Nanostructured dioxomolybdenum(VI) catalyst for the liquid-phase epoxidation of olefins

Article abstract of DOI:10.1002/ejic.200901093

A new organosilica hybrid mesoporous material (HM) containing phosphane oxide surface groups was prepared via the "dual" organosilane templated co-condensation of 4-bis-(triethoxysilyl)benzene and (diphenylphosphanyl)propyltriethoxysilane, followed by post-synthesis oxidation of the phosphane surface groups. The HM material was used to immobilise the [Mo(O)₂Cl₂] complex. The resultant material (HM-Mo) is an active and selective catalyst in the liquid-phase epoxidation of olefins [cis-cyclooctene, (R)-(+)-limonene, trans-2-octene, 1-octene] with tBuOOH at 55°C and gives comparable epoxide yields to those observed for the free complex [MoCl₂(O)₂{OP(CH₂CH₃)(Ph) ₂}₂] (1), under similar reaction conditions: the crystal structure of complex 1 remains the same after reaction with tBuOOH. The oxidant is essentially consumed in the HM-Mo or 1 catalysed epoxidation of the olefin. In the case of limonene (possesses endo- and exocyclic C=C bonds), the molar ratio of 1,2-epoxy-p-menth-8-ene/1,2-8,9-diepoxy-p-menthane at ca. 90% conversion (reached after 24 h reaction) is ca. 5 and 10 for 1 and HM-Mo, respectively, indicating a favourable regioselectivity towards the epoxidation of the endocyclic C1-C2 double bond, especially in the case of the heterogeneous catalyst.

Full text of DOI:10.1002/ejic.200901093

FULL PAPER
DOI: 10.1002/ejic.200901093
Nanostructured Dioxomolybdenum(VI) Catalyst for the Liquid-Phase
Epoxidation of Olefins
Alichandra Castro,[a] João C. Alonso,[b] Anabela A. Valente,*[a] Patrícia Neves,[a]
Paula Brandão,[a] Vitor Félix,[a] and Paula Ferreira*[b]
Keywords: Molybdenum / Mesoporous materials / Heterogeneous catalysis / Epoxidation
A new organosilica hybrid mesoporous material (HM) con-
taining phosphane oxide surface groups was prepared via
the “dual” organosilane templated co-condensation of 4-bis-
(triethoxysilyl)benzene and (diphenylphosphanyl)propyltri-
ethoxysilane, followed by post-synthesis oxidation of the
phosphane surface groups. The HM material was used to im-
mobilise the [Mo(O)2Cl2] complex. The resultant material
(HM-Mo) is an active and selective catalyst in the liquid-
phase epoxidation of olefins [cis-cyclooctene, (R)-(+)-limon-
ene, trans-2-octene, 1-octene] with tBuOOH at 55 °C and
gives comparable epoxide yields to those observed for the
free complex [MoCl2(O)2{OP(CH2CH3)(Ph)2}2] (1), under sim-
ilar reaction conditions: the crystal structure of complex 1 re-
mains the same after reaction with tBuOOH. The oxidant is
essentially consumed in the HM-Mo or 1 catalysed epoxid-
ation of the olefin. In the case of limonene (possesses endo-
and exocyclic C=C bonds), the molar ratio of 1,2-epoxy-p-
menth-8-ene/1,2–8,9-diepoxy-p-menthane at ca. 90% con-
version (reached after 24 h reaction) is ca. 5 and 10 for 1 and
HM-Mo, respectively, indicating a favourable regioselectiv-
ity towards the epoxidation of the endocyclic C1–C2 double
bond, especially in the case of the heterogeneous catalyst.
used as catalysts in industrial liquid-phase epoxidation pro-
cesses.[6] Molybdenum(VI) complexes of the type [MoO-
(O2)2(OPR3)(H2O)] (R = alkyl ligands) are of commercial
interest as homogeneous epoxidation catalysts (patented by
BASF[7]): complexes of this type were first introduced as
epoxidation catalysts with hexamethylphosphoramide in-
stead of (OPR3) in pioneering studies carried out by the
group of Mimoun.[8]
Introduction
Organic–inorganic hybrid materials are particularly at-
tractive for preparing a wide range of multifunctional mate-
rials with potential commercial applications. They are pre-
pared by the combination of organic and inorganic compo-
nents, allowing a synergy of the thermal stability and chem-
istry of ceramic materials with the relatively easy process-
ability and flexibility of organic compounds and poly-
mers.[1] In recent years, ordered mesoporous silicon-con-
taining hybrid materials (MesoSiHyb) prepared via co-
condensation of organosilanes in the presence of a template
have become attractive catalyst supports partly due to their
good mechanical strength, thermal and chemical stability
(e.g., in oxidation environment), versatility (the physico-
chemical properties can be modified/fine-tuned through
changes in compositions, architectures and functionalities
during the process of material synthesis), high specific sur-
face area and large pore sizes.[2]
Molybdenum(VI) complexes of the type [Mo(O)2Cl2]
and related complexes bearing organic ligands have been
used as catalysts for various organic transformations,[9] in-
cluding the selective epoxidation of olefins with tert-butyl
hydroperoxide (tBuOOH).[10–18] However, these metal com-
plexes are homogeneous catalysts, and it is desirable to het-
erogenise them, because important advantages of hetero-
geneous catalysts are the easier product–catalyst separation
and catalyst reuse. For liquid-phase heterogeneous catalytic
processes, the catalyst stability towards solvation/leaching
of active species is one of the major concerns. Complexes
of the type [Mo(O)2Cl2L], where L = O- or N-containing
ligand(s), have been covalently anchored onto micelle-tem-
plated ordered mesoporous silicas (e.g., MCM-41, MCM-
48, SBA-15) via one of the following routes: (i) by directly
adding the prepared metal complex to the silica support or
(ii) pretreating the silica with L and subsequent introduc-
tion of the metal precursor (so-called tethering ap-
proach).[19–22] Although this group of catalysts is quite ef-
fective in the epoxidation of olefins with tBuOOH, it pres-
ents at least some of the following disadvantages: (i) hetero-
geneous distribution of active species, (ii) different active
The metal-catalysed epoxidation of olefins into epoxides
is an important reaction in organic synthesis, as epoxides
are useful intermediates that can be transformed into a vari-
ety of other compounds.[3–6] Molybdenum complexes are
[a] Department of Chemistry, CICECO, University of Aveiro,
3810-193 Aveiro, Portugal
E-mail: atav@ua.pt
[b] Department of Ceramics and Glass Engineering, CICECO,
University of Aveiro,
3810-193 Aveiro, Portugal
Fax: +351-234401470
E-mail: pcferreira@ua.pt
Eur. J. Inorg. Chem. 2010, 1405–1412
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1405

A. A. Valente, P. Ferreira et al.
FULL PAPER
sites formed via the reaction of the molybdenum precursor lengths and angles subtended at the molybdenum centre
with the (un)functionalised silica support (e.g., the metal compare well with those found for the related complex
precursor may react with silanol or L surface groups), [Mo(O)2Cl2(OPMePh2)2].[29]
(iii) partial metal leaching during catalysis. Improvements
concerning the stability towards metal leaching and cata-
lytic activity have been accomplished through treating the
mesoporous silica surface with (CH3)3SiCl, which enhances
the surface hydrophobicity.[21,23,24] Fairly good catalyst sta-
bility has been reported for polystyrene resin functionalised
with phosphane groups, and they have been subsequently
grafted with [Mo(CO)6] and used in the catalytic epoxid-
ation of aromatic and aliphatic terminal alkenes with
tBuOOH [most likely the supported metal species are oxi-
dised to molybdenum(VI)].[25] Alternatively, efficient solid
epoxidation catalysts have been prepared by supporting
molybdenum(VI) complexes on MesoSiHyb materials via
ligand exchange reactions.[26,27]
Recently, some of us reported on the catalytic perform-
ance of a mesostructured hybrid materials prepared by in-
troducing [Mo(O)2Cl2] in a preformed mesoporous silica
Figure 1. Molecular diagram showing the overall geometry of com-
plex [MoCl2(O)2{OP(CH2CH3)(Ph)2}2] (1) with labelling scheme
adopted.
functionalised with phosphane oxide spacer ligands:[28] in
general, epoxide selectivity was excellent, albeit the occur-
rence of Mo and P leaching during catalysis and a concomi-
tant decrease in catalytic activity of the used catalyst was
observed, which could be related to the diversity of surface
metal species. In this work, a new hybrid mesoporous mate-
rial (HM) has been prepared by co-condensation of the or-
ganosilanes 1,4-bis(triethoxysilyl)benzene (BTEB – carries
the organic component of the matrix) and (diphenylphos-
phanyl)propyltriethoxysilane (TEPPP – for pending phos-
phane groups) in the presence of Pluronics P123 as a tem-
plate, followed by template removal (through chemical
treatment) and post-synthesis oxidation of the phosphane
surface groups into phosphane oxide. The resultant HM
material is used as a support for grafting [Mo(O)2Cl2], giv-
ing a new catalyst (HM-Mo) for the liquid-phase epoxid-
ation reaction of olefins with tBuOOH. The investigated
olefins are cis-cyclooctene, (R)-(+)-limonene, trans-2-octene
and 1-octene, and the catalytic reactions were carried out
at 55 °C, without a cosolvent. For comparative studies the
complex [MoCl2(O)2{OP(CH2CH3)(Ph)2}2] (1) was pre-
pared and tested as a catalyst, under similar conditions. The
crystal structure of complex 1 is also reported.
Mesostructured Material HM-Mo
The co-condensation of BTEB and TEPPP in the pres-
ence of a template, followed by the removal of the latter
and, finally, the oxidation of the surface phosphane surface
groups into phosphane oxide (complete as monitored by
31P CP MAS NMR), gives HM (Scheme 1). The complex
[Mo(O)2Cl2] was grafted onto HM to give HM-Mo with
0.47 mmolg–1 of Mo and a P/Mo mol ratio of 1.8.
Results and Discussion
Preparation and Characterisation of the Catalysts
Complex 1
The complex [MoCl2(O)2{OP(CH2CH3)(Ph)2}2] (1) is
obtained by treating [Mo(O)2Cl2] with two equivalents of
the phosphane oxide ligand CH3CH2(Ph)2PO. The molecu-
lar structure of 1 excluding the hydrogen atoms (Figure 1)
has a distorted octahedral coordination sphere with equato-
rial coordination of the molybdenum centre composed of
four oxygen atoms from two O=Mo groups (oxo) and two
diphenylphosphane oxide ligands (dppeO). The bond Scheme 1. Schematic synthesis of heterogeneous catalyst.
1406 Eur. J. Inorg. Chem. 2010, 1405–1412
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Nanostructured Dioxomolybdenum(VI) Epoxidation Catalyst for Olefins
The powder XRD patterns of both HM and HM-Mo
evidence diffraction at 2θ at ca. 0.91°, indicating some
mesoporous order, which is supported by TEM showing a
hexagonal arrangement of mesoporous (Figure 2 for HM-
Mo). The typical short range order usually observed when
phenylene groups are present in the material[30] was not ob-
served, suggesting that the molecular scale periodicity is
perturbed by the introduction of the bulky phosphane oxide
ligands. This may also explain the unclear presence of the
(110), (200), (210) and (300) reflections. Another explana-
tion for the lack of molecular scale periodicity is the synthe-
sis under acidic conditions, which is not favourable for the
formation of a periodic arrangement of the phenylene-silica
moieties as reported by Inagaki and co-workers.[31]
Figure 3. Nitrogen adsorption–desorption isotherms and pore-size
distribution curve of HM (open square) and HM-Mo (open tri-
angle).
Table 1. Texture parameters of the hybrid materials.
[a]
[b]
[c]
Sample
SBET [m2 g–1] VP [cm3 g–1]
dp [nm]
HM
HM-Mo
437
335
0.51
0.43
6.1
5.8
[a] BET specific surface area, relative pressure range 0.03–0.15.
[b] Specific total pore volume by using the Gurvitsch equation for
relative pressure of ca. 0.98. [c] Maximum of the pore-size distribu-
tions calculated by the BJH method by using the nitrogen adsorp-
tion data.
Figure 2. Powder X-ray diffraction pattern and TEM image (inset)
of HM-Mo.
The HM and HM-Mo materials exhibit irreversible type
IV nitrogen adsorption–desorption isotherms (characteris-
tic of mesoporous solids) with relatively sharp capillary
condensation steps in the relative pressure range 0.5–0.62
(Figure 3).[32] The nitrogen uptake of HM-Mo is slightly
lower than that of HM, leading to a concomitant reduction
in BET specific surface area and total pore volume
(Table 1). The pore-size distribution (PSD) curves are rela-
tively narrow with a maximum dp of 6.1 and 5.8 nm for
HM and HM-Mo, respectively. The pore-wall thickness cal-
culated as [(2d100/Ί3) – dp], where the first term represents
the lattice parameter considering a hexagonal unit cell, is
5.1 nm for HM. These results, together with the powder
XRD and TEM data, indicate that the (fairly ordered)
mesoporous structure of HM is preserved during the post-
synthesis grafting procedure and suggest that the metal
complex is grafted on the internal surface of HM.
Figure 4. 29Si CP MAS NMR spectra of the HM materials: (a)
HM and (b) HM-Mo.
The 29Si CP MAS NMR spectra of HM and HM-Mo
display three signals at δ = –69.9, –77.0 and –80.0 ppm, as-
signed to T1, T2 and T3 organosilica species [Tm = RSi(OSi)m-
The 13C CP MAS NMR spectrum of HM-Mo shows res-
(OH)3–m], respectively (Figure 4). No Qn species were iden- onances at δ = 15.6 (CH3) and 57.8 ppm (CH2) due to (rem-
tified in the spectra, indicating that silanol surface groups iniscent) [SiO–CH2CH3] groups (Figure 5). The resonances
are negligible (or inexistent) and that the Si–C bond is pre- at δ = 62.5 and 30.0 ppm are assigned to the first and sec-
served (e.g., resistant to hydrolysis) during the (post-)syn- ond (counting from the left) methylene groups in
thesis processes.
[PCH2CH2CH2Si], respectively. One could expect the signal
Eur. J. Inorg. Chem. 2010, 1405–1412
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1407

A. A. Valente, P. Ferreira et al.
FULL PAPER
related with the methylene group bonded to the silicon products formed in 73 and 27% selectivity, respectively, at
atom in [PCH2CH2CH2Si] to appear at higher field, which 10% conversion (reached after 24 h; Figure 7). Another
is probably overlapped with that of the methyl group in control experiment consisted of performing the reaction of
[SiO–CH2CH3]. The peak at δ = 73 ppm is assignable to the Cy8 in the presence of HM (without Mo), which is very
residual template.
sluggish, giving 7% conversion after 24 h: the main product
was cyclooctene oxide (75% selectivity) and cyclooctane-
1,2-diol was the byproduct (Figure 7). Iodometric titrations
were performed for the reaction of tBuOOH in the presence
of 1 or HM-Mo (without olefin), which indicated that the
“nonproductive” decomposition of tBuOOH is negligible.
Figure 5. 13C CP MAS NMR spectra of the HM materials: (a) HM
and (b) HM-Mo. The symbol * denotes spinning sidebands.
The 31P CP MAS spectra of HM and HM-Mo show a
peak at δ = 36.9 ppm, assigned to PV species (Figure 6). A
small shoulder at ca. δ = 21.0 ppm can be observed in both
spectra and may be attributed to a side reaction during the
synthesis procedure to give PV species of the type (EtO)3-
Si(CH2)3(Ph)2P(-O-[SiO2])2 (this phosphorus atom is not
available to coordinate the molybdenum).[33]
Figure 7. Reaction of cis-cyclooctene with tBuOOH at 55 °C in the
presence of complex 1 [run 1 (᭺); run 2, (+)], OP(Ph)2CH2CH3
(᭛), HM-Mo [run 1 (᭝); run 2 (ϫ)] or HM (–); leaching test per-
formed for HM-Mo (catalyst filtration at 10 min; *).
Catalytic Performance of Complex 1
The reaction of Cy8 with tBuOOH in the presence of
complex 1 gives quantitative epoxide yield after 24 h
(Table 2). The catalytic activity of 1 is comparable to that
reported in the literature for [MoCl2(O)2{OP(Ph)2CH3}2],
used as a catalyst in the same reaction, under similar condi-
tions: TOF calculated for a 30 min reaction is 150 and
179 molmolMo–1 h–1 for [MoCl2(O)2{OP(Ph)2CH3}2] and 1,
respectively.[34] Relatively high catalytic activity in olefin
epoxidation with tBuOOH was also reported in the case of
the complex [Mo(O)2C12L], where L is a 3-(diethoxyphos-
phoryl) derivative of camphor.[35]
In the case of complex 1, after 10 min, 86% conversion
was reached, and afterwards the reaction slowed down con-
siderably (Figure 7). During the course of the catalytic reac-
tion tert-butyl alcohol (tBuOH), a byproduct associated to
the consumption of tBuOOH, may act as an inhibitor for
the coordination of the oxidant to the metal centre.[36–41]
Figure 6. 31P CP MAS NMR spectra of the HM materials: (a) HM
and (b) HM-Mo. The * symbol denotes spinning sidebands.
Catalysis Studies
The catalytic performance of HM-Mo was investigated On the basis of the mechanistic considerations reported in
in the epoxidation of cis-cyclooctene (Cy8) with tBuOOH the literature for complexes possessing the [Mo(O)2]2+ moi-
at 55 °C without a cosolvent. For comparative purposes, the ety and Lewis base N,O,S ligands, used as catalysts in the
reaction was carried out in the presence of complex 1, un- same reaction, the oxidant coordinates to the metal centre
der similar reaction conditions. A control experiment was to give an (electrophilic) active oxidising specie responsible
performed by using OP(Ph)2CH2CH3 (completely soluble for the oxygen atom transfer to the (nucleophilic) ole-
in the reaction medium) instead of complex 1, which gives fin.[42–45] Electrophilic attack of the oxidant on the C=C
cyclooctene oxide and cyclooctane-1,2-diol as the only bond has also been reported for complexes of the type
1408
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 1405–1412

Nanostructured Dioxomolybdenum(VI) Epoxidation Catalyst for Olefins
Table 2. Reaction of olefins with tBuOOH in the presence of dif- and dried at room temperature (the resultant compound is
ferent catalysts over 4/24 h at 55 °C.[a]
denoted 1rec_crystal). Single-crystal XRD showed that the
Olefin
Catalyst
Conv.[b]
[%]
Sel.[c]
[%]
Yield[d]
[%]
structure of 1rec_crystal is similar to that of complex 1:
these results were reproducible for different crystals in the
sample. The FTIR spectrum of 1rec_crystal is very similar
to that of 1 (Figure 8). Hence, complex 1 is stable, at least
to a certain extent, under the applied oxidising conditions.
cis-Cyclooctene
1
HM-Mo
HM
97/100
98/100
4/7
100/100
100/100
100/75
100/100
84/79[e]
85/86[e]
100/100
100/100
100/100
100/100
97/100
98/100
4/5
none
1
HM-Mo
1
HM-Mo
1
HM-Mo
1/4
1/4
(R)-(+)-Limonene
1-Octene
86/93
80/91
28/48
24/45
57/79
59/80
72/73
68/78
28/48
24/45
57/79
59/80
Catalytic Performance of HM-Mo
The HM-Mo catalyst gives quantitative epoxide yield
within 6 h (Figure 7). Similar results (100% epoxide yield
at 4 h) were recently reported for complexes of the type
[MoO(O2)2L] {L = (3-triethoxysilylpropyl)-[3-(2-pyridyl)py-
razol-1-yl]acetamide} supported on phenylene-bridged
mesoporous organosilicas, used as catalyst in the same reac-
tion, and chloroform (b.p. 61.2 °C) as solvent, under re-
flux.[48] Initial TOF (calculated for 10 min reaction) is
541 molmolMo–1 h–1 for HM-Mo, which is comparable with
that observed for complex 1 (519 molmolMo–1 h–1), suggest-
ing that the nature of the active species involved may be
similar. The lower conversions reached at the beginning of
the reaction in the presence of HM-Mo in relation to that
observed for complex 1 may be, at least partly, due to the
lower amount of molybdenum species loaded into the reac-
tion vessel in the former case (similar conversions are
reached after 90–120 min).
To assess the homo/heterogeneous nature of the catalytic
reaction in the presence of HM-Mo a leaching test was per-
formed by filtering off (at 55 °C) the reaction solution after
10 min, through a 0.2 µm PVDF w/GMF Whatman mem-
brane, and leaving it to react for a further 50 min at 55 °C.
No increase in olefin conversion was observed after the fil-
tration step, whereas in the presence of HM-Mo conversion
increased by ca. 35%. These results suggest that the olefin
epoxidation reaction with HM-Mo as catalyst is hetero-
geneous in nature. The FTIR spectra of the fresh and reco-
vered solids were identical (not shown). The recovered HM-
Mo catalyst (washed with n-hexane and dried at room tem-
perature) was reused in a second 24 h run at 55 °C. After
reaction, the solid catalyst was separated by centrifugation,
thoroughly washed with n-hexane and dried at room tem-
perature overnight. Epoxide selectivity was always 100%.
The kinetic curves for the two 24-h runs are roughly coinci-
dent (Figure 7), and ICP-AES analyses of the recovered so-
lid indicated no measurable Mo or P leaching. High catalyst
stability has also been reported by the group of Thiel for
(i) complexes of the type [MoO(O2)2] supported on hybrid
supports synthesised by sol–gel copolymerisation of a che-
late ligand with TEOS,[26,27] and (ii) the complex [MoO-
(O2)2L] {L = (3-triethoxysilylpropyl)-[3-(2-pyridyl)pyrazol-
1-yl]acetamide} supported on a hydrophobic phenylene-
bridged mesoporous organosilica.[48] The catalytic perform-
ance of HM-Mo is superior to that recently reported for a
catalyst prepared by tethering [Mo(O)2Cl2] onto a pre-
formed mesoporous silica functionalised with phosphane
oxide spacer ligands:[28] Mo and P leaching and a concomi-
tant decrease in catalytic activity were observed for the used
catalyst. The higher catalyst stability of HMo in compari-
trans-2-Octene
[a] Reaction conditions: olefin (1.8 mmol), tBuOOH (2.75 mmol),
complex 1 (18 µmol) or HM-Mo (17.3 mg, 8.1 µmol Mo). [b] Olefin
conversion after 4/24 h. [c] Selectivity to the corresponding epoxide
(limonene oxide in the case of limonene) after 4/24 h. [d] Epoxide
yield after 4/24 h. [e] Selectivity to 1,2–8,9-diepoxy-p-menthane af-
ter 4/24 h is 9/16% and 7/8% for 1 and HM-Mo, respectively.
[MoO(O2)2(OPR3)] (R = alkyl), albeit involving different
transition states.[46,47]
After the first 24-h run, a second run was carried out by
recharging the reaction vessel with oxidant and substrate
(in the same amounts as those used in the first run), and
by monitoring the catalytic reaction for a further 24 h.
Quantitative epoxide yield was obtained in two consecutive
24-h runs, albeit the reaction was slower in the second run
(at least until 6 h), which may be due to dilution effects and/
or the presence of tBuOH at the beginning of the second
run. These two factors may be avoided by using a hetero-
geneous catalyst and by separating it from the reaction
products prior to its reuse (see next section).
A separate (larger scale) experiment was performed by
mixing complex 1 with tBuOOH (1:20 molar ratio) without
olefin for 3 h at 55 °C. After cooling to room temperature,
n-hexane was added until a precipitate was obtained, which
was separated by centrifugation, washed with n-hexane and
dried at room temperature. Subsequently, the solid was dis-
solved in dichloromethane, and this solution was left under
static conditions for approximately one month to give crys-
tals that were separated by filtration, washed with n-hexane
Figure 8. FTIR spectra of 1 and 1rec_crystal (compound recovered
after treatment under catalytic conditions).
Eur. J. Inorg. Chem. 2010, 1405–1412
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1409

A. A. Valente, P. Ferreira et al.
FULL PAPER
son to the catalyst prepared through the tethering synthetic at room temperature to give similar catalytic results as those
approach may be related to its hydrophobic, uniform, sin- observed for the fresh catalyst. The heterogeneous nature
gle-site features.
of the catalytic reaction was assessed through a leaching
The HM-Mo catalyst was further tested in the reactions test carried out at the reaction temperature. In the case of
of (R)-(+)-limonene, 1-octene and trans-2-octene with the limonene reaction, regioselectivity (endo/exo C=C
tBuOOH at 55 °C without a cosolvent: the same reactions epoxidation, at ca. 90% conversion) to 1,2-epoxy-p-menth-
were carried out in the presence of complex 1 for compari- 8-ene was higher for HM-Mo than for complex 1, which
son (Table 2). Interestingly the homogeneous and hetero- may be partly due to confinement effects of the nanostruc-
geneous catalysts give similar conversions and epoxide tured catalyst.
yields at 4/24 h for all the olefins, under the applied reaction
Overall, it seems that HM-Mo is an interesting, uniform,
conditions. The linear olefins 1-octene and trans-2-octene single-site, mesostructured epoxidation catalyst. It may be
give the respective epoxides as the only products. The epox- interesting to further investigate the effects of the aliphatic
ides yields at 4 h are 24–28 and 57–59% for 1-octene and chain length and the substituents in the phenyl groups of
trans-2-octene, respectively, indicating that the latter is more the phosphane oxide ligand on the catalytic performance
reactive than the former. The higher reactivity of the more and the examination of the reactivity of different substrates.
substituted (electron richer) olefin than the less substituted On the other hand, the use of HM and related materials as
one is congruent with the above mechanistic considerations: catalyst supports may be extendable to a plethora of homo-
as a result of the electrophilic nature of oxygen transfer, geneous catalysts used in different organic reactions (e.g.,
more nucleophilic olefins tend to react faster. The reaction asymmetric catalysis, hydrogenations). Other promising
of limonene gives mainly 1,2-epoxy-p-menth-8-ene (P1) fields of application include nanomedicine – hundreds of
formed in 68–72% yield after 4 h. Byproducts include organophosphorus drugs are currently known.
mainly 1,2-8,9-diepoxy-p-menthane (P2), p-menta-1,8-dien-
7-ol and 4-isopropyl-1-methylcyclohexane-1,2-diol. The
molar ratio P1/P2 at ca. 90% conversion (reached at 24 h
Experimental Section
Materials and Characterisation Techniques: Starting reagents were
acquired from Aldrich, and commercial, extra-dried solvents were
obtained from Acros Organics (H2O Ͻ 0.003%, AcrosSeal®) and
used as received. The catalysts were prepared by using standard
Schlenk techniques under a nitrogen atmosphere. Powder X-ray dif-
fraction (PXRD) data were collected with a Philips XЈPert MPD
diffractometer (Cu-Kα X-radiation, λ = 1.54060 Å) fitted with a
graphite monochromator and a flat-plate sample holder, in a
Bragg–Brentano para-focusing optics configuration. Samples were
step-scanned in 0.02° 2θ steps with a counting time of 2 s per step;
reference material 675 (National Bureau of Standards) was used as
internal standard. TEM images were recorded with a 300 eV Hita-
chi H9000-NA instrument. Nitrogen adsorption isotherms at
–196 °C were measured by using a Gemini 2375 Micromeritics In-
strument (Eng. M. C. Costa, University of Aveiro). The mesostruc-
tured materials were pretreated at 150 °C prior to analysis. Micro-
analyses for C and H were carried out at the Department of Chem-
istry, University of Aveiro (M. M. Marques). The P and Mo con-
tents were determined by ICP-AES at the Central Laboratory for
Analysis, University of Aveiro (L. Carvalho). Infrared spectra (KBr
pellets) were measured by using a Mattson 7000 FT spectrometer.
13C, 29Si and 31P solid-state NMR spectra were recorded at 100.62,
79.49 and 161.98 MHz, respectively, with a (9.4-T) Bruker Avance
400P spectrometer. 29Si MAS NMR spectra were recorded with 40°
pulses, spinning rates 5.0–5.5 kHz and 60-s recycle delays. 29Si CP
MAS NMR spectra were recorded with 4-µs 1H 90° pulses, 8 ms
contact time, a spinning rate of 5 kHz and 4-s recycle delays. For
31P MAS NMR spectra were recorded with 45° pulses, spinning
rate of 15 kHz, 1H decoupling and a 60 s recycle delay. Chemical
shifts are quoted in parts per million from phosphoric acid (85%).
13C CP MAS NMR spectra were recorded with 4.5-µs 1H 90°
pulses, 2 ms contact time, a spinning rate of 7 kHz and 4-s recycle
delays. Chemical shifts are quoted in ppm from TMS. 13C spectra
were also recorded in the solid state at 125.76 MHz with a Bruker
Avance 500 spectrometer.
reaction) is ca. 5 and 10 for 1 and HM-Mo, respectively,
indicating a favourable regioselectivity towards the epoxid-
ation of the endocyclic C1–C2 double bond. The signifi-
cantly higher P1/P2 molar ratio observed for the supported
catalyst in comparison to complex 1 may be due to confine-
ment effects.
Conclusions
A nanostructured catalyst (HM-Mo) was prepared by
grafting of the molybdenum precursor [Mo(O)2Cl2] to a
novel hybrid mesoporous support containing phosphane
oxide surface groups (HM). The HM material is prepared
by the “dual” organosilane co-condensation method, fol-
lowed by template (Pluronics P123) removal and post-syn-
thesis oxidation of the phosphane surface templated groups
into phosphane oxide. The two organosilanes are 1,4-bis-
(triethoxysilyl)benzene, which carries the organic compo-
nent of the matrix, and (diphenylphosphanyl)propyltri-
ethoxysilane, which accounts for pending phosphane
groups.
The HM-Mo material is an active and selective catalyst
in the liquid-phase epoxidation of cis-cyclooctene (100%
epoxide yield at 24 h), (R)-(+)-limonene (78% epoxide
yield), 1-octene (45% epoxide yield) and trans-2-octene
(80% epoxide yield) with tBuOOH at 55 °C without a co-
solvent. Similar results were obtained for complex 1, under
similar reaction conditions. Single-crystal XRD data for
complex 1 before and after contact with tBuOOH are sim-
ilar. For complex 1 and HM-Mo, tBuOOH is essentially
consumed in the epoxidation process. In the case of HM-
Mo, no P and Mo leaching was detected and the catalyst
X-ray Crystallography: Crystals of 1 with suitable quality for single-
could be recovered by simple filtration, washing and drying crystal X-ray diffraction analysis were grown from a dichlorometh-
1410 Eur. J. Inorg. Chem. 2010, 1405–1412
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Nanostructured Dioxomolybdenum(VI) Epoxidation Catalyst for Olefins
ane solution. The X-ray data were collected with a CCD Bruker
APEX II at 150(2) K by using graphite monochromated Mo-Kα
radiation (λ = 0.71073 Å). The crystal was positioned at 35 mm
from the CCD and the spots were measured by using a counting
time of 20 s. Data reduction, including a multiscan absorption cor-
rection, were carried out by using the SAINT-NT from Bruker
AXS. The structure was solved by using SHELXS-97[49] and re-
fined by using full-matrix least-squares in SHELXL-97.[50] The C–
H hydrogen atoms were included at calculated positions, whereas
the hydrogen atoms bonded to water molecules were located from
final difference Fourier maps. Anisotropic thermal parameters were
used for all non-hydrogen atoms, whereas the hydrogen atoms were
refined with isotropic parameters equivalent to 1.2 times those of
the atom to which they were attached. 19040 reflections were col-
lected and subsequently merged to 6702 unique reflections with a
Rint of 0.0258. The final refinement of 336 parameters converged
to final R and Rw indices R1 = 0.0367 and wR2 = 0.0933 for 5405
reflections with IϾ2σ(I) and R1 = 0.0487 and wR2 = 0.1002 for all
hkl data. Molecular diagrams presented are drawn with graphi-
cal package software PLATON.[51] Crystal data for 1:
C28H30Cl2MoO4P2, Mr = 659.33; monoclinic, space group P21/n,
Z = 4, a = 9.4338(2) Å, b = 17.8462(4) Å, c = 17.5228(3) Å, β =
103.3970(10)°, V = 2869.82(10) Å3, ρcalcd. = 1.526 Mgm–3, Mo-Kα
= 0.787 mm–1. CCDC-744897 contains the supplementary crystal-
lographic data for this paper. These data can be obtained free of
charge from the Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/ data_request/cif.
-CH2Si) ppm. 31P NMR (300 MHz, CDCl3, r.t.): δ = –16.8 ppm.
IR (KBr pellet): ν = 3051 [ν(C–H)Ph], 1363–1481 [ν(C–C)Ph], 1294
˜
[νp(CH)], 696 [νop(CH)], 1074 [ν(Si–O)] cm–1.
Step 3: The mesoporous hybrid support was prepared by hydrolysis
and co-condensation of BTEB and TEPPP in the presence of Plu-
ronics P123 (EO20PO70EO20) as template. In a typical synthesis,
the template (0.75 g, 1.32ϫ10–4 mol) was dissolved in a mixture of
distilled water (29 g, 1.599 mol) and HCl (37 wt.-% aq.; 0.20 mL,
2.25ϫ10–3 mol). After addition of the premixed organosilanes,
BTEB (0.893 g or 2.22ϫ10–3 mol) and TEPPP (0.096 g or
2.46ϫ10–4 mol), the reaction mixture was kept at 50 °C for 22 h
under vigorous stirring. After hydrothermal treatment for 24 h at
100 °C in a teflon-lined stainless steel autoclave, the obtained pre-
cipitate was filtered, washed with distilled water and dried in air
overnight. Removal of the template was accomplished by extraction
of the as-synthesised solid (0.5 g) with a mixture of 37% HCl
(4.5 g) in ethanol (125 mL). The phosphane surface groups were
oxidised by using a described literature method[54] to accomplish
the HM material.
HM-Supported MoVI Catalyst (HM-Mo): The HM material
(0.175 g) was impregnated with a solution of [Mo(O)2Cl2] (0.040 g,
2.011ϫ10–4 mol) in dichloromethane (5 mL). After stirring the
mixture overnight at room temperature, HM-Mo was separated by
filtration, washed three times with dichloromethane and dried un-
der vacuum at room temperature.
Catalysis: The liquid-phase olefin epoxidation reactions were car-
ried out in air and autogenous pressure, in batch microreactors
equipped with a magnetic stirrer and a sampling valve, and im-
mersed in an oil bath heated at 55 °C. Typically, the reaction vessels
were loaded with the olefin (1.8 mmol), tBuOOH (5.5  in decane,
Preparation of the Catalysts
[MoCl2(O)2{OP(CH2CH3)(Ph)2}2] (1): Complex 1 was synthesised
by following a similar procedure to that described in the literature
by Hursthouse et al.[52] Phenylphosphane oxide (0.500 g,
2.75 mmol) and complex
1 (18 µmol) or HM-Mo (17.3 mg,
2.17 mmol) was added to a solution of [Mo(O)2Cl2] (0.200 g,
1.00 mmol) in dichloromethane (5 mL) under a nitrogen atmo-
sphere. The solution was stirred for 2 h and then left at 5 °C for
8 d to give a solid containing colourless transparent crystals
(0.653 g, yield: 99 wt.-%). The solid was separated by filtration,
washed with n-hexane and dried at room temperature.
C28H30Cl2MoO4P2 (659.33): calcd. C 51.01, H 4.59; found C 50.28,
8.1 µmol Mo): the amount of HM-Mo was chosen to have a rea-
sonable catalyst bulk density and that of complex 1 to facilitate
comparison with previously reported oxomolybdenum(VI) com-
plexes used as epoxidation catalysts. The reactions were carried out
without a cosolvent. The choice of the oxidant resides in the fact
that it is known that [Mo(O)2Cl2]-type complexes are not very
water stable because of the hydrolysis of the halide ligand in the
presence of an excess amount of water.
1
H 4.66. H NMR (300 MHz, CDCl3, r.t.): δ = 7.30–7.76 (m, 10 H,
Ph), 2.45 (9, 3JH,H = 6.00 Hz, 4 H, PCH2-), 1.13 (t, 3JH,H = 6.00 Hz,
6 H, -CH2CH3) ppm. 31P NMR (300 MHz, CDCl3, r.t.): δ =
The course of the reactions was monitored by using a Varian 3900
GC equipped with a capillary column (SPB-5, 20 mϫ0.25 mm)
and a flame-ionisation detector. For quantification of the products,
nonane or undecane was used as an internal standard (withdrawn
samples were added to diluted standard solutions): %(olefin con-
version at time t) = [(initial concentration) – (concentration at time
t)]/(initial concentration)ϫ100. The reaction products were iden-
tified by GC–MS [Trace GC 2000 Series (Thermo Quest CE Instru-
ments)/DSQ II (Thermo Scientific)] by using He as carrier gas.
49 ppm. IR (KBr pellet): ν = 901, 944 (νMo=O), 1173, 1146 (νP=O),
˜
321 (νMo–Cl), 3057 [ν(C–H)Ph], 1232–1482 [ν(C–C)Ph], 1238 [νp(CH)],
1030 [νs(CC)], 719 [νop(CH)] cm–1.
Preparation of the Hybrid Mesoporous Material (HM)
Step 1: The 1,4-bis(triethoxysilyl)benzene (BTEB) precursor was
prepared by following a procedure previously described.[31]
Step 2: The (diphenylphosphanyl)propyltriethoxysilane precursor
(TEPPP) was prepared by a modified procedure of that reported
elsewhere:[53] A solution of potassium diphenylphosphide (0.5  in
THF, 20 mL) was added dropwise to a solution of 1-chloro-3-(tri-
ethoxysilyl)propane (2.40 mL, 0.01 mol) in THF (20 mL). The mix-
ture was stirred for 3 h at 15 °C and left at room temperature over-
night. The work up of the reactions consisted of the evaporation
of THF, addition of pentane to precipitate the salt and dissolution
of the products. After filtration and evaporation of pentane, the
resulting yellow oil was distilled under reduced pressure (1 Torr) at
240 °C. C21H31O3PSi (390.53): calcd. C 64.59, H 8.00; found C
64.02, H 7.89. 1H NMR (300 MHz, CDCl3, r.t.): δ = 7.20–7.33 (m,
10 H, Ph), 3.79 (q, 3JH,H = 7.06 Hz, 6 H, OCH2CH3), 2.06 (t, 3JH,H
= 7.70 Hz, 2 H, PCH2-), 1.44–1.57 (m, 2 H, CH2CH2CH2), 1.10 (t,
Acknowledgments
The authors are grateful to the Programa Operacional Ciência e
Inovação/2010, the Fundo Europeu de Desenvolvimento Regional
and the Fundação para a Ciência e a Tecnologia for financial sup-
port (POCI/CTM/ 55648/2004 and PPCDT/CTM/55648/2004).
[1] F. Hoffmann, M. Cornelius, J. Morell, M. Fröba, Angew.
Chem. Int. Ed. 2006, 45, 3216–3251.
[2] J. H. Clark, D. J. Macquarrie, Chem. Commun. 1998, 853–860.
[3] M. Dusi, T. Mallat, A. Baiker, Catal. Rev. Sci. Eng. 2000, 42,
213–278.
3
3JH,H = 7.06 Hz, 9 H, OCH2CH3), 0.74 (t, JH,H = 7.93 Hz, 2 H,
Eur. J. Inorg. Chem. 2010, 1405–1412
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1411

A. A. Valente, P. Ferreira et al.
FULL PAPER
[4] I. W. C. E. Arends, R. A. Sheldon, Appl. Catal. A 2001, 212,
175–187.
[5] D. E. De Vos, M. Dams, B. F. Sels, P. A. Jacobs, Adv. Synth.
Catal. 2003, 345, 457–473.
[6] J.-M. Brégeault, Dalton Trans. 2003, 3289–3302.
[7] a) M. Schulz, J. H. Teles, J. Sundermeyer, G. Wahl, (BASF AG)
DE Patent 195.33.31.4, 1995; b) M. Schulz, J. H. Teles, J. Sun-
dermeyer, G. Wahl, (BASF AG) WO Patent 10054, 1995.
[8] a) H. Mimoun, I. Seree de Roch, L. Sajus, Bull. Soc. Chim. Fr.
1969, 1481–1492; b) H. Mimoun, I. Seree de Roch, L. Sajus,
Tetrahedron 1970, 26, 37–50.
[9] K. Jeyakumar, D. K. Chand, J. Chem. Sci. 2009, 121,111–123.
[10] C. Freund, W. Herrmann, F. E. Kühn, Top. Organomet. Chem.
2007, 22, 39–77.
[30] a) S. Inagaki, S. Guan, T. Ohsuna, O. Terasaki, Nature 2002,
416, 304–307; b) N. Bion, P. Ferreira, A. A. Valente, I. S. Gonç-
alves, J. Rocha, J. Mater. Chem. 2003, 13, 1910–1913.
[31] a) Y. Goto, S. Inagaki, Chem. Commun. 2002, 2410–2411; b)
Y. Goto, K. Okamoto, S. Inagaki, Bull. Chem. Soc. Jpn. 2005,
78, 932–936.
[32] S. J. Gregg, K. S. W. Sing, Adsorption, Surface Area and Poros-
ity, 2nd ed., Academic Press, London, 1982, ch. 2, p. 150.
[33] J. Blümel, Inorg. Chem. 1994, 33, 5050–5056.
[34] A. Jimtaisong, R. Luck, Inorg. Chem. 2006, 45, 10391–10402.
[35] R. Clarke, D. J. Cole-Hamilton, J. Chem. Soc., Dalton Trans.
1993, 1913–1914.
[36] W. A. Herrmann, J. J. Haider, J. Fridgen, G. M. Lobmaier, M.
Spiegler, J. Organomet. Chem. 2000, 603, 69–79.
[11] S. M. Bruno, C. C. L. Pereira, M. S. Balula, M. Nolasco, A. A. [37] F. E. Kühn, A. M. Santos, A. D. Lopes, I. S. Gonçalves, J. E.
Valente, A. Hazell, M. Pillinger, P. R. Claro, I. S. Gonçalves, J.
Mol. Catal. A 2007, 261, 79–87.
Rodríguez-Borges, M. Pillinger, C. C. Romão, J. Organomet.
Chem. 2001, 621, 207–217.
[38] S. M. Bruno, B. Monteiro, M. S. Balula, F. M. Pedro, M. Abr-
antes, A. A. Valente, M. Pillinger, P. Ribeiro-Claro, F. E. Kühn,
I. S. Gonçalves, J. Mol. Catal. A 2006, 260, 11–18.
[12] S. M. Bruno, M. S. Balula, A. A. Valente, F. A. A. Paz, M. Pill-
inger, C. Sousa, J. Klinowski, C. Freire, P. R. Claro, I. S. Gonç-
alves, J. Mol. Catal. A 2007, 270, 185–194.
ˇ
[13] Z. Petrovski, M. Pillinger, A. A. Valente, I. S. Gonçalves, A.
[39] I. S. Gonçalves, A. M. Santos, C. C. Romão, A. D. Lopes, J. E.
Rodríguez-Borges, M. Pillinger, P. Ferreira, J. Rocha, F. E.
Kühn, J. Organomet. Chem. 2001, 626, 1–10.
Hazell, C. C. Romão, J. Mol. Catal. A 2005, 227, 67–73.
[14] F. E. Kühn, A. M. Santos, A. D. Lopes, I. S. Gonçalves,
J. E. R. Borges, M. Pillinger, C. C. Romão, J. Organomet.
Chem. 2001, 621, 207–217.
[40] S. Gago, J. E. Rodríguez-Borges, C. Teixeira, A. M. Santos, J.
ˇ
Zhao, M. Pillinger, C. D. Nunes, Z. Petrovski, T. M. Santos,
[15] S. Gago, J. E. R. Borges, C. Teixeira, A. M. Santos, J. Zhao,
F. E. Kühn, C. C. Romão, I. S. Gonçalves, J. Mol. Catal. A
2005, 236, 1–6.
ˇ
M. Pillinger, C. D. Nunes, Z. Petrovski, T. M. Santos, F. E.
Kühn, C. C. Romão, I. S. Gonçalves, J. Mol. Catal. A 2005,
236, 1–6.
[41] J. Zhao, X. Zhou, A. M. Santos, E. Herdtweck, C. C. Romão,
F. E. Kühn, Dalton Trans. 2003, 3736–3742.
[16] F. E. Kühn, E. Herdtweck, J. J. Haider, W. A. Herrmann, I. S.
Gonçalves, A. D. Lopes, C. C. Romão, J. Organomet. Chem.
1999, 583, 3–10.
[17] F. E. Kühn, A. D. Lopes, A. M. Santos, E. Herdtweck, J. J.
Haider, C. C. Romão, A. G. Santos, J. Mol. Catal. A 2000, 151,
147–160.
[18] F. E. Kühn, A. M. Santos, A. D. Lopes, I. S. Gonçalves, E.
Herdtweck, C. C. Romão, J. Mol. Catal. A 2000, 164, 25–38.
[19] K. R. Jain, F. E. Kühn, Dalton Trans. 2008, 2221–2227.
[20] C. D. Nunes, A. A. Valente, M. Pillinger, A. C. Fernandes,
C. C. Romão, J. Rocha, I. S. Gonçalves, J. Mater. Chem. 2002,
12, 1735–1742.
[42] J. A. Brito, M. Gómez, G. Muller, H. Teruel, J.-C. Clinet, E.
Duñach, M. A. Maestro, Eur. J. Inorg. Chem. 2004, 4278–4285.
[43] A. Al-Ajlouni, A. A. Valente, C. D. Nunes, M. Pillinger, A. M.
Santos, J. Zhao, C. C. Romão, I. S. Gonçalves, F. E. Kühn, Eur.
J. Inorg. Chem. 2005, 1716–1723.
[44] F. E. Kühn, M. Groarke, É. Bencze, E. Herdtweck, A. Prazeres,
A. M. Santos, M. J. Calhorda, C. C. Romão, I. S. Gonçalves,
A. D. Lopes, M. Pillinger, Chem. Eur. J. 2002, 8, 2370–2383.
[45] L. F. Veiros, A. Prazeres, P. J. Costa, C. C. Romão, F. E. Kühn,
M. J. Calhorda, Dalton Trans. 2006, 1383–1389.
[46] D. V. Deubel, J. Sundermeyer, G. Frenking, Eur. J. Inorg. Chem.
2001, 1819–1827 and references cited therein.
[21] C. D. Nunes, M. Pillinger, A. A. Valente, J. Rocha, A. D.
Lopes, I. S. Gonçalves, Eur. J. Inorg. Chem. 2003, 3870–3877.
[22] P. Ferreira, I. S. Gonçalves, F. E. Kühn, A. D. Lopes, M. A.
Martins, M. Pillinger, A. Pina, J. Rocha, C. C. Romão, A. M.
Santos, T. M. Santos, A. A. Valente, Eur. J. Inorg. Chem. 2000,
2263–2270.
[23] M. Jia, W. R. Thiel, Chem. Commun. 2002, 2392–2393.
[24] M. Jia, A. Seifert, W. R. Thiel, Chem. Mater. 2003, 15, 2174–
2180.
[25] S. Tangestaninejad, M. H. Habibi, V. Mirkhani, M. Mogh-
adam, G. Grivani, Inorg. Chem. Commun. 2006, 9, 575–578.
[26] M. Jia, A. Steifert, M. Berger, H. Giegengack, S. Schulze, W. R.
Thiel, Chem. Mater. 2004, 16, 877–882.
[27] M. Jia, A. Steifert, W. R. Thiel, J. Catal. 2004, 221, 319–324.
[28] A. Castro, J. C. Alonso, P. Neves, A. A. Valente, P. Ferreira,
Eur. J. Inorg. Chem. 2010, 602–607.
[47] A. Hroch, G. Gemmecker, W. R. Thiel, Eur. J. Inorg. Chem.
2000, 1107–1114.
[48] S. Shylesh, M. Jia, A. Seifert, S. Adappa, S. Ernsta, W. R.
Thiel, New J. Chem. 2009, 33, 717–719.
[49] G. M. Sheldrick, Acta Crystallogr., Sect. A 1990, 46, 467–473.
[50] G. M. Sheldrick, SHELXL-97, Computer Program for Crystal
Structure Refinement, University of Göttingen, Göttingen,
Germany, 1997.
[51] A. L. Spek, PLATON, A Multipurpose Crystallographic Tool,
Utrecht University, Utrecht, Netherlands, 2008.
[52] M. B. Hursthouse, W. Levason, R. Ratnani, G. Reid, Polyhe-
dron 2004, 23, 1915–1921.
[53] a) R. J. Angelici, H. Gao, US Pat. 5 789 333, 1998; b) M.
Capka, Synth. React. Inorg. Met.-Org. Chem. 1977, 7, 347–354.
[54] J. P. Bezombes, C. Chuit, R. J. P. Corriu, C. Reyé, J. Mater.
Chem. 1999, 9, 1727–1734.
[29] F. R. Fronczek, R. L. Luck, G. Wang, Inorg. Chim. Acta 2003,
342, 247–254.
Received: November 10, 2009
Published Online: February 12, 2010
1412
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 1405–1412