Grafting of molecularly ordered mesoporous phenylene-silica with molybdenum carbonyl complexes: Efficient heterogeneous catalysts for the epoxidation of olefins

Article abstract of DOI:10.1002/adsc.201000042

Arenetricarbonyl complexes, or the general formula -C₆H ₄Mo(CO)₃^-, were incorporated into crystal-like mesoporous phenylene-silica by liquid-phase deposition of molybdenum hexacarbonyl [Mo(CO)₆]. By adjusting the reaction conditions, different molybdenum loadings of 1.5 and 5.9 wt% were obtained, which correspond to 3% and 14% of the phenylene contents. The texture properties of the materials as well as the nature of the surface-fixed complexes were characterized by powder X-ray diffraction, transmission electron microscopy (TEM), N₂ adsorption, FT-IR, UV-vis and MAS (¹³C, ²⁹Si) NMR spectroscopy. The derivatized organosilicas were examined as catalyst precursors for the liquid-phase epoxidation of cis-cyclooctene, 1-octene, trans-2-octene and (R)-(+)-limonene at 55°C, using tertbutyl hydroperoxide as the oxidant. For each olefin the corresponding epoxide was the only product detected. In the case of cyclooctene, the intrinsic reaction rates per surface molybdenum atom were similar for both Mo loadings (TOF-1150 mol mol_Mo^-1 h^-1), suggesting that the resultant materials act as single site epoxidation catalysts. Leaching tests and metal analyses of reaction solutions showed that the catalytic activity stemmed from the immobilized species and not from the leaching of active species into solution. The oxidation of limonene gave limonene oxide as the only product in 95% yield at 3 h, which reveals an outstanding regioselectivity to the epoxidation of the endocyclic double bond.

Full text of DOI:10.1002/adsc.201000042

FULL PAPERS
DOI: 10.1002/adsc.201000042
Grafting of Molecularly Ordered Mesoporous Phenylene-Silica
with Molybdenum Carbonyl Complexes: Efficient Heterogeneous
Catalysts for the Epoxidation of Olefins
a
a
a
b
Ana C. Coelho, Salete S. Balula, Sofia M. Bruno, Jo¼o C. Alonso,
a,c
b
a
a
Nicolas Bion, Paula Ferreira, Martyn Pillinger, Anabela A. Valente,
Jo¼o Rocha, and Isabel S. GonÅalves *
a
a,
a
Department of Chemistry, CICECO, University of Aveiro, 3810-193 Aveiro, Portugal
Fax : (+351)-234-370084; e-mail: igoncalves@ua.pt
Department of Ceramics and Glass Engineering, CICECO, University of Aveiro, 3810-193 Aveiro, Portugal
Laboratoire de Catalyse en Chimie Organique - UMR6503, Avenue du Recteur Pineau, 86022 Poitiers Cedex, France
b
c
Received: January 18, 2010; Revised: June 9, 2010; Published online: July 2, 2010
Abstract: Arenetricarbonyl complexes, or the gener- the corresponding epoxide was the only product de-
al formula ꢀC H Mo(CO) ꢀ, were incorporated into tected. In the case of cyclooctene, the intrinsic reac-
6
4
3
crystal-like mesoporous phenylene-silica by liquid- tion rates per surface molybdenum atom were simi-
phase deposition of molybdenum hexacarbonyl lar for both Mo loadings (TOF~1150 mol
ꢀ
1
ꢀ1
[
Mo(CO) ]. By adjusting the reaction conditions, dif- molMo h ), suggesting that the resultant materials
6
ferent molybdenum loadings of 1.5 and 5.9 wt% act as single site epoxidation catalysts. Leaching tests
were obtained, which correspond to 3% and 14% of and metal analyses of reaction solutions showed that
the phenylene contents. The texture properties of the the catalytic activity stemmed from the immobilized
materials as well as the nature of the surface-fixed species and not from the leaching of active species
complexes were characterized by powder X-ray dif- into solution. The oxidation of limonene gave limo-
fraction, transmission electron microscopy (TEM), nene oxide as the only product in 95% yield at 3 h,
13
29
N adsorption, FT-IR, UV-vis and MAS ( C, Si) which reveals an outstanding regioselectivity to the
2
NMR spectroscopy. The derivatized organosilicas epoxidation of the endocyclic double bond.
were examined as catalyst precursors for the liquid-
phase epoxidation of cis-cyclooctene, 1-octene, trans- Keywords: epoxidation; heterogeneous catalysis; mo-
2
-octene and (R)-(+)-limonene at 558C, using tert- lybdenum; periodic mesoporous organosilicas; sup-
butyl hydroperoxide as the oxidant. For each olefin ported catalysts
[
7–10]
Introduction
work. In
a
few selected cases [R=C H ,
6 4
[7,13,14]
[
11,12]
(
C H ) ,
CH=CH,
and 2,6-naphthylene ), the self-organization
by the hydrolysis and condensation of bridged tri- of bis-silylated organic precursors has been exploited
alkoxysilanes of the type (R’O) SiꢀRꢀSi
(OR’) in the to prepare PMOs with molecular scale ordering of
CH=CHꢀC H ꢀCH=
6
4
2
6
4
[
15,16]
[17]
Periodic mesoporous organosilicas (PMOs) prepared CH,
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
ACHTUNGTRENNUNG
3
3
[
18,19]
presence of surfactant templates have a uniform dis- the organic groups within the pore walls.
The first
tribution of organic groups both within the pore walls report of an ordered mesoporous material possessing
[
1–6]
and on the pore channels surface.
Like the well a crystal-like periodic pore wall structure came from
known micelle-templated silicas, these materials typi- Inagaki et al., who used 1,4-bis(triethoxysilyl)benzene
cally exhibit well defined pore structures with narrow (BTEB) as the precursor to a phenylene-silica hybrid
[
7]
pore size distributions, high surface areas and high mesoporous material.
Powder X-ray diffraction
pore volumes. A large variety of PMOs bearing or- (XRD) and transmission electron microscopy (TEM)
ganic groups (e.g., ethylene, vinylene, ethynylene, showed that this material has a unique surface struc-
phenylene, biphenylene, thiophenylene and ferroceny- ture with alternating hydrophobic phenylene and hy-
[
4–6]
lene) have been reported.
Most of these materials drophilic silica layers with a periodicity of 7.6 ꢀ. The
have amorphous pore wall structures, that is, the or- pore diameter could be varied between 32 and 39 ꢀ
ganic groups are randomly arranged within the frame- by variation of the length of the hydrocarbon chain
Adv. Synth. Catal. 2010, 352, 1759 – 1769
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1759

Ana C. Coelho et al.
FULL PAPERS
(
C to C ) of the alkyltrimethylammonium surfactant ing via ligands, and examined as precursors to hetero-
14 18
[8]
used.
geneous catalysts for the epoxidation of olefins with
[
26]
PMOs with crystal-like pore walls offer new oppor- TBHP as the oxidant.
The tricarbonyl complexes
tunities for various kinds of modification by subse- undergo oxidative decarbonylation in situ with TBHP
quent derivatization. In particular, grafted guest spe- to give supported oxomolybdenum(VI) species such
cies for catalytic or optoelectronic applications may as Cp’MoO X and/or Cp’MoO(O )X.
2
2
[
10,20]
be spatially organized along the wall surface.
In the present work, phenylene-bridged PMO has
Thiel and co-workers reported that oxodiperoxomo- been functionalized with Mo(CO) moieties by con-
3
lybdenum complexes of the type (LꢀL)MoO(O )
tacting the organosilica with refluxing solutions of
heterogenized on phenylene-bridged mesoporous or- Mo(CO)6 in 1,2-dichloroethane. FT-IR, solid state
2
2
ganosilicas showed a large (up to 10-fold) increase in NMR, UV-vis, N adsorption, powder XRD and TEM
2
catalytic activity and a high stability in liquid-phase analyses have been carried out, and two materials
olefin epoxidation reactions, with aqueous tert-butyl with different Mo loadings have been examined as
hydroperoxide (TBHP) or H O as the oxidant, com- precursors to solid catalysts for the liquid-phase oxi-
2
2
pared to corresponding systems based on convention- dation of a variety of olefins at 558C, using TBHP as
[
21]
al ordered mesoporous silica (MCM-41) supports.
the oxidant.
This effect was attributed to the increased hydropho-
bicity of the framework walls due to the presence of
phenylene bridges in defined positions. The authors Results and Discussion
proposed that the unique hydrophobic pores could fa-
cilitate the adsorption of olefins close to the active Catalyst Preparation and Characterization
sites, and/or reduce the adsorption of the more polar
epoxide and by-products (tert-butyl alcohol or H O).
Phenylene-bridged PMO (PMO-ph) was prepared by
2
To date, only a few studies have addressed the the literature method using octadecyltrimethylammo-
post-synthetic chemical modification of the organic nium bromide and 1,4-bis(triethoxysilyl)benzene. The
groups in molecularly ordered PMOs. Bromination exposure of this material to Mo(CO) was carried out
6
experiments performed for vinylene- and phenylene- in two ways (Scheme 1). In the first (giving material
divinylene-bridged PMOs showed that up to 35% of 2a), PMO-ph was treated consecutively (ꢂ3) with re-
the organic groups were accessible for functionaliza- fluxing solutions of Mo(CO) in 1,2-dichloroethane
6
[14,16]
tion.
Crystal-like mesoporous phenylene-silica (DCE). After each 5 h reaction, the solution was fil-
can be sulfonated in one step (by treatment with tered off and replaced by a fresh one. In the second
fuming sulfuric acid or chlorosulfonic acid) or aminat- method (giving 2b), PMO-ph was treated once for
ed in two steps (by nitration with HNO -H SO fol- 24 h with a large excess of Mo(CO)6 in refluxing
3
2
4
[
7,22,23]
lowed by treatment with SnCl -HCl).
The DCE. Both 2a and 2b were repeatedly washed with
2
amino-functionalized material was found to be a reus- DCE and vacuum-dried to remove unreacted
able base catalyst for the Knçevenagel condensation Mo(CO) . Despite the fact that the overall excess of
6
[22]
ꢀ1
reaction.
Thiel and co-workers successfully pre- Mo(CO)6 used for 2a (2.84 mmolg ) was greater
ꢀ1
pared a bifunctional mesoporous organosilica catalyst, than that for 2b (2.28 mmolg ), ICP-AES analyses
in which sulfonic acid groups were located on hydro- revealed molybdenum loadings of 1.48 wt% for 2a
phobic phenylene layers and propylamine groups and 5.90 wt% for 2b, which correspond to ca. 3% and
were attached to hydrophilic silica layers. The co- 14% of the phenylene contents. The higher loading is
operative behaviour of the bifunctional catalyst was comparable with the 15% conversion obtained by Ka-
demonstrated by using it in the catalytic conversion megawa et al. after treatment of phenylene-bridged
[
23]
[
24]
of benzaldehyde dimethyl acetal into 2-nitrovinylben- PMO with Cr(CO) vapour at 1258C. As will be de-
6
zene, which involves an acid-catalyzed deprotection scribed below, nitrogen adsorption studies did not in-
step followed by a base-catalyzed nitroaldol reaction. dicate any pore blocking effects for 2a vs. 2b. The
Kamegawa et al. reported the incorporation of longer reaction time of 24 h for 2b is probably the
arene
Mo] by treatment of phenylene-bridged PMO with
Cr(CO) or Mo(CO) vapour, achieving 15% conver- ph, 2a and 2b. The pattern of PMO-ph shows one
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
tricarbonyl complexes [ꢀC H M(CO) ꢀ, M=Cr, main reason why a higher Mo loading was achieved.
6
4
3
Figure 1 shows the powder XRD patterns of PMO-
6
6
sion of the phenylene groups, but no catalytic results strong low-angle peak at 47.5 ꢀ, which is assigned to
[24]
were reported. Such materials are potentially inter- the (100) reflection of a two-dimensional hexagonal
esting as (pre)catalysts for a variety of organic trans- symmetry (p6mm) lattice with a lattice constant a=
[25]
formations. In particular, cyclopentadienylmolybde- 54.8 ꢀ. The broad, weak shoulder on this peak
num tricarbonyl complexes of the type Cp’Mo(CO) X toward higher angles (3–48 2q) comprises the overlap-
3
(
X=Cl, alkyl) have been immobilized in ordered ping (110) and (200) reflections. In addition to the
mesoporous silicas by either direct grafting or tether- low-angle peaks, three medium-range reflections at
1760
asc.wiley-vch.de
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 1759 – 1769

Grafting of Molecularly Ordered Mesoporous Phenylene-Silica with Molybdenum Carbonyl Complexes
Figure 1. Powder XRD patterns of (a) PMO-ph, (b) 2a and
c) 2b.
(
Scheme 1. Preparation of 2a and 2b.
7.6, 3.8 and 2.5 ꢀ are observed, which arise from the
molecular-scale periodicity in the PMO-ph pore walls
[
7,8]
along the channel direction.
After treatment of
PMO-ph with Mo(CO) , the four main reflections are
6
still observed at about the same 2q values, which indi-
cates retention of the mesoporous structure as well as
the molecular-scale periodicity. Compared with PMO-
ph, the (100) reflection is much weaker for 2a and Figure 2. Representative TEM image of material 2a taken
2
b, which could indicate a loss of structural order with the incident beam perpendicular to the channel axis.
with respect to the pore channels. Such a change was
not supported by TEM as shown by the representa-
tive image in Figure 2 in which it is possible to ob-
serve ordered mesochannel arrays running throughout silica matrix and the pore-filling material probably
the sample. The pore channels are straight and paral- makes a major contribution to the decrease in intensi-
[
28,29]
lel to each other. As discussed previously for Mo ty of the (100) reflection.
Concerning the
complexes incorporated to similar loadings within the medium-range reflections, the XRD pattern of 2a is
[
27]
channels of ordered mesoporous silicas, a reduction unchanged from that of PMO-ph, while for 2b there
in the X-ray scattering contrast between the organo- is a noticeable intensity decrease, which may be relat-
Adv. Synth. Catal. 2010, 352, 1759 – 1769
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1761

Ana C. Coelho et al.
FULL PAPERS
Figure 3. Nitrogen sorption isotherms at ꢀ1968C and pore
size distribution curves (inset) for PMO-ph (*), 2a (~) and
2b (&).
ed with the more extensive derivatization of phenyl-
ene groups.
PMO-ph exhibits a type IV nitrogen adsorption iso-
therm (measured at ꢀ1968C) based on the IUPAC
[30]
classification,
where the nitrogen adsorption in-
creases as the relative pressure values (p/p ) tend to
0
unity (Figure 3). A step appears in the relative pres-
sure range of 0.2–0.45, which may be attributed to ca-
pillary condensation in mesopores. The pore size dis-
tribution is relatively narrow and is centred around a
pore width (d) of about 40 ꢀ, suggesting that the mes-
oporous structure of the PMO is fairly uniform. These
features are preserved after the post-synthesis grafting
2
9
13
Figure 4. Solid-state Si (top) and C (bottom) CP MAS
NMR spectra for material 2a. * denotes spinning sideband.
treatments. The nitrogen uptake and S
decrease in
BET
2
ꢀ1
2
ꢀ1
[8]
the order PMO-ph (644 m g )> 2a (613 m g )> 2b ph (not shown), exhibiting peaks at ca. ꢀ80.5, ꢀ71.5
2
ꢀ1
3
2
1
(
538 m g ). The higher S
of 2a in relation to 2b is and ꢀ62 ppm attributed to T , T and T organosilica
m n
BET
consistent with the lower loading of metal complex in species [T =RSi ACTHUNGTRNEUN(G OSi) (OH) ], respectively. No Q
m 3-m
the former case. Based on the molybdenum loadings, signals corresponding to silicon atoms connected en-
3
ꢀ1
the Gurvitsch specific pore volume of 0.588 cm g
tirely to OSi and/or OH units were detected, which
3
(
p/p ffi0.98) for PMO-ph, and an estimate of 90 ꢀ for indicates that all silicon atoms are covalently bonded
0
the van der Waals volume of the fragment within the PMO network and no SiꢀC bond cleavage
[31]
13
Mo(CO)3, the specific pore volumes for 2a and 2b took place during the chemical modification. The
C
3
ꢀ1
are predicted to be 0.56 and 0.49 cm g , respectively, CP MAS NMR spectrum displays a main peak at
which are comparable to the Gurvitsch specific pore 136.8 ppm due to the benzene ring carbon atoms and
volumes calculated from the adsorption isotherms a very weak peak at ca. 224 ppm which, by reference
0.560 cm g for 2a and 0.458 cm g for 2b). Hence, to the solution C NMR spectrum of the model com-
it seems that the introduction of the metal complexes plex (h -C H CH )Mo(CO) (1) (d=221.17 ppm), is
3
ꢀ1
3
ꢀ1
13
(
6
6
5
3
3
did not cause significant (if any) pore blockage in the assigned to the CO carbon atoms of arenetricarbonyl
mesostructures.
complexes [ꢀC H Mo(CO) ꢀ].
6
4
3
29
13
Solid-state Si and C CP MAS NMR spectra for
Further evidence for the presence of (arene)-
material 2a are shown in Figure 4; similar spectra Mo(CO) moieties in materials 2a and 2b comes from
3
were obtained for 2b and are not shown here. The an analysis of the FT-IR (KBr) and diffuse reflectance
29
Si NMR spectrum is very similar to that for PMO- UV-vis spectra. Unmodified PMO-ph exhibits IR
1762
asc.wiley-vch.de
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 1759 – 1769

Grafting of Molecularly Ordered Mesoporous Phenylene-Silica with Molybdenum Carbonyl Complexes
ꢀ1
bands in the region 400–1700 cm very similar to
those described by Hoffmann et al. for a phenylene-
bridged PMO prepared using BTEB as the precur-
[32]
sor.
Toward higher frequencies the spectrum dis-
plays absorption bands assigned to combination
modes originating from the aromatic groups (1830
ꢀ
1
[33]
ꢀ1
and 1934 cm ), nCH (3019 and 3062 cm ) and n
OH
ꢀ
1
of hydrogen-bonded silanol groups (3200–3700 cm ).
The only important differences between this spectrum
and those for 2a and 2b are the appearance in the
latter of a fairly narrow peak at 1978 cm and a
broad peak at 1860–1920 cm (Figure 5). Kamegawa
et al. obtained almost identical results after treatment
ꢀ
1
ꢀ
1
of phenylene-bridged PMO with M(CO) vapour at
6
[
24]
either 758C (M=Mo) or 1258C (M=Cr). By refer-
ence to the spectrum of (C H )Cr(CO) physically ad-
6
6
3
sorbed on the surface of an ordered mesoporous silica
MCM-41), these authors assigned the narrow band
(
to A and the broad band to split (overlapping) E vi-
1
brational modes involving carbonyl groups of arene-
Figure 6. Diffuse reflectance UV-vis spectra of PMO-ph
tricarbonyl complexes [ꢀC H Cr(CO) ꢀ]. The spectra
6
4
3
6
AHCTUNGTRNENG(U - · - ·), (h -C H CH )Mo(CO) (1) (—), 2a (· · · · ·) and 2b
6 5 3 3
of 2a and 2b can be interpreted in the same way. A
relevant reference from the literature concerns a
hybrid organic-inorganic solid prepared by the hydro-
AHCTUNGTRENNUNG( - - - -).
6
lytic sol-gel polymerization of [h -1,4-C H (Si-
6
4
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(OMe) ) ]Cr(CO) , for which two n(CO) bands at broad band at roughly the same position for complex
3
2
3
ꢀ
1
[34]
1
896 and 1971 cm were observed.
1 may be due to a local transition within the phenyl
Figure 6 shows the diffuse reflectance UV-vis spec- ligand and/or charge transfer from molybdenum to
6
tra of PMO-ph, materials 2a and 2b, and (h - the carbonyl groups, in accordance with the calculated
6
C H CH )Mo(CO) (1). PMO-ph exhibits an intense and experimental electronic spectra of (h -
6
5
3
3
[
35]
band at ca. 270 nm, which is assigned to an aromatic C H )Cr(CO) . The broad absorption between 310
6
6
3
[
24]
p-p* transition of network phenylene moieties.
A
and 490 nm probably involves overlapping metal-to-
ligand charge transfer (from molybdenum to the ring
and, to a lesser extent, to the carbonyl groups) and
[
24,35,36]
ligand field bands.
For example, in addition to a
band at 286 nm, absorption bands at 326 and 370 nm
6
were
reported
for 36]the
complex
(h -
[
Me SiSiMe C H )Mo(CO) .
Materials 2a and 2b
3
2
6
5
3
show two bands at about 280 and 326 nm, and a
shoulder around 380 nm. Hence, as reported for
PMO-phCr(CO) prepared by chemical vapour depo-
3
[
24]
sition, the UV-vis spectra of 2a and 2b involve the
superimposed absorption of both PMO-ph and
AHCTUNGTNERN(GU arene)Mo(CO) , fully supporting the formation of
3
ꢀ
C H Mo(CO) ꢀ complexes.
6
4
3
Catalysis
In the presence of materials 2a and 2b, the liquid
phase epoxidation of cis-cyclooctene with anhydrous
TBHP at 558C without a co-solvent gave cyclooctene
oxide quantitatively in 30 min (Table 1); turnover fre-
quencies (TOFs, calculated at 5 min reaction) were
ꢀ
1
ꢀ1
ꢀ1 ꢀ1
1
116 mol molMo
Figure 5. FT-IR spectra in the range of 1700–2200 cm of for 2b. For comparison, the reaction in homogeneous
a) PMO-ph, (b) 2a and (c) 2b. phase using complex 1 (using the same Mo:olefin:oxi-
h
for 2a and 1152 mol molMo
h
ꢀ
1
(
Adv. Synth. Catal. 2010, 352, 1759 – 1769
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1763

Ana C. Coelho et al.
FULL PAPERS
Table 1. Olefin epoxidation with anhydrous TBHP in the
presence of complex 1, materials 2a and 2b, without a co-
solvent at 558C.
[a]
Sample
Olefin
Conversion [%]
5
15
30
60
180
1
cyclooctene
limonene
cyclooctene
limonene
100
–
93
–
30
5
96
–
–
–
99
–
56
7
–
–
99
–
89
92
22
–
–
100
–
95
100
37
–
98
100
82
84
13
100
83
2a
2
1
-octene
-octene
2b
cyclooctene
limonene
99
–
89
95
[
a]
Olefin conversion at different reaction times (min). The
selectivity to the corresponding epoxide (1,2-epoxide in
the case of limonene) was always 100%.
dant molar ratio) gave 100% conversion within 5 min.
Blank experiments performed without a catalyst or in
the presence of PMO-ph gave negligible conversion
Figure 7. Epoxidation of cis-cyclooctene with TBHP at
after 24 h, indicating that the active species contains 558C: anhydrous TBHP [1 (^), 2a (ꢂ), 2b (&)]; commercial
molybdenum. The absence of induction periods indi- TBHP (used as received) - complex 1 (* run 1, * run 2), 2a
cates that the active species are formed rapidly and/or (+) and 2b (~
are quite active. The very similar results for 2a and 2b
).
(
using the same Mo:olefin:oxidant molar ratio) show
that the intrinsic reaction rate per surface molybde- the presence of the grafted AlMCM materials slowed
num atom is similar for Mo loadings in the range of down markedly during the course of the reaction such
1.5–6 wt%, suggesting that the resultant materials act that 24 h were required to completely convert the
as single site epoxidation catalysts. From a practical substrate to the epoxide. This behaviour was attribut-
viewpoint this is important because it means that sim- ed to the interaction of the catalysts with the by-prod-
ilar catalytic results can be achieved using a much uct tert-butyl alcohol. In another example, catalysts
lower amount of the PMO-ph support (the amount of prepared by the immobilization of [(—)-menthylCp]-
2
b charged was about one quarter of that for 2a). Io- Mo(CO) Cl on AlMCM-41 and AlMCM-48 proved to
3
dometric titrations revealed no significant “non-pro- be quite active for cyclooctene epoxidation with
ductive” decomposition of TBHP at 30 min reaction TBHP at 558C (TOFs=360–420 mol molMo h ), but
ꢀ
1
ꢀ1
[38]
(
epoxide yield was 100%), i.e., the oxidant was essen- 1,2-cyclooctanediol was formed as a by-product.
tially consumed in the catalytic epoxidation of the
olefin.
The catalyst stability for 1 was examined by re-
charging the microreactor at 3 h with cis-cyclooctene
When the epoxidation of cyclooctene was carried and commercial TBHP (used as received) in amounts
out using commercial TBHP as received (<4 wt% equal to those used initially. As shown in Figure 7, the
water), the reaction was slower, although the epoxide initial reaction rate in the second run (TOF=603 mol
ꢀ
1
ꢀ1
selectivity was always 100% (Figure 7). TOFs for 2a molMo h ) is about three times higher than that in
ꢀ
1
ꢀ1
ꢀ1 ꢀ1
and 2b were 613 and 572 mol molMo h , and the ep- the first run (200 mol molMo h ), despite the fact
oxide yields at 30 min were 63%, 83% and 81% for 1, that in the second run the reaction mixture is more
2a and 2b. Hence, water seems to inhibit the reaction, dilute in catalyst and in substrate, and contains a
possibly by competing with the oxidant for coordina- large amount of tert-butyl alcohol (which could inhibit
tion to the metal centre. Overall the catalytic perfor- the reaction by competing with the oxidant for coor-
mance of the PMO-phMo(CO) materials compares dination to the metal centre). The main reason for
3
favourably with related examples from the literature. this behaviour must be that the oxidative decarbony-
For example, materials prepared by the direct grafting lation of the pre-catalyst 1 [to give a molybdenum-
of the pre-catalyst CpMo(CO) CH on ordered meso-
ACHTUNGTRENNU(GN VI) catalyst] is irreversible and reaches completion
3
3
porous aluminosilicas (AlMCM-41 and AlMCM-48) during the first run. Accordingly, the initial TOF for
exhibited similarly high levels of initial activity and run 1 is an underestimate since the number of moles
selectivity when tested for cyclooctene epoxidation of active species (formed by oxidative decarbonyla-
[
37]
with TBHP at 558C. However, the reaction rate in tion) is most likely less than that charged to the reac-
1764
asc.wiley-vch.de
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 1759 – 1769

Grafting of Molecularly Ordered Mesoporous Phenylene-Silica with Molybdenum Carbonyl Complexes
tion vessel. Based on the similar conversions at 3 h
for runs 1 (94%) and 2 (96%), it seems that the cata-
lytic system formed from 1 is fairly stable.
In order to test the stability and the catalytic reusa-
bility of material 2a, the catalyst was recycled for two
additional runs, applying cis-cyclooctene as the sub-
strate and commercial TBHP (used as received) as
the oxidant. After the first catalytic run, the solid was
separated from the reaction solution, washed several
times with n-hexane, dried at room temperature and
reused in another catalytic cycle. The conversions at
3
9
h were similar for all three runs (99%, 97% and
5% for runs 1, 2 and 3, respectively), and ICP-AES
analyses for Mo in the reaction solution indicated no
measurable loss of Mo during catalysis. Similar results
were observed for 2b: 93–100% epoxide yield in three
runs. Evidence for the loss of CO ligands from the
supported complexes during the first run was ob-
tained from the FT-IR spectra of the recovered solids,
which showed the disappearance of the n bands ini-
CO
Figure 8. Epoxidation of cis-cyclooctene with commercial
TBHP (used as received) at 558C using material 2a in the
presence of ethanol (ꢂ), n-decane (*), toluene (+), acetoni-
ꢀ
1
tially present at ca. 1900 and 1978 cm . An experi-
ment to further assess the heterogeneous nature of
the catalytic reaction was performed by removing the
solid from the reaction medium after 5 min by filter-
ing at the reaction temperature of 558C through a
trile ( ) and 1,2-dichloroethane (&) co-solvents.
~
0
.2 mm PVDF w/GMF Whatman membrane. The fil- leading to polarization/inductive effects on the activi-
trate was then tested for catalytic activity for a further ty/stability of the metal species.
h reaction time. No additional conversion was ob- The catalytic performance of material 2a was fur-
2
served during this period, compared with a further ther evaluated in the epoxidation of (R)-(+)-limo-
conversion of 45% when no filtration was performed. nene, trans-2-octene and 1-octene with anhydrous
These experiments prove that the catalytic activity TBHP at 558C, without a co-solvent. For the cyclic
stems from the surface-fixed species and not from the and linear C olefins, conversion at 30 min followed
8
leaching of active species into solution at the reaction the order: cis-cyclooctene (100%)>2-octene (84%)@
temperature.
1-octene (13%), and the corresponding epoxides were
The reactions discussed above were carried out the only products (Table 1). Hence, catalytic activity
without additional solvent. In order to assess the in- is higher in the epoxidation of olefins which are more
fluence of the solvent on the reaction of cis-cyclooc- substituted.
tene with commercial TBHP (used as received) at
The selective epoxidation of limonene is important
558C using material 2a, the reaction was performed in because limonene oxide (1,2-epoxy-p-menth-8-ene) is
the presence of 1,2-dichloroethane (DCE), toluene, n- a key raw material for the synthesis of fine chemicals,
decane, acetonitrile or ethanol (Figure 8). The epox- pharmaceuticals, fragrances, perfumes, food additives,
[
39,40]
ide was always the only reaction product. Initial and polymers.
Conventional manufacturing pro-
ꢀ
1
ꢀ1
TOFs (mol molMo h ) followed the order DCE cesses based on stoichiometric peracid (RCO H)
3
(
(
1075)>acetonitrile (820)ffitoluene (789)>n-decane routes are becoming increasingly unacceptable in the
217)>ethanol (26). No clear relationship exists be- context of green chemistry (since equivalent amounts
tween the dielectric properties of these solvents and of acid waste are produced), and can suffer from the
the catalytic activity, which may be at least partly be- non-selective formation of mono- and diepoxides as
[
41–43]
cause of the major differences in polarity between the well as cleaved products.
This has motivated the
olefin and the oxidant. The detrimental effect of etha- search for alternative methods based on catalytic ep-
nol on the reaction rate may be related with its polar oxidation using either homogeneous or (preferably)
[
39,41,43–45]
protic nature interfering in an analogous way to that heterogeneous catalysts.
The oxidation of (R)-
described above for tert-butyl alcohol. On the other (+)-limonene with anhydrous TBHP using materials
hand, the use of acetonitrile, an aprotic coordinating 2a or 2b as the heterogeneous pre-catalysts gave limo-
solvent, led to a quite reasonable epoxide yield of nene oxide in 95% yield at 3 h reaction and no by-
85% after 30 min. The co-solvent probably plays a products were detected, which reveals an outstanding
complex role in these reactions by influencing the ad- regioselectivity to the epoxidation of the endocyclic
sorption/diffusion of the reagents/products and by double bond (Table 1). Comparable results were ob-
Adv. Synth. Catal. 2010, 352, 1759 – 1769
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1765

Ana C. Coelho et al.
FULL PAPERS
Table 2. Comparison of results for the oxidation of (R)-(+)-limonene over solid catalysts.
[a]
Catalyst
Mcat:Lim:Ox
molar ratio
Temp.
[8C]
Time [h]
Conv.
[%]
Select.
[%]
By-products
Ref.
[b]
[c]
2
a or 2b
1:100:154
0.1:200:300
1:68:205
55
80
40
55
3
95
81
78
78
95
97
62
64
100
80
91
–
This work
[
[
[
46]
47]
48]
(
Mo)SiO2
24
24
6
24
24
8
8,9-epoxide, diepoxide
diepoxide+diol
–
diol
(
Mo)MCM-41
Mo-MCM-41
1:100:158
100
53 [
55
d]
[54]
[55]
[55]
Ti-SBA-15
Mn
Mn
1:30:47
80
25
25
100
–
[
d]
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(TPP)Cl-SBA-15
1:4688
1:750
95
100
8,9-epoxide
–
[
d]
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(salen)Cl-SBA-15
8
[
a]
(
Mo)SiO =microporous molybdosilicate; (Mo)MCM-41=Mo(VI)-containing MCM-41 mesoporous silica; Mo-MCM-
2
4
1=catalyst prepared by deposition of MoO Cl AHCTUNGTNERNU(G THF) onto MCM-41; Ti-SBA-15=catalyst prepared by deposition of
2 2 2
TiCl onto SBA-15 mesoporous silica; Mn ACHUTNGRENNU(G TPP)Cl-SBA-15 and Mn ACHTUNGTRENUNNG( salen)Cl-SBA-15=catalysts prepared by grafting
4
manganese complexes of tetraphenylporphyrin or salen ligands onto organo-functionalized SBA-15.
[
[
[
b]
c]
d]
Metal:limonene:TBHP molar ratio.
Selectivity to 1,2-epoxide (limonene oxide).
Oxidant=air.
tained for complex 1. When TBHP was used as re- sibly because cyclooctene is more reactive than limo-
ceived and not dried prior to use, the limonene oxide nene (Table 1), thereby masking the “catalyst deacti-
yields at 3 h remained very high (at least 95% for 2a vation” phenomenon. Possibly, less active/inactive
and 2b). These results are superior to those previously metal species are gradually formed. An approach to
reported for solid catalysts based on molybdenum- compensate catalyst deactivation is to increase the re-
[
46–48]
containing silicas or silicates (Table 2).
Even if action time (epoxide yield at 7 h for runs 3, 4 and 5
we widen the scope of the comparison to include was 84%, 73% and 64%, respectively).
other types of solid catalysts, the catalytic perfor-
mance of 2a and 2b for the epoxidation of limonene
stands up very well. For example, various types of ti- Conclusions
tanium-containing silicas have been tested as catalysts
for limonene oxidation with oxidants such as H O , We have shown that the treatment of phenylene-
2
2
TBHP and cumeme hydroperoxide, but mixtures of bridged periodic mesoporous silica with solutions of
products (monoepoxides, diepoxide, carveol and/or Mo(CO) allows arene molybdenum tricarbonyl com-
6
[
49–53]
carvone, diol) tend to be formed.
One exception plexes to be incorporated at loadings that depend on
concerns titanium-containing SBA-15 catalysts, which the reaction conditions. These materials give out-
were 100% selective to epoxide formation for all standing epoxide selectivity in the liquid-phase cata-
three of the above oxidants (Table 2); however, a rel- lytic reaction of olefins with TBHP, under quite mild
atively high reaction temperature of 808C and a reac- reaction conditions (558C and approximately atmos-
tion time of 24 h were required to reach an epoxide pheric pressure). No metal leaching was detected and
[
54]
yield of almost 100% (with TBHP as the oxidant).
the catalytic reaction is heterogeneous in nature. The
The selective aerial oxidation of limonene at ambient epoxidation of (R)-(+)-limonene in the absence of a
temperature in the presence of manganese salen or co-solvent gave 95% limonene oxide yield at 3 h. In
porphyrin complexes grafted on organo-functional- catalyst recycling experiments carried out with cyclo-
ized SBA-15 was recently reported by Srinivas and octene as the substrate, quite similar epoxide yields
[55]
co-workers, with the best results being 95–100% li- were obtained at 3 h in consecutive runs. For limo-
monene oxide selectivity at 62–64% conversion (8 h, nene, slowing down of the reaction was observed in
Table 2).
recycling runs, although conversion continued to in-
Additional catalyst recycling tests were performed crease with time, without changes in product selectivi-
for 2b in the epoxidation of limonene with anhydrous ty. The difficulty in characterizing the surface metal
TBHP. The conversion of limonene at 3 h decreased species (due to the low metal loadings and/or intense/
in recycling runs without affecting product selectivity overlapping signals arising from the PMO support)
(
95, 80, 69, 57 and 48% at 3 h in runs 1–5). These re- makes the understanding of catalyst deactivation phe-
sults contrast with those for cyclooctene as substrate nomena quite challenging. In the future, EXAFS and
which gave very similar yields in recycling runs), pos- XPS techniques may be helpful in this respect.
(
1766
asc.wiley-vch.de
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 1759 – 1769

Grafting of Molecularly Ordered Mesoporous Phenylene-Silica with Molybdenum Carbonyl Complexes
Experimental Section
882 (m), 825 (w), 791 (m), 778 (m), 617 (s), 590 (s), 530 (m),
01 (s), 480 (m), 460 (m), 424 (m), 352 (w), 330 (w), 301
5
ꢀ
1
1
(
w), 269 cm (s); H NMR (300.13 MHz, 258C, CDCl ): d=
Characterization Methods
3
5
.73 (t, 2H, arene), 5.44 (dd, 2H, arene), 5.37 (tt, 1H,
1
3
Microanalyses for C and H, and ICP-AES determination of
Mo were performed at the University of Aveiro. Powder
XRD data were collected on a Philips X’Pert MPD diffrac-
tometer (Cu-Ka X-radiation, l=1.54060 ꢀ) fitted with a
graphite monochromator and a flat plate sample holder, in a
Bragg–Brentano para-focusing optics configuration. Samples
were step-scanned in 0.028 2q steps with a counting time of
arene), 2.25 (s, 3H, CH ); C NMR (75.47 MHz, 258C,
3
CDCl ): d=221.17 (CO), 113.25 (C CH ), 96.19 (arene),
ꢀ
3
3
1
3
9
3.96 (arene), 90.89 (arene), 21.10 (CH ); C CP MAS
3
NMR: d=219.1 (CO), 116.3 (C CH ), 99.6 (arene), 97.7
ꢀ
3
(
arene), 22.2 (CH ).
3
2
s per step. TEM images were recorded on a 300 eV Hita-
chi H9000-NA instrument. The BET specific surface areas
SBET, p/p in the range 0.03–0.13) were calculated on the
PMO-phMo(CO) (2)
3
(
Prior to derivatization, physisorbed water was removed
from PMO-ph by heating the material at 1208C under re-
duced pressure for 2 h.
0
basis of nitrogen adsorption (at ꢀ1968C) data as described
[56]
previously.
The pore size distributions were determined
from the adsorption branches of the isotherms using the
Barret, Joyner and Halenda (BJH) model with the modified
Kelvin equation and a correction for the statistical film
2a: A solution of Mo(CO) (0.50 g, 1.89 mmol) in 1,2-di-
6
chloroethane (10 mL) was added to a suspension of PMO-
ph (2.0 g) in 1,2-dichloroethane (35 mL). After refluxing the
mixture for 5 h, the solution was filtered off and the solid
washed with CH Cl (4ꢂ30 mL), and vacuum-dried (first
[57,58]
thickness on the pore walls.
The statistical film thickness
was calculated using the Harkins–Jura equation. FT-IR spec-
tra of KBr pellets were obtained on an FT-IR Mattson-7000
2
2
cycle). A further two cycles were performed using freshly
ꢀ
1
infrared spectrophotometer with 2 cm resolution. Diffuse
reflectance UV-vis spectra were measured on a JASCO V-
prepared solutions of Mo(CO) in 1,2-dichloroethane. Ele-
6
mental analysis found: C 35.30, H 3.01, Mo 1.48%; IR
(KBr): n=3449 (s), 3061 (m), 3016 (m), 1979 (w), 1933 (w),
1631 (m), 1384 (m), 1161 (vs), 1072 (vs), 1021 (s), 922 (m),
560 instrument using MgO as a reference. Solid-state magic-
angle spinning (MAS) NMR spectra were recorded at
2
9
13
ꢀ1
7
9.49 MHz for Si and 125.76 MHz for C on Bruker
809 (m), 755 (m), 559 (s), 520 (s), 434 (m), 388 cm (m).
2
9
Avance 400 and 500 spectrometers, respectively. Si cross-
2b: A solution of Mo(CO) (0.15 g, 0.57 mmol) in 1,2-di-
6
polarization (CP) MAS NMR spectra were recorded with a
chloroethane (15 mL) was added to a suspension of PMO-
ph (0.25 g) in 1,2-dichloroethane (15 mL). After refluxing
the mixture for 24 h, the solution was filtered off and the
solid washed with CH Cl (2ꢂ25 mL), and vacuum-dried.
1
4
5
.0 ms H 908 pulse, a contact time of 8 ms, a spinning rate of
1
3
kHz and 5 s recycle delays. C CP MAS NMR spectra
1
were recorded with 4.5 ms H 908 pulses, a contact time of
ms, a spinning rate of 7.0 kHz and 4 s recycle delays.
2
2
2
Elemental analysis found: C 32.08, H 3.34, Mo 5.90%; IR
KBr): n=3424 (s), 3061 (m), 3016 (m), 1977 (w), 1935 (w),
(
1
8
628 (m), 1384 (m), 1160 (vs), 1071 (vs), 1020 (s), 916 (m),
08 (m), 754 (m), 560 (s), 518 (s), 431 (m), 388 cm (m).
Materials
ꢀ
1
All air-sensitive operations were carried out using standard
Schlenk techniques under a nitrogen atmosphere. Octade-
cyltrimethylammonium bromide (ODTMA) and Mo(CO)6
were obtained from Aldrich and used as received. Solvents
were dried by standard procedures, distilled under nitrogen
and stored over 4 ꢀ molecular sieves. BTEB was prepared
by a Barbier–Grignard reaction and purified by distillation
under reduced pressure.
a crystal-like wall structure (PMO-ph) was prepared under
basic conditions using ODTMA as the structure directing
Catalytic Studies
The liquid-phase catalytic oxidations of a variety of olefins
[cis-cyclooctene, (R)-(+)-limonene, trans-2-octene and 1-
octene] were carried out at 558C, using 5–6M TBHP in
decane (Sigma–Aldrich, <4% water) as the oxidant. Unless
otherwise stated, the TBHP was dried using activated mo-
lecular sieves (4 ꢀ). Throughout the text this solution is re-
ferred to as anhydrous TBHP, while the undried solution is
referred to as commercial TBHP (used as received). All of
the reactions were carried out under air (autogeneous pres-
sure) and stirred magnetically in a closed borosilicate 5 mL
reaction vessel immersed in an oil bath thermostated at
[8,59]
The phenylene-silica PMO with
[8]
agent and BTEB as the silica source. The surfactant was
removed from the as-synthesized material by solvent extrac-
tion with an acidic solution. Powder XRD (hkl in parenthe-
ses): 2q (8)=1.86 (100 overlapping with 110 and 200),
p
11.60, 23.40, 35.40; d100 =47.5 ꢀ; a=2d100/ 3=54.8 ꢀ.
5
58C. Typically, 1% molar ratio of molybdenum catalyst/
6
substrate and a substrate/oxidant molar ratio of 0.65 were
used (assuming 5.5M TBHP, based on iodometric titration).
The reactions were carried out either with no co-solvent or
with 2 mL (for 1.7 mmol substrate) of an organic solvent
chosen from 1,2-dichloroethane, toluene, n-decane, acetoni-
trile and ethanol. The course of the reactions was monitored
using a Varian 3800 GC equipped with a capillary column
(SPB-5, 20 mꢂ0.25 mm) and a flame ionization detector.
The products were identified by GC-MS (HP 5890 Series II
GC, HP 5970 Series Mass Selective Detector) using He as
the carrier gas.
(
h -C H CH )Mo(CO) (1)
6
5
3
3
Mo(CO) (1.00 g, 3.79 mmol) was dissolved in refluxing tolu-
ene (30 mL). After 12 h, the excess toluene was removed
under reduced pressure and the resultant yellow solid was
purified by sublimation of excess Mo(CO) ; yield: 0.55 g
53%); anal. calcd. for C H MoO (272.11): C 44.14, H
.96; found: C 44.07, H, 3.17; IR (KBr): n=3093 (w), 3078
m), 2924 (m), 1960 (vs), 1859 (vs), 1839 (vs), 1529 (m),
447 (s), 1404 (m), 1380 (s), 1301 (m), 1262 (w), 1204 (w),
154 (m), 1144 (m), 1059 (m), 1034 (m), 1013 (w), 988 (m),
6
6
(
2
(
1
1
10 8 3
Adv. Synth. Catal. 2010, 352, 1759 – 1769
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1767

Ana C. Coelho et al.
FULL PAPERS
Acknowledgements
[26] J. R. Kain, F. E. Kꢃhn, Dalton Trans. 2008, 2221.
[
27] M. Abrantes, S. Gago, A. A. Valente, M. Pillinger, I. S.
GonÅalves, T. M. Santos, J. Rocha, C. C. Rom¼o, Eur. J.
Inorg. Chem. 2004, 4914, and references cited therein.
The authors are grateful to the Fundażo para a CiÞncia e a
Tecnologia (FCT), POCI 2010, OE and FEDER (Projects
POCI/CTM/55648/2004,
PPCDT/CTM/55648/2004
and
[28] B. Marler, U. Oberhagemann, S. Voltmann, H. Gies,
PTDC/QUI/71198/2006), and to RNME (Project FCT:
REDE/1509/RME/2005) for financial support. The FCT is
acknowledged for a doctoral grant to A.C.C. and a post-doc-
toral grant to S.M.B.
Microporous Mater. 1996, 6, 375.
[29] W. Hammond, E. Prouzet, S. D. Mahanti, T. J. Pinna-
vaia, Microporous Mesoporous Mater. 1999, 27, 19.
[30] S. J. Gregg, K. S. W. Sing, Adsorption, Surface Area and
Porosity, 2nd edn., Academic Press, London, 1982.
[
[
[
[
[
[
[
[
[
[
[
[
[
[
31] Y. H. Zhao, M. H. Abraham, A. M. Zissimos, J. Org.
Chem. 2003, 68, 7368.
32] F. Hoffmann, M. Gꢃngerich, P. J. Klar, M. Frçba, J.
Phys. Chem. C 2007, 111, 5648.
33] B. Onida, L. Borello, C. Busco, P. Ugliengo, Y. Goto, S.
Inagaki, E. Garrone, J. Phys. Chem. B 2005, 109, 11961.
34] G. Cerveau, R. J. P. Corriu, C. Lepeytre, Chem. Mater.
References
[
[
[
[
[
1] S. Inagaki, S. Guan, Y. Fukushima, T. Ohsuna, O. Tera-
saki, J. Am. Chem. Soc. 1999, 121, 9611.
2] T. Asefa, M. J. MacLachlan, N. Coombs, G. A. Ozin,
Nature 1999, 402, 867.
3] B. J. Melde, B. T. Holland, C. F. Blanford, A. Stein,
Chem. Mater. 1999, 11, 3302.
4] W. J. Hunks, G. A. Ozin, J. Mater. Chem. 2005, 15,
1
997, 9, 2561.
35] D. G. Carroll, S. P. McGlynn, Inorg. Chem. 1968, 7,
285.
36] H. Sun, Z. Chen, X. Zhou, Inorg. Chim. Acta 2003,
55, 404.
1
3
716.
3
5] F. Hoffmann, M. Cornelius, J. Morell, M. Frçba,
37] A. Sakthivel, J. Zhao, M. Hanzlik, F. E. Kꢃhn, Dalton
Trans. 2004, 3338.
38] M. Abrantes, A. Sakthivel, C. C. Rom¼o, F. E. Kꢃhn, J.
Organomet. Chem. 2006, 691, 3137.
39] J. L. F. Monteiro, C. O. Veloso, Top. Catal. 2004, 27,
Angew. Chem. 2006, 118, 3290; Angew. Chem. Int. Ed.
2
006, 45, 3216.
[
[
[
[
6] P. Van Der Voort, C. Vercaemst, D. Schaubroeck, F.
Verpoort, Phys. Chem. Chem. Phys. 2008, 10, 347.
7] S. Inagaki, S. Guan, T. Ohsuna, O. Terasaki, Nature
1
69.
2
002, 416, 304.
40] C. M. Byrne, S. D. Allen, E. B. Lobkovsky, G. W.
Coates, J. Am. Chem. Soc. 2004, 126, 11404.
41] G. Grigoropoulou, J. E. Clark, Tetrahedron Lett. 2006,
8] N. Bion, P. Ferreira, A. Valente, I. S. GonÅalves, J.
Rocha, J. Mater. Chem. 2003, 13, 1910.
9] M. P. Kapoor, Q. Yang, S. Inagaki, Chem. Mater. 2004,
4
7, 4461.
42] S. Sakaguchi, Y. Nishiyama, Y. Ishii, J. Org. Chem.
996, 61, 5307.
1
6, 1209.
[
[
10] B. Camarota, S. Mann, B. Onida, E. Garrone, Chem-
PhysChem 2007, 8, 2363.
11] M. P. Kapoor, Q. Yang, S. Inagaki, J. Am. Chem. Soc.
1
43] A. Bordoloi, F. Lefebvre, S. B. Halligudi, J. Mol. Catal.
A: Chem. 2007, 270, 177.
2
002, 124, 15176.
44] L. D. Pinto, J. Dupont, R. F. de Souza, K. Bernardo-
Gusm¼o, Catal. Commun. 2008, 9, 135, and references
cited therein.
[
[
12] Y. Yang, A. Sayari, Chem. Mater. 2007, 19, 4117.
13] Y. Xia, W. Wang, R. Mokaya, J. Am. Chem. Soc. 2005,
1
27, 790.
[
45] S. G. Casuscelli, M. E. Crivello, C. F. Perez, G. Ghione,
E. R. Herrero, L. R. Pizzio, P. G. Vꢄzquez, C. V.
Cꢄceres, M. N. Blanco, Appl. Catal. A: Gen. 2004, 274,
[
[
14] Y. Xia, R. Mokaya, J. Phys. Chem. B 2006, 110, 3889.
15] A. Sayari, W. Wang, J. Am. Chem. Soc. 2005, 127,
1
2194.
16] M. Cornelius, F. Hoffmann, M. Frçba, Chem. Mater.
005, 17, 6674.
17] N. Mizoshita, Y. Goto, M. P. Kapoor, T. Shimada, T.
Tani, S. Inagaki, Chem. Eur. J. 2009, 15, 219.
18] M. P. Kapoor, S. Inagaki, Bull. Chem. Soc. Jpn. 2006,
1
15.
[
[
[
[
[
[
46] U. Arnold, R. Serpa da Cruz, D. Mandelli, U. Schu-
chardt, J. Mol. Catal. A: Chem. 2001, 165, 149.
47] P. C. Bakala, E. Briot, L. Salles, J.-M. Brꢅgeault, Appl.
Catal. A: Gen. 2006, 300, 91.
48] B. Monteiro, S. S. Balula, S. Gago, C. Grosso, S. Fig-
ueiredo, A. D. Lopes, A. A. Valente, M. Pillinger, J. P.
LourenÅo, I. S. GonÅalves, J. Mol. Catal. A: Chem.
2009, 297, 110.
2
7
9, 1463.
[19] S. Fujita, S. Inagaki, Chem. Mater. 2008, 20, 891.
[20] Q. Yang, M. P. Kapoor, S. Inagaki, J. Am. Chem. Soc.
2
002, 124, 9694.
[
[
[
21] S. P. Shylesh, M. Jia, A. Seifert, S. Adappa, S. Ernst,
W. R. Thiel, New J. Chem. 2009, 33, 717.
22] M. Ohashi, M. P. Kapoor, S. Inagaki, Chem. Commun.
[49] R. Hutter, T. Mallat, A. Baiker, J. Catal. 1995, 153, 177.
[50] E. Gianotti, C. Bisio, L. Marchese, M. Guidotti, N.
Ravasio, R. Psaro, S. Coluccia, J. Phys. Chem. C. 2007,
111, 5083.
[51] M. Guidotti, I. Batonneau-Gener, E. Gianotti, L. Mar-
chese, S. Mignard, R. Psaro, M. Sgobba, N. Ravasio,
Microporous Mesoporous Mater. 2008, 111, 39.
2
008, 841.
23] S. Shylesh, A. Wagener, A. Seifert, S. Ernst, W. R.
Thiel, Angew. Chem. 2010, 122, 188; Angew. Chem. Int.
Ed. 2010, 49, 184.
[
[
24] T. Kamegawa, T. Sakai, M. Matsuoka, M. Anpo, J. Am.
[52] M. V. Cagnoli, S. G. Casuscelli, A. M. Alvarez, J. F.
Bengoa, N. G. Gallegos, N. M. Samaniego, M. E. Crivel-
lo, G. E. Ghione, C. F. Pꢅrez, E. R. Herrero, S. G. Mar-
chetti, Appl. Catal. A: Gen. 2005, 287, 227.
Chem. Soc. 2005, 127, 16784.
25] J. G. Vitillo, E. Groppo, S. Bordiga, S. Chavan, G. Ric-
chiardi, A. Zecchina, Inorg. Chem. 2009, 48, 5439.
1768
asc.wiley-vch.de
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 1759 – 1769

Grafting of Molecularly Ordered Mesoporous Phenylene-Silica with Molybdenum Carbonyl Complexes
[
53] D. Marino, N. G. Gallegos, J. F. Bengoa, A. M. Alvarez,
M. V. Cagnoli, S. G. Casuscelli, E. R. Herrero, S. G.
Marchetti, Catal. Today 2008, 133, 632.
[56] C. D. Nunes, M. Pillinger, A. A. Valente, I. S. Gon-
Åalves, J. Rocha, P. Ferreira, F. E. Kꢃhn, Eur. J. Inorg.
Chem. 2002, 1100.
[
[
54] F. Chiker, F. Launay, J. P. Nogier, J. L. Bonardet, Green
Chem. 2003, 5, 318.
55] L. Saikia, D. Srinivas, Catal. Today 2009, 141, 66.
[57] M. Kruk, M. Jaroniec, Langmuir 1997, 13, 6267.
[58] M. Kruk, V. Antochshuk, M. Jaroniec, J. Phys. Chem. B
1999, 103, 10670.
[
59] K. J. Shea, D. A. Loy, O. Webster, J. Am. Chem. Soc.
1
992, 114, 6700.
Adv. Synth. Catal. 2010, 352, 1759 – 1769
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
1769