Chiral cationic [Cp′Mo(CO)2(NCMe)]+ species-catalyst precursors for olefin epoxidation with H2O 2 and tert-butyl hydroperoxide

Article abstract of DOI:10.1002/ejic.201001065

A novel cyclopentadienyl ligand bearing a chiral oxazoline pendant group (Cp^ox) has been prepared. Its coordination to molybdenum and tungsten afforded optically pure (R)-Cp^oxM(I·³-C ₃H₅)(CO)₂ (M = Mo, W) in which the pendant oxazoline fragment is not coordinated to the metal center. Reaction of (R)-Cp^oxMo(I·³-C₃H ₅)(CO)₂ with tetrafluoroboric acid gives the bidentate I·⁵-cyclopentadienyloxazoline complex [Cp ^oxMo(CO)₂(NCMe)]BF₄ in which the oxazoline is coordinated through the N-atom to the molybdenum center. Their catalytic performance in the epoxidation of cis-cyclooctene, (R)-limonene, and trans-β-methylstyrene with H₂O₂ and tert-butyl hydroperoxide (TBHP) as oxidants has been studied. (R)-Cp ^oxW(I·³-C₃H₅)(CO) ₂ displayed high catalytic activity achieving quantitative conversion of cyclooctene epoxide in 2 h with H₂O₂. [Cp ^oxMo(CO)₂(NCMe)]⁺ has been shown to be an efficient catalyst with TBHP and H₂O₂, reaching quantitative conversions of the corresponding epoxide in 30 min and 11 h, respectively. ESI-MS studies of the reaction of [Cp^oxMo(CO) ₂(NCMe)]⁺ with H₂O₂ and TBHP revealed the in situ formation of the corresponding peroxido [Cp ^oxMo(O₂)O]⁺ and dioxido [Cp ^oxMoO₂]⁺ species, respectively. Further oxidation of these complexes resulted in the loss of the cyclopentadienyloxazoline ligand. Based on spin trap experiments, the involvement of both carbon-and oxygen-centred radicals in the olefin epoxidation catalyzed by [Cp^oxMo(CO)₂(NCMe)]⁺ has been proved. Copyright

Full text of DOI:10.1002/ejic.201001065

FULL PAPER
DOI: 10.1002/ejic.201001065
Chiral Cationic [CpЈMo(CO)2(NCMe)]+ Species – Catalyst Precursors for
Olefin Epoxidation with H2O2 and tert-Butyl Hydroperoxide
Patrícia M. Reis,[a] Carla A. Gamelas,[b] José A. Brito,[a,c] Nathalie Saffon,[d]
Montserrat Gómez,[c] and Beatriz Royo*[a]
Keywords: Molybdenum / Tungsten / Cyclopentadienyl / Oxazoline / Epoxidation
A novel cyclopentadienyl ligand bearing a chiral oxazoline
pendant group (Cpox) has been prepared. Its coordination to
molybdenum and tungsten afforded optically pure (R)-
CpoxM(η3-C3H5)(CO)2 (M = Mo, W) in which the pendant ox-
azoline fragment is not coordinated to the metal center. Reac-
tion of (R)-CpoxMo(η3-C3H5)(CO)2 with tetrafluoroboric acid
gives the bidentate η5-cyclopentadienyloxazoline complex
achieving quantitative conversion of cyclooctene epoxide in
2 h with H2O2. [CpoxMo(CO)2(NCMe)]+ has been shown to
be an efficient catalyst with TBHP and H2O2, reaching quan-
titative conversions of the corresponding epoxide in 30 min
and 11 h, respectively. ESI-MS studies of the reaction of
[CpoxMo(CO)2(NCMe)]+ with H2O2 and TBHP revealed the
in situ formation of the corresponding peroxido [CpoxMo-
[CpoxMo(CO)2(NCMe)]BF4 in which the oxazoline is coordi- (O2)O]+ and dioxido [CpoxMoO2]+ species, respectively. Fur-
nated through the N-atom to the molybdenum center. Their
catalytic performance in the epoxidation of cis-cyclooctene,
(R)-limonene, and trans-β-methylstyrene with H2O2 and tert-
butyl hydroperoxide (TBHP) as oxidants has been studied.
(R)-CpoxW(η3-C3H5)(CO)2 displayed high catalytic activity
ther oxidation of these complexes resulted in the loss of the
cyclopentadienyloxazoline ligand. Based on spin trap experi-
ments, the involvement of both carbon- and oxygen-centred
radicals in the olefin epoxidation catalyzed by [CpoxMo(CO)2-
(NCMe)]+ has been proved.
alysts.[5] To avoid this problem, Kühn and coworkers have
developed a series of catalytically active ansa-bridged cyclo-
pentadienyl compounds of molybdenum.[6] However, some
of these complexes show low stability under oxidative con-
ditions leading to a significant decomposition of the cata-
lyst during the catalytic cycle. With the aim of stabilizing
cyclopentadienyl ansa-bridged complexes, we recently pre-
pared a series of ansa-(Cp-NHC)Mo(CO)2I (Cp = η5-C5H4,
η5-C5Me4, η5-C5(CH2Ph)4, NHC = N-heterocyclic carbene)
derivatives containing a cyclopentadienyl group function-
alized with a NHC, which were applied as olefin epoxid-
ation catalysts.[7] We envisioned that the strong M–NHC
bond would help to stabilize this type of complex. However,
poor catalytic activity and stability under oxidative condi-
tions were displayed by (Cp–NHC)Mo(CO)2I com-
pounds.[7,8] Based on these results, we decided to develop a
novel chelate cyclopentadienyl ligand capable of both stabi-
lizing molybdenum complexes under oxidative conditions
and introducing chirality into the system. Herein, we report
the preparation of a cyclopentadienyl ring bearing a pen-
dant oxazoline group. Oxazolines have been used as stabiliz-
ing agents for many transition metals based on their robust-
ness under a variety of reaction conditions.[9] Upon coordi-
nation of the oxazoline ring through the N atom, the stereo
directing substituent is situated in close proximity to the
metal center controlling the space available for the sub-
strate. Several chiral dioxidomolybdenum(VI) complexes
containing oxazoline ligands have been applied as epoxid-
Introduction
The excellent catalytic performance of CpMoO2X (Cp =
η5-C5R5; X = Cl, Br, CH3) derivatives and their carbonyl
precursors CpMo(CO)3X in the epoxidation of olefins with
TBHP is well established.[1] The catalytic activity of
CpMoO2Cl derivatives, first reported by Trost and
Bergman in 1991,[2] has recently been the subject of inten-
sive studies. Olefin epoxidation with a variety of cyclopen-
tadienylmolybdenum complexes of the general formula
CpMoO2X containing differently substituted cyclopen-
tadienyl rings has been reported.[3,4] Attempts to introduce
chirality into these systems through cyclopentadienyl
groups containing linked chiral ligands suffers from a major
drawback that free rotation of the cyclopentadienyl ring
yields negligible asymmetric induction when applied as cat-
[a] Instituto de Tecnologia Química e Biológica da Universidade
Nova de Lisboa,
Av. da República, EAN, 2780-157 Oeiras, Portugal
Fax: +351-214411277
E-mail: broyo@itqb.unl.pt
[b] Escola Superior de Tecnologia, Instituto Politécnico de Setúbal,
2910-761 Setúbal, Portugal
[c] Université Paul Sabatier, Laboratoire Hétérochimie Fondamen-
tale et Appliquée, UMR CNRS 5069,
118, route de Narbonne, 31062 Toulouse cedex 9, France
[d] Université de Toulouse, UPS, Structure Fédérative Toulousaine
en Chimie Moléculaire, FR2599,
118, route de Narbonne, 31062 Toulouse cedex, France
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201001065.
666
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 666–673

Chiral Cationic [CpЈMo(CO)2(NCMe)]+
ation catalysts.[10] However, little success has been achieved mixture of endo and exo isomers.[13] Nevertheless, the exo
until now in asymmetric epoxidation, probably due to labile configuration predominates in the solid state as shown by
behavior of the oxazoline moiety bound to dioxido-molyb- X-ray crystallographic studies.[14] This contrasts with that
denum(VI) fragments.[10b] We envisaged that functionaliza- seen for 3 as both isomers crystallize from dichloromethane,
tion of the oxazoline ligand with a cyclopentadienyl ring which has been established by X-ray crystallography (Fig-
might provide a bidentate ligand capable of stabilizing che- ure 1, see below for discussion).
late molybdenum complexes and preventing oxazoline de-
coordination. It is well known that the excellent spectator
ligand qualities of cyclopentadienyl rings, which usually
bind very strongly to transition metal centers, give high
thermal stabibilty.[11] To the best of our knowledge, chelate
oxazoline-derived cyclopentadienyl systems are not known.
Here we report the synthesis of the new cyclopentadieny-
loxazoline hybrid ligand 2 (Cpox) and its coordination to
molybdenum and tungsten. The chiral η3-allyl complexes
(R)-CpoxM(η3-C3H5)(CO)2 (M = Mo, W) and ionic
Scheme 1. Synthesis of compound 2.
[CpoxMo(CO)2(NCMe)]BF4 have been fully characterized
and applied as catalysts in olefin epoxidation with TBHP
and H2O2.
Results and Discussion
The cyclopentadienyl group bearing an oxazoline pen-
dant ligand 2 (Cpox) was obtained as shown in Scheme 1
by deprotonation at the methylene group of 2-methyl-(4R)-
phenyl-4,5-dihydrooxazole (1)[12] with nBuLi, followed by
reaction with 6,6-diphenylfulvene and treatment with meth-
anol. The product was isolated as an orange oil in high
1
yield (88%) and characterized by H and 13C NMR spec-
troscopy and mass spectrometry. Reaction of 2 with nBuLi
and subsequent treatment with Mo(η3-C3H5)Cl(CO)2-
(NCMe)2 afforded the η3-allyl dicarbonyl complex (R)-
CpoxMo(η3-C3H5)(CO)2 (3) as a crystalline yellow solid in
quantitative yield (Scheme 2). Compound 3 was charac-
Scheme 2. Synthesis of molybdenum and tungsten complexes 3–5.
terized by means of IR and NMR spectroscopy (1H, 13C,
95Mo), elemental analysis, and an X-ray single-crystal dif-
fraction study. Details on the NMR characterization of this
complex are giving in the Supporting Information [includ-
ing complete assignments of 13C NMR resonances based
on heteronuclear multiple quantum coherence (HMQC)
data]. On the basis of NMR spectroscopic data the two
isomers, endo and exo, were shown to be present in solution
1
(Scheme 3). The H NMR spectrum of 3 displayed a single
set of broadened resonances for the allyl protons at room
temperature due to exo–endo fluxional interconversion; two
broad signals at δ = 0.73 and 2.48 ppm were assigned to
Scheme 3. endo and exo isomers.
The analogous reaction of W(η3-C3H5)Cl(CO)2-
the allyl anti (Ha) and syn (Hs) protons, respectively. The (NCMe)2 with 2, previously deprotonated with nBuLi, gave
signal of Hc (corresponding to the central proton of the (R)-CpoxW(η3-C3H5)(CO)2 (4), which was isolated as a
allyl group) overlaps with the signal of the CH2 protons crystalline yellow solid in high yield. Its 1H NMR spectrum
of the linker between the cyclopentadienyl and oxazoline recorded at room temperature showed broad resonances for
fragments. At low temperature (263 K), each of the broad the allyl anti and syn protons at δ = 0.87 and δ = 2.43
signals splits into two signals of the same intensity at δ = ppm, respectively, suggesting the presence of exo and endo
0.67 and δ = 0.71 ppm for Ha and two resonances at δ = conformers of 4 in solution. Lowering the temperature of
2.38 and δ = 2.45 ppm for Hs, indicating that the two iso- the NMR probe to 263 K, causes splitting of the broad res-
mers coexist in solution in a 1:1 ratio. This situation is com- onance at δ = 2.43 ppm into two multiplets at δ = 2.42 and
mon in molybdenum complexes of general type CpMo(η3- 2.25 ppm (in 1:1 ratio); however, the signal corresponding
C3H5)(CO)2, which exist in solution as an approximate 1:1 to the allyl anti protons remained as a broad multiplet. Fur-
Eur. J. Inorg. Chem. 2011, 666–673
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
667

B. Royo et al.
FULL PAPER
The shift to higher frequency of the CO stretching band in
5 (compared with that of 3) is in agreement with the cat-
ionic nature of compound 5. These values are in agreement
with other carbonyl stretching bands reported in the litera-
ture for related complexes of general type [CpMo(CO)2-
(NCMe)2]BF4.[16] The high resolution ESI mass spectrum
of 5 gave the molecular peak corresponding to [M –
NCMe]+ at m/z = 544.0883. The BF4 anion was detected in
the negative mode of the ESI mass spectrum at m/z = 87
with the characteristic isotopic pattern of boron. Attempts
to synthesise the analogous tungsten cationic complex
[CpoxW(CO)2(NCMe)]BF4 were made following a similar
synthetic method to that used for 5. The reaction of 4 with
HBF4 yielded a red oil which contained the ionic complex
[CpoxW(CO)2(NCMe)]BF4 [ESI-MS analysis displayed a
molecular peak at m/z = 630 assigned to [CpoxW(CO)2]+;
but its NMR spectrum revealed that this compound was
contaminated with unidentified products. Therefore, the
tungsten complex [CpoxW(CO)2(NCMe)]BF4 was not used
in catalysis.
Figure 1. ORTEP diagram of compound 3 with 50% probability
ellipsoids. H atoms are omitted for clarity. Selected bond lengths
distances [Å] and angles [°]: Mo(1)–C(4) 1.935(4), Mo(1)–C(5)
1.946(3), Mo(1)–C(1) 2.282(19), Mo(1)–C(2) 2.319(6), Mo(1)–C(3)
2.331(18), Mo(1)–Cp*centroid 2.026, C(4)–Mo(1)–C(5) 75.60(15),
C(5)–Mo(1)–C(1) 79.2(5), C(4)–Mo(1)–C(3) 81.1(14).
Complex 3 crystallized from concentrated ether/hexane
solutions affording single crystals suitable for X-ray diffrac-
ther temperature decrease leads to the precipitation of 4 tion studies. The structure and selected bond lengths and
from the solution. angles are shown in Figure 1. The structure shows a pseu-
Both exo and endo isomers for 3 and 4 were not distin- dotetrahedral coordination around the MoII center with η3-
guished by their IR spectra recorded from KBr pellets. Two allyl, η5-substituted cyclopentadienyl, and two carbonyl li-
characteristic carbonyl absorption bands at ν(CO) 1941 and gands. The crystal contains both the exo and endo isomers
1855 cm–1 with a shoulder at 1846 cm–1 for 3 and 1932 and in a ratio of ca. 1:1. The average bond distance of molybde-
1842 cm–1 with a shoulder at 1833 cm–1 for 4 were observed. num to the five-membered cyclopentadienyl ring is 2.026 Å,
As expected, the CO absorption bands for the tungsten and the Mo–CO bond lengths are 1.946 and 1.935 Å. These
complex were found at lower frequencies than those of re- bond lengths are in close accordance to those observed for
lated molybdenum complex 3, due to the electron-rich na- dicarbonyl(η3-allyl)(η5-cyclopentadienyl)molybdenum(II)
ture of W compared to Mo. These data correlate with those complexes.[14] The pendant oxazoline arm is situated away
of corresponding CpM(η3-C3H5)(CO)2 complexes (M = from the metal center; the stereocenter (C27) shows R abso-
Mo, W).[13d,15]
lute configuration.
Reaction of 3 with HBF4 in dichloromethane at room
temperature followed by addition of acetonitrile, yielded the
ionic complex [CpoxMo(CO)2(NCMe)]BF4 (5) which was
isolated as a brown solid in high yield (Scheme 2). Com-
pound 5 has been identified by NMR and IR spectroscopy
Compounds 3 and 5 as Pre-Catalysts in Olefin Epoxidation
with TBHP
1
and HRMS spectrometry. Its H NMR spectrum indicates
We first investigated the catalytic activity of 3 and 5 with
intramolecular coordination of the oxazoline ring to the cyclooctene at 55 °C in chloroform using a catalyst:sub-
metal center through the N atom. The methylene protons strate:oxidant ratio of 1:100:200. Catalytic data are summa-
of the group adjacent to the oxazoline ligand are nonequiv- rized in Table 1. Compounds 3 and 5 were found to be
alent (AB system), displaying two doublets at δ = 3.90 and highly efficient catalysts for the epoxidation of cyclooctene.
3.75 ppm with JH–H = 16 Hz, suggesting the coordination They showed similar activities, achieving quantitative sub-
of the oxazoline group, whereas in complex 3, these CH2 strate conversion towards 1,2-epoxycyclooctane in 30 min
protons displayed a single resonance at δ = 3.78 ppm. The for the ionic complex 5, and in 45 min for the η3-allyl com-
singlet at δ = 1.95 ppm proved the coordination of acetoni- pound 3 (Figure 2, entries 4 and 1 in Table 1). Reduction
trile. Additional evidence for coordination of the oxazoline of the amount of 5 to 0.2 mol-% did not affect the yield
ligand was provided by the 13C NMR spectrum, which dis- (90%); a turnover frequency (TOF) of 5421 mol mol-
played two resonances for the carbonyl ligands at δ = 248.7
–1 h–1 was obtained (TOF calculated at 5 min). Four con-
cat
and 248.5 ppm, indicating that they are cis-oriented. The secutive catalytic runs were carried out without noticeable
IR spectrum of 5 in the solid state exhibited two carbonyl loss of activity of 5. The catalytic activity of 3 and 5 is
stretching bands at νsym(CO) at 1996 cm–1 and νasym(CO) comparable to that observed for complexes of the general
1904 cm–1. The band due to the symmetric CO stretching formula CpMo(CO)3X (X = Cl, Br, CH3) reported in the
vibration was found to be more intense than the asymmetric literature.[1–3] While we were preparing the submission of
one, corroborating the cis orientation of the CO ligands. this manuscript, Gonçalves and coworkers reported the
668
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 666–673

Chiral Cationic [CpЈMo(CO)2(NCMe)]+
catalytic activity of CpЈMo(η3-C3H5)(CO)2 (CpЈ = η5- stability of Cp-oxido-molybdenum catalysts during epoxid-
C5H5; η5-C5H4Me; η5-C5Me5) in olefin epoxidation with ation catalysis remains uncertain.[4] In order to gain an in-
TBHP.[17]
sight into the nature of the species formed during the course
of the reaction, the catalytic precursor 5 was treated with a
10-fold excess of TBHP in acetonitrile at room temperature,
and the reaction was followed by ESI-MS. After addition
of TBHP, three new peaks at m/z = 520, 408, and 392 imme-
diately appeared in the mass spectrum along with the peak
at m/z = 544, which corresponds to 5. The intensity of the
three new peaks increased in the first minute, while the in-
tensity of the peak corresponding to 5 decreased. After ap-
proximately 2 min of reaction, the peaks at m/z = 520 and
544 completely disappeared from the spectrum, while the
peaks at m/z = 408 and 392 remained. The peak at m/z
= 520 displayed the characteristic pattern arising from the
natural abundance of molybdenum isotopes and it was as-
signed to the dioxido-molybdenum species [CpoxMoO2]+
(6). Similar ionic dioxido-molybdenum species have been
isolated from the reaction of [CpMo(CO)2(NHC)-
(NCMe)]+ with TBHP.[8] The peaks at m/z = 408 and 392
correspond to a species that does not contain molybdenum,
and on the basis of the m/z values they have been assigned
to the cyclopentenone (7) and cyclopentadienyloxazoline li-
gand (2) (Scheme 4, mass spectra can be found in the Sup-
porting Information). At the end of the reaction of 5 with
TBHP, a pale yellow solid precipitated from the acetonitrile
solution. Attempts to characterize this yellow solid were
unsuccessful.
Table 1. Olefin epoxidation catalyzed by 3 and 5 with TBHP.[a]
Entry
Catalyst Olefin
Time
Yield [%][b]
1
2
3
4
5
6
3
3
3
5
5
5
cyclooctene
(R)-limonene
45 min
1 h
96
100
58[c]
98
trans-β-methylstyrene 16 h
cyclooctene
(R)-limonene
30 min
1 h
100
trans-β-methylstyrene 16 h
14[c]
[a] All reactions were carried out using a catalyst:substrate:oxidant
ratio 1:100:200 in CHCl3 at 55 °C. [b] Yield determined by GC. [c]
Reactions carried out at room temperature.
Figure 2. Kinetic profiles for the epoxidation of cyclooctene with
TBHP at 55 °C in the presence of complexes 3 (᭿) and 5 (᭹) (cata-
lyst load 1 mol-%).
The high catalytic performance of 3 and 5 extends to the
oxidation of (R)-(+)-limonene with TBHP achieving 100%
conversion to limonene oxide in 1 h. The reaction was fully
selective to the oxide giving a mixture of the corresponding
cis/trans epoxide in a ratio of 50:50 for 3 and 60:40 for 5.
Next, we investigated the epoxidation of trans-(β)-methyl-
styrene, a model substrate for trans-olefins, in order to ex-
plore the capability of these catalytic precursors to promote
asymmetric induction. In a typical catalytic run, one equiv-
alent of substrate was mixed with 1 mol-% catalyst and two
equivalents of TBHP. The reaction was carried out at room
temperature in order to optimize conditions for asymmetric
epoxidation, since higher temperatures are usually regarded
as detrimental for catalyst enantioselectivity. Poor conver-
sion of trans-β-methylstyrene was obtained after 16 h (38%
conversion) in the presence of a catalytic amount of 5, and
the reaction was not selective to the epoxide (14% yield,
entry 6 in Table 1) giving nearly racemic methylstyrene ox-
ide (Յ 5% ee). Using 3, the reaction was fully selective to
Scheme 4. Reaction of compound 5 with TBHP monitored by ESI-
MS analysis.
Similar ESI-MS analyses were carried out in the presence
methylstyrene epoxide (58% yield in 16 h, entry 3 in of cyclooctene in a 5:TBHP:olefin ratio of 1:10:10 in order
Table 1) without asymmetric induction (ee Յ 5%). to explore if the presence of the olefin would prevent the
An intramolecular coordination of the oxazoline moiety decomposition of the dioxido species 6. We found that a
during the catalytic process was expected to effect an asym- similar reaction took place in the presence of the olefin.
metric induction. However, negligible enantiomeric excess The addition of TBHP to a mixture of 5 and cyclooctene
was observed, even at low temperature. This behavior could in acetonitrile rapidly produced 6, 7, and 2 (Scheme 4). In
be due to the decoordination of the oxazoline moiety dur- view of these results, two main conclusions can be made: i)
ing the catalytic reaction.
the first step in the epoxidation reaction is the in situ forma-
Despite intensive studies on molybdenum cyclopen- tion of the dioxido-molybdenum compound 6, which is gen-
tadienyl compounds as catalysts for olefin epoxidation, the erated by complete oxidative decarbonylation of the carb-
Eur. J. Inorg. Chem. 2011, 666–673
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
669

B. Royo et al.
FULL PAPER
onyl metallic precursor; and ii) complex 6 reacts further los- Compounds 3–5 and CpM(CO)3Cl (M = Mo, W) as Pre-
ing the cyclopentadienyl fragment (Cpox).
Catalysts in Olefin Epoxidation with H2O2
The loss of the cyclopentadienyl ligand from CpMoO2Cl
compounds leading to the in situ generation of a much
more efficient molybdenum (per)oxido species has been al-
ready detected in the epoxidation of cis-cyclooctene with
THBP in the presence of CpArMoO2Cl (CpAr = η5-
C5Ph5).[4] Based on NMR and kinetic data, Colbran and
coworkers demonstrated that at the end of the catalytic re-
action the only molybdenum-containing species present
were formed by loss of the cyclopentadienyl ligand. Loss of
the cyclopentadienyl-bearing oxazoline ligand would ex-
plain the lack of asymmetric induction found in our experi-
ments. It should be noted that in a recent mechanistic study
reported by Kühn and coworkers on epoxidation catalyzed
by CpMoO2Me,[18] the decomposition of the catalyst to a
more active catalyst not containing the cyclopentadienyl
group has not been reported. Based on our findings, we
believe the warning given by Colbran et al.[4] that the loss
of the cyclopentadienyl ring must be taken into account
when these systems are applied as catalysts in epoxidation
reactions.
Despite the fact that a radical mechanism has not been
proposed in previous studies on epoxidation catalyzed by
cyclopentadienylmolybdenum complexes,[18,19] we per-
formed radical trapping experiments in order to confirm
the nonexistence of radicals in the epoxidation catalyzed by
our [CpoxMo(CO)2(NCMe)]+/TBHP system. Under a nitro-
gen atmosphere, addition of 20 equiv. per catalyst of
Ph2NH, an efficient scavenger of alkylperoxyl radicals,[20]
inhibited cyclooctene oxidation. A 10% yield of the corre-
sponding epoxide was obtained after 24 h of reaction, while
96% yield in 45 min was obtained in the absence of radical
scavenger. Other radical scavengers such as 2,2,6,6-tetrame-
thypiperidine-1-oxyl (TEMPO) also slowed down the reac-
tion to a great extent (Figure 3). In addition, 2,6-di-tert-
butyl-4-methylphenol (BHT) and CBrCl3 diminished the re-
action rate in the first 40 min, reaching quantitative conver-
sion of the corresponding epoxide in 90 min (Figure 3).
These results indicate that free radical species are involved
in the epoxidation reaction catalyzed by 5.
Based on our recent findings on the capability of
CpMo(CO)3Cl complexes to catalyze the oxidation of sul-
[21]
fides with H2O2
and the ability of molybdenum η3-allyl
carbonyl complexes to perform olefin epoxidation with
H2O2,[22] we decided to explore the catalytic performance
of 5 in the epoxidation of cis-cyclooctene with H2O2 (30%
in H2O) at 70 °C. The use of hydrogen peroxide is an at-
tractive option on environmental and economic grounds.
The by-product generated is water and the aqueous solution
is inexpensive and easy to handle. The catalytic reaction
was performed using acetonitrile as a co-solvent to give a
homogeneous mixture in which complex 5, substrate, and
oxidant were completely dissolved. Blank runs performed
in the absence of catalyst showed no epoxide formation un-
der the conditions applied. Compound 5 exhibited moder-
ate catalytic activity, reaching quantitative conversion in
11 h (92% conversion). The catalytic reaction was entirely
selective for epoxidation with 1,2-epoxycyclooctane the
only product obtained (Table 2, entry 2). Complex 3 also
catalyzed the epoxidation of cis-cyclooctene with H2O2. De-
spite the slower reaction, the epoxide selectivity was 100%
achieving 77% of epoxide yield in 20 h (Table 2, entry 1).
This is remarkable, since cyclopentadienyl compounds of
molybdenum have been reported as inactive catalysts for
olefin epoxidation using H2O2 as oxidant.[2,3b,5b,23] Never-
theless, the catalytic tests described in the literature were
performed at 55 °C, without evaluation of the effect of tem-
perature. For comparative purposes, we studied the per-
formance of CpMo(CO)3Cl in the epoxidation of cis-cyclo-
octene with H2O2 at 70 °C in acetonitrile. The catalytic
epoxidation reaction yielded 1,2-epoxycyclooctane as the
only product in a 92% yield after 10 h (Table 2, entry 5).
Similar experiments carried out at 55 °C gave negligible ep-
oxide yields using different catalytic precursors, 8% yield
of 1,2-epoxycyclooctane for CpMo(CO)3Cl and 9% for 5
(Table 2, entries 6 and 3, respectively), indicating that the
effect of the temperature is crucial in the epoxidation of
olefins with H2O2. In view of these results, we decided to
explore the catalytic performance of 4 under similar condi-
tions, as according to some literature reports, tungsten com-
plexes afford better results with H2O2 compared to molyb-
denum species.[24] Complex 4 showed excellent catalytic ac-
Table 2. Catalytic epoxidation catalysed by 3–5, CpMo(η3-
C3H5)(CO)2, and CpW(η3-C3H5)(CO)2 with H2O2.[a]
Entry Catalyst
T [°C]
t [h]
Yield [%]
[b]
1
2
3
4
5
6
7
3
70
70
55
70
70
55
70
20
11
20
2
10
20
5
77
92
9
88
92
8
5
5
4
CpMo(CO)3Cl
CpMo(CO)3Cl
CpW(η3-C3H5)(CO)2
Figure 3. Kinetic profiles for the epoxidation of cyclooctene with
TBHP at 55 °C using 5 as catalyst in the presence of several spin
92
traps: without spin trap (᭡), CBrCl3 (᭿), BHT (♦), TEMPO (᭹), [a] All reactions were carried out with a catalyt:substrate:oxidant
and Ph2NH (—).
ratio 1:100:200 in acetonitrile. [b] Yield determined by GC.
670
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 666–673

Chiral Cationic [CpЈMo(CO)2(NCMe)]+
tivity in the epoxidation of cyclooctene with H2O2 affording cals in the catalytic epoxidation of olefins, as the addition
quantitative conversion to 1,2-epoxycyclooctane (88% of C- and O-centred radical traps such as CBrCl3, Ph2NH,
yield) in 2 h (Table 2, entry 4). The unsubstituted cyclopen- BHT, and TEMPO to the catalytic reaction mixture sup-
tadienyl derivative CpW(η3-C3H5)(CO)2 displayed similar press the formation of the corresponding epoxide (Fig-
catalytic activity when applied as catalyst for cis-cyclooc- ure 5). These findings provide strong evidence for a radical
tene epoxidation with H2O2 achieving quantitative conver- mechanism involved in the epoxidation of olefins catalyzed
sion to the corresponding epoxide in 5 h (Table 2, entry 7). by this system.
The course of the epoxidation reactions with H2O2 and cat-
alysts 3–5 and CpW(η3-C3H5)(CO)2 is shown in Figure 4
and additional data are summarized in Table 2. The cata-
lytic activity of [Cp*2M2O5] (M = Mo, W) in the epoxid-
ation of cyclooctene using aqueous H2O2 as oxidant has
recently been reported by Poli and coworkers.[19]
Figure 5. Kinetic profiles for the epoxidation of cyclooctene with
H2O2 at 70 °C using 5 as catalyst in the presence of several spin
traps: without spin trap (♦), TEMPO (᭿), CBrCl3 (᭡), and BHT
(—).
Conclusions
Figure 4. Kinetic profiles for the epoxidation of cyclooctene with
H2O2 at 70 °C in the presence of complexes 3 (᭹), 4 (᭡), 5 (᭿),
In this work, we have prepared the novel cyclopen-
tadienyl ligand 2 (Cpox) functionalized with an oxazoline
ring, which has been coordinated to molybdenum and tung-
sten affording the corresponding metal complexes (R)-
CpoxM(η3-C3H5)(CO)2 [M = Mo (3), W (4)]. The pure chi-
ral cationic complex of molybdenum [CpoxMo(CO)2-
(NCMe)]BF4 (5) has been isolated and fully characterized.
Compound 5 is a highly active catalyst in olefin epoxidation
with TBHP and a moderate catalyst with H2O2. Our find-
ings based on ESI-MS analysis corroborate the instability
of cyclopentadienyl-oxido-molybdenum(VI) catalysts dur-
ing the epoxidation reaction already observed by Colbran
and coworkers, and suggest that different reaction pathways
can compete in the epoxidation of olefins catalyzed by
[CpoxMo(CO)2(NCMe)]+ systems. We have disclosed the in-
volvement of both C- and O-centred radicals in the epoxid-
ation reaction catalyzed by the [CpoxMo(CO)2(NCMe)]BF4
as shown by experiments with radical traps. As far as we
know, this is the first time that experimental results prove
the involvement of radicals in the epoxidation reaction cata-
lyzed by cyclopentadienyl molybdenum complexes. These
findings might explain the lack of asymmetric induction
displayed by these systems.
and CpW(η3-C3H5)(CO)2 (♦).
The reaction of 5 with a 10-fold excess of H2O2 was mon-
itored by ESI-MS analysis. After addition of 10 equiv. of
H2O2 to an acetonitrile solution of 5, three new peaks at
m/z = 536, 408, and 392 immediately appeared in the spec-
trum. The peak at m/z = 536 has been assigned to the ox-
ido(peroxido)molybdenum complex [CpoxMo(O2)O]+ (8)
(assigned on the basis of the m/z value and its isotopic dis-
tribution), the peak at m/z = 408 corresponds to 7 (already
detected in the reaction of 5 with TBHP, see above), and
the peak at m/z = 392 corresponds to 2 (Scheme 5). After
some minutes of reaction, the peak at m/z = 536 completely
disappeared from the spectrum and the only peaks that re-
mained were those at m/z = 408 and 392. Despite the fact
that Cp*Mo(O2)OCl has been shown to be totally inactive
in olefin epoxidation,[2] subsequent studies by Kühn have
shown that related complexes with different ligands such as
CpMoO(O2)(CH3) are catalytically active.[18]
Experimental Section
General Procedures: Mo(η3-C3H5)Cl(CO)2(NCMe)2,[13d] W(η3-
C3H5)Cl(CO)2(NCMe)2,[15] and 6,6-diphenylfulvene[25] were pre-
pared following literature procedures. (4R)-2-Methyl-4-phenyl-4,5-
dihydrooxazole (1) was prepared following the method described
by Meyers,[12] starting from (R)-(–)-2-phenylglycinol and triethyl
orthoacetate. All other reagents were used as received from com-
Scheme 5. Reaction of 5 with H2O2 monitored by ESI-MS analysis.
A further insight into the mechanism of the catalytic re-
action with H2O2 came from radical trapping experiments,
suggesting the involvement of both C- and O-centred radi- mercial suppliers and used without further purification. 1H and
Eur. J. Inorg. Chem. 2011, 666–673
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
671

B. Royo et al.
FULL PAPER
13C NMR spectra were recorded with Bruker Avance III 400 MHz.
Infrared spectra were recorded from samples as KBr pellets using a
Mattson 7000 FTIR spectrometer. Electrospray mass spectra were
recorded with a Micromass Quatro LC instrument, nitrogen was
employed as the drying and nebulizing gas. Optical rotations were
determined with a Perkin–Elmer 241 polarimeter. Elemental analy-
ses were performed in our laboratories at ITQB. Complex 5 did
not give reproducible elemental analysis do to the retention of frac-
tional amounts of solvent. In lieu of acceptable elemental analysis
for compound 5 HRMS (ESI-TOF) were recorded.
Cp), 6.80 (m, 2 H, Ph), 7.12–7.28 (m, 13 H, Ph) ppm. 13C NMR
(100 MHz, CD2Cl2): δ = 237.7 (CO), 166.5 (NCO), 147.3 (Cipso
Ph), 142.2 (Cipso Ph), 129.1(CPh), 128.7 (CPh), 128.5 (CPh), 128.1
(CPh), 127.8 (CPh), 124.5 [C5(Cp)-C], 93.5 (Cp), 93.2 (Cp), 92.8
(Cp), 92.6 (Cp), 74.5 (O–CH2–CHPh), 69.6 (O–CH2–CHPh), 67.9
(Ccentral allyl), 51.6 (CPh2 linker), 41.3 (Canti/syn allyl), 40.7 (CH2 linker
)
ppm. 95Mo NMR (26 MHz, CD Cl ): δ = 92.3 ppm. IR (KBr): ν
˜
2
2
= 1941 (vs), 1855 (vs), 1846 (sh) [ν(CO)] cm–1.
Synthesis of 4: A solution of 2 (145 mg, 0.37 mmol) in THF
(30 mL) was treated with nBuLi (0.24 mL of 1.6 m in hexane,
0.38 mmol) at –60 °C. The mixture was stirred for 15 min followed
by warming to room temperature and further stirred for 1 h. To
the resulting mixture, a solution of W(η3-C3H5)Cl(CO)2(NCMe)2
(147 mg, 0.37 mmol) in THF (5 mL) was added at –10 °C, and the
reaction mixture was warmed to room temperature and stirred
overnight. All volatiles were removed under vacuum and the re-
maining solid was extracted into diethyl ether yielding 4 as a yellow
solid (149 mg, 54%). C33H29NO3W(OC4H8) (743): calcd. C 59.73,
H 5.02, N 1.88; found C 59.40, H 5.20, N 1.64. [α]2D0 = –6.0 (c =
0.2, CH2Cl2). The NMR assignments below were determined with
the help of 1H-13C-HMQC. 1H NMR (400 MHz, CD2Cl2): δ = 0.87
(broad m, 2 H, Ha allyl), 2.43 (broad m, 2 H, Hs allyl), 3.40 (m, 1
H, Hm allyl), 3.54 (m, 1 H, O–CHAHB–CHPhN), 3.77 (s, 2 H,
CH2 linker), 4.29 (m, 1 H, O–CHAHB–CHPhN), 4.93 (m, 1 H, O–
CH2–CHPhN), 5.38 (s, 2 H, Cp), 5.78 (s, 2 H, Cp), 6.78 (m, 2 H,
Ph), 7.13–7.33 (m, 13 H, Ph) ppm. 13C NMR (100 MHz, CD2Cl2):
δ = 226.3 (CO), 166.2 (NCO), 147.0 (Cipso Ph), 142.3 (Cipso Ph),
129.1 (CPh), 128.9 (CPh), 128.7 (CPh), 128.4 (CPh), 127.9 (CPh),
122.5 [C5(Cp)–C], 92.4 (Cp), 92.3 (Cp), 91.7 (Cp), 74.3 (O–CH2–
CHPh), 69.6 (O–CH2–CHPh), 61.1 (Ccentral allyl), 51.8 (CPh2 linker),
X-ray Diffraction Studies: X-ray data for crystal 3 were collected at
low temperature (193 K) using an oil-coated shock-cooled crystal
with a Bruker-AXS APEXII diffractometer with Mo-Kα radiation
(λ = 0.71073 Å). The structures were solved by direct phase deter-
mination (SHELXS-97)[26] and refined for all non-hydrogen atoms
by full-matrix least-square methods on F2 and subject to aniso-
tropic refinement.[27]
Crystal data for 3: C33H29MoNO3, M = 583.51, orthorhombic,
space group P212121, a = 10.0441(2) Å, b = 11.4927(2) Å, c =
23.2420(2) Å,
V
=
2682.91(8) Å3,
Z
=
4, crystal size
0.38ϫ0.28ϫ0.06 mm3, 50594 reflections collected (6615 indepen-
dent, Rint = 0.0609), 371 parameters, R1 [IϾ2σ(I)] = 0.0314, wR2
[all data] = 0.0717, largest diff. peak and hole: 0.532 and
–0.354 eÅ–3.
CCDC-793815 contains the supplementary crystallographic 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.
Synthesis of 2: A hexane solution of nBuLi (1.9 mL of 1.6 m in
hexane, 3.04 mmol) was added dropwise to a solution of 1 (473 mg,
2.93 mmol) in dried THF (20 mL) at –60 °C. The mixture was
stirred for 30 min, 6,6-diphenylfulvene (675 mg, 2.93 mmol) was
added and the reaction mixture was allowed to reach room tem-
perature with stirring for 3 h. Methanol (2 mL) was added, and
the volatiles were evaporated. The crude oil was purified by flash
chromatography (hexane/ethyl acetate, 4:1) affording 2 as an
40.9 (CH2 linker), 32.9 (Canti/syn allyl) ppm. IR (KBr): ν = 1932 (vs),
˜
1842 (vs), 1833 (sh) [ν(CO)] cm–1.
Synthesis of 5: A solution of 3 (474 mg, 0.72 mmol) in CH2Cl2
(20 mL) was treated with HBF4 (0.44 mL, 54% in diethyl ether,
1.74 mmol) at room temperature. The mixture was stirred for
15 min followed by addition of acetonitrile (10 mL), and the reac-
tion mixture was stirred for further 2 h. All volatiles were removed
under vacuum and the remaining solid was washed with toluene
and extracted into dichloromethane yielding 5 as a brown solid
(290 mg, 60%). The NMR assignments below were determined
orange oil (101 mg, 88%). [α]2D0 = +21.2 (c = 0.9, diethyl ether). H
1
NMR (400 MHz, CD2Cl2): δ = 7.37–6.85 (m, 15 H, CHPh), 6.47–
6.28 (m, 4 H, CHCp), 4.89 (m, 1 H, NCHPh), 4.30 (m, 1 H, OCH2),
3.66 (m, 1 H, OCH2), 3.56 (s, 2 H, CH2 linker), 3.04 (s, 1 H, CHCp
)
1
1
with the help of H-13C-HMQC. H NMR (400 MHz, CD3CN): δ
= 1.95 (s, 3 H, NCMe), 3.75 (d, JH-H = 16 Hz, 1 H, CHAHB linker),
3.90 (d, JH-H = 16 Hz, 1 H, CHAHB linker), 4.47 (m, 1 H, O–
CHAHB–CHPhN), 5.14 (m, 1 H, O–CHAHB–CHPhN), 5.26 (m, 1
H, O–CH2–CHPhN), 5.81 (m, 1 H, Cp), 5.85 (m, 2 H, Cp), 5.90
(m, 1 H, Cp), 6.82 (m, 2 H, Ph), 7.21–7.46 (m, 13 H, Ph) ppm.
13C NMR (100 MHz, CD3CN): δ = 248.7 (CO), 248.5 (CO), 178.8
(NCO), 145.8 (Cipso Ph), 144.8 (Cipso Ph), 135.5–130.2 (CPh), 123.7
[C5(Cp)–C], 102.3 (Cp), 99.0 (Cp), 96.9 (Cp), 95.3 (Cp), 81.8 (O–
CH2–CHPh), 61.4 (O–CH2–CHPh), 52.8 (CPh2 linker), 39.3
(CH2 linker) ppm. 95Mo NMR (26 MHz, CD2Cl2): δ = 40.1 ppm.
ppm. 13C NMR (100 MHz, CD2Cl2): δ = 166.1 (OCN), 145.6 (CPh),
134.6 (CCp), 132.9 (CCp), 129.7 (CPh), 129.14 (CPh), 127.7 (CPh),
126.7 (CPh), 126.3 (CPh), 74.3 (OCH2), 69.5 (NCHPh), 42.46 (CCp),
40.98 (CH2 linker), 39.0 (CPh2 linker) ppm. MS (EI): m/z (%) = 392
[M + H+].
Synthesis of 3: A solution of 2 (256 mg, 0.65 mmol) in THF
(30 mL) was treated with nBuLi (0.41 mL of 1.6 m in hexane,
0.68 mmol) at –60 °C. The mixture was stirred for 15 min followed
by warming to room temperature and further stirred for 1 h. To
the resulting mixture, a solution of Mo(η3-C3H5)Cl(CO)2(NCMe)2
(203 mg, 0.65 mmol) in THF (5 mL) was added at –30 °C, and the
reaction mixture was warmed to room temperature and stirred
overnight. All volatiles were removed under vacuum and the re-
maining solid was extracted into diethyl ether yielding 3 as a yellow
HRMS (EI): m/z = 544.0883 [M – NCMe]+. IR (KBr): ν = 1996
˜
(vs), 1904 (vs) [ν(CO)] cm–1.
Olefin Epoxidation Experiments with TBHP: The catalytic reac-
tions were performed in a reaction vessel equipped with a magnetic
stirrer immersed in an oil bath at the appropriate temperature. A
catalyst:olefin:TBHP ratio of 1:100:200 was used, with 2 mL of
CHCl3. Olefin, chloroform, mesitylene (as internal standard), and
solid (426 mg, 89%). [α]2D0
= +18.1 (c = 0.2, CH2Cl2).
C33H29MoNO3(OC4H8) (657): calcd. C 67.56, H 5.67, N 2.13;
found C 67.70, H 5.65, N 2.01. The NMR assignments below were
determined with the help of 1H-13C-HMQC (400 MHz, CD2Cl2):
δ = 0.73 (broad m, 2 H, Ha allyl), 2.48 (broad m, 2 H, Hs allyl), the catalyst were placed into the reaction vessel, and TBHP was
3.55 (m, 1 H, O–CHAHB–CHPhN), 3.67 (m, 3 H, Hm allyl, added to the mixture. The course of the reaction was monitored by
CH2 linker), 4.28 (m, 1 H, O–CHAHB–CHPhN), 4.89 (m, 1 H, O–
quantitative GC analysis. Samples taken were diluted with CH2Cl2
CH2–CHPhN), 5.17 (m, 2 H, Cp), 5.29 (s, 1 H, Cp), 5.62 (s, 1 H, and treated with MgSO4 and MnO2 to remove water and destroy
672
www.eurjic.org
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2011, 666–673

Chiral Cationic [CpЈMo(CO)2(NCMe)]+
[8] During the preparation of this manuscript, stability studies of
CpMo(NHC)(CO)2X complexes under oxidative conditions
have been described: S. Li, C. W. Kee, K.-W. Huang, T. S. A.
Hor, J. Zhao, Organometallics 2010, 29, 1924.
the excess peroxide. The resulting slurry was filtered, and the fil-
trate was injected into a GC column. The conversion of the olefin
and the formation of the corresponding epoxide were calculated
from calibration curves (r2 = 0.999) recorded prior to the reaction.
[9] M. Gómez, G. Muller, M. Rocamora, Coord. Chem. Rev. 1999,
Olefin Epoxidation Experiments with H2O2: The catalytic reactions
were performed in a reaction vessel equipped with a magnetic stir-
rer and immersed in an oil bath at the appropriate temperature. A
catalyst:olefin:H2O2 ratio of 1:100:200 was used, with 2 mL of
NCMe. Olefin, acetonitrile, mesitylene (as internal standard), and
the catalyst were placed into the reaction vessel, and H2O2 was
added to the mixture. The course of the reaction was monitored by
quantitative GC analysis as described above.
193–195, 769.
[10] a) M. Gómez, S. Jansat, G. Muller, G. Noguera, H. Teruel, V.
Moliner, E. Cerrada, M. Hursthouse, Eur. J. Inorg. Chem. 2001,
1071; b) J. A. Brito, M. Gómez, G. Muller, M. Teruel, J.-C.
Clinet, E. Duñach, M. A. Maestro, Eur. J. Inorg. Chem. 2004,
4278; c) J. A. Brito, H. Teruel, G. Muller, S. Massou, M.
Gómez, Inorg. Chim. Acta 2008, 361, 2740; d) F. E. Kühn,
A. M. Santos, A. D. Lopes, I. S. Gonçalves, J. E. Rodríguez-
Borges, M. Pillinger, C. C. Romão, J. Organomet. Chem. 2001,
621, 201; e) P. Neves, S. Gago, C. C. L. Pereira, S. Figuereido,
A. Lemos, A. D. Lopes, I. S. Gonçalves, M. Pillinger, C. M.
Silva, A. A. Valente, Catal. Lett. 2009, 132, 94.
Radical Trap Experiments: The catalytic reactions were performed
in a reaction vessel equipped with a magnetic stirrer immersed in
an oil bath at the appropriate temperature. A 5:olefin:oxidant ratio
of 1:100:200 was used with 2 mL of CHCl3 (for TBHP) and NCMe
(for H2O2). Olefin, chloroform, mesitylene (as internal standard),
and spin trap in a ratio 1.3:1 relative to substrate, and the catalyst
were placed into the reaction vessel, and the oxidant was added to
the mixture. The course of the reaction was monitored by quantita-
tive GC analysis as described above.
[11]
Selected reviews on cyclopentadienyl rings bearing functional
groups: a) U. Siemeling, Chem. Rev. 2000, 100, 1495; b) H.
Butenschön, Chem. Rev. 2000, 100, 1527.
[12]
[13]
A. I. Meyers, G. Knaus, K. Kamata, M. E. Ford, J. Am. Chem.
Soc. 1976, 98, 567.
a) R. B. King, Inorg. Chem. 1966, 5, 2242; b) J. W. Faller, M. J.
Incorvia, Inorg. Chem. 1968, 7, 840; c) J. W. Faller, A. Jakubow-
ski, J. Organomet. Chem. 1971, 31, C75; d) J. W. Faller, C.-C.
Chen, M. J. Mattina, A. Jakubowski, J. Organomet. Chem.
1973, 52, 361; e) J. W. Faller, B. C. Whitmore, Organometallics
1986, 5, 752.
Supporting Information (see footnote on the first page of this arti-
cle): Mass spectra, NMR spectra, details on the crystal structure
determination.
[14]
a) W. E. Vanarsdale, J. K. Kochi, J. Organomet. Chem. 1986,
317, 215; b) J. W. Faller, D. F. Chodosh, D. Katahira, J. Or-
ganomet. Chem. 1980, 187, 227; c) S. Was, M. J. Begley, P.
Mountford, J. Organomet. Chem. 1995, 489, C28; d) M. J. Be-
gley, P. Mountford, P. J. Stewart, D. Swallow, S. Wan, J. Chem.
Soc., Dalton Trans. 1996, 1323; e) M. Green, T. D. McGrath,
R. L. Thomas, A. P. Walker, J. Organomet. Chem. 1997, 532,
61; f) H. G. Alt, H. E. Engelhardt, B. Wrackmeyer, R. D. Rog-
ers, J. Organomet. Chem. 1989, 379, 289.
Acknowledgments
The authors gratefully acknowledge financial support from the
Portuguese Fundação para a Ciência e a Tecnologia (FCT): project
PTDC/QUI-QUI/098682/2008), grant SFRH/BPD/20655/2004 (to
P. M. R.), grant SFRH/BD/30917/2006 (to J. A. B.), REDE/1517/
RMN/2005, and REDE71504/REM/2005. M. C. Almeida and Dr.
A. Coelho are acknowledged for providing data from the Elemental
Analyses and Mass Spectrometry Services at ITQB. The authors
also wish to thank Marc Vedrenne from University Paul Sabatier,
Toulouse, France, for 95Mo NMR spectra.
[15]
[16]
I. S. Gonçalves, C. C. Romão, J. Organomet. Chem. 1995, 486,
155.
a) S. A. Benyunes, A. Binelli, M. Green, M. J. Grimshire, J.
Chem. Soc., Dalton Trans. 1991, 895; b) I. J. S. Fairlamb, J. M.
Lynam, I. E. Taylor, A. C. Whitwood, Organometallics 2004,
23, 4964.
[17]
P. Neves, C. C. L. Pereira, F. A. A. Paz, S. Gago, M. Pillinger,
C. M. Silva, A. A. Valente, C. C. Romão, I. S. Gonçalves, J.
Organomet. Chem. 2010, 695, 2311.
[1] For recent reviews, see: a) F. E. Kühn, A. M. Santos, M. Abr-
antes, Chem. Rev. 2006, 106, 2455; b) C. Freud, M. Abrantes,
F. E. Kühn, C. C. Romão, J. Organomet. Chem. 2006, 691,
3718.
[2] M. K. Trost, R. G. Bergman, Organometallics 1991, 10, 1172.
[3] a) M. Abrantes, A. M. Santos, J. Mink, F. E. Kühn, C. C.
[18] A. M. Al.-Ajulouni, D. Veljanosvki, A. Capapé, J. Zhao, E.
Herdtweck, M. J. Calhorda, F. E. Kühn, Organometallics 2009,
28, 639.
Romão, Organometallics 2003, 22, 2112; b) A. M. Martins, [19] Ch. Dinoi, M. Ciclosi, E. Manoury, L. Maron, L. Perrin, R.
C. C. Romão, M. Abrantes, M. C. Azevedo, J. Cui, A. R. Dias,
T. Duarte, M. A. Lemos, T. Lourenço, R. Poli, Organometallics
2005, 24, 2582; c) J. Zhao, A. M. Santos, E. Herdtweck, F. E.
Kühn, J. Mol. Catal. A 2004, 222, 265; d) A. A. Valente, J. D.
Seixas, I. S. Goncalves, M. Abrantes, M. Pillinger, C. C.
Romão, Catal. Lett. 2005, 101, 127.
Poli, Chem. Eur. J. 2010, 16, 9572.
[20] P. A. MacFaul, D. D. M. Wayner, K. U. Ingold, Acc. Chem.
Res. 1998, 31, 159.
[21] C. A. Gamelas, T. Lourenço, A. P. da Costa, A. L. Simplício,
B. Royo, C. C. Romão, Tetrahedron Lett. 2008, 49, 4708.
[22] V. V. K. M. Kandepi, J. M. S. Cardoso, B. Royo, Catal. Lett.
2010, 136, 222.
[23] A. A. Valente, J. D. Seixas, I. S. Goncalves, M. Abrantes, M.
Pillinger, C. C. Romão, Catal. Lett. 2005, 101, 127.
[24] A. V. Biradar, M. K. Dongare, S. B. Umbakar, Tetrahedron
Lett. 2009, 50, 2885.
[25] J. L. Kice, F. M. Parham, J. Am. Chem. Soc. 1958, 80, 3792.
[26] G. M. Sheldrick, SHELXL-97, Program for Crystal Structure
Refinement, University of Göttingen, Göttingen, Germany,
1997.
[27] G. M. Sheldrick, Acta Crystallogr., Sect. A 1990, 46, 467.
Received: October 6, 2010
[4] M. Pratt, J. B. Harper, S. B. Colbran, Dalton Trans. 2007, 2746.
[5] a) M. Abrantes, A. Sakthivel, C. C. Romão, F. E. Kühn, J. Or-
ganomet. Chem. 2006, 691, 3137; b) M. Abrantes, F. A. Alme-
ida Paz, A. A. Valente, C. C. L. Pereira, S. Gago, A. R. Rodrig-
ues, J. Klinowski, M. Pillinger, I. S. Gonçalves, J. Organomet.
Chem. 2009, 694, 1826.
[6] a) J. Zhao, E. Herdtweck, F. E. Kühn, J. Organomet. Chem.
2006, 691, 2199; b) J. Zhao, K. R. Jain, E. Herdtweck, F. E.
Kühn, Dalton Trans. 2007, 5567; c) A. Capapé, A. Raith, F. E.
Kühn, Adv. Synth. Catal. 2009, 351, 66.
[7] V. V. K. M. Kandepi, A. P. da Costa, E. Peris, B. Royo, Organo-
metallics 2009, 28, 4544.
Published Online: January 3, 2011
Eur. J. Inorg. Chem. 2011, 666–673
© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
673