Behaviour of dimeric methylrhenium(VI) oxides in the presence of hydrogen peroxide and its consequences for oxidation catalysis

Article abstract of DOI:10.1039/b608219j

Avoiding the use of toxic methyltin precursors to synthesize methyltrioxorhenium (MTO) and its mono- and bis-peroxo derivatives, applicable as oxidation catalysts, dimethyl zinc might be considered a promising alternative alkylating agent. However, the methylrhenium(vi) dimers, formed as reduction products alongside MTO during the reaction of dimethyl zinc with Re₂O₇, are not as straightforwardly transformed into the epoxidation catalysts as MTO itself in the presence of excess H ₂O₂. In the case of red (μ-oxo) bis[trimethyloxorhenium(vi)], the main reaction product with H₂O ₂ is the catalytically inactive trimethyldioxorhenium(vii). In the case of bis[dimethyl(μ-oxo)oxorhenium(vi)], slow conversion to the monomeric mono- and bis-peroxo congenes of MTO occurs. Furthermore, part of the Re(vi) starting complex is transformed into inactive perrhenate. While bis[dimethyl(μ-oxo)oxorhenium(vi)] might be applied (also in a mixture with MTO) as an oxidation catalyst precursor, (μ-oxo)bis[trimethyloxorhenium(vi)] can be applied as a useful precursor for the synthesis of trimethyldioxorhenium(vii), which was previously only accessible by less convenient synthetic pathways. the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006.

Full text of DOI:10.1039/b608219j

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
Behaviour of dimeric methylrhenium(VI) oxides in the presence of
hydrogen peroxide and its consequences for oxidation catalysis
Alexandra M. J. Rost,a Andrea Scherbaum,a Wolfgang A. Herrmann*a and
Fritz E. Ku
¨
hn*ab
Received (in Montpellier, France) 9th June 2006, Accepted 20th July 2006
First published as an Advance Article on the web 14th August 2006
DOI: 10.1039/b608219j
Avoiding the use of toxic methyltin precursors to synthesize methyltrioxorhenium (MTO) and its
mono- and bis-peroxo derivatives, applicable as oxidation catalysts, dimethyl zinc might be
considered a promising alternative alkylating agent. However, the methylrhenium(VI) dimers,
formed as reduction products alongside MTO during the reaction of dimethyl zinc with Re2O7,
are not as straightforwardly transformed into the epoxidation catalysts as MTO itself in the
presence of excess H2O2. In the case of red (m-oxo)bis[trimethyloxorhenium(VI)], the main reaction
product with H2O2 is the catalytically inactive trimethyldioxorhenium(VII). In the case of
bis[dimethyl(m-oxo)oxorhenium(VI)], slow conversion to the monomeric mono- and bis-peroxo
congenes of MTO occurs. Furthermore, part of the Re(VI) starting complex is transformed into
inactive perrhenate. While bis[dimethyl(m-oxo)oxorhenium(VI)] might be applied (also in a mixture
with MTO) as an oxidation catalyst precursor, (m-oxo)bis[trimethyloxorhenium(VI)] can be applied
as a useful precursor for the synthesis of trimethyldioxorhenium(VII), which was previously only
accessible by less convenient synthetic pathways.
boiling THF,4 the reaction with dimethylzinc takes place at
Introduction
dry ice temperature (À78 1C).8a At higher temperatures,
dimethylzinc reduces Re2O7, at least partially, to the oxidation
state Re(VI),4,8a forming bis[dimethyl(m-oxo)oxorhenium(VI)]
(2) and (m-oxo)bis[trimethyloxorhenium(VI)] (3) as the by-
products or main products. Since the latter two compounds
show some catalytic activity in olefin epoxidation reactions,9 it
might be assumed that they are transformed into the same
mono- and bis-peroxo species10 as is MTO in the presence of
excess H2O2. However, this would help avoid the necessity of
preparing (or even isolating) MTO as a precursor for the
active species in oxidation catalysis. Dimethylzinc could be
utilized instead of tin precursors in the alkylation of the
rhenium species, and the oxidizing agent would oxidize the
methylrhenium precursors to a uniform catalyst. It has been
observed in olefin epoxidation reactions with organomolybde-
num oxides that their organomolybdenum carbonyl precur-
sors can be directly added to the oxidation mixture (in this case
tert-butyl hydroperoxide acts as oxidizing agent), thus avoid-
ing the necessity of preparing the more sensitive oxides,
forming the same catalyst.11
Industrial transformations of olefins into epoxides involve
catalysts that utilize organic peroxides, hydrogen peroxide or
oxygen as oxidants. However, for the epoxidation and oxida-
tion of fine chemicals stoichiometric reactions are still com-
monly used.1 An important discovery in this field arose with
the finding of Herrmann et al. that methyltrioxorhenium (1,
MTO) and some of its derivatives act as efficient catalysts for
olefin epoxidation.2 MTO was first described in 1979 by Beatty
and Jones,3 but its applicability as a catalyst was only detected
nearly a decade later, following a much improved synthesis.4
Due to the broad applicability of MTO in a variety of catalytic
reactions,2,5 several modified synthetic approaches have been
reported.6 Nevertheless, an important drawback of these
syntheses that still remains is the use of highly toxic tetra-
methyltin or the closely related and also toxic methyltributyl-
tin as methylating agents. However, it is known that several
derivatives of MTO are available via bis(alkyl)zinc pre-
cursors.2,5f,7
Furthermore, it has been noted that mixtures of MTO and
methylrhenium(VI) oxides are available when reacting di-
methylzinc with Re2O7 at low temperatures.4,8 The choice of
the best alkylation agent in reactions with Re2O7 depends
largely on the stability of the product and the ease of alkyl
transfer. While tetramethyltin only reacts with Re2O7 in
In this work we examine the behaviour of the dimeric Re(VI)
compounds in the presence of H2O2 and their applicability as
epoxidation catalysts in comparison to the well established
MTO.
Experimental
a Lehrstuhl fur Anorganische Chemie der Technischen Universitat
¨
¨
Munchen, Lichtenbergstraße 4, D-85747 Garching bei Munchen,
¨
¨
General
Germany. E-mail: wolfgang.herrmann@ch.tum.de. E-mail:
fritz.kuehn@ch.tum.de; Fax: +49 (0)89 289 13473
NMR spectra (1H and 13C) were recorded on a Bruker DPX
400 instrument. Chemical shifts are given in ppm. The spectra
are referenced to the residual protons (1H) or to the 13C signals
b
´
Department of Chemistry, Instituto Tecnologico e Nuclear (ITN),
Estrada Nacional No. 10, 2686-953 Sacavem, Portugal. E-mail:
´
fekuhn@itn.pt; Fax: +351 219946185
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
New J. Chem., 2006, 30, 1599–1605 | 1599

View Article Online
(13C) of the solvent. Elemental analyses were determined at the
from 500 to 190 nm. The total time of the measurements and
the time between the cycles was varied, depending on the
compounds.
Microanalytical Laboratory at TU Munchen. CI-MS spectra
¨
were measured on a Finnigan MAT 90 mass spectrometer.
Thermogravimetric mass spectra (TG-MS) analysis measure-
ments were conducted with a Netzsch TG209 system; typically
about 10 mg of each sample was heated from 35 to 1000 1C at
10 K minÀ1. GC spectra were measured either on a Hewlett-
Packard gas chromatograph HP 6890 equipped with a mass
selective detector HP or a Hewlett-Packard gas chromato-
graph HP 5890 using a FID detector. The yields of the
products of oxidation catalysis were calculated according to
calibrations conducted prior to the measurements. UV-vis
spectra were recorded on a Jasco V-550 spectrophotometer.
IR spectra were acquired using a Jasco FT/IR-460 Plus
spectrometer.
Oxidation catalysis
Epoxidation of cis-cyclooctene. The catalytic activity of
compounds 1–3 was tested at room temperature by examining
the epoxidation of cis-cyclooctene (0.8 g, 7.3 mmol). Mesity-
lene (1 g, 8.3 mmol) was used as the internal standard and
H2O2 (35 wt%; 1.41 mL, 14.6 mmol) as the oxidising agent.
The oxidation of cis-cyclooctene yields quantitatively epoxy-
cyclooctene. A catalyst : oxidant : substrate ratio of 1 : 200 :
100 was used. The samples were first analyzed by GC-MS after
5 min and then every 30 min for 7 h. The reaction was
terminated after 23 h.
Synthesis
Oxidation of benzyl alcohol. The catalytic behavior was
additionally examined in the oxidation of benzyl alcohol
(0.52 mL, 5 mmol) at room temperature. Diethylene glycol
dimethyl ether was used as the internal standard, H2O2 (35
wt%; 1.5 mL, 15 mmol) as the oxidising agent, LiBr (44 mg,
0.5 mmol) as the co-catalyst and TEMPO (40 mg, 0.25 mmol)
as the mediator. The oxidation of benzyl alcohol in this four-
component system yields selectively benzyl aldehyde. A cata-
lyst : oxidant : substrate ratio of 1 : 300 : 100 was used. 0.1
equiv. of H2O2 was added per minute and the reaction was
terminated after 24 h. The reaction product was identified and
quantified by GC-MS.
All experiments were carried out under an oxygen- and water-
free argon atmosphere using standard Schlenk techniques. All
solvents used in the inert atmosphere conditions were purified
by standard procedures and distilled under a nitrogen atmo-
sphere directly prior to use. Compounds 1–3 were prepared
according to literature procedures.6a,8a The observed 1H and 13
C
NMR data, elemental analyses and mass spectra confirmed that
the compounds were pure and identical to previously collected
data.4,6a,8a All other chemicals were obtained from commercial
sources and used without further purification.
Preparation of trimethyldioxorhenium(VII) (4)
Results and discussion
To a cooled solution (0 1C) of 3 (100 mg, 0.20 mmol) in 5 mL
diethyl ether was added H2O2 (35 wt% in water, 41.9 mL, 0.8
mmol). The solution was stirred for 20 min at 0 1C and then
warmed to room temperature. The color of the solution
changed from deep red to yellow, indicating the completion
of the reaction. The solution was washed twice with 2 mL
water and then dried over Na2SO4. Removal of the solvent in
vacuo yielded a red oil, which was purified by trap-to-trap
fractionation in vacuo.
Synthesis
The preparation of compounds 2 and 3 via the processes
described in the literature4,8a is quite difficult and time con-
suming, compared to the preparation of 1. Usually in such
preparations, all three compounds 1–3 are formed. However,
the nature of the main product depends strongly on the
reaction conditions applied. When reacting Re2O7 with 1
equivalent of (CH3)2Zn, 1 is obtained as main product; when
applying 3 equivalents of (CH3)2Zn below 0 1C, the formation
of compound 2 is favoured; with 4 equivalents of (CH3)2Zn
above 0 1C, compound 3 is obtained as the major product
(Scheme 1). The purification of the three products requires
several sublimation, extraction and chromatography steps.
While compound 2 can be purified by extraction with n-hexane
and several subsequent sublimation steps, the separation of
compound 3 is more difficult due to the pronounced sensitivity
of the latter complex to traces of water and oxygen, which
convert it easily into compound 2 and other products (see
below). Accordingly, the isolation of compound 3 has to be
conducted with the rigorous exclusion of moisture and air in
extremely dry solvents. Pure (m-oxo)bis[trimethyloxo-
rhenium(VI)] (3) is obtained after extraction with n-pentane,
sublimation, column chromatography (0 1C, RP18 silica gel,
n-pentane) and if necessary, a further sublimation step.
1
Yield: 84%. IR (in n-pentane) n(ReQO): 1001s, 963vs. H
NMR (400.1 MHz, CDCl3): dH 2.49 (s, 2 H, CH3), 2.04 (s, 1 H,
CH3). 1H NMR (400.1 MHz, d8-THF): dH 2.41 (s, 2 H, CH3),
1.99 (s, 1 H, CH3). 13C NMR (100.1 MHz, CDCl3) dC 26.76,
29.69 (CH3).
Reactions of compounds 1–3 with H2O2
1H NMR studies. To a solution of 1–3 (0.063 mmol) in
0.5 mL d8-THF was added 2 equiv. H2O2 (35 wt%) per Re.
The reaction was monitored by recording 1H NMR spectra
(298 K, 32 scans) at various times. All spectra were referenced
to the peak of d8-THF at dH = 1.73.
UV-vis studies. Measurements were carried out in silica cells
with a path length of 1 cm and a volume of 1 mL. The data for
the kinetic investigations were acquired from freshly prepared
solutions comprising 7.3 Â 10À5 mol of the rhenium com-
pound in 23.5 mL THF with 7.3 Â 10À5 mol of H2O2 (35
wt%). To obtain an appropriate concentration, 0.3 mL of the
solution was diluted with 0.6 mL THF and transferred im-
mediately to the spectrophotometer. The scanning range was
TG-MS studies
The TG-MS plots of complexes 1–3 are given in Fig. 1. The
obtained TG curve of 1 is similar to the published results.12
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
1600 | New J. Chem., 2006, 30, 1599–1605

View Article Online
Fig. 1 TG spectra of compounds 1–3.
Compound 1 starts subliming below 100 1C and the sublima-
tion process is fast. At 150 1C, only 16% of the original mass is
left. The sublimation of compound 2 starts at 130 1C and
continues until 218 1C. Complex 3 is stable up to 112 1C. In the
case of compound 2, there is only one remarkable weight loss
step, with a 73% reduction in the original mass. Compound 3
is the only one of the three examined rhenium complexes that
presents visible decomposition steps while being heated to
200 1C. In the first step, until 135 1C, a mass loss of 6% is
observed. This is equivalent to the mass of two of its methyl
groups. The remaining methyl group is lost before reaching
185 1C. In the last decomposition step, which continues over a
larger temperature range, a mass loss of 7% is observed,
possibly due to the loss of oxygen ligands.
Scheme 1 Reaction of Re2O7 with dimethylzinc.
Reactions of compounds 1–3 with H2O2
1
Kinetic H NMR studies. The reaction of compound 1 with
H2O2 (1 : H2O2 = 1 : 2; 2 equiv. H2O2 per Re) in d8-THF
affords four CH3 signals in the 1H NMR spectrum (see Fig. 3),
ascribable to MTO (d = 2.1), CH3Re(O2)O2 (5) (d = 2.3),
CH3Re(O2)2O Á H2O (6) (d = 2.6) and the decomposition
product CH3OH (d = 3.2). In correspondence with published
results, the reaction reached equilibrium after 1 h under the
applied conditions.15 As expected, during this time, the peaks
originating from 1 (MTO) and, after a while, 5 decreased,
while the signals originating from 6 and methanol increased
(see Scheme 2). This behavior is in full accordance with the
well documented literature data.15 Adding a large excess of
H2O2 lead to the fast formation of 6 and only a small amount
of methanol, as also reported in the literature.15
Additionally, changes can be deduced from the DSC plot
(Fig. 2). The observed exothermic reaction of 1, with a peak at
63 1C, indicates that the sublimation is already starting at
40 1C under the conditions applied. The following endother-
mic step, with a minimum at 116 1C, is very likely to be
associated with the dimerization of 1.13 No weight is lost
during this process. Compound 2 starts subliming at higher
temperatures in an exothermic step and subsequently seems to
dimerize at 127 1C, possibly leading to a tetrameric structure.14
After an exothermic step at 63 1C, compound 3 probably
dimerizes. No mass loss is observed during these two steps.
The following endothermic steps indicate the beginning of the
decomposition of compound 3. After the loss of two methyl
groups, a minimum at 156 1C is reached. Decomposition
continues with a significant exothermic step at 177 1C, as-
signed to the loss of the remaining methyl groups.
The reaction of 2 with H2O2 (2 : H2O2 = 1 : 4; 2 equiv. H2O2
per Re) was also monitored by 1H NMR spectroscopy (see Fig.
4). During the reaction, four CH3 signals are observed (see Fig.
4): dH(CH3) = 2.3 (5), 2.6 (6), 2.7 (2) and 3.2 (CH3OH from
partial decomposition). About 1 h after the addition of H2O2,
the very slowly emerging peaks of 5 (at 2.3 ppm) and 6
Fig. 3 1H NMR kinetic study of the reaction of compound 1 with
H2O2.
Fig. 2 DSC plots of compounds 1–3.
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
New J. Chem., 2006, 30, 1599–1605 | 1601

View Article Online
Scheme 2 Reaction of compounds 1–3 with H2O2.
(at 2.6 ppm) become visible. The peak of methanol at 3.2 ppm
increases steadily during the 20 h reaction time. The peak due to
compound 2 (at 2.7 ppm) decreases insignificantly during the
first 8 h of the reaction. Only after this time do the peaks of
compounds 5, 6 and concomitantly formed methanol become
more pronounced. After an 18 h reaction time, the peaks of 5
and 6 begin to decrease slowly. These experimental results are
indicative of the cleavage of the Re–Re bond, the loss of two
methyl groups per molecule of 2, and the formation of 5 and 6
(see Scheme 2). Methane, ethane, ethene and propane are
detected by GC-MS during the reaction in the gas phase. Traces
of these also appear in the NMR spectrum but these peaks are
not depicted in Fig. 4. The formation of these hydrocarbons
probably results from the radical cleavage of the Re–C bonds.16
Only one methyl group remains at the Re center. The oxidative
degradation of 2 also leads to a change in the oxidation state of
Re (from Re(VI) to Re(VII)), resulting in the formation of 5 and
6, analogously to the reaction observed for compound 1 with
H2O2. Due to the slow oxidative degradation of 2, the partial
decomposition of 5, yielding methanol, is more pronounced
than in the case of compound 1. Only part of the applied
quantity of 2 is converted into 5 and 6 because some H2O2 is
decomposed by the radicals created during the cleavage of the
Re–methyl bonds. Adding a large excess of H2O2 leads to a
faster formation of 6 and a smaller amount of methanol.
However, when compared to the reaction of MTO, the latter
reaction is still significantly slower and a more pronounced
formation of methanol from partial decomposition is observed.
In presence of an alkene, e.g., cis-cyclooctene, the resulting
amount of methanol formed is only one-third of that found in
the reaction in the absence of an alkene (as shown in Fig. 4).
Thus, the addition of an alkene significantly suppresses the
formation of methanol. This behavior suggests that methanol
formation results from the decomposition of 5 and 6, and not
from other processes, e.g., the cleavage of two methyl groups
per molecule of 2 during the reaction of the dimeric methyl-
rhenium(VI) oxide with H2O2.
During the reaction of compound 3 with H2O2 (see Fig. 5),
signals at d = 2.44 and 2.46 (originating from compound 3), at
d = 2.7 (originating from compound 2), and at d = 2.0 and
2.41 are apparent. Immediately after the addition of H2O2,
pronounced signals at d = 2.0 and 2.41, and a minor peak for
compound 2 emerge. As for the peaks observed for compound
3, displaying a 2 : 1 (d 2.44 : d 2.46) ratio, the new peaks display
a 2 : 1 ratio (d 2.41 : d 2.0). The deep red color of the reaction
mixture turns quickly to yellow. Within minutes, the signals of 3
disappear entirely. However, in contrast to MTO, where a fast
epoxidation reaction takes place when cyclooctene is provided,
in case of 3, the reaction is quite sluggish. A newly formed
compound, responsible for the peaks at d = 2.0 and 2.41, can
be isolated from the cyclooctene-free reaction mixture. The
application of several spectroscopic methods (see experimental
section) leads to its identification as trimethyldioxorhenium (4).
This compound can be synthesized from 3 in good yields by the
Fig. 4 1H NMR kinetic study of the reaction of compound 2 with
H2O2.
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
1602 | New J. Chem., 2006, 30, 1599–1605

View Article Online
Fig. 6 The formation of compound 6 from the reaction of 2 with
H2O2 (difference spectrum to emphasize the changes); see text for more
details.
Fig. 5 1H NMR kinetic study of the reaction of compound 3 with
the disappearance of the red-coloured 3. Additionally, a
maximum at l = 350 nm increases.
H2O2.
addition of H2O2. Prior to this work, the only known synthesis
of compound 4 was the oxidation of 3 with trimethyl amine
N-oxide.8a,17 The synthetic pathway found during this work
seems to be easier than the one described previously in the
literature. While, in the presence of H2O2, the tiny amounts of
compound 2, formed during the fast reaction of compound 3
with H2O2, behave as described above, compound 4 does not
react with H2O2, even when the latter is present in a large
excess. Neither 5 or 6 are formed starting from isolated
compound 4. However, this observation sheds new light on
the applicability of compound 3 as an oxidation catalyst.
Obviously, the main product from the oxidation of 3 with
H2O2 is compound 4, not being transformable into 5 or 6 by
the reaction of 4 with H2O2. Moreover, compound 4 seems to
be virtually inert to H2O2, even when exposed to it for many
hours. Due to the formation of only tiny amounts of 2 during
the reaction of 3 with H2O2, which forms very slowly the active
species 5 and 6, no methanol originating from the deactivation
could be detected by NMR, but traces of the latter were
detected by GC-MS.
Oxidation catalysis
Epoxidation of cis-cyclooctene. The catalytic activities of 1–3
were tested on the epoxidation of cis-cyclooctene (see Fig. 8).
In comparison to compounds 2 and 3, MTO showed 100%
olefin conversion after a reaction time of 150 min. After the
same time, 2 (as a catalyst precursor) achieved only a conver-
sion of 37%. However, after 420 min, it also reached complete
conversion. Compound 3 lead to an olefin conversion of only
10% after 270 min. In the case of 1, the quickest conversion
rate was obtained during the first hour, while 2 and 3 did not
increase the conversion rate during the first 4 h of the reaction.
Only 12% conversion was obtained after 24 h with compound
3. The notably higher catalytic activity of 1 in olefin epoxida-
tion reactions can be explained by the formation of the bis-
peroxocomplex 6, which represents the catalytically active
species. Due to its smaller steric bulk and higher oxidation
state in comparison to 2 and 3, the catalyst precursor 1 forms
the active species much faster in presence of H2O2. This
phenomenon is supported by kinetic UV-vis and 1H NMR
studies. After 250 min, 2 has been converted mainly into the
catalytically active species, the formation reaction accelerating
significantly and finally reaching 100% conversion. Never-
theless, the formation of the catalytically active species 6 takes
much longer than in the case of compound 1.
UV-vis studies. The reaction kinetics of the rhenium com-
pounds with H2O2 in water in the absence of olefin was
additionally examined by UV-vis spectroscopy.
The complexes were combined in a 1 : 100 ratio with respect
to the oxidant to more closely mimic catalytic conditions.
Compound 1 forms the active species fastest among the
examined rhenium complexes. After 1 h, the characteristic and
very pronounced maximum for compound 6 was observed at
350 nm.9 It remained unchanged for 2 h, before 6 started
slowly decomposing in the absence of olefin.
During the reaction of compound 2 with H2O2, the visible
yellow color seemed to remain unchanged. However, in re-
ference to pure compound 2, the changes were easily detect-
able in the UV-vis spectrum (see Fig. 6), indicating the
formation of yellow compound 6 from the yellow compound
1
2, as observed in the H NMR spectra (see above). A mini-
mum at 350 nm was established during the reaction time.
Treating compound 3 with H2O2 leads to visible changes
in the UV-vis spectrum (see Fig. 7). The maximum at l =
275 nm decreased during the reaction time of 10 h, indicating
Fig. 7 UV-vis spectra of compound 3, treated with hydrogen
peroxide (see text).
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
New J. Chem., 2006, 30, 1599–1605 | 1603

View Article Online
5% of 2 was also obtained. The 1H NMR spectra also proved
that 4 does not react with excess H2O2 and does not form
peroxo species. In this case, the catalytic activity originates—
as mentioned above—from the conversion of by-product 2 into
5 and 6. Due to the very low concentration of compound 2, the
conversion increased only slowly until a 10% aldehyde yield
was reached. The stagnation of the conversion at this value
after 240 min is a consequence of the partial decomposition of
the Re precursor to methanol and perrhenate during the slow
formation of the active catalyst species 5 and 6.
The oxidation of benzyl alcohol with compound 2 as the
catalyst precursor reached completeness after 48 h. Compared
to compound 1, the reaction rate of 2 was much lower. This is
a consequence of the significantly slower formation of the
catalytic active species 5 and 6 from 2 due to the higher steric
bulk of the latter, the necessary cleavage of the dirhenium
complex and the change in oxidation state having already been
mentioned above. When, after one catalytic cycle with com-
plete conversion of benzyl alcohol, new substrate and oxidant
are added, total conversion was again reached after 240 min.
Due to the fact that the catalytically active species 5 and 6 have
already been formed in the first cycle, the reaction in the
second cycle is significantly faster than in the first, but not as
fast as in the case of compound 1. During the formation of the
catalytically active species from precursor 2 in the first cycle,
part of the monoperoxo complex 5 is deactivated to methanol
and perrhenate, leaving less catalytically active species for the
second cycle and therefore reducing the reaction rate in
comparison to catalyst precursor 1.
Fig. 8 Time–conversion diagram of the epoxidation of cis-cyclo-
octene with 1 (m), 2 (’) and 3 (K) as catalyst precursor.
Oxidation of benzyl alcohol. The catalytic behavior of 1–3
was additionally examined in the oxidation of benzyl alcohol
(see Fig. 9). Compared to the alkylrhenium(VI) oxides 2 and 3,
MTO (1) showed excellent performance. Complete conversion
was already achieved after 120 min with catalyst 1. After the
same reaction time, compound 2 reached only a conversion of
50% and compound 3 reached a conversion of merely 8%.
With compound 2 as catalyst precursor, the conversion was
complete after 48 h. Compound 3 reached only 10% conver-
sion after the same time, and during the following 24 h no
additional conversion increase was observed. In order to test
whether all the oxidant had been reacted or whether deactiva-
tion took place, again 300 mol% of H2O2 was added in a
second step. No further conversion could be observed in the
Conclusions
MTO is transformed much faster into the catalytically active
species of olefin epoxidation than the dimeric Re(VI) deriva-
tives 2 and 3. While compound 2 is slowly reacting to give the
same active species as MTO itself, compound 3 mainly reacts
to form the catalytically inactive trimethyldioxorhenium(VII).
This complex does not react with hydrogen peroxide at all.
Only a small proportion of compound 3 reacts to yield the
catalytically active mono- and bis-peroxo species 5 and 6 via
compound 2. Both compounds 2 and 3 are also partially
transformed into methanol and perrhenate in the presence of
H2O2, additionally reducing their catalytic applicability.
These results demonstrate that the Re(VI) compounds 2 and
particularly 3 are not suitable replacements for MTO as
catalyst precursors in oxidation reactions. Despite being only
available via toxic tin precursors, MTO seems currently to be
the only easily accessible, high oxidation state methylrhenium
oxide applicable to oxidation reactions. However, the synth-
esis of trimethyldioxorhenium is straightforwardly achieved
starting from compound 3, thus presenting a good alternative
to the known literature preparation of this compound.
1
case of compound 3. The results of the H NMR and UV-vis
kinetic studies of the reaction of 3 with H2O2 showed that
trimethyldioxorhenium(VII) (4) was mainly formed and about
Acknowledgements
The authors are grateful to the Margarethe-Ammon-Stiftung
(PhD grant for A. R.) and the Bayerische Forschungsstiftung
(PhD grant for A. S.) for financial support.
Fig. 9 Time–conversion diagram of the oxidation of benzyl alcohol
with 1 (m), 2 (’) and 3 (K) as catalyst precursor.
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
1604 | New J. Chem., 2006, 30, 1599–1605

View Article Online
9 W. A. Herrmann, R. W. Fischer, M. U. Rauch and W. Scherer, J.
Mol. Catal., 1994, 86, 243.
References
1 R. A. Sheldon, in Applied Homogeneous Catalysis with Organome-
tallic Compounds, ed. B. Cornils and W. A. Herrmann, Wiley-
VCH, Weinheim, 2nd edn, 2002, vol. 1, pp. 412–427.
2 (a) W. A. Herrmann and F. E. Kuhn, Acc. Chem. Res., 1997, 30,
¨
169; (b) C. C. Romao, F. E. Kuhn and W. A. Herrmann, Chem.
¨
Rev., 1997, 97, 3197.
3 R. Beatty and J. P. Jones, Inorg. Chem., 1979, 18, 2318.
4 W. A. Herrmann, J. G. Kuchler, J. K. Felixberger, E. Herdtweck
and W. Wagner, Angew. Chem., Int. Ed. Engl., 1988, 27, 394.
5 (a) A. M. J. Rost, H. Schneider, J. P. Zoller, W. A. Herrmann and
10 (a) W. A. Herrmann, R. W. Fischer, W. Scherer and M. U. Rauch,
Angew. Chem., Int. Ed. Engl., 1993, 32, 1157; (b) A. M. Al-Ajlouni
and J. H. Espenson, J. Am. Chem. Soc., 1995, 117, 9243; (c) J. H.
Espenson, Chem. Commun., 1999, 479; (d) W. Adam, C. R. Saha-
Moller and O. Weichold, J. Org. Chem., 2000, 65, 5001; (e) P.
¨
Gisdakis, I. V. Yudanov and N. Rosch, Inorg. Chem., 2001, 40,
¨
Gisdakis and N. Rosch, Eur. J. Org. Chem., 2001, 4, 719; (f) P.
¨
3755; (g) C. Valentin, R. Randolfi, P. Gisdakis and N. Rosch, J.
¨
Am. Chem. Soc., 2001, 123, 2365.
11 (a) M. Abrantes, A. M. Santos, J. Mink, F. E. Kuhn and C. C.
¨
Romao, Organometallics, 2003, 22, 2112; (b) J. Zhao, A. M.
F. E. Kuhn, J. Organomet. Chem., 2005, 690, 4712; (b) F. E. Kuhn
¨
and A. M. Santos, Mini-Rev. Org. Chem., 2004, 1, 55; (c) A. M.
¨
Santos, E. Hertweck and F. E. Kuhn, J. Mol. Catal. A: Chem.,
¨
A. M. Afonso, L. C. Branco, I. S. Gonc¸ alves, M. Pillinger and C.
¨
2004, 222, 265; (c) F. E. Kuhn, J. Zhao, M. Abrantes, W. Sun, C.
Santos, F. M. Pedro, A. A. Yogalekar, I. S. Lucas, C. C. Romao
and F. E. Kuhn, Chem.–Eur. J., 2004, 10, 6313; (d) F. M. Pedro, S.
¨
Hirner and F. E. Kuhn, Tetrahedron Lett., 2005, 46, 7777; (e) F. E.
¨
Kuhn, J. Zhao and W. A. Herrmann, Tetrahedron: Asymmetry,
C. Romao, Tetrahedron Lett., 2005, 46, 47; (d) F. E. Kuhn, A. M.
¨
Santos and W. A. Herrmann, Dalton Trans., 2005, 2483; (e) F. E.
Kuhn, A. M. Santos and M. Abrantes, Chem. Rev., 2006, 106,
¨
2455.
¨
2005, 16, 3469; (f) F. E. Kuhn, A. Scherbaum and W. A. Herr-
¨
mann, J. Organomet. Chem., 2004, 689, 4149.
6 (a) W. A. Herrmann, F. E. Kuhn, R. W. Fischer, W. R. Thiel and
¨
C. C. Romao, Inorg. Chem., 1992, 31, 4431; (b) W. A. Herrmann,
W. R. Thiel, F. E. Kuhn, R. W. Fischer, M. Kleine, E. Herdtweck
¨
12 W. A. Herrmann, W. M. Wachter, F. E. Kuhn and R. W. Fischer,
¨
J. Organomet. Chem., 1998, 553, 443.
´
13 P. Gisdakis, N. Rosch, E. Bencze, J. Mink, I. S. Gonc¸ alves and F.
¨
14 S. Kostlmeier, G. Pacchioni, W. A. Herrmann and N. Rosch, J.
¨
E. Kuhn, Eur. J. Inorg. Chem., 2001, 981.
and W. Scherer, Inorg. Chem., 1993, 32, 5188; (c) W. A. Herrmann,
R. M. Kratzer and R. W. Fischer, Angew. Chem., Int. Ed. Engl.,
1997, 36, 2652; (d) W. A. Herrmann and R. M. Kratzer, Inorg.
Synth., 2002, 33, 110.
7 (a) W. A. Herrmann, C. C. Romao, R. W. Fischer, P. Kiprof and
´
C. Meric de Bellefon, Angew. Chem., Int. Ed. Engl., 1991, 30, 185;
¨
Organomet. Chem., 1996, 514, 111.
¨
15 (a) M. M. Abu-Omar, P. J. Hansen and J. H. Espenson, J. Am.
Chem. Soc., 1996, 118, 4966; (b) S. Yamazaki, J. H. Espenson and
P. Huston, Inorg. Chem., 1993, 32, 4683; (c) P. J. Hansen and J. H.
Espenson, Inorg. Chem., 1995, 34, 5839; (d) J. H. Espenson, O.
Pestovski, P. Huston and S. Staudt, J. Am. Chem. Soc., 1994, 116,
2869.
(b) W. A. Herrmann, F. E. Kuhn, C. C. Romao, H. Tran Huy, M.
¨
Wang, R. W. Fischer, W. Scherer and P. Kiprof, Chem. Ber., 1993,
126, 45.
8 (a) W. A. Herrmann, C. C. Romao, P. Kiprof, J. Behm, M. R.
Cock and M. Taillefer, J. Organomet. Chem., 1991, 413, 11; (b) F.
16 R. W. Fischer, PhD Thesis, TU Munchen, 1994.
¨
17 A. Haaland, W. Scherer, H. V. Volden, H. P. Verne, O. Gropen,
G. S. McGrady, A. J. Deuns, G. Dierker, W. A. Herrmann,
P. W. Roesky and M. R. Geisberger, Organometallics, 2000,
19, 22.
E. Kuhn, J. Mink and W. A. Herrmann, Chem. Ber./Recl., 1997,
¨
130, 295; (c) J. M. Huggins, D. R. Randolph and L. Lebioda, J.
Organomet. Chem., 1986, 312, C15.
ꢀc
This journal is the Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2006
New J. Chem., 2006, 30, 1599–1605 | 1605