealibrary.com on
Generation of Singlet Molecular Oxygen from H O with
2
2
Molybdate-Exchanged Layered Double Hydroxides: Effects
of Catalyst Composition and Reaction Conditions
�
�
F. M. P. R. van Laar, D. E. De Vos, F. Pierard,† A. Kirsch-De Mesmaeker,†
1,
�
L. Fiermans,‡ and P. A. Jacobs
�
Centre for Surface Chemistry and Catalysis, K. U. Leuven, Kardinaal Mercierlaan 92, B-3001 Heverlee, Belgium; †Physical Organic
Chemistry, CP160/08, Universit e´ Libre de Bruxelles, 50 Av. F.D. Roosevelt, B-1050 Brussels, Belgium; and ‡Department
of Solid State Sciences, Ghent University, Krijgslaan 281/S1, B-9000 Gent, Belgium
Received June 12, 2000; revised August 14, 2000; accepted August 28, 2000
The literature contains elaborate studies of photosensiti-
(
Mg, Al)-layered double hydroxides (LDHs) with different Mg/Al zation for the generation of singlet molecular oxygen. How-
ratios in the octahedral layer were prepared via the coprecipitation ever, due to the rather low stability of the “old generation”
method and were exchanged with varying amounts of molybdate.
The composition of the LDH supports and the state of the molyb-
date in these materials were studied with X-ray diffractometry,
X-ray photoelectron spectroscopy and Raman. As was proven by
dyes (such as methylene blue, rose bengal) under photoox-
idative conditions (3–7), there has been an intensive search
for more stable sensitizers like halogenated porphyrins (8,
9
) and for alternative “dark” systems. These dark systems
NIR luminescence, aqueous H2O2 is converted at the surface of
are based on the conversion of hydrogen peroxide into wa-
1
these catalysts into excited, singlet molecular dioxygen ( O2). The
1
1
ter and singlet molecular oxygen (10). O2 can be gener-
singlet dioxygen diffuses away from the O2 generating centers and
�
ated by stoichiometric reaction of H2O2 with ClO , but the
can perform selective oxygenations, such as endoperoxidations of
dienes, or hydroperoxidations of olefins. Catalyst composition and
reaction produces salt waste and the rate of this reaction
reaction conditions (temperature, solvent) have major effects on the is not easily controlled. Moreover, depending on the pH
1
catalyzed O2 generation and the subsequent substrate oxygenation. and on the substrate molecule, the use of the electrophilic
In order to maximize the efficiency of the H2O2 use and to limit the hypochlorite may lead to undesired halogenations. Early
epoxidation side reaction, it is advisable to limit the amount of ex-
reports from the 1970s mention that some transition met-
changed molybdate to � 0.2 mmol per g and to use an LDH with
Mg0.9Al0.1 composition in the octahedral layer. The latter material
provides the optimum basicity for the catalytic activity of the ex-
IV
VI
als, such as Ti and Mo , may catalytically generate one
molecule of singlet molecular oxygen from two molecules
hydrogen peroxide (11, 12). Later, screening of the periodic
table revealed four groups of inorganic compounds capa-
ble of generating singlet molecular oxygen from hydrogen
peroxide (13).
changed molybdate.
�c 2001 Academic Press
Key Words: molybdate; layered double hydroxides; hydrogen per-
oxide; singlet dioxygen; peroxidation.
Several studies have been devoted to the use of homo-
geneous molybdate catalysts. High alkalinity, with a pH
between 10 and 11, is required for a fast and efficient
singlet molecular oxygen generation. The reaction pro-
ceeds efficiently only in pure water, in aqueous MeOH
and EtOH. For reactions with highly apolar substrates,
INTRODUCTION
Selective oxyfunctionalization of organic compounds by
singlet molecular oxygen has received much attention since
the 1960s, because of the unique reactivity of O2 and its
1
1
a reverse microemulsion may be used, in which O2 is
highly selective reaction with electron-rich organic sub-
1
generated in the aqueous droplets, while the oxygenation
takes place in the continuous organic phase (14). How-
ever, the system has several drawbacks: if catalytic amounts
of molybdate are used, soluble base is needed to adjust
the pH and relatively large solvent volumes are used.
Only one truly immobilized molybdate system has been
reported by McGoran and Wyborney (15), with molyb-
date immobilization on an anion exchanging Amberlyst
resin.
strates such as olefins. Typical O2 products include enyl
hydroperoxides, dioxetanes and endoperoxides (1). Reac-
tions of O2 thus essentially differ in regio- and chemose-
lectivity from those of radical or electrophilic agents such
as RO or ROO radicals or (in)organic peracids (2).
1
�
�
1
To whom correspondence should be addressed. Fax: 32 16 32 1998.
E-mail: pierre.jacobs@agr.kuleuven.ac.be.
139
0021-9517/01 $35.00
Copyright �c 2001 by Academic Press
All rights of reproduction in any form reserved.

1
40
VAN LAAR ET AL.
We have recently reported on a molybdate-exchanged the pH to 10, and adding 1.5 g of LDH. For Mg0.7Al0.3–
layered double hydroxide (LDH), which is a heterogeneous MopTOS (MoO4 25% , pTOS 75% ) 4.5 mmol pTOS and
1
catalyst for conversion of H2O2 into O2 (16). The use of 1.125 mmol Na2MoO4 � 2H2O were used. The exchange is
�
such a heterogeneous catalyst allows working with catalytic typically performed at pH 10, for 18 h at 65 C, subsequently
amounts of molybdate without extra dissolved base, and for 18 h at room temperature, all under N2 atmosphere.
should allow more freedom in the choice of an appropriate
solvent for the substrate molecule. LDHs are simple inor-
ganic anion exchangers, which are analogous to the mineral
Catalyst Characterization
The lattice structure of the LDHs was determined us-
ing powder X-ray diffraction with a Siemens D 5000 matic
diffractometer and Ni-filtered CuK� radiation (8048 eV).
Hydrotalcite [Mg6Al2(OH)16](CO3) � 4H2O (17, 18). The
focus of the present work is to study the behavior of these
molybdate LDH catalysts as a function of their composi-
The surface composition of the LDH samples was ana-
tion and of the reaction conditions. A first, physicochemical
lyzed by X-ray photoelectron spectroscopy (XPS) with a
part, studies the changes of the catalyst when the Mg/Al ra-
Perkin Elmer PHI 5500 ESCA system, using monochro-
tio in the octahedral layer or the Mo content are varied. The
matic AlK� radiation (1486.6 eV). The integrated areas of
catalyst is studied with ex situ techniques as well as under
the O 1s, Mg 2s, Al 2s, S 2s, and Mo 3d3/2–Mo 3d5/2 lines
catalytic conditions. Particular attention is devoted to the
were used to calculate the atomic ratio of the elements. The
characterization of the basic properties of the host material
via several complementary methods (19, 20). Next, some
XPS binding energies were calibrated with the C 1s line at
2
86.4 eV as a reference. For samples containing sulfur, an
parameters that affect the catalytic properties of the LDH
catalyst are studied in test reactions. Important variables
are Mo content, Mg/Al ratio in the support, hydrophilic-
ity of the catalyst, temperature, and solvent. Relations are
sought between the properties of the LDH (e.g., basicity)
and reaction characteristics such as product selectivity. This
optimization leads to the design of improved reaction con-
ditions that are applied in the selective oxyfunctionalization
of organic molecules such as dienes.
overlap between the S 2s and Mo 3d5/2 lines is observed.
In these cases, deconvolution was performed by calculat-
ing the S 2s area from the S 2p area. Moreover, the Mo
3
p peak was used in order to check the quantification of
the Mo 3d lines. Bulk elemental analyses of Mg, Al, and
Mo were carried out by emission spectroscopy with induc-
tively coupled plasma (ICP), after dissolution of the LDH
samples in 14% HNO3. FT-Raman spectra were recorded
with a Bruker IFS 100 spectrometer; samples were ana-
lyzed using a CW Nd : YAG laser at 150–450 mW. Scan-
ning electron micrographs (SEM) were taken on a Philips
XL 33.
EXPERIMENTAL
Catalyst Preparation
UV–vis diffuse reflectance spectra (DRS) were recorded
on a Cary 5 spectrophotometer, using anhydrous BaSO4 as
the standard over the whole spectral range. Near infrared
Layered double hydroxides are easily synthesized by co-
precipitation at low supersaturation. All syntheses and ex-
changes are performed, according to literature procedures
at a constant pH of 10 � 0.5 and under nitrogen atmo-
sphere to avoid CO2 contamination (17). In order to obtain
MgxAl(1� x)A(1� x)(OH)2 � mH2O, with A a monovalent an-
(
NIR) luminescence (IRL) spectra were recorded with a
liquid N2 cooled Ge detector of North Coast (EO 817-L)
connected with a lock-in amplifier and muon filter. The Ge
detector was attached to an Edinburgh fluorimeter (CD
900). Experiments to determine the basic properties of the
layered double hydroxides with the use of indicators were
performed as follows: 0.15 g LDH was vacuum dried at
room temperature for 24 h and was mixed with 2 ml indica-
tor solution (0.1 mg bromothymol blue, phenolphthalein or
alizarin per ml dry toluene). Suspensions containing bro-
mothymol blue and phenolphthalein were titrated with a
�
�
ion such as Cl or NO , equal volumes of decarbonated so-
3
lutions of MgCl2 � 6H2O or Mg(NO3)2 � 6H2O (xM, 200 ml)
and AlCl3 � 6H2O or Al(NO3)3 � 9H2O ((1� x) M, 200 ml)
are slowly added to 80 ml H2O. To maintain the pH at 10,
addition of 1.5 M NaOH is required. The LDH is obtained
after 18 h aging at room temperature, centrifuged, washed,
and freeze–dried. Overnight exchange of 1.5 g LDH in
1
00 ml 2.8–11.2 mM solutions of Na2MoO4 � 2H2O (pH 10,
0
.01 M solution of benzoic acid in toluene. Samples with ad-
room temperature, N2 atmosphere) yields the hydrophilic
molybdate exchanged Mg, Al-LDH materials (LDH-Mo).
In a similar procedure, the organic anion p-tolu-
enesulfonate (pTOS) is coexchanged with molybdate on
the LDH to yield a more hydrophobic molybdate catalyst,
further denoted as LDH-MopTOS. Mg0.7Al0.3–MopTOS
sorbed alizarin were filtered, vacuum dried, and analyzed
with DRS.
Catalytic Reactions and Characterization
of Reaction Products
(
MoO4 12.5% , pTOS 87.5% of the anion exchange capac-
Hydrogen peroxide concentrations were derived from
IV
ity) is obtained by mixing 9 mmol pTOS and 0.5625 mmol cerimetrical titrations, using a 0.1 N Ce solution. Samples
Na2MoO4 � 2H2O in 150 ml decarbonated water, correcting were acidified with HCl prior to titration.

1
CATALYST COMPOSITION EFFECTS IN LDH-Mo CATALYZED O2 GENERATION
141
Reaction mixtures were analyzed on a GC equipped with
TABLE 1
a 50-m CP-Sil 5 column. Samples were taken after complete
hydrogen peroxide conversion (catalyst becomes white
again). Mixtures containing enyl hydroperoxides were re-
duced with an excess triphenylphosphine (PPh3) prior to
GC analysis; the (exothermic) reduction is complete after a
few minutes. Product identity was checked with GC-MS on
a MD 800 mass spectrometer of Fisons in combination with
known fragmentation patterns(21) or with the use ofphoto-
sensitization. Characteristic fragment ions of some impor-
tant products are endoperoxide from 1,3-cyclohexadiene,
m/z = 112, 83, 80, 68, 55; endoperoxide from 2,3-dimethyl-
Layered Double Hydroxide Supports: Chemical Composition
and XRD Characterization
x = Al/(Al + Mg)
Sample,
nominal
composition
XRD
analysis
phase
Lattice
Bulk, Surface,
parameter,
a = 2d[110] ( A˚ )
ICP
XPS
Mg0.7Al0.3–LDH
Mg0.8Al0.2–LDH
Mg0.9Al0.1–LDH
0.32
0.22
0.13
0.32
0.28
0.17
Hydrotalcite
Hydrotalcite
Hydrotalcite +
Brucite
3.050
3.080
3.094
1
,3-butadiene, m/z = 114, 96, 85, 82, 67; endoperoxide from
of the samples decreases from its initial value for Brucite
a = 3.146 A˚ ) as the Al content increases (Table 1). This
contraction of the hydrotalcite unit cell can be rationalized
cis,cis-1,3-cyclooctadiene, m/z = 140, 122, 111, 97, 79; as-
(
caridole (endoperoxide from �-terpinene), m/z = 168, 136,
1
21, 93, 79.
Products
Products were purchased in the highest purities available
2
+
by the isomorphic substitution of the larger Mg cation by
3
+
2+
the smaller Al cation (ionic radius of Mg = 65 pm and
of Al3 = 50 pm) (22).
+
For the samples with an Al content lower than 30% , XPS
detects a surface enrichment of Al. This seems in line with
a gradual phase separation as the composition of the ma-
terial deviates from the limiting hydrotalcite composition
and used asobtained:bromothymolblue, hydrochloricacid,
and dimethylformamide (DMF) from Merck, phenolph-
thalein and AlCl3 � 6H2O from UCB, alizarin, hydrogen per-
oxide (35% in H2O), Ce(SO4)2 � 4H2O, Mg(NO3)2 � 6H2O,
NaOH, p-toluenesulfonic acid, 1-methyl-1-cyclo-hexene,
(
Mg0.75Al0.25–LDH). As indicated by the XRD data, brucite
is formed as a separate phase; the XPS data suggest that
for Mg0.8Al0.2–LDH and Mg0.9Al0.1–LDH an Al-rich LDH
phase is formed on top of the brucite phase, resulting in an
apparent surface enrichment with Al.
As shown in Fig. 1, the difference between a hydrophilic
LDH-Mo and an expanded, hydrophobic LDH-MopTOS
can easily be investigated with XRD. The decrease of the
� value of the d003 line corresponds to an increase of the
basal spacing from 7.6 A˚ up to 17.25 A˚ for the pTOS pil-
lared LDH. This value is in good agreement with the value
�
-terpinene, 1,3-cyclohexadiene, 1,4-dioxane, and tetra-
hydrofuran (THF) from Acros, benzoic acid, Na2MoO4 �
H2O, (NH4)6Mo7O24 � 4H2O, 2,3-dimethyl-2-butene, 2,3-
dimethyl-1,3-butadiene, cis,cis-1,3-cyclooctadiene, and (-)-
2
�
-citronellol from Aldrich, Al(NO3)3 � 9H2O, bengal rose,
and 2-methyl-2-pentene from Fluka, toluene, acetonitrile,
and acetone from Biosolve LTD, MgCl2 � 6H2O from VEL,
ethanol from BDH, and methanol and 2-propanol from
Riedel-de-Ha e¨ n.
2
RESULTS AND DISCUSSION
Structural Characterization of the Layered
Double Hydroxides
The nominal chemical composition, the bulk and sur-
face composition, and the crystalline phases in the as-
synthesized layered double hydroxides with varying Mg/Al
ratio are shown in Table 1. For all samples, the XRD pat-
terns clearly show an LDH crystalline phase, with Hydrotal-
cite structure (Fig. 1). When the aluminum content is low
(
10% of the octahedrally coordinated atoms; Mg0.9Al0.1-
LDH), a Brucite phase (Mg(OH)2) is detected, with char-
�
acteristic reflections at 18.5, 38, 50.8, 58.8, and 62.2 2�.
Further lowering the Al content of the support is not rel-
evant, since the specific anion exchange capacity of these
mixed hydrotalcite–brucite samples becomes quite small.
The broad diffraction lines of the layered double hydrox-
ides indicate the presence of rather small crystallites, which
was confirmed with scanning electron micrographs. Crystal-
FIG. 1. XRD patterns of (bottom) hydrophilic Mg0.7Al0.3–LDH-Mo
2�
MoO , 25% of AEC; 75% NO ) and (top) Mg0.7Al0.3–LDH-MopTOS
4
�
(
3
2
�
lite size is estimated at 100–150 nm. The a lattice parameter (MoO , 25% of AEC; 75% pTos).
4

1
42
VAN LAAR ET AL.
TABLE 2
2
4
�
2�
Mo and S Content of MoO or MoO4 , pTOS Exchanged Layered Double Hydroxides
(Mg0.7Al0.3–LDH) as Determined by Bulk Analysis or XPS
Mo/Al ratio
S/Al ratio
(XPS)
Nominal composition of AEC (% )
Bulk (ICP)
Surface (XPS)
2
4
�
�
�
�
1
2
3
4
5
6
12.5% MoO
25% MoO
87.5% NO
75% NO3
0.064
0.104
0.143
0.255
—
0.053
0.105
0.138
0.19
0.07
0.104
—
—
—
—
0.86
0.79
3
2
4
�
�
2
4
�
37.5% MoO
62.5% NO
3
2
4
�
�
50% MoO
50% NO3
2
4
12.5% MoO
87.5% pTOS
75% pTOS
2
4
�
25% MoO
—
calculated in the assumption that the anions are perpendic- cannot be excluded that the Raman observation of a minor
ularly oriented between the layers (23). Such a perpendic- amount of polymerized Mo is due to local overheating of
ular orientation is probable given the high surface charge the sample in the laser beam. Anyhow, in view of the very
of the layered double hydroxides. Moreover, the XPS data limited changes in d-spacing in the X-ray diffractograms, it
Table 2, entries 5 and 6) indicate that apart from molyb- is excluded that the major part of Mo is present as Mo7O6
�
,
(
2
4
date, only para-toluenesulfonate is present as a counter an- since the limited interlayer spacing does not allow stacking
ion; in these LDH-MopTOS, which were prepared starting of these bulky Mo7O6 anions between the LDH sheets.
�
2
4
from a chloride exchanged LDH, no Cl could be detected
by XPS.
Summarizing, both XPS and Raman convincingly prove
that, even for samples with a Mo content up to 50% of the
anion exchange capacity, the major part of molybdenum is
present as non-polymerized MoO2 . This is in accordance
�
4
Characterization of Immobilized Molybdenum
with the conditions of the molybdate ion exchanges, which
In order to properly evaluate the catalytic data, it is im- were performed at a pH of 10; note that heptamolybdate
portant to study the state of the exchanged Mo, particu- is rather formed and stable in more acidic solutions (pH
larly when the molybdate concentration on the catalyst is about 4).
increased. Comparison of XPS and bulk elemental analy-
sis demonstrates that molybdenum is homogeneously dis-
persed throughout the support (Table 2, entries 1–4); there
Catalyst Characterization in Reaction Conditions
is no evidence for accumulation of Mo as clusters on the
Time-resolved UV–vis diffuse reflectance spectroscopy
outer surface. The XPS Mo 3d3/2 and Mo 3d5/2 lines are (DRS) in combination with infrared luminescence (IRL)
located at 235.50 and 232.35 eV, respectively, which is char- can be used to investigate respectively the formation of the
acteristic for MoO2 (24). No other Mo lines, such as of
�
1
O2 precursors and of singlet molecular oxygen itself. In
4
1
heptamolybdate (Mo 3d3/2 and Mo 3d5/2 located at 237.2 molybdenum-catalyzed reactions, O2 can be released from
and 234.3 eV) were observed. di-, tri-, and tetraperoxomonomolybdate species (27–30).
As the molybdenum-oxygen vibrations in the infrared While the triperoxomolybdate MoO(O ) ishard to detect
region (800–1000 cm ) severely overlap with the absorp- with UV–vis spectroscopy, Mo-NMR/kinetic experiments
tions of the support, Raman spectroscopy was applied. All on concentrated molybdate solutions indicate that partic-
molybdate-exchanged layered double hydroxides show a ularly this triperoxo species easily forms singlet molecu-
2
�
2
3
�
1
95
�
1
1
clear vibration at 896 cm . The latter is assigned to the lar oxygen, while O2 release is slower from di- and tetra-
Mo–O symmetricalstretchingvibration ofMoO2 . Onlyfor peroxomolybdates. Exposure of the solid LDH-Mo catalyst
�
4
the sample with the highest molybdate loading (Mg0.7Al0.3– to a hydrogen peroxide solution results in a brick-red ma-
LDH exchanged with MoO2 , 50% of the AEC) a supple- terial. DRS study of this material shows the appearance
�
4
�
1
� 1
mentary band is observed at 940 cm . Based on literature, of two distinct new absorptions, one around 32,000 cm
�
1
this band may be identified as the most intense band of (310 nm) and the other around 22,000 cm (455 nm)
polymerized Mo species such as heptamolybdate (25, 26). (Fig. 2). Based on solution UV–vis data (31–33), these
Weaker bands, e.g., the Mo–O–Mo � and � vibrations of absorptions can be ascribed to the yellow diperoxomolyb-
Mo7O6 located at 564 and 219 cm respectively, could not date MoO2(O2) (310 nm) and the tetraperoxomolybdate
�
� 1
2�
2
4
2
clearly be discerned in the spectrum. The failure of XPS to Mo(O2) 24 (455 nm). The simultaneous presence of the
�
detect other species than MoO2 might be due to the rela- di- and tetraperoxomolybdate indicates that the triperoxo-
�
4
tively low sensitivity of the technique; on the other hand, it molybdate must be present as an intermediate species. The

1
CATALYST COMPOSITION EFFECTS IN LDH-Mo CATALYZED O2 GENERATION
143
FIG. 2. Diffuse reflectance spectra as a function of time of Mg0.7Al0.3–
LDH-MoO4 (AEC 37.5% ), wetted with a dilute H2O2 solution in 1,4-
dioxane, K-M, Kubelka-Munk.
1
FIG. 3. Infrared luminescence of O2 in the catalytic disproportiona-
tion of H2O2 by LDH-MoO4 (37.5% of AEC) in aqueous H2O2–dioxane.
Besides the observation of the polyperoxo precursors,
�
1
shoulder observed around 27,500 cm (365 nm) may ten-
tatively be assigned to this triperoxo species, in agreement
with the proposals of Aubry and Cs a´ nyi (29, 31). Hence, the
1
O2 can also be directly observed with IRL (infrared lu-
1
minescence). The near infrared detection of O2 is based
on the weak, spin forbidden relaxation of the excited, sin-
1
species that are known to generate O2 in solution, are also
1
1
glet molecular oxygen [ O2, 1g] to its ground state, triplet
present at the surface of the LDH-Mo catalyst. The interac-
tions of LDH-Mo with hydrogen peroxide are summarized
in Scheme 1.
3
3
�
molecular oxygen [ O2, 6 ] (27, 28). Figure 3 shows this
g
emission around 1276 nm for a LDH-Mo catalyst in a
dioxane–H2O2 solution.
Catalyst Heterogeneity
In order to verify the true heterogeneity of the cata-
lyst, an experiment was performed in which the catalyst
was repeatedly filtered off from a hydrogen peroxide con-
taining suspension and later resuspended into the solution
(
Fig. 4). After catalyst removal no further hydrogen perox-
ide conversion was monitored in time. Further evidence for
catalyst heterogeneity was obtained in an attempt to mea-
sure the 95Mo liquid NMR spectra for the molybdate or
peroxomolybdate species of the catalyst in contact with hy-
drogen peroxide. No Mo signal was detected from the sus-
pension or the clear supernatant even after extensive accu-
mulations. This suggests that Mo is confined to the support
and therefore not measurable in a liquid NMR experiment.
Finally, elemental analyses of reaction solutions showed
that molybdate concentrations in the reaction liquor were
SCHEME 1. Possible interactions of molybdate with H2O2 at the sur-
face of an LDH.

1
44
VAN LAAR ET AL.
prior to these XPS analyses, samples were calcined in or-
der to avoid interference from nitrate–oxygen atoms (34,
5). Such a combined approach overcomes the limitations
3
associated with each of the methods as such. Results are
summarized in Table 3.
Acid titration of the basic sites in the presence of a pH
indicator, such as bromothymol blue or phenolphthalein,
provides a first indication of increasing total basicity upon
decreasing aluminum content in Mg,Al-LDHs. Due to
stacking of the layered double hydroxides in toluene, it is
likely that only the external surfaces are titrated. Both the
number of weaker basic sites (as measured in the titration
with bromothymol blue) and the number of the stronger
base sites (measured with phenolphthalein as indicator) in-
crease steadily from Mg0.7Al0.3–LDH to Mg0.9Al0.1–LDH.
Measurement of the reflectance spectrum of LDH-
adsorbed alizarin (1,2-dihydroxyanthraquinone) confirms
this trend. Alizarin is red at pH 10.1 and turns to violet at
pH 12.1. The maximum absorption shifts bathochromically
with decreasing Al content, indicating a more pronounced
basicity at lower aluminum content. However, these mea-
surements may be influenced by electrostatic interactions
between the conjugated system of the dye and the charged
support.
FIG. 4. Heterogeneity test in the LDH-Mo catalyzed H2O2 decompo-
sition (H2O2 molarity as a function of time). The catalyst was repeatedly
removed from the reaction suspension by centrifugation and filtration (R)
and subsequently added again to the reaction filtrate (A). Reaction con-
ditions: 200 ml methanol, 450 mmol H2O2 (35 wt% in H2O), 30 mmol
�
�
-terpinene, 0.3 mmol MoO2 on Mg0.7Al0.3–LDH (AEC = 0.375), RT.
4
below the ICP detection limit (10 ppb), proving that less
than 0.05% of the total Mo comes to the solution.
Therefore, a supplementary analysis was performed by
measuring the XPS O 1s binding energy, which is directly
Basic Properties of the Layered Double Hydroxides
The literature shows that the catalytic properties of related to the basic strength of the oxygen atoms. A decreas-
homogeneous molybdate are largely dependent not only ing O 1s binding energy is related to increasing electron pair
on the hydrogen peroxide concentration, but also on the donation. As expected, the XPS analyses display the low-
pH. In this study the basic properties of the LDHs were est binding energy for the sample with the lowest aluminum
analyzed via three approaches: (i) in vacuo dried samples content. Finally, in order to probe the basic properties of
were mixed with a dye solution in toluene, followed by a the LDHs also in fully hydrated conditions, the pH values
benzoic acid titration (19, 20); (ii) alizarin, which is a pH of aqueous suspensions containing equal amounts of LDH
sensitive dye, was adsorbed on samples dried in vacuo, and were measured. The pH increases from 9.1 for Mg Al
0.7
0.3
its diffuse reflectance spectrum was measured; (iii) XPS was to 10.5 for Mg Al –LDH. Based on all these indications,
0.9
0.1
used to determine the binding energy of the O 1s electrons; the supports with the highest Mg content clearly provide
TABLE 3
Characterization of Basic Properties of the LDH Supports
mol basic sites per g LDHb
LDH
nominal
composition
pH of
suspension
in H2Oa
Bromothymol
Blue
(pKa = 7.1)
DRSc
Alizarin
(nm)
XPSd
BE
(eV)
Phenolphthalein
(pKa = 9.3)
�
�
�
4
4
4
� 4
Mg0.7Al0.3–LDH
Mg0.8Al0.2–LDH
Mg0.9Al0.1–LDH
9.1
9.7
10.5
1.33 � 10
1.80 � 10
2.74 � 10
0.73 � 10
518
539
549
531.4
531.2
531.1
� 4
1.03 � 10
� 4
1.83 � 10
a
Suspension of 1.5 g LDH in 100 ml deionized H2O.
b
0
.15 g vacuum dried LDH, suspended in 2 ml indicator solution (0.1 mg/ml), is titrated with 0.01 M
benzoic acid.
c
Wavelength of maximum absorption in diffuse reflectance spectroscopy: 0.15 g of vacuum dried
LDH is mixed with 2 ml alizarin in toluene solution (0.1 mg/ml). Prior to analyses, samples are vacuum
dried.
d
�
XPS analyses of calcined samples (calcination at 850 C).

1
CATALYST COMPOSITION EFFECTS IN LDH-Mo CATALYZED O2 GENERATION
145
1
the most basic environment to the molybdate anions and and H2O2/Mo ratio, at which the rate of O2 generation and
1
this may be reflected in the catalytic properties.
the yield of O2 products are maximal (29, 30, 38). If the pH
is too high (e.g., 12.5), H2O2 may disproportionate sponta-
neously, or the peroxomolybdates are hydrolyzed by excess
Catalytic Reactions: Effects of Catalyst Composition
and Reaction Conditions
�
OH . In a too large excess of H2O2, tetraperoxomolybdate
2
4
�
MoO(O ) is formed, and this complex only slowly re-
2
In this section, various factors that influence the sin-
glet oxygenation with the molybdate LDH catalysts are
investigated. Important output variables are (i) product
selectivity and (ii) product yield, which at full conversion
of H2O2 essentially reflects the efficiency of the oxidant
use. Product selectivity is of particular interest, in order to
1
1
leases O2 (Scheme 1). The best results for O2 oxygena-
tion are generally obtained at a pH of about 10 and with a
sufficient H2O2/Mo ratio to promote the formation of the
triperoxomolybdate MoO(O ) , which is the most labile
O2 precursor. When either the H2O2/Mo ratio or the pH
are lowered, epoxidation gradually predominates over O2
formation. In such conditions, the average number of per-
oxo groups per Mo is lower than three. In literature, this
epoxidation has been ascribed to the tetraperoxodimolyb-
date Mo2O3(O2)4 , which is in equilibrium with the diper-
oxomolybdate MoO2(O2)2 (39). A similar preference for
2�
2
3
1
1
1
discriminate between epoxidation, O2 oxygenation, and
radical-induced oxidations. In the latter context, the use of
1
-methyl-1-cyclohexene as a diagnostic substrate allows a
good discrimination between hydroperoxides formed via
2�
1
O2 or via radicals (36, 37). These enyl hydroperoxides are
2�
reduced with PPh3 after full consumption of the hydrogen
peroxide, and the GC analysis of the allylic alcohols 1 to 3
in Table 4 results in characteristic distribution patterns. Par-
ticularly the fraction of allylic alcohol 2 is important: it is
formed with at least 50% selectivity in a photosensitized re-
action (e.g., with bengal rose, Table 4, entry 1), but is almost
epoxidation is expected for other species with a low number
of peroxide groups per Mo; for instance, the monoperoxo-
molybdate MoO3(O2) cannot release O2, but may well
effect epoxidation (Scheme 1).
2�
1
1. Molybdate loading of the catalyst. Table 4 shows re-
absent in a free radical reaction (Table 4, entry 2). The dis- sults for different Mo concentrations on a fixed amount
1
crimination between epoxidation and O2 oxygenation can of LDH. The product distributions for the oxygenation of
be conveniently studied with either 2,3-dimethyl-2-butene 1-methyl-1-cyclohexene are in agreement with oxygena-
1
or 1-methyl-1-cyclohexene.
tion by O (entries 3–5); radical oxidation or epoxida-
2
For a good understanding of the results with the het- tion are negligible. At first sight, it may seem of interest to
erogeneous catalyst, it should be noted that for homoge- have as many active sites as possible, in order to speed up
neous MoO2 catalysis, there are optimum values of pH the reaction. However, Table 4 indicates that the highest
�
4
TABLE 4
Oxidation of 1-Methyl-1-Cyclohexene with H2O2 and LDH-Mo: Effect of Molybdate Loading and
Hydrophobicity/Hydrophilicity of the Catalyst
Product yields (% )a
Entry
Epoxide
AA 1
AA 2
AA 3
b
1.
2.
3.
4.
5.
6.
7.
Sensitization (Bengal Rose)
—
—
31
39
50
7
19
54
Radical oxidation (Fe /H2O2)c
37.5% AEC (28 �mol MoO4)
25.0% AEC (18 �mol MoO4)
12.5% AEC (10 �mol MoO4)
12.5% AEC MoO4, 87.5% AEC pTOS
25.0% AEC MoO4, 75.0% AEC pTOS
2
+
0.6
0.6
0.9
4.4
4.9
5.2
5.8
8.9
10.3
9.3
11.6
12.2
19.1
22.0
21.5
2.3
2.9
6.4
5.7
3.9
Note. Reaction conditions: 2.5 mmol 1-methyl-1-cyclohexene, 2 aliquots of 2.5 mmol H2O2 (35% in H2O), 0.05 g Mg0.7Al0.3–
�
LDH in 3 ml 1,4-dioxane, 20 C.
a
Product yields on substrate basis. AA, allylic alcohol. At complete H2O2 consumption, samples were taken and reduced
with PPh3, then analyzed with GC.
b
1
Reference reaction: O2 oxygenation.
c
Reference reaction: free radical oxidation (FeSO4 � 7H2O + H2O2). Italic numbers in entries 1 and 2 give the relative
ratios of the allylic alcohols.

1
46
VAN LAAR ET AL.
for complete H2O2 consumption is about 50% longer than
with a LDH-Mo with the same Mo content. The obvious
explanation is that with the hydrophobic catalyst, the sur-
face concentration of the hydrophobic substrate increases,
while the lower H2O2 concentration at and around the Mo
sites decreases the rate of H2O2 disproportionation. Sec-
ond, it is clear that use of pTOS as a charge-compensating
�
anion instead of NO3 increases the overall yield of oxy-
genated products (Table 4, entries 6 and 7 vs entries 4 and
5
), which means that H2O2 is used more efficiently. Finally,
to a large extent, the higher yield of oxygenated products
is due to an increased epoxide formation. This seems re-
lated to the lower surface H2O2 concentration, which favors
epoxidizing species, such as monoperoxo or possibly diper-
oxo species, over the triperoxomolybdate, which releases
SCHEME 2. Generation of singlet molecular oxygen by the catalytic
LDH-MoO4/H2O2 system and its potential fates: (A) physical quenching
1
O2 fastest (cf. Scheme 1).
1
by the solvent (Solv), (B) self-annihilation of two O2 molecules, (C) for-
mation of an exciplex with a substrate molecule (S), leading to physical
quenching or chemical reaction (peroxidation), (D) physical quenching of
O2 by the solid support.
3. Mg–Al ratio in the octahedral layer of the LDH. The
basic properties of the layered double hydroxide are of key
importance for the catalytic activity of the molybdenum
peroxo species. Adjusting the local pH can be achieved via
altering the molar ratio of Mg and Al in the LDH. The
physicochemical evidence presented in the previous sec-
tions proves that the alkalinity of the LDH-Mo catalysts
increases when the content of the more acidic component
1
eventual product yield at complete H2O2 consumption is
obtained with the same weight of LDH catalyst contain-
ing less Mo (entry 5). A possible explanation is that for
the LDH catalyst with high Mo loading, the high rate of
(
Al) is lowered.
1
O2 production causes a depletion of the rather hydropho-
Catalytictestswere performed with different Mg/Alcom-
bic substrate 1-methyl-1-cyclohexene around the surface,
positions, but keeping the weight percentage of Mo in the
catalyst constant (Table 5). The catalyst with the lowest
Al content gives the best catalytic results: the eventual hy-
droperoxide yield is highest, while a clear decrease in the
epoxide selectivity is observed in comparison with the more
Al-rich catalysts. As pointed out previously, epoxidizing
species, e.g., monoperoxomolybdate, are not only favored
at lower H2O2/Mo ratios, but also at lower pH. Hence it
1
1
occupied by the O2 generating centers. Since O2 has a
low lifetime in water-containing solutions, it may be deac-
tivated by solvent quenching (Scheme 2, A) or by quench-
ing on the solid material (Scheme 2, D) before it can react
with the substrate S to form an oxygenated product SO2
(
Scheme 2, C).
An alternative rationalization may be O2– O2 annihi-
1
1
lation (Scheme 2, B). This phenomenon is known for the
hypochlorite–hydrogen peroxide reaction, and arises when
1
truly large amounts of O2 are formed per unit of time and
TABLE 5
volume. A similar situation may arise for the LDH cata-
lyst with the highest Mo loading and O2 generation rate.
Oxidation of 2,3-Dimethyl-2-Butene with H2O2 and LDH–Mo at
Complete H2O2 Conversion: Effects of Varying the LDH Chemical
1
However, we did not succeed in detecting the 663.4 nm lu- Composition
minescence, resulting from the simultaneous decay of two
excited O2 molecules to the ground state (1). Summarizing,
Product yield (% )a
1
a rather low molybdate concentration in the catalyst (typi-
cally 0.2 mmol per g) seems most appropriate for efficient
1
oxygenation with the O2 generated from H2O2.
Enylhydro-
peroxide
2
. Hydrophobic versus hydrophilic catalysts. Exchange
Epoxide
of LDH catalysts with hydrophobic anions such as pTOS
is a straightforward way of increasing the hydrophobicity
of the catalyst and may even allow the use of such cata-
lysts in a biphasic emulsion in the absence of an organic
solvent. The use of catalysts containing both molybdate
and p-toluenesulfonate (LDH-MopTOS) was evaluated in
the reaction of 1-methyl-1-cyclohexene, in dioxane as a sol-
vent. First, with LDH-MopTOS, the reaction time needed
Mg0.7Al0.3–LDH (10 �mol MoO4)
Mg0.8Al0.2–LDH (10 �mol MoO4)
Mg0.9Al0.1–LDH (10 �mol MoO4)
Recylce 1 of Mg0.9Al0.1–sample
Recylce 2 of Mg0.9Al0.1–sample
4.7
2.9
2.3
2.4
2.3
19.6
32.4
47.9
46.8
47.1
Note. Reaction conditions: 2.5 mmol 2,3-dimethyl-2-butene, 2 aliquots
�
of 2.5 mmol H2O2 (35% in H2O), 0.05 g LDH in 3 ml 1,4-dioxane, 20 C.
a
Product yields on substrate basis, after complete H2O2 consumption.

1
CATALYST COMPOSITION EFFECTS IN LDH-Mo CATALYZED O2 GENERATION
147
�
seems that with Mg0.9Al0.1–LDH, which is the most basic
A clear reaction optimum is observed around 35 C. The
1
support, the formation of such species is sufficiently sup- decreasing yield of O2 oxygenates with increasing reaction
�
pressed to avoid significant epoxidation.
temperature from 35 C upward can be caused by several
It was not attempted to use a support with an even lower elements. First, the O2 lifetime decreases with increasing
aluminum content than 10% of the octahedral ions. First, environment temperature (40). Second, a too fast hydro-
more and more brucite will be present in the materials at gen peroxide conversion into O2 is likely to result in self-
higher Mgconcentration, and second, the decrease ofthe Al annihilation of O2 (Scheme 2, B), or in an oxygenation that
1
1
1
content also lowers the anion exchange capacity of the is limited by the depletion of the substrate around the cata-
LDH, which may ultimately lead to bad retention of the lyst particles. Use of hydrogen peroxide at more elevated
molybdate on the catalyst.
temperatures may entail risks of creating free radicals,
Catalyst stability and heterogeneity were confirmed for a particularly if traces of metals are mixed with peroxide so-
Mg0.9Al0.1–Mo catalyst in a recycling experiment. Both the lutions. However, the allylic alcohol distribution obtained
�
fresh catalyst and the recycled catalyst yield similar oxy- in the reaction of 1-methyl-1-cyclohexene at 50 C is still in
genation yields (Table 5).
1
full agreement with that of a true O2 reaction. At this point,
the reason for the decreasing oxygenate yields at tempera-
tures lower than 35 C is unclear. Anyway, in order to have
4
. Temperature. In addition to molybdate loading, ba-
�
sicity, and hydrophilicity/hydrophobicity of the catalyst,
temperature provides another parameter to control the
chemical reaction rate. Figure 5 presents for reactions at
optimum rate and oxygenate yield, a reaction temperature
�
of 35 C seems most appropriate.
1
Table 6 shows O2 oxidation reactions for some differ-
ent substrates. In these reactions, oxygenate yield and se-
lectivity for the O2 products are optimized via choice of
appropriate reaction conditions, namely 35 C, and a hy-
1
different temperatures the final O2-oxygenate yields from
1
-methyl-1-cyclohexene and 2,3-dimethyl-2-butene, upon
1
complete consumption of H2O2.
�
drophilic LDH support with a sufficiently alkaline charac-
ter (Mg0.9Al0.1) and with low Mo content (0.2 mmol per g).
Since only two equivalents of H2O2 were used per equiv-
alent of organic substrate, a 100% yield can be obtained
1
only if there is no physical quenching whatsoever of O2 by
the solvent, the substrate, or the support. Efficient endoper-
oxidations ([4� + 2�]-cycloadditions) are observed for 1,3-
dienes, with yields up to 77% . Higher yields are achieved
for more electron-rich substrates such as �-terpinene in
comparison with the unsubstituted 1,3-cyclohexadiene.
Molecules with an extended ring are known to be less sus-
1
ceptible to O2, which explains the decreased reaction yield
for 1,3-cyclooctadiene (1). The trends observed in the “ene
reactions” are similar to those observed in the endoper-
oxidations: higher peroxidation yields are obtained for the
more electron-rich alkenes such as 2,3-dimethyl-2-butene
than for the more electron-poor �-citronellol. Thus it seems
1
that the susceptibility of the molecules to O2 in these het-
erogeneously catalyzed reactions parallels their reactivity
in homogeneous conditions; there is no clear evidence for
�
an effect of the support on the reaction at 35 C between
the organic substrate and O2.
1
5
. Solvent effects. Table 7 shows rates for dispropor-
tionation of hydrogen peroxide by LDH-MoO4 in different
solvents. Unlike the substrate oxygenation, these dispro-
portionation rates cannot be affected by solvent quenching
1
of O2 and thus directly reflect the solvent effects on the
FIG. 5. Temperature effect on the oxidation of 1-methyl-1-cyclo- activity of the heterogeneous Mo catalyst. The observed
hexene or 2,3-dimethyl-2-butene. Yields of hydroperoxides, reduced to
solvent effects may originate in the formation of the perox-
1
allylic alcohols ( O2 yield) at complete consumption of H2O2 are plot-
omolybdate species, or in the subsequent disproportiona-
ted vs reaction temperature. Conditions: 2.5 mmol olefin, H2O2 (5 mmol,
1
2
4
�
tion of these species into O ; even differences between the
added in two portions), 0.05 g Mg0.9Al0.1–LDH-MoO4 (10 �mol MoO );
2
3
ml 1,4-dioxane.
dispersibility of the catalyst in the various solvents may be

1
48
VAN LAAR ET AL.
TABLE 6
1
Oxygenation of Various Substrates with O2 Generated by the LDH–Mo/H2O2
�
System at 35 C
Substrate
1O2 yield
Epoxide yield
<
1%
�
-Terpinene
77%
2
,3-Dimethyl-1,3-butadiene
62%
61%
5%
1%
1
,3-Cyclohexadiene
cis, cis-1,3-Cyclooctadiene
54%
66%
4%
4%
2
,3-Dimethyl-2-butene
25%
37%
62%
2
-Methyl-2-pentene
1%
1
8%
24%
50%
7%
1
-Methyl-1-cyclohexene
<1%
a
(
-)-�-Citronellol
rose oil 25%
1%
Note. Reaction conditions: 1.25 mmol substrate, 2 aliquots of 1.25 mmol H2O2 (35% in H2O),
2
�
�
.5 ml 1,4-dioxane, 0.025 g Mg0.9Al0.1–MoO4 (5 �mol MoO ), 35 C. Sampling after full hydro-
4
1
gen peroxide conversion, about 8 h.
a
Obtained after PPh3 reduction and acidification of the sample with HCl.
�
important. The rate is highest in water, but this solvent is as These reactions were performed at 35 C, with a catalyst
such obviously unsuitable for oxygenation of organic com- with a relatively small Mo concentration and a high Mg con-
pounds. Of the four other, H2O2 miscible organic solvents tent. For all reactions, the hydroperoxide (or allylic alcohol)
1
in Table 7, the rate is lowest for the reaction in methanol.
distribution patterns are in satisfactory agreement with O2
Table 8 describes a series of eight solvents, the yields and oxygenation. While the hydroperoxide selectivity is gen-
selectivities of the reactions with 1-methyl-1-cyclohexene. erally excellent, small but significant increases in epoxide

1
CATALYST COMPOSITION EFFECTS IN LDH-Mo CATALYZED O2 GENERATION
149
TABLE 7
vents does not reveal a clear correlation (1, 43). At least this
proves that with our optimized catalyst (e.g., low Mo con-
tent, high Mg content) and conditions (e.g., sufficient sub-
Rate Constant for Hydrogen Peroxide
Decomposition on a LDH–Mo Catalyst in
Different Solvents
1
strate concentration), major deactivation of O2 by the sol-
�
vent is avoided. A better correlation is obtained with the �
�
1
� 1
s
Solvent
Rate, M (mol Mo)
solvent parameter, the so-called “dipolarity–polarizability”
parameter. For instance, the highest hydroperoxide yields
are obtained in acetonitrile and DMF, which happen also
Acetone
,4-Dioxane
Ethanol
Methanol
Water
0.323
0.312
0.46
0.241
1.071
1
�
to have the highest � values of the solvents in Table 8.
This correlation may be rationalized in various ways. First,
it has been proposed that solvents with a high � value can
better stabilize the [substrate-O2] exciplex, which proba-
�
1
Note. Reaction conditions: 0.2 g Mg0.7Al0.3–
LDH-Mo (MoO : 37.5% of AEC), 50 mmol
H2O2 (35% in H2O), 25 ml solvent, RT.
2
4
�
bly has some charge transfer character, and from which the
oxygenated products are eventually formed (43, 44). In an
alternative hypothesis, an increasing solvent dipolarity may
selectivity are observed for methanol and acetonitrile. In improve the dispersibility of the hydrophilic catalyst in the
the case of acetonitrile, base-catalyzed epoxidation may oc- organic medium; in such a case, a positive effect of the sol-
cur via formation of a peroxyimidic acid from acetonitrile vent polarity on the oxygenate yield is expected as well.
and H2O2 (41, 42); however, the coproduct of such an oxi- Further subtle differences between the solvents may be as-
�
1
dation, acetamide, was not detected in the reaction mixture, cribed to a combination of � and O2 lifetime effects; for
which rules out that solvent oxidation is playing a role. In instance, in the comparison between acetonitrile and DMF,
the case of methanol, competition between the solvent and the lower lifetime in DMF is compensated by its higher �
�
H2O2 for coordination on the molybdate is a likely reason value, and eventually about the same oxygenate yields are
for the increased epoxide formation (2). Even if the red obtained. Overall, in view of the high hydroperoxide yield
color of the suspension of the catalyst in methanol qualita- and selectivity, and because its relatively low boiling point
tively indicates that even tetraperoxomolybdate species are facilitates workup, 1,4-dioxane is an excellent choice.
formed, the competing effect of methanol might increase
the fraction of lower peroxomolybdate species, which are
capable of epoxidation. A slight inhibitory effect of MeOH
in the formation of the peroxomolybdates also successfully
CONCLUSIONS
Molybdate exchanged LDHs are excellent and truly het-
1
accounts for the lower reaction rates in methanol (Table 7). erogeneous catalysts for conversion of H2O2 into O2. For
It was attempted to relate the variations of the allylic various Mo contents of the catalyst, and for different Mg/Al
alcohol yields in Table 8 to properties of the solvent. Com- ratios in the octahedral layer, Mo is predominantly present
1
parison with the known lifetimes (�1) of O2 in all these sol- as monomeric molybdate. In contact with H2O2, singlet
TABLE 8
Solvent Effects on the Oxidation of 1-Methyl-1-Cyclohexene by the LDH–Mo Catalyst, and Comparison
1
�
with the O2 Lifetime (�1) and the Solvent � Parameter
�
(�s)a
1
Yield AA,b
%
Yield
epoxide, %
AA 1b
%
AA 2b
%
AA 3b
%
�a
b
Solvent
�
Acetonitrile
DMF
Acetone
Dioxane
Methanol
THF
62
12
50
30
9.5
22
0.75
0.88
0.71
0.55
0.60
0.58
0.54
0.48
68
66
55
50
50
44
40
40
0.6
0.2
0.2
0.2
0.8
0.2
0.5
0.4
36
39
38
37
33
39
36
34
47
46
47
48
48
46
47
49
17
15
16
15
18
15
17
16
Ethanol
12
22.1
2
-propanol
Note. Reaction conditions: 1.25 mmol 1-methyl-1-cyclohexene, 2 aliqouts of 1.25 mol H2O2 (35% in H2O), 1.5 ml solvent,
2
�
�
.025 g Mg0.9Al0.1–MoO4 (5 �mol MoO ), 35 C. Sampling after complete hydrogen peroxide conversion. Samples reduced with
4
0
PPh3.
a
1
�
�1, O2 lifetime; � , solvent dipolarity-polarizability parameter, Ref. 23.
b
Yields are given on substrate basis. AA, allylic alcohol; see Table 4. The distribution over AA1, AA2 and AA3 is given in%
of the total amount AA.

1
50
VAN LAAR ET AL.
oxygen is formed at the surface, as is evidenced for instance
Guerzo, A., Pierard, F., and Kirsch-De Mesmaeker, A., Chem. Com-
mun. 267 (1998).
17. Cavani, F., Trifir o` , F., and Vaccari, A., Catal. Today 11, 173 (1991).
8. Drezdzon, M. A., U.S. patent 4774212 (1988).
9. Tanabe, K., Misono, M., Ono, Y., and Hattori, H., Stud. Surf. Sci. Catal.
1
by the NIR luminescence spectra. Next, O2 diffuses away
1
from the O2 generating centers, in order to react with di-
1
1
enes or alkenes. A high dipolarity/polarizability of the sol-
vent has a clearly positive effect on these oxygenations. In
51, (1989).
order to maximize the efficiency of H2O2 use, and in or- 20. Tanabe, K., “Solid Acids and Bases, Their Catalytic Properties.”
Kodansha International, Tokyo, 1970.
1. Ando, W., “Organic Peroxides.” Wiley, New York, 1992.
2. Miyata, S., Clays Clay Miner. 28, 50 (1980).
3. Drezdzon, M. A., Inorg. Chem. 27, 4628 (1988).
4. Wagner, C. D., Riggs, W. M., Davis, L. E., Moulder, J. F., and
Muilenberg, G. E., “Handbook of X-ray Photoelectron Spectroscopy.”
Perkin Elmer, Palo Alto, CA, 1979.
5. Gardner, E., and Pinnavaia, T. J., Appl. Catal. A 67, 65 (1998).
6. Niemantsverdriet, J. W., “Spectroscopy in Catalysis.” VCH, 1995.
7. Foote, C. S., and Niu, Q. J., Inorg. Chem. 31, 3472 (1992).
der to minimize the epoxidation, it is advisable to limit the
2
2
2
2
Mo content of the catalyst, to use a hydrophilic molybdate-
exchanged LDH, and to work with LDHs with a high Mg
content in the octahedral layer. As was proved by various
procedures to study the surface basicity, the latter provide
a more basic microenvironment to the catalytic molybdate
anions.
2
2
2
REFERENCES
28. B o¨ hme, K., and Brauer, H. D., Inorg. Chem., 31, 3468 (1992).
29. Nardello, V., Marko, J., Vermeersch, G., and Aubry, J. M., Inorg. Chem.
1
2
. Frimer, A. A., “Singlet Oxygen.” Vols. I–IV. CRC Press, Boca Raton,
FL, 1985.
. Sheldon, R. A., and Kochi, J. K., “Metal-Catalyzed Oxidations of
Organic Compounds.” Academic Press, San Diego, 1981.
. Midden, R. A., and Wang, S. Y., J. Am. Chem. Soc. 105, 4129 (1983).
. Blossey, E. C., Neckers, D. C., Thayer, A. L., and Schaap, A. P., J. Am. 32. Cs a´ nyi, L. J., Transition Met. Chem. 14, 298 (1989).
Chem. Soc. 95, 5820 (1973). 33. Cs a´ nyi, L. J., Transition Met. Chem. 15, 371 (1990).
. Schaap, A. P., Thayer, A. L., Blossey, E. C., and Neckers, D. C., J. Am. 34. Okamoto, Y., Ogawa, M., Maezawa, A., and Imanaka, T., J. Catal. 112,
34, 4950 (1995).
30. Nardello, V., Bouttemy, S., and Aubry, J. M., J. Mol. Cat. A 117, 439
(1997).
31. Cs a´ nyi, L. J., Horv a´ th, I., and Galb a´ cs, Z. M., Transition Met. Chem.
14, 90 (1989).
3
4
5
6
7
Chem. Soc. 97, 3741 (1975).
. Neckers, D. C., Blossey, E. C., and Schaap, A. P., U.S. patent 4315998
427 (1988).
35. Hattori, H., Chem. Rev. 95, 537 (1995).
(
1982).
36. van Laar, F. M. P. R., De Vos, D. E., Vanoppen, D. L., Pierard, F.,
Brodkorb, A., Kirsch-De Mesmaeker, A., and Jacobs, P. A., in “Pro-
ceedings, 12th International Zeolite Conference (M. M. J. Treacy,
B. K. Marcus, M. E. Bisher, and J. B. Higgins, Eds.) MRS (1999) 1213.
37. Pettit, T. L., and Fox, M. A., J. Phys. Chem. 90, 1353 (1986).
38. Aubry, J. M., and Cazin, B., Inorg. Chem. 27, 2013 (1988).
. Nowakowska, M., Sustar, E., and Guillet, J. E., Photochem. Photobiol.
A 80, 369 (1994).
. Quast, H., Dietz, T., and Witzel, A., Liebigs. Ann. 1495 (1995).
. DiMagno, S. G., Dussault, P. H., and Schultz, J. A., J. Am. Chem. Soc.
8
9
1
18, 5312 (1996).
1
0. Gekhman, A. E., Moiseeva, N. I., and Moiseev, I. I., Russ. Chem. Bull. 39. Campbell, N. J., Dengel, A. C., Edwards, C. J., and Griffith, W. P.,
4
4, 584 (1995).
J. Chem. Soc. Dalton Trans. 1203 (1989).
1
1
1. Bacon, R. G. R., and Kuan, L. C., Tetrahedron Lett. 3397 (1971).
2. Hayashi, Y., Shioi, S., Togami, M., and Sakan, T., Chem. Lett. 651 41. Payne, G. B., Tetrahedron 18, 763 (1962).
40. Schmidt, R., and Brauer, H.-D., J. Am. Chem. Soc. 109, 6976 (1987).
(
1973).
42. Ueno, S., Yamaguchi, K., Yoshida, K., Ebitani, K., and Kaneda, K.,
Chem. Commun. 295 (1998).
1
1
1
1
3. Aubry, J. M., J. Am. Chem. Soc. 107, 5844 (1985).
4. Aubry, J. M., and Bouttemy, S., J. Am. Chem. Soc. 119, 5286 (1997).
5. McGoran, E. C., and Wyborney, M., Tetrahedron Lett. 30, 783 (1989).
6. van Laar, F., De Vos, D., Vanoppen, D., Sels, B., Jacobs, P. A., Del 44. Clennan, E. L., Tetrahedron 47, 1343 (1991).
43. Aubry, J. M., Mandard-Cazin, B., Rougee, M., and Bensasson, R. V.,
J. Am. Chem. Soc. 117, 9159 (1995).