Preparation and structural characterization of polymer-supported methylrhenium trioxide systems as efficient and selective catalysts for the epoxidation of olefins

Article abstract of DOI:10.1021/jo011033f

Novel heterogeneous compounds of methylrhenium trioxide (MTO) were prepared with poly(4-vinylpyridine) and polystyrene as polymeric supports. The wide-angle X-ray diffraction (WAXS.) analysis, performed by the application of the difference method, showed, in a representative case of the poly(4-vinylpyridine)/MTO derivatives, a slightly distorted octahedral conformation on the metal's primary coordination sphere. The Re-O and Re-C bond distances were not influenced by the polymeric nature of the ligand, while the Re-N bond distance was abnormally shorter than those previously observed for homogeneous MTO/L_n complexes, showing a strong coordination of the rhenium atom to the support. A set of scanning electron microscopy (SEM) photographs showing the morphology of the surface of particles of poly(4-vinylpyridine)/MTO and polystyrene/MTO systems are reported. The reticulation grade of the polymer was a crucial factor for the morphology of the particles surface. Poly(4-vinylpyridine) 2% cross-linked systems were characterized by particles with very irregular shape and surface. Poly(4-vinylpyridine) 25% cross-linked systems showed particles with regular spherical shape, which morphology was similar to microcapsules obtained with polystyrene. All novel MTO compounds were efficient and selective heterogeneous catalysts for the epoxidation of olefins using environmentally friendly H₂O₂ as oxygen atom donor. The catalyst activity was maintained for at least five recycling experiments.

Full text of DOI:10.1021/jo011033f

J . Org. Chem. 2002, 67, 1323-1332
1323
P r ep a r a tion a n d Str u ctu r a l Ch a r a cter iza tion of
P olym er -Su p p or ted Meth ylr h en iu m Tr ioxid e System s a s Efficien t
a n d Selective Ca ta lysts for th e Ep oxid a tion of Olefin s
Raffaele Saladino,*,† Veronica Neri,† Anna Rita Pelliccia,† Ruggero Caminiti,‡ and
Claudia Sadun‡
Dipartimento A.B.A.C, Istituto Nazionale di Fisica della Materia, Universita` della Tuscia, Via S. Camillo
de Lellis s.n.c., 01100, Viterbo, Italy, and Dipartimento di Chimica, Istituto Nazionale di Fisica della
Materia, Universita` degli studi di Roma “La Sapienza”, p.le Aldo Moro 5, 00185 Rome, Italy
saladino@unitus.it
Received October 29, 2001
Novel heterogeneous compounds of methylrhenium trioxide (MTO) were prepared with poly(4-
vinylpyridine) and polystyrene as polymeric supports. The wide-angle X-ray diffraction (WAXS.)
analysis, performed by the application of the difference method, showed, in a representative case
of the poly(4-vinylpyridine)/MTO derivatives, a slightly distorted octahedral conformation on the
metal’s primary coordination sphere. The Re-O and Re-C bond distances were not influenced by
the polymeric nature of the ligand, while the Re-N bond distance was abnormally shorter than
those previously observed for homogeneous MTO/Ln complexes, showing a strong coordination of
the rhenium atom to the support. A set of scanning electron microscopy (SEM) photographs showing
the morphology of the surface of particles of poly(4-vinylpyridine)/MTO and polystyrene/MTO
systems are reported. The reticulation grade of the polymer was a crucial factor for the morphology
of the particles surface. Poly(4-vinylpyridine) 2% cross-linked systems were characterized by
particles with very irregular shape and surface. Poly(4-vinylpyridine) 25% cross-linked systems
showed particles with regular spherical shape, which morphology was similar to microcapsules
obtained with polystyrene. All novel MTO compounds were efficient and selective heterogeneous
catalysts for the epoxidation of olefins using environmentally friendly H2O2 as oxygen atom donor.
The catalyst activity was maintained for at least five recycling experiments.
In tr od u ction
insertion into C-H bonds.10 Among them, the epoxidation
of olefins has been extensively studied.11The role of
MTO in these oxidations has been understood by the
characterization of its reaction products with hydrogen
peroxide (H2O2). The active catalytic forms are a mono-
peroxo metal [MeRe(O)2(O2)] (I) and a bisperoxo metal
[MeRe(O)(O2)2] (II) complexes and/or their adducts with
solvent molecules.12 A detailed structure of the 1:1 (II)/
H2O/diglyme (diethylene glycol dimethyl ether) adduct,
in which two η2-coordinated peroxo ligands are nearly
coplanar with the Re-C bond, has been rigorously
established by single-crystal X-ray analysis.12 Theoretical
studies have been performed to elucidate the reactivity
of I and II toward olefins and the geometrical features
of the transition state of the epoxidation.13
In the past few years methylrhenium trioxide
(CH3ReO3, MTO)1 has become an important catalyst
for its high efficiency and selectivity in a variety of
synthetic transformations.2 The advantages of MTO in
oxidative reactions have been well-documented in the
oxidation of olefins,3 alkynes,4 aromatic derivatives,5
sulfur compounds,6 phosphines,7 organonitrogen com-
pounds,8 Baeyer-Villiger rearrangment,9 and oxygen
* Corresponding author. Phone 39(0)761357284. Fax 39(0)761357242.
† Universita` della Tuscia.
‡ Universita` degli studi di Roma “La Sapienza”.
(1) Beattie, I. R.; J ones, P. J . Inorg. Chem. 1979, 18, 2318.
(2) Roma˜o C. C.; Ku¨hn, F. E.; Herrmann W. A. Chem. Rev. 1997,
97, 3197-3246.
(3) Herrmann W. A.; Fischer, R. W.; Marz, D. W. Angew. Chem.,
Int. Ed. Engl. 1991, 30, 1638.
(4) Zhu, Z.; Espenson, J . H. J . Org. Chem. 1995, 60, 7728.
(5) (a) Adam, W.; Herrmann W. A.; Lin J .; Saha-Moller C. R. J . Org.
Chem 1994, 59, 8281-8283. (b) Adam, W.; Herrmann W. A.; Lin J .;
Saha-Moller C. R.; Fischer, R. W., Correira, J . D. G. Angew. Chem.,
Int. Ed. Engl. 1994, 33, 2475-2477. (c) Adam, W.; Herrmann W. A.;
Saha-Moller C. R.; Shimizu M. J . Mol. Catal. A. 1995, 97, 15-20. (d)
Saladino, R.; Neri, V.; Mincione, E.; Marini, S.; Coletta, M.; Fiorucci,
C.; Filippone, P. J . Chem. Soc., Perkin Trans. 1 2000, 581-586.
(6) (a) Huston, P.; Espenson, J . H.; Bakac, A. Inorg. Chem. 1993,
32, 4517. (b) Brown, K. N.; Espenson, J . H. Inorg. Chem. 1996, 35,
7211.
Although effective for numerous substrates, the most
important drawback of this reaction is the concomitant
ring opening of the oxiranyl ring to form trans-1,2-diols.14
(9) (a) Herrmann W. A.; Fischer, R. W., Correira, J . D. G. J . Mol
Catal. 1994, 94, 213-223. (b) Bernini, R.; Mincione, E.; Cortese, M.;
Aliotta, G.; Saladino, R. Tetrahedron Lett. 2001, 42/ 32, 5401-5404.
(10) (a) Murray, R. W.; Iyanar, K.; Chen, J .; Wearing, J . T.
Tetrahedron Lett. 1995, 36, 6415. (b) Schuchardt, U.; Mandelli, D.;
Shul′pin, G. B. Tetrahedron Lett. 1996, 37, 6487-6490.
(11) Al-Ajlouni, A. M.; Espenson, J . H. J . Org. Chem. 1996, 61,
3969-3976.
(12) Herrmann W. A.; Fischer, R. W.; Scherer, W.; Rauch, M. U.
Angew. Chem., Int. Ed. Engl. 1993, 32, 1157-1160.
(13) (a) Wu, Y.-D.; Sun, J . J . Org. Chem. 1998, 63, 1752-1753. (b)
Gisdakis, P.; Antonczak, S.; Ko¨stlmeier, S.; Herrmann W. A.; Ro¨sch,
N. Angew. Chem., Int. Ed. 1998, 37, 2211.
(7) Abu-Omar, M. M.; Espenson, J . H. J . Am. Chem. Soc. 1995, 117,
272.
(8) (a) Zhu, Z.; Espenson, J . H. J . Org. Chem. 1995, 60, 1326. (b)
Murray, R. W.; Iyanar, K.; Chen, J .; Wearing, J . T. Tetrahedron Lett.
1996, 37, 805-808. (c) Goti, A.; Nannelli, L. Tetrahedron Lett. 1996,
37, 6025-6028.
10.1021/jo011033f CCC: $22.00 © 2002 American Chemical Society
Published on Web 01/22/2002

1324 J . Org. Chem., Vol. 67, No. 4, 2002
Saladino et al.
Two general procedures have been developed to avoid this
problem. The first procedure requires an anhydrous reac-
tion medium, which is usually obtained by the use of
nonaqueous media and dried H2O2 in tert-butyl alcohol,3
or urea/hydrogen peroxide adduct (UHP) in CH2Cl2.15 In
the case of UHP, the helical urea channels may serve as
an host confined stabilizing environment for MTO.16
Recently, the MTO/UHP protocol has been applied for
the synthesis of sensitive 5,6-oxiranyl-5,6-dihydrouracils,
which are of particular biological significance because
responsible for the formation of protein-nucleic acids
cross-links.17 The second procedure requires the use of
low molecular weight compounds bearing one or more
nitrogen atoms as mediators of the oxidation.18 Lewis
base adducts of MTO, of general formula MTO/Ln (where
L ) ligand, n ) 1 or 2), prepared by reaction with
pyridine,19 pyridine derivatives,20 quinuclidine,21 aniline,
toluidine, and pyrazole,22 decrease the formation of diols
mainly as a consequence of the reduced Brønsted and
Lewis acidity of the catalyst system. The structures of
these MTO/Ln complexes have been characterized by
means of single-crystal X-ray, neutron, or electron dif-
fraction. All show a slightly distorted tetrahedral geom-
etry with the methyl group and the ligand in a trans-
configuration to each other in the apical positions.2,22
Moreover, it was found that a biphasic system and excess
of pyridine not only prevent the formation of diols but
also increase the reaction rate in comparison to MTO.19,23
Herrmann and co-workers extended in a patent24 the
“mediator” concept by the preparation of heterogeneous
MTO compounds, of general formula (polymer)f /(MTO)g
(the f/g quotient expresses the ratio by weight of the two
components), in which MTO was supposed to be bonded
to the support by coordination with only one nitrogen
atom in analogy with the homogeneous systems. Unre-
ticulated poly(4-vinylpyridine), poly(2-vinylpyridine), poly-
(vinylpyrrolidone), poly(acrylamide), and nylon 6 were
used as organic supports. The heterogeneous rhenium
compounds were applied for the epoxidation of cyclohex-
ene, methyl oleate, and allyl alcohol with dried H2O2 in
tert-butyl alcohol. Unfortunately, low yields of cyclohex-
ene oxide and 1,2-epoxypropane-3-ol (27% and 20%,
respectively) were obtained.24 The structures of these
(polymer)f /(MTO)g systems were not completely charac-
terized (only the elemental analysis was reported), and
useful structure activity relationships are still lacking.
Apart from a silica supported MTO complex of γ-(2,2′-
dipiridyl)aminopropylpolysiloxane,25 and a NaY zeolite/
MTO supercage system,26 no further data are available
in the literature about heterogeneous MTO catalysts. The
heterogeneization of MTO by using a polymeric support
is an important tool because it allows an easier recovery
of the catalyst and, sometimes, may improve the reactiv-
ity.27
We report here the preparation of a novel family of
poly(4-vinylpyridine)/MTO systems by the use of poly(4-
vinylpyridine) 25% cross-linked (with divinylbenzene)
(PVP-25%), and poly(4-vinylpyridine-N-oxide) 2% cross-
linked (PVPN-2%) as supports. Poly(4-vinylpyridine) 2%
cross-linked (PVP-2%) was also used as a reference
system. Since the reactivity of polymer catalysts having
coordinate bonds can be low, an unprecedented prepara-
tion of microencapsulated MTO,28 with polystyrene 2%
cross-linked (with divinylbenzene) (PS-2%) or a mixture
of PS-2% and PVP-2% (optimal ratio 5:1), is also studied.
The novel (polymer)f /(MTO)g systems have been charac-
terized by infrared spectroscopy and scanning electron
microscopy (SEM). In a representative case, the PVP-
2%/MTO complex, a detailed structure of the local
geometry of the rhenium atom was obtained for the first
time by an Energy-Dispersive X-ray diffraction (EDXD)
analysis. The PVP-2%/MTO, PVP-25%/MTO, PVPN-2%/
MTO, and PS-2%/MTO catalysts have been used for the
efficient and selective epoxidation of olefins with H2O2
(35% water solution) as environmentally friendly oxidant.
The role on the reaction pathway of several parameters,
such as the loading factor, the reticulation grade of the
support, the shape and dimension of particles, the
coordinate versus coating immobilization, and the nature
of the heterogeneous atom bonded to rhenium, is also
described.
P r ep a r a tion a n d Ch a r a cter iza tion of P oly(4-vi-
n ylp yr id in e)/MTO a n d P olystyr en e/MTO Ca ta lyst
System s. PVP-2% 1, PVP-25% 2, and PS-2% 7 were
obtained from a commercial source (Aldrich). PVPN-2%
3 was obtained by oxidation of 1 with an excess of
3-chloroperbenzoic acid (m-CPBA) in ethanol (EtOH) at
room temperature.29 The preparation of PVP-2%/MTO 4,
PVP-25%/MTO 5, and PVPN-2%/MTO 6 complexes was
performed by a modification of the procedure described
by Herrmann.24 Compounds 1, 2, 3 were suspended in
ethanol at room temperature, and to these solutions was
added the appropriate amount of powdered MTO under
nitrogen atmosphere (as represented in the schematic
drawing in Scheme 1).
(14) Gansa¨uer, A. Angew. Chem., Int. Ed. Engl. 1997, 36, 2591-
2592.
(15) Adam, W.; Mitchell, C. M. Angew. Chem., Int. Ed. Engl. 1996,
35, 533-535.
The mixture was stirred for 1 h at room temper-
ature and then cooled to 0 °C. During this period
polymers changed from colorless to bright yellow, with
the exception of 3 that shows only a weak color. The
mixture was filtered and the solid residue washed with
ethyl acetate several times. No traces of MTO were found
in the organic phase. Under otherwise specified, poly(4-
vinylpyridine)/MTO complexes were prepared with a
loading factor (that is mmol of MTO for 1 g of support)
(16) Adam, W.; Mitchell, C. M.; Saha-Moller C. R.; Weichold, O. J .
Am. Chem. Soc. 1999, 121, 2097-2103.
(17) Saladino, R.; Carlucci, P.; Danti, M. C.; Crestini, C.; Mincione,
E. Tetrahedron 2000, 56, 10031-10037.
(18) Herrmann W. A.; Fischer, R. W.; Rauch, M. U.; Scherer, W. J .
Mol. Catal. 1994, 86, 243-266.
(19) Herrmann W. A.; Ku¨hn, F. E.; Mattner, M. R.; Artus, G. R. J .;
Geisberger, M.; Correira, J . D. G. J . Organomet. Chem. 1997, 538, 203.
(20) (a) Cope`ret, C.; Adolfsson, H.; Sharpless, K. B. J . Chem. Soc.,
Chem. Commun. 1997, 1915. (b) Herrmann W. A.; Kratzer, R. M.; Ding,
H.; Glas, H.; Thiel, W. R. J . Organomet. Chem. 1998, 555, 293.
(21) Herrmann W. A.;.Correira, J . D. G.; Rauch, M. U.; Artus, G.
R.; Ku¨hn, F. E. J . Mol. Catal. 1997, 118, 33.
(25) Wang, T.-J .; Li D.-C.; Bai, J .-H.; Huang, M.-Y.; J iang, Y.-Y. J .
Macromol. Sci., Pure Appl. Chem. 1998, A35, 531-538.
(26) Adam, W.; Saha-Mo¨ller, C. R.; Weichold, O. J . Org. Chem. 2000,
65, 2897-2899.
(27) Thomas, J . M.; Thomas W. J . Principles and Practice of
heterogeneous catalysis; VCH: New York, 1997.
(22) Ku¨hn, F. E.; Santos, A. M.; Roesky, P. W.; Herdtweck, E.;
Scherer, W.; Gisdakis, P.; Yudanov, I. V.; Di Valentin, C.; Ro¨sch, N.
Chem. Eur. J . 1999, 5, 3603-3615.
(23) Rudolph, J .; Reddy, K. L.; Chiang, J . P.; Sharpless, K. B. J .
Am. Chem. Soc. 1997, 119, 6189.
(24) Herrmann, W. A.; Fritz-Meyer-Weg, D. M.; Wagner, M.; Kuchler,
J . G.; Weichselbaumer, G.; Fischer, R. U.S. Patent 5,155,247 (Oct. 13,
1992).
(28) Donbrow, M. Microcapsules and Nanoparticles in Medicine and
Pharmacy; CRC Press: Boca Raton, FL, 1992.
(29) Ochiai, E. Aromatic Amine Oxides; Elsevier: Amsterdam, 1967.

Heterogeneous Catalysts of Methylrhenium Trioxide
J . Org. Chem., Vol. 67, No. 4, 2002 1325
Ta ble 1. ν(ReO) Vibr a tion a l F r equ en cies of P VP /MTO
Sch em e 1
Com p lexes a n d P S/MTO Micr oca p su lesa
compound
ν(ReO) cm-1
PVP-2%/MTO 4
963, 923
964, 932
924
953, 911
960, 921
PVP-25%/MTO 5
PVPN-2%/MTO 6
PS-2%/MTO 9
PS-2%(PVP-2%)/MTO 10
a
Data recovered for solid KBr samples at room temperature.
polybutadiene, only 11% of MTO was observed in the
polyethylene system as revealed by elemental analysis.32
FT-IR spectroscopy appeared to be a useful tool to
evaluate the electronic properties of the ReO3 fragment
in poly(4-vinylpyridine)/MTO complexes and polystyrene/
MTO microcapsules since the ν(ReO) vibrational frequen-
cies are highly characteristic of alkyltrioxorhenium (VII)
compounds. On the basis of the theoretical valence-bond
resonance structures for the Re-O bond, the ν(ReO)
vibration is expected to shift to lower frequencies in the
presence of electron-donating ligands with respect to
electron-withdrawing ligands.2 The assignment of the
vibration modes of the ReO3 fragment in several R-ReO3
and R-ReO3/ligand(s) has been previously reported in
the literature,33 also by inelastic neutron scattering (INS)
and high-resolution FT-Raman analyses.34 To the best
of our knowledge, there are no reports dealing with the
FT-IR data of polymer-supported MTO systems. Table 1
gives the values recovered for ν(ReO) stretching vibration
frequencies of compounds 4, 5, 6, 9, and 10, as revealed
by solid-state FT-IR (KBr) analyses.
Sch em e 2
The ν(ReO) frequencies of polymer-supported MTO
systems display values that are lower than that found
for MTO, equal to 998 cm-1 for the symmetric stretching
vibration mode, and 959 cm-1, for the asymmetric
stretching vibration mode.34 These values are close to
that reported for homogeneous low molecular weight
MTO complexes. On the basis of these data it is reason-
able to suggest that also in the case of polymer-supported
MTO systems, the π-bonded hydrocarbon ligands and
Lewis bases ligands lower the Re-O bond order, probably
as a result of their electron-donating abilities. Moreover,
compound 9 shows ν(ReO) frequencies that are lower
with respect to 4-6, in accord with the order of increasing
Re-O bond order previously described for Cp*ReO3
(Cp* ) cyclopentadienyl), MTO/bipyridine, PhReO3,
and free MTO.35 PS-2%(PVP-2%)/MTO 10 displays an
intermediate value.
A set of scanning electron microscopy (SEM) photo-
graphs showing the morphology of the surface of particles
of poly(4-vinylpyridine)/MTO and polystyrene/MTO sys-
tems are reported in Figures 1-5. Compound 4 shows
particles of different shapes, with very irregular surfaces,
and characterized by a scaffold of an average size of the
order of 100 µm (Figure 1). At a largest magnification,
equal to 0.5. Compounds 4, 5, and 6 were used without
any further purification. The preparation of PS-2%/MTO
9 and PS-2%/PVP-2%/MTO 10 was performed by the
microencapsulation technique.28 Polystyrene microcap-
sules have been used for coating and isolating Lewis acids
on the basis of physical envelopment by the polymer and
on electron interactions between the π electrons of the
benzene ring and the metal center.30 The size of micro-
capsule achievable has been reduced from a few microme-
ters to nanometers.28 Compound 7, or a mixture of 7 and
1 (ratio 7:1 ) 5:1), namely compound 8, was suspended
at room temperature in tetrahydrofuran (THF, 5 mL),
which is a solvent with high swelling properties toward
polystyrene (9.9 cal cm-3).31 Under these experimental
conditions MTO might enter the enlarged mesh structure
of the polymer. To this mixture the appropriate amount
of powdered MTO was added as a solid core under
nitrogen atmosphere at room temperature (as repre-
sented in the schematic drawing in Scheme 2).
Grumes were found in suspension in the mixture, and
hexane (10 mL) was added to harden the capsule walls.
The mixture was stirred during 1 h, and in this period
the mixture changed from colorless to bright blue. No
traces of MTO were found in the organic phase after the
workup of the reaction. The polystyrene/MTO microcap-
sules were prepared with a loading factor equal to 0.5.
Compounds 9 and 10 were used without any further
purification. Control experiment using polybutadiene or
polyethylene as polymeric matrixes were performed.
While 95% of MTO (100% ) the amount of MTO mi-
croencapsulated on polystyrene) was immobilized by
(31) Polo, E.; Amadelli, R.; Parassiti, V.; Maldotti, A. Inorg. Chim.
Acta 1992, 192, 1.
(32) Even if we have not studied in detail the oxidation state of the
rhenium atom into microcapsules, the possibility of electron transfer
processes to Re(VI), which is consistent with the blue color of the
catalyst and with the significantly lower value of the R-O IR
frequencies, cannot be completely ruled out.
(33) Mink, J .; Keresztury, G.; Stirling, A.; Herrmann, W. A. Spec-
trosc. Chim. Acta 1994, 50A, 2039.
(34) Parker, S. F.; Herman, H. Spectrochimica Acta 2000, 56A,
1123-1129.
(35) Herrmann, W. A.; Roesky, P. W.; Scherer, W.; Kleine, M.
Organometallics 1994, 13, 4536.
(30) Kobayashi, S.; Nagayama, S. J . Am. Chem. Soc. 1998, 120,
2985-2986.

1326 J . Org. Chem., Vol. 67, No. 4, 2002
Saladino et al.
F igu r e 1. Scanning electron microscopy (SEM) micrograph
of PVP-2%/MTO catalyst.
the surface appears defined by two different kinds of
morphology: a partially fused smooth sheet-paving, and
a chaotic aggregate of fusiform small “rounded-up”cap-
sules adhered to each other, probably due to the small
size of the core. Control experiments were performed with
the aim to evaluate if the incorporation of MTO itself
causes a change in the morphology of the polymer. The
comparison between the electron microscopy photographs
of 4 and PVP-2% alone not show appreciable differences.
These data is in accord with the low value of the loading
factor used in the preparation of the polymer.
A different morphology was observed for compound 5.
In this case, particles are characterized by a very regular
spherical shape, with an average value of diameter of the
order of 500 µm (Figure 2, panel a). A low number of
irregular fragments was observed, which are probably
formed by a mechanical damage of particles during the
preparation of the sample. At a largest magnification the
particles show a dense sheet paving, apparently formed
by a compact aggregate of small grumes, which is
partially covered by irregular craters. When a cross-
section of particles was observed at large magnification,
the core appeared to be formed by several superimposed
layers, in which sheaves of fusiform fibers move from the
center of particle to the surface. Channels, cavities, and
small irregular grumes are distributed around and over
the fibers (Figure 2, panel b).
F igu r e 2. Panel a: Scanning electron microscopy (SEM)
micrograph of PVP-25%/MTO catalyst. Panel b: cross-section
at largest magnification.
The morphology of 6 appears to be similar to that
previously observed for 4. Particles are of different
shapes, irregular surface, and with an average scaffold
dimension of the order of 200 µm (Figure 3). Two main
surface motifs are clearly evident; a compact and smooth
sheet-paving and several “rounded-up” capsules partially
fused on the sheet-paving of the particle.
As expected, compound 9, which is a microencapsu-
lated MTO system, is composed by regular spherical
microcapsules, with an average value of diameter of the
order of 100 µm (Figure 4). Also in this case, a low
number of irregular fragments was observed.
Note that 10 is characterized by regular spherical
microcapsules (Figure 5). Because of 4 shows particles
of very irregular morphology, the spherical shape of
particles of 10 is probably due to the efficient microen-
capsulation of the PVP-2% and MTO by PS-2%.
In the poly(4-vinylpyridine) family, a low value of the
reticulation grade (2% cross-link with divinyl benzene)
affords particles that are of different shapes and very
F igu r e 3. Scanning electron microscopy (SEM) micrograph
of PVPN-2%/MTO catalyst.
irregular surfaces. In the presence of an increased value
of the reticulation grade (25% cross-link with divinyl
benzene), the transformation of irregular “spongy” par-
ticles to regular spherical particles is observed. Prelimi-

Heterogeneous Catalysts of Methylrhenium Trioxide
J . Org. Chem., Vol. 67, No. 4, 2002 1327
applied also to solid systems,41 for treating experimental
WAXS data, assuming that polymer structure is not
altered by methylrhenium trioxide (MTO) insertion,
when 4 was prepared.
In this way, the radial distribution function of PVP-
2%, the reference system, can be subtracted from the
radial distribution function of 4. By doing so, the contri-
butions of MTO-polymer and MTO-MTO interactions
remain.
The difference method of neutron diffraction and
isotopic substitution (NDIS) has been applied by R. H.
Tromp and G. W. Neilson42 to aqueous (D2O) poly-
(styrene-sulfonate) ion-exchange resins with Ni(II) and
Li(I).
The X-ray diffraction experiments were carried out
with a noncommercial X-ray energy scanning diffrac-
tometer.43,44 The diffractometer consists of a Seifert X-
ray generator equipped with a water-cooled tungsten
X-ray source, with 3.0 kW maximum power. The brems-
strahlung component of the X-ray source was used. A
Germanium solid-state detector (SSD) connected to a
multichannel analyzer carried out the diffraction spectra
collection by an electronic chain. A collimating system
positioned in front of and behind the sample focuses the
X-ray beam. Two-step motors for moving the arms
supporting the source and the detector allow the angular
positioning. An adjustable sample holder was used to
position the sample in the optical center of the diffrac-
tometer.
F igu r e 4. Scanning electron microscopy (SEM) micrograph
of PS-2%/MTO catalyst.
The spectra were recorded at different fixed θ to obtain
the complete q range. The function q can be written
q ) 4π sin θ/λ ) Ek sin θ
where q is expressed in (Å-1) and E in (keV), and the
value of the constant k is 1.014.
Operating conditions were as follows: high voltage
supply: ) 50 kV, current ) 35 mA, total power ) 1750
W; energy range ) 12.4 ÷ 50 kV; measurement angles
(Θ) ) 21.0°, 10.5°, 8.0°, 3.5°, 2.0°, 1.0°, 0.5°; scattering
F igu r e 5. Scanning electron microscopy (SEM) micrograph
of PS-2%/PVP-2%/MTO catalyst.
parameter range q ) 0.110 ÷ 16.0 Å-1
.
nary results obtained by an energy-dispersive X-ray
(EDX) map of 4 showed that MTO is located efficiently
over the polymer surface. Thus, it is reasonable to suggest
that the polymer morphology may, at least in part, affect
the behavior of the catalysts, mostly in the case of
catalysts that are different only for the value of the
reticulation grade. Polystyrene forms microcapsules that
are characterized by an average value of the diameter
that is smaller than that determined for the PVP-25%/
MTO system. Finally, 10 microcapsules were obtained
whatever the reticulation grade of the polyvinyl pyridine
component.
It is necessary correct the experimental data45,46,47
normalizing to the incident radiation intensity and
dividing by X-ray absorption and polarization coefficients.
Other corrections are the escape peaks suppression and
elimination of contributions due to the Mylar cell and
inelastic scattering from the observed intensities I (E,
Θ).
(37) Atzei, D.; De Filippo, D.; Rossi, A.; Caminiti, R.; Sadun, C. Inorg.
Chim. Acta 1996, 248 (2), 203.
(38) Atzei, D.; Sadun, C.; Pandolfi, L. Spectrochimica Acta 2000,
56A, 531.
En er gy-Disp er sive X-r a y Diffr a ction An a lysis.
Wide-angle X-ray diffraction (WAXS) was used to deter-
mine the local geometry of the rhenium atom in com-
pound 4. This analysis is particularly well suited for
providing insight into the way metal ions interact with
their surroundings. The coordination geometry of several
metal ions in amorphous state has been studied using
this technique.36-38 The major difficulties in determining
the coordination structure of a metal ion with a polymer
system lie in its highly complex amorphous structure.
Here we used the difference method,39,40 successfully
(39) Gontrani, L.; Caminiti, R.; Bencivenni, L.; Sadun, C. Chemical
Phys. Lett. 1999, 301 (1-2), 131-137.
(40) Caminiti, R.; Carbone, M.; Panero, S.; Sadun, C. J . Phys. Chem.
1999, 103 (4), 10348-10355.
(41) Atzei, D.; Ferri, T.; Sadun, C.; Sangiorgio, P.; Caminiti, R. J .
Am. Chem. Soc. 2001, 123 (11), 2552-2558.
(42) Tromp, R. H.; Neilson, G. W. J . Phys. Chem. 1996, 100, 7380-
7383.
(43) (a) Caminiti, R.; Sadun, C.; Rossi, V.; Cilloco, F. Felici, R.
Presented at the XXV Italian Congress on Physical Chemistry,
Cagliari, Italy, 1991. (b) Caminiti, R.; Sadun, C.; Rossi, V.; Cilloco, F.
Felici, R. Patent No. RM/93 01261484 J une 23, 1993.
(44) Ballirano, P.; Caminiti, R.; Ercolani, C.; Maras, A.; Orru`, A. J .
Am. Chem. Soc. 1998, 120 (49), 12798-12807.
(45) Nishikawa, K.; Iijima, T. Bull. Chem. Soc. J pn. 1984, 57, 1750.
(46) Fritsch, G.; Keimel, D. A. Mater. Sci. Eng. 1991, 134 A, 888.
(47) Carbone, M.; Caminiti, R.; Sadun, C. J . Mater. Chem. 1996, 6,
1709.
(36) Atzei, D.; De Filippo, D.; Rossi, A.; Caminiti, R.; Sadun, C.
Spectrochimica Acta 1995, 51 (1) A, 11.

1328 J . Org. Chem., Vol. 67, No. 4, 2002
Saladino et al.
F igu r e 6. Experimental structure functions in the qi(q)M(q)
form: (a) ReO3CH3/PVP-2%, (b) PVP-2%.
F igu r e 7. Experimental radial distribution functions in the
Diff(r) ) D(r) - 4πr2F0 form: (a) ReO3CH3/PVP-2%, (b) PVP-
2%.
Atomic scattering factors, fh(s), were taken from In-
ternational Tables.48 The static structure function
polymer structure, but there are some minor differences
in the main short-range peaks, due to the metal coordi-
nation sphere, which will be described in detail.
The structures of MTO in solid crystal state have been
studied elsewhere.49,50 A complete structural study of the
PVP-2%/MTO complex by X-ray diffraction would be very
difficult and time-consuming. This is because the first
coordination distances of the metal ions fall in the range
of the second neighbors of the polymer. They are not
dependent on amorphous resin structure, whereas the
distances of the second and third coordination sphere are
closely related to the resin structure. Thus, it would be
necessary to determine the complete polymer structure
in order to identify which atoms belong to the second
coordination sphere of the metal.
We have assumed that MTO only induce minor modi-
fications in the polymer structure because no substantial
differences in the structure functions of PVP-2% were
detected. Thus, we are quite confident in the validity of
the difference method. The difference method that was
previously applied to the liquid systems, recently, was
also applied to solid systems.41 In a liquid system, the
addition of a small amount of solute to obtain a diluted
solution does not alter its long- and medium-range
structure. Thus, the solute-solute and solute-solvent
interactions can be derived by subtracting the structure
function of the solvent from that of the solution. In the
present case, the PVP-2% has been considered as a
solvent. Therefore, we have subtracted a weighted amount
of the experimental PVP-2% structure function to the
experimental curves of 4. In so doing we can focus our
attention on all new interactions present, namely MTO-
polymer and MTO-MTO interactions. The use of the
difference method was necessary because of the great
complexity of the polymer structure and of the small
proportion of MTO compared to the resin (0.5 loading
factor).
i(q) ) Icoh(E) - ∑f 2h(q)
was then calculated.
The radial distribution D(r) is
Smax
D(r) ) 4πr2F0 + 2rπ-1
qI(q)M(q)sin(rq)dq
∫
0
where q is the scattering parameter, F0 the sample’s
average electronic density (F0 ) (∑hnh fh(0))2V-1), V the
stoichiometric unit volume chosen and M(q) a modifica-
tion function defined by
f 2Re(0)/f 2Re(q) exp(-0.01q2)
The experimental radial distribution functions are
shown also as Diff(r) ) D(r) - 4πr 2F0. Theoretical peaks
were calculated by Fourier transformation of the theo-
retical intensities for the pair interactions,
iij ) ∑fifj sin(rij q)(rij q)-1 exp(-1/2σ2ij q2)
using the same sharpening function as for the experi-
mental data and assuming the root-mean-square varia-
tion over distance to be σij. The equipment and technique
used are described in detail elsewhere.43,46
The experimental structure functions show the typical
behavior of an amorphous sample. The curves are very
similar, exhibiting a similar oscillation period and am-
plitude. This suggests that the MTO inclusions do not
modify the short-range order of polymer structure. The
functions, in the range 0-16 Å-1, for PVP-2% and 4, are
shown in Figure 6.
The experimental radial distribution function of PVP-
2%, in the Diff(r) form, in the range 0-16 Å (Figure 7)
shows peaks at about 1.45 and 2.45 Å and two large
oscillations at greater distances.
The first peaks at 1.45 and 2.45 Å can be attributed to
interactions in the polymer molecule. They represent the
distances between the directly bound atoms and between
two nondirectly bound atoms. These nonbonded dis-
tances, like those between atoms 1-3 in the polymer
chain or in the benzene ring of the polymer matrix, do
not depend on molecular conformation.
In Figure 7, it is also shown the experimental radial
distribution function in the Diff(r) form for compound 4.
The presence of MTO does not affect the long-range
We have subtracted the weighted PVP-2% distribution
radial function, D(r) curve, Figure 8b, from the respective
D(r) curve of 4 (Figure 8, panel a). The resulting curve
isolates the MTO-polymer and MTO-MTO interactions,
and it is devoid of contributions of the polymer amor-
phous structure. The difference curves show three main
peaks located at about 1.75, 2.35, and 3.20 Å (Figure 8,
panel b).
(49) Takacs, J .; Kiprof, P.; Riede, J .; Herrmann, W. A. Organome-
tallics 1990, 9, 782-787.
(50) Roma˜o, C. C.; Khu¨n, F. E.; Herrmann, W. A. Chem. Rev. 1997,
3197-3246.
(48) International Tables for X-ray Crystallography; Kynoch Press:
Birmingham, 1974; Vol. 4.

Heterogeneous Catalysts of Methylrhenium Trioxide
J . Org. Chem., Vol. 67, No. 4, 2002 1329
F igu r e 8. (a) Experimental radial distribution function D(r) of ReO3CH3/PVP-2%. (b) Weighted experimental radial distribution
function D(r) of PVP-2%. (c) Difference curve for the two radial distribution functions, D(r)ReO
- D(r)PVP-2% [solid],
CH /PVP-2%
3
theoretical molecular peak shapes for the shown model [red circle], difference curve for the radial d3istribution function and the
theoretical molecular peak shapes [blue square]. (d) Balls and sticks view of the PVP-2% coordination to ReO3CH3.
First, we analyzed the first peaks of the curves. They
provide information on the metal’s first neighbor interac-
tions, i.e., on the bond distances, on the coordination
numbers, and on the interaction types, metal-oxygen or
metal-nitrogen. We found that the first peak at 1.75 Å
was fitted by three metal-oxygen interactions, of 1.70
Å. The second peak is due to the interactions Re-N and
Re-CH3, which respectively fall at 2.35 and 2.10 Å. The
experimental peak is fitted by two metal-nitrogen
contributions and one metal-carbon. Six atoms assuming
an octahedral conformation surround the metal. So far
the interactions between the atoms belonging to the first
coordination sphere of the metal contribute to the third
peak at 3.20 Å. The theoretical peak shape function for
this model is compared with the experimental difference
radial distribution function in Figure 8c.
The values of the Re-O and Re-C bond distances
found for 4 are in accord with the mean values previously
reported for MTO/Ln complexes.49,50 Thus, the Re-O and
Re-C bond distances are nearly not influenced by the
polymeric nature of the ligand. On the other hand, the
Re-N bond distances were found abnormally shorter
than that observed for MTO/Ln complexes showing a
strong coordination of the rhenium atom by the two
nitrogens. The model presented in Figure 8 refers to one

1330 J . Org. Chem., Vol. 67, No. 4, 2002
Saladino et al.
Ta ble 2. Dista n ce (Å) of th e Octa h ed r a l Ar r a n gem en ts
of Re
Ta ble 4. P oly(4-vin ylp yr id in e)/MTO a n d P olystyr en e/
MTO Ca ta lyzed Oxid a tion of Olefin sa
Re-O1
Re-O2
Re-O3
Re-N1
Re-N2
Re-C
1.70
1.70
1.70
2.35
2.35
2.10
epoxide 1,2-diol
time conversion yield
yield
(%)b,c
entry
olefin
catalyst (h)
(%)
(%)b
1
2
3
4
5
6
7
8
9
cis-cyclohexene
cis-cyclohexene 4/UHPd 1.0
cis-cyclohexene
cis-cyclohexene
cis-cyclohexene 4e
cis-cyclohexene 5e
cis-cyclohexene
cis-cyclooctene
cis-cyclooctene
4
1.0
78
47
94
>98
80
>98
>98
>98
>98
96
91
89
9
0
5
6
1.5
0.5
0.5
0.5
1.5
0.5
0.5
0.5
0.5
2.0
6.5
95
0
82
18
0
0
45
0
0
Ta ble 3. Root Mea n Squ a r e (r m s) σp q Va lu es Used in th e
>98
>98
55
>98
>98
>98
>98
70
>98
88
80
75
>98
54
78
>98
46
>98
>98
>98
Mod el of P VP -2%MTO 4
range (r, Å)
σpq
9
4
5
6
9
4
5
6
9
4
5
6
9
0 < r e 2.00
2.00 < r e 3.5
3.50 < r e 7.00
7.00 < r e 13.00
r > 13.00
0.050
0.080
0.150
0.200
0.300
10 cis-cyclooctene
11 cis-cyclooctene
12 styrene
13 styrene
14 styrene
0
0
94
>98
>98
85
60
95
>98
>98
>98
68
>98
>98
>98
93
30
0
12
20
25
0
46
22
0
54
0
18
Sch em e 3
15 styrene
2.0
12.0
3.5
7.0
1.0
16 R-methylstyrene
17 R-methylstyrene
18 R-methylstyrene
19 R-methylstyrene
20 R-methylstyrene 9/UHPd 2.0
21 trans-stilbene
22 trans-stilbene
23 trans-stilbene
24 trans-stilbene
4
5
6
9
6.0
6.0
1.0
1.0
of the possible configurations that the coordinating atoms
can take around the metal.
The distances between Re atom and ligands in octa-
hedral arrangement are reported in Table 2
The same mean square deviation was assigned calcu-
lating the theoretical radial distribution function for each
distance falling within a certain range, in the presented
model. Five fixed root-mean-square values (rms σpq) were
used. The σpq are given in Table 3
0
0
a
Under otherwise specified, all the reactions were performed
in CH2Cl2/CH3CN (5 mL, 1:1 v/v) at room temperature, with H2O2
(30% aqueous solution), using a value of the catalyst loading factor
b
of 1.0. Refers to isolated materials. Yields normalized to 100%
conversion. c No pinacol rearrangement products or products of
other possible secondary reactions were observed by GC-MS and
d
1H NMR spectrum of the crude reaction mixture. Reaction
performed in CH2Cl2. e Loading factor 1.0.
The analysis of the first coordination sphere of the
metal does not allow us to determine the relative posi-
tions of the methyl group with respect to the oxygen and
nitrogen atoms and of the oxygen and nitrogen atoms to
each other. Besides, the two nitrogen atoms can belong
to the same polymer chain or to two polymer chains. This
analysis cannot discriminate the position of the methyl
group because of the small electron differences between
this group and the oxygen and nitrogen atoms. The
analysis of the whole structure of the compound 4 can
discriminate if one or two polymer chains are involved
in the complex, and if the two nitrogen atoms are
positioned perpendicular or parallel, one with respect to
the other. The whole structure analysis of the complex 4
will be published elsewhere since it exceeds the scope of
this work.
P olym er -Su p p or ted MTO-Ca ta lyzed Ep oxid a tion
of Olefin s. Epoxidations with hydrogen peroxide (H2O2)
were investigated using cis-cyclohexene, cis-cyclooctene,
styrene, R-methyl styrene, and trans-stilbene, as repre-
sentative model substrates. As a general procedure, the
olefin (1 mmol) to be oxidized and H2O2 (1.2 mmol, 30%
aqueous solution) were added to a suspension of the
catalyst (loading factor 0.5 or 1.0) in CH2Cl2/CH3CN (5
mL, 1:1 v/v), and the mixture was stirred at room
temperature (Scheme 3). Catalysts were recovered by
filtration at the end of the reaction after washing with
CH3CN and were used in successive oxidations under
identical conditions (vide infra). In some cases the
reaction was performed using urea/hydrogen peroxide
adduct (UHP) as oxygen atom donor, and CH2Cl2 as
solvent. In the absence of the catalyst, less than 2%
conversion of substrates took place under otherwise
identical conditions. Results are reported in Table 4. All
poly(4-vinylpyridine)/MTO and polystyrene/MTO systems
Ta ble 5. Sta bility of P oly(4-vin ylp yr id in e)/MTO a n d
P olystyr en e/MTO Ca ta lysts in th e Ep oxid a tion of
cis-Cycloh exen ea
conversion (%)b
run no. run no. run no. run no. run no.
entry catalyst
1
2
3
4
5
1
2
3
4
4
5
6
9
76(91)c 80(86)
94(95) 91(>95) 89(92)
>98(82) 96(80)
>98(55) 97(53)
74(83)
72(90)
93(94)
75(92)
95(93)
94(81) >98(81) >98(84)
95(54) 88(55) 77(53)
a
The reactions were performed in CH2Cl2/CH3CN (5 mL, 1:1
v/v) at room temperature, with H2O2 (30% aqueous solution), using
a value of the catalyst loading factor of 1.0. Determined by GLC
analysis using a SPB column (25 m × 0.30 mm and 0.25 mm film
thickness), and an isothermal temperature profile of 80 °C for the
first 2 min, followed by a 10 °C/min temperature gradient to 200
°C for 10 min. c Values of the epoxide yield are given in paren-
theses and are normalized to 100% of conversion.
b
showed a high catalyst activity and selectivity for the
formation of epoxides. Only in a few cases were trans-
1,2-diols observed as side-products. The stability of the
catalysts was determined, in some representative ex-
amples, by recycling experiments (Table 5).
The oxidation of cis-cyclohexene with the 4/H2O2
system gave cyclohexene oxide in acceptable conversion
and yield, in the presence of a very low amount of trans-
1,2-cyclohexanediol (Table 4, entry 1). As reported in
Table 5 (entry 1), a slight decrease in activity occurred
during five recycling experiments, showing that 4 is a
stable catalyst.
This result is not completely in agreement with data
previously reported by Herrmann and co-workers for the
oxidation of cis-cyclohexene with heterogeneous MTO and

Heterogeneous Catalysts of Methylrhenium Trioxide
J . Org. Chem., Vol. 67, No. 4, 2002 1331
anhydrous H2O2 in tert-butyl alcohol, in which case only
a low yield of epoxide (27%) was described (the conversion
of the reaction and the yield of the diol are not reported
by the authors).24 A low conversion was obtained under
similar experimental conditions using UHP as oxygen
atom donor, and CH2Cl2 as solvent (Table 4, entry 2). In
this case no traces of trans-1,2-diol were recovered by GC-
duced, probably because of a partial loss of MTO from
the microcapsules (Table 5, entry 4).
In the oxidation of cis-cyclooctene, both poly(4-vinylpy-
ridine)/MTO and polystyrene/MTO systems were shown
to be active and selective catalysts. The conversions of
the reactions were found always higher than 90%, and
the cyclooctene oxide was the only recovered product
(Table 4, entries 8-11).
1
MS and H NMR spectrum of the crude reaction mixture.
Probably, the heterogeneous character of UHP is respon-
sible of the reduced activity of the catalyst. When the
oxidation of cis-cyclohexene was performed with 5, a high
reaction conversion was obtained, and the epoxide was
recovered in 95% yield (Table 4, entry 3). Note that with
the poly(4-vinylpyridine) support, an enhancement of the
reticulation-grade of the polymer increased both the
efficiency and the selectivity of the reaction. Only a slight
decrease in activity occurred during five recycling experi-
ments (Table 5, entry 2). The oxidation of cis-cyclohexene
with 6 and H2O2 showed a conversion higher than the
previously obtained with 4, a yield of epoxide higher than
80%, and an appreciable amount of diol as side-product
(Table 4, entry 4). The role of pyridine N-oxide as
mediator in the MTO-catalyzed epoxidation is still subject
of a debate. It is well-known that the equilibrium
constant for the interaction between pyridine N-oxide and
To probe the generality of this procedure, we studied
the synthesis of some sensitive epoxides, starting from
styrene, R-methyl styrene, and trans-stilbene. The oxida-
tion of styrene with 4 and 5 yielded the quantitative
conversion of the substrate. Only 5 showed a high
selectivity, while in the case of 4 the 1-phenyl-1,2-
ethanediol was recovered in 30% yield (Table 4, entries
12 and 13). The same reaction performed with 6 and 9
as catalysts afforded the styrene oxide in acceptable
conversion and yield. The diol was still obtained in 12%
and 20% yields, respectively (Table 4, entries 14 and 15).
A similar reaction pathway was observed for the high
reactive R-methylstyrene, in which case 5 gave both a
quantitative conversion and yield of R-methylstyrene
oxide, while 4, 6, and 9 showed a high reactivity but not
a similar high selectivity (Table 4, entries 16-19).
Finally, the catalysts transformed the trans-stilbene into
the corresponding epoxide with high conversion and
selectivity, with the exception of 4, in which case only a
modest selectivity was obtained (Table 4, entries 21-24).
It is important to note, that in the case of the catalysts
based on the poly(4-vinylpyridine) as support, the self-
oxidation of the free pyridine rings present on the
polymer remains a potential problem. Sharpless and co-
workers reported that the oxidation of alkenyl substi-
tuted pyridines shows a different reaction pathway if the
alkenyl moiety is conjugated or not conjugated to the
aromatic ring. In particular, in the case of conjugated
pyridine derivatives, the N-oxidation pathway dominates
and the corresponding epoxides were not detected. By
contrast, the 4-(3′-cyclohexenyl)pyridine undergoes fast
epoxidation while not N-oxidation was observed.37 On the
basis of these data, it is reasonable to suggest that in
the oxidations of cis-cyclohexene and cis-cyclooctene, the
self-oxidation of the PVP-2% and PVP-25% supports
should not be an important side-process. On the other
hand, with styrene, R-methylstyrene, and trans-stilbene,
the largest reaction times were observed; thus, the
possibility of the self-oxidation process cannot further
ruled out. Regardless of whether or not the self-oxidation
is a side process, Table 5 shows that PVP-2%/MTO and
PVP-25%/MTO catalysts are stable enough to perform
at least five recycling experiments with similar conver-
sion and selectivity. Thus, if pyridine N-oxide moieties
are formed, they did not interfere with the oxidation of
the olefin at the rhenium active center.
1
MTO, determined by H NMR analysis in CD3NO2, is of
the same order (K ) 210 ( 4 L mol-1) than that recovered
for pyridine (K ) 200 ( 6 L mol-1).51 Moreover, the
pyridine N-oxide/MTO complex shows the highest value
of the energy for the transition state in a frontal attack
of the ligand-bisperoxo complex on ethylene as a model
olefin (19.7 kcal mol-1).22 At first glance, the pyridine-
assisted epoxidation of olefins with MTO was found to
be retarded by a large excess of pyridine, and this result
was attributed to a detrimental effect of pyridine N-oxide
on the stability of MTO.23 On the other hand, Sharpless
and co-workers showed that in the absence of olefin, MTO
is an excellent and quite stable catalyst for the oxidation
of pyridine.51 Thus, the low stability of the MTO in the
pyridine-mediated system becomes relevant only in the
presence of the olefin. In our case, 6 appears to be enough
stable (Table 5, entry 3) to perform the epoxidation of
cis-cyclohexene with high efficiency and selectivity. Even
if we have not studied in detail the mechanism of the
reaction, it is reasonable to suggest that the transition
from an homogeneous system, pyridine N-oxide/MTO, to
6, has a dramatic effect on the stability of the catalyst in
the presence of the olefin.52
In the oxidation of cis-cyclohexene with 4 and 5, a
modification of the loading factor of the catalyst from 0.5
to 1 displayed no particular advantages for the conversion
and yield of epoxide, even if the reaction time was
reduced to half (Table 4, entries 5 and 6). The microen-
capsulated system 9 showed a high activity, but a very
low selectivity, and the trans-1,2-cyclohexanediol was
recovered in almost 1:1 ratio with the epoxide (Table 4,
entry 7). These data further confirm the importance of
pyridine or pyridine N-oxide as Lewis base ligands to the
rhenium atom for the selectivity of the epoxidation.
Moreover, even if only a very slight decrease in activity
occurs during the first four recycling experiments, in the
fifth recycling step the conversion was appreciably re-
Con clu sion s
Poly(4-vinylpyridine)/MTO and polystyrene/MTO sys-
tems are efficient and selective catalysts for the conver-
sion of olefins to the corresponding epoxides, using
environmentally friendly, easily available, and low cost
H2O2 as oxygen atom donor. The efficient oxygen transfer
activity of these systems, under relatively mild experi-
mental conditions, may be attributed to the retained
reactivity of the monoperoxo metal [MeRe(O)2(O2)] and
a bisperoxo metal [MeRe(O)(O2)2] intermediates even in
(51) Wang, W.-D.; Espenson, J . H. J . Am. Chem. Soc. 1998, 120,
11335-11341.
(52) Cope`ret, C.; Adolfsson, H.; Khuong, T.-A. V.; Yudin, A. K.;
Sharpless, K. B. J . Org. Chem. 1998, 63, 1740-1741.

1332 J . Org. Chem., Vol. 67, No. 4, 2002
Saladino et al.
uncorrected. Gas chromatography and gas chromatography-
mass spectroscopy (GC-MS) of the reaction products were
performed using a SPB column (25m × 0.30 mm and 0.25 mm
film thickness) and an isothermal temperature profile of 80
°C for the first 2 min, followed by a 10 °C/min temperature
gradient to 200 °C for 10 min. The injector temperature was
200 °C. Chromatography grade helium was used as the carrier
gas. Mass spectra were recorded with an electron beam of 70
eV. For the scanning electron microscopy (SEM) photographs,
samples were spitter-coated with gold (20 nm). Chromato-
graphic purifications were performed on columns packed with
silica gel, 230-400 mesh, for flash technique. All reagents and
solvents were of highest grade commercially available and used
purified or freshly distilled as required by literature proce-
dures.
P r ep a r a tion of P oly(4-vin ylp yr id in e)/MTO Ca ta lysts.
To a suspension of 600 mg of the appropriate resin in 4 mL of
ethanol was added 77 mg (0.3 mmol) of MTO, and the mixture
was stirred for 1 h using a magnetic stirrer. The solvent was
removed by filtration, and the solid residue was washed five
times with 20 mL of ethyl acetate each time and finally dried
under high vacuum. Coocervates were found to envelop the
solid core dispersed in the medium, and 5.0 mL of hexane were
added to harden the capsule walls. The suspension was stirred
for 1 h at this temperature and then slowly cooled to 0 °C. A
yellow powder remained. In each case, MTO had completely
became bound to the polymer. This result was confirmed by
spectroscopic analysis of the residue obtained after evaporation
of the organic layers. The catalyst was used without any
further purification.
P r ep a r a tion of P olystyr en e/MTO Ca ta lysts. Polysty-
rene or the mixture poly(4-vinylpyridine)/polystyrene 600 mg
(1:5 ratio) was suspended in 5.0 mL of THF at 25 °C, and to
this solution was added 77 mg of MTO (0.3 mmol) as a solid
core. The suspension was stirred for 1h at this temperature
and then slowly cooled to 0 °C. Coocervates were found to
envelop the solid core dispersed in the medium, and 5.0 mL
of hexane were added to harden the capsule walls. The mixture
was stirred for 1 h at 25 °C. The solvent was removed by
filtration, and the solid residue was washed five times with
20 mL of ethyl acetate each time and finally dried under high
vacuum. A blue powder remained. In each case, MTO had
completely became bound to the polymer. This result was
confirmed by spectroscopic analysis of the residue obtained by
evaporation of the organic layers. The catalyst was used
without any further purification.
Oxid a tion of Olefin s. Gen er a l P r oced u r e. To the sus-
pension of 146 mg of the appropriate catalyst in 5.0 mL of
CH3CN/CH2Cl2 (ratio ) 2 /1 v/v) at 25 °C were added the
olefin (1 mmol) and the oxygen atom donor (2.0 mmol), H2O2
(2.0 mmol, 35% water solution), or urea hydrogen peroxide
adduct (2.0 mmol, UHP). The reaction was monitored by gas-
chromatography, using dodecane as an internal standard. The
suspension was filtered and the recovered catalyst washed five
times with 10 mL of ethyl acetate each time. After drying
under high vacuum, the catalyst was used for further reaction
to evaluate its stability. The filtrate was treated with a low
amount of manganese dioxide (MnO2) at 25 °C to decompose
the excess of primary oxidant and filtered. The solvent was
dried with Na2SO4. After the remotion of the solvent, the crude
product was analyzed by gas chromatography-mass spectros-
copy and purified by preparative thin-layer chromatography
to obtain the epoxides from acceptable to good yields. Products
gave spectrocopic data consistent with the literature.
the presence of the polymeric organic support. The UHP
may be considered as an alternative to H2O2 for the
synthesis of sensitive epoxides, even if lower reaction
conversions are observed, probably because of the hetero-
geneous character of the oxidant. Independently from the
system used in the oxidation, values of the loading factor
of the catalyst higher than 0.5 did not give an appreciable
increase of the reaction conversion or of the epoxide yield.
Catalysts based on poly(4-vinylpyridine) are stable sys-
tems for at least five recycling experiments. By contrast,
PS-2%/MTO showed a slight decrease in activity during
the first four recycling experiments, while in the fifth
recycling step the conversion was appreciably reduced.
Thus, the physical envelopment of the benzene ring of
polystyrene did not completely ensure the stability of the
catalyst. In the case of PVP-2%/MTO the wide-angle
X-ray diffraction (WAXS.) analysis performed by the
application of the difference method showed the first
reported geometry of a MTO heterogeneous species. The
complex is characterized by a slightly distorted octahe-
dral conformation on the metal’s primary coordination
sphere. The Re-O and Re-C bond distances are nearly
not influenced by the polymeric nature of the ligand. On
the other hand, the Re-N bond distances were found
abnormally shorter than those observed for homogeneous
MTO/Ln complexes showing a strong coordination of the
rhenium atom. These data are in accord with the high
stability observed for this catalyst in recycling experi-
ments. The reticulation grade of the polymer is a crucial
factor for the catalysts based on poly(4-vinylpyridine),
either for the morphology of the surface of particles and
for the reaction pathway. Poly(4-vinylpyridine) 2% cross-
linked systems, such as PVP-2%/MTO and PVPN-2%/
MTO, are characterized by particles with irregular
surfaces. On the other hand, the most reticulated PVP-
25%/MTO shows particles with a regular spherical shape,
the morphology of which is very similar to the microcap-
sules of PS-2%/MTO. Moreover, PVP-25%/MTO is the
best catalyst system, giving the epoxides of conjugated
and nonconjugated olefin derivatives in high conversions
and yields. PVPN-2%/MTO, which is probably character-
ized by a pyridine N-oxide moiety as anchorage site for
the rhenium atom, shows a reactivity and selectivity that
is similar to that of PVP-2%/MTO, with the exception of
conjugated olefin derivatives, in which cases a highest
selectivity was observed. The loss of the detrimental
effect of the pyridine N-oxide moiety on the stability of
the catalyst, previously reported for the homogeneous
system MTO/pyridine, is a new important clue. PS-2%/
MTO is the less selective catalyst of all of the systems
used. Design of other supported MTO compounds with
high selectivity and activity is being undertaken in our
laboratory.
Exp er im en ta l Section
NMR spectra were recorded on a 200 MHz spectrometer and
are reported in δ values. Infrared spectra were recorded on a
FT-IR spectrophotometer using KBr and NaCl plates. Melting
points were obtained on a Reichert Kofler apparatus and are
J O011033F