Cyclopentadienyl molybdenum dicarbonyl η3-allyl complexes as catalyst precursors for olefin epoxidation. Crystal structures of Cp′Mo(CO)2(η3-C3H5) (Cp′ = η5-C5H4Me, η5-C5Me5)

Article abstract of DOI:10.1016/j.jorganchem.2010.07.001

The complexes Cp′Mo(CO)₂(η³-C ₃H₅) [Cp′ = η⁵-C₅H ₅ (1), η⁵-C₅H₄Me (2), η⁵-C₅Me₅ (3)] have been prepared, structurally characterised by X-ray diffraction (2, 3), and tested as catalyst precursors for the epoxidation of olefins at 55 °C. Complex 1 gave a turnover frequency (TOF) of 310 mol mol_Mo^-1 h^-1 in the epoxidation of cis-cyclooctene with tert-butylhydroperoxide (TBHP, in decane) as oxidant, and 1,2-epoxycyclooctane was obtained quantitatively within 6 h. A similar result was obtained for complex 2, while the TOF for 3 was about one order of magnitude lower, suggesting a possible activity dependence on the ring substituents. For 1 the use of 1,2-dichloroethane as solvent increased the initial reaction rate to 361 mol mol_Mo^-1 h^-1, with no decrease in epoxide selectivity. Under these conditions the reaction rates for other olefins increased in the order 1-octene < trans-2-octene < cyclododecene < (R)-(+)-limonene < cis-cyclooctene, and, with the exception of limonene, the corresponding epoxide was the only product. For 1 the selective epoxidation of cis-cyclooctene could also be achieved in aqueous solution, using TBHP or H₂O₂ as oxidants, which gave epoxide yields of 99% and 27% at 24 h, respectively. The possibility of facilitating catalyst recycling by using ionic liquids as solvents was investigated.

Full text of DOI:10.1016/j.jorganchem.2010.07.001

Journal of Organometallic Chemistry 695 (2010) 2311e2319
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Cyclopentadienyl molybdenum dicarbonyl
h
3-allyl complexes as catalyst
precursors for olefin epoxidation. Crystal structures of Cp0Mo(CO)2(
h
3-C3H5)
(Cp0 ¼
h
5-C5H4Me,
h
5-C5Me5)
Patrícia Neves a, Cláudia C.L. Pereira a b, Filipe A. Almeida Paz a, Sandra Gago a, Martyn Pillinger a,
,
Carlos M. Silva a, Anabela A. Valente a , Carlos C. Romão d, Isabel S. Gonçalves a
a Department of Chemistry, CICECO, University of Aveiro, 3810-193 Aveiro, Portugal
,
c
,
,
*
*
b Unit of Chemical and Radiopharmaceutical Sciences (UCRS), Nuclear and Technological Institute, Estrada Nacional 10, 2686-953 Sacavém, Portugal
c Instituto de Tecnologia Química e Biológica da Universidade Nova de Lisboa, Av. da República, Estação Agronómica Nacional, 2780-157 Oeiras, Portugal
d Alfama Ltd., Taguspark, Núcleo Central 267, 2740-122, Porto Salvo, Portugal
a r t i c l e i n f o
a b s t r a c t
The complexes Cp0Mo(CO)2(h3-C3H5) [Cp0 ¼
h h h
5-C5H5 (1), 5-C5H4Me (2), 5-C5Me5 (3)] have been prepared,
Article history:
Received 19 May 2010
Received in revised form
29 June 2010
Accepted 1 July 2010
Available online 5 August 2010
structurally characterised by X-ray diffraction (2, 3), and tested as catalyst pꢁre1cursors for the epoxidation of
olefins at 55 ꢀC. Complex 1 gave a turnover frequency (TOF) of 310 mol molMo ꢁ1 in the epoxidation of cis-
cyclooctene with tert-butylhydroperoxide (TBHP, in decane) as oxidant, and 1,2-epoxycyclooctane was
obtained quantitatively within 6 h. A similar result was obtained for complex 2, while the TOF for 3 was
about one order of magnitude lower, suggesting a possible activity dependence on the ring substituents. For
h
ꢁ1
1 the use of 1,2-dichloroethane as solvent increased the initial reaction rate to 361 mol mol
h
ꢁ1, with no
Mo
Keywords:
decrease in epoxide selectivity. Under these conditions the reaction rates for other olefins increased in the
order 1-octene < trans-2-octene < cyclododecene < (R)-(þ)-limonene < cis-cyclooctene, and, with the
exception of limonene, the corresponding epoxide was the only product. For 1 the selective epoxidation of
cis-cyclooctene could also be achieved in aqueous solution, using TBHP or H2O2 as oxidants, which gave
epoxide yields of 99% and 27% at 24 h, respectively. The possibility of facilitating catalyst recycling by using
ionic liquids as solvents was investigated.
Molybdenum
Cyclopentadienyl ligands
Oxidative decarbonylation
Homogeneous catalysis
Olefin epoxidation
Ó 2010 Elsevier B.V. All rights reserved.
1. Introduction
of the Cp0M(CO)2( 3-C3H5) moiety as a labelling group in bio-
h
organometallic chemistry [7].
The allyl complexes Cp0M(CO)2(h3-C3H5) [Cp0 ¼ Cp (
h
5-C5H5) or
Since 2003 there has been growing interest in the use of
molybdenum carbonyl complexes as precursors to MoVI catalysts for
oxidation reactions [8e13], with most of the research being
focussed on tricarbonyl complexes [8,12,13]. The oxidative decar-
substituted Cp; M ¼ Cr, Mo, W] have for some time been of
considerable theoretical and experimental interest [1e6]. From
a synthetic standpoint, these complexes serve as precursors to
several mixed ring and indenyl analogues of molybdenocene and
bonylation of Cp0Mo(CO)3Cl (Cp0 ¼
h
5-C5R5; R ¼ H, Me, CH2Ph) with
tungstenocene [5]. For example, protonation of Cp0Mo(CO)2(h3
-
excess tert-butylhydroperoxide (TBHP) in n-decane provides
a convenient pathway to generating the dioxomolybdenum(VI)
complexes Cp0MoO2Cl [8a]. Like the isoelectronic derivatives
with general formula MoO2X2L (X ¼ Cl, Br, Me; L ¼ bidentate Lewis
base ligand), the Cp0MoO2X complexes (X ¼ Cl, alkyl) catalyse the
epoxidation of olefins with TBHP as the oxidising agent
[8ae8c,8h,8l,14,15]. With R ¼CH2Ph (Bz) the catalytic activity for cis-
cyclooctene epoxidation at 55 ꢀC surpassed that of the well-known
MeReO3/H2O2 system. Similar catalytic results can be obtained by
using the parent tricarbonyl complexes directly as catalyst precur-
sors. Depending on the nature of X, R, and the reaction conditions,
the MoVI species formed by in situ oxidation may include Cp0MoO2X
C3H5) with HBF4 followed by cyclopentadiene coordination gives
h4-cyclopentadiene complexes. Compounds with the CpCp0Mo
moiety are then available through oxidative, reductive, or photo-
chemical pathways [5a,5b]. The MoII allyl complexes are also useful
in organic synthesis because functional groups adjacent to the h3
-
allyl group can be transformed with excellent stereoselectivity and
regioselectivity [6]. Another potential application concerns the use
* Corresponding authors. Fax: þ351 234 370084.
and [Cp0MoO2]2( -O), the peroxo complexes Cp0MoO(O2)X and
m
E-mail addresses: atav@ua.pt (A.A. Valente), igoncalves@ua.pt (I.S. Gonçalves).
0022-328X/$ e see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2010.07.001

2312
P. Neves et al. / Journal of Organometallic Chemistry 695 (2010) 2311e2319
[Cp0MoO(O2)]2(
m
-O), and the anionic trioxo complex [Cp0MoO3]ꢁ
4.49%. Selected FTIR (CH2Cl2, cmꢁ1): 1944vs, 1858vs [ (CO)]. 1H
n
[16]. Experimental and theoretical studies have indicated that the
complexes activate TBHP for oxygen atom transfer to an olefin by
forming a sterically crowded active intermediate containing a coor-
NMR (CDCl3, 300 MHz, room temperature, d ppm): 5.15 (s, 2H,
C5H4Me), 5.09 (s, 2H, C5H4Me), 3.76 (m, 1H, allyl-H2), 2.66 (d, 2H,
allyl-H1,3), 1.87 (s, 3H, Me), 0.93 (d, 2H, allyl-H1,3). 13C NMR (CDCl3,
dinated tert-butylperoxido ligand [8h,8l,15]. Molybdenum
h
3-allyl
75 MHz, room temperature, d ppm): 237.5 (CO), 92.7 (C5H4Me), 88.2
dicarbonyl complexes of the type [Mo(
h
3-allyl)X(CO)2Ln] (X ¼ Cl, Br;
(C5H4Me), 69.2 (allyl-C2), 41.0 (allyl-C1/3), 13.5 (C5H4Me). The atom
numbering of the allyl group for the NMR assignments is shown in
Chart 1.
L ¼ MeCN or bidentate diimine ligand) have also been examined as
precursors to reactive epoxidation catalysts [9]. Oxido-bridged
dimers of the type [MoO2(NeN)]2(m-O)2 were proposed as the active
species.
2.2.2. Cp*Mo(CO)2(
Reagents, solvent, yield: LiCp* (0.69 g, 4.83 mmol), (h
3-C3H5)Mo
h
3-C3H5) (3)
As part of our ongoing research into molybdenum carbonyl
complexes as precursors to oxomolybdenum(VI) catalysts, we now
(CO)2(NCMe)2Cl (1.50 g, 4.83 mmol), toluene, 58% (0.92 g). Anal.
Calcd for C15H20MoO2 (328.25): C, 54.88; H, 6.14. Found: C, 54.72; H,
wish to report on the synthesis of the complexes Cp0Mo(CO)2(h3
-
C3H5) [Cp0 ¼ Cp,
h
5-C5H4Me (CpMe),
h
5-C5Me5 (Cp*)] (Chart 1) and
6.00%. Selected FTIR (CH2Cl2, cmꢁ1): 1933vs, 1846vs [
NMR (CDCl3, 300 MHz, room temperature, d
n
(CO)]. 1H
ppm): 2.80 (m, 1H,
allyl-H2), 2.12 (d, 2H, allyl-H1,3), 1.79 (s, 15H, Me), 0.90 (d, 2H, allyl-
their catalytic performance in the epoxidation of several olefins
with TBHP or H2O2 as oxidising agents. The X-ray crystal structures
of the CpMe and Cp* analogues are described.
H
1,3). 13C NMR (CDCl3, 75 MHz, room temperature,
d ppm): 240.3
(CO), 103.4 (C5Me5), 75.3 (allyl-C2), 43.2 (allyl-C1/3), 10.5 (Me).
2. Experimental
2.1. General considerations
2.3. X-ray crystallography
All preparations and manipulations were carried out using
standard Schlenk techniques under argon. Solvents were dried by
standard procedures, distilled under argon, and kept over 4 A
Suitable single-crystals of 2 and 3 were manually harvested
from the crystallisation vials and mounted on Hampton Research
CryoLoops using FOMBLIN Y perfluoropolyether vacuum oil (LVAC
25/6) purchased from Aldrich with the help of a Stemi 2000
stereomicroscope equipped with Carl Zeiss lenses [18]. Data were
ꢀ
molecular sieves. Microanalyses for CHN were performed at the
ITQB, Oeiras, Portugal (by Z. Tavares). 1H and 13C NMR spectra were
measured with a Bruker CXP 300 instrument. Chemical shifts are
quoted in parts per million from tetramethylsilane. Infrared spectra
were recorded on a Unican Mattson Mod 7000 FTIR spectropho-
tometer using KBr pellets and/or in a CH2Cl2 solution. The
collected on a Bruker X8 Kappa APEX II CCD area-detector diffrac-
ꢀ
tometer (Mo Ka graphite-monochromated radiation,
l
¼ 0.7107 A)
controlled by the APEX2 software package [19], and equipped with
an Oxford Cryosystems Series 700 cryostream monitored remotely
using the software interface Cryopad [20]. Images were processed
using the software package SAINTþ [21], and data were corrected
for absorption by the multi-scan semi-empirical method imple-
mented in SADABS [22]. Structures were solved using the Patterson
synthesis algorithm implemented in SHELXS-97 [23], which
allowed the immediate location of the central molybdenum centre
for each complex. All of the remaining non-hydrogen atoms were
directly located from difference Fourier maps calculated from
successive full-matrix least squares refinement cycles on F2 using
SHELXL-97 [24]. Non-hydrogen atoms were successfully refined
using anisotropic displacement parameters.
compounds (
C3H5) (1) [5b] were prepared as described in the literature.
h -
3-C3H5)Mo(CO)2(NCMe)2Cl [17] and CpMo(CO)2(h3
2.2. General procedure for the preparation of Cp0Mo(CO)2( 3-C3H5)
h
(Cp0 ¼ CpMe, Cp*)
Solid (h3-C3H5)Mo(CO)2(NCMe)2Cl and LiCp0 were weighed in
a Schlenk tube and cooled to ꢁ80 ꢀC. Precooled THF or toluene
(30 mL) was added and the mixture was slowly warmed to room
temperature. Stirring was continued for 18 h, after which time the
solution was evaporated to dryness. The residue was extracted with
hexane at 40 ꢀC for a few hours. The resultant yellow extract
solution was concentrated, and a yellow powder was separated,
which was recrystallised from hexane/Et2O at ꢁ30 ꢀC.
Hydrogen atoms attached to carbon were located at their ide-
alised positions using appropriate HFIX instructions in SHELXL: 43
for the aromatic hydrogen atoms associated with the Cp rings; 13
and 23 for the central and peripheral hydrogen atoms of the h
3-allyl
2.2.1. (CpMe)Mo(CO)2(
h
3-C3H5) (2)
ligands; 137 for the substituent methyl group of CpMe, and 127 for
the same groups belonging to Cp*. These hydrogen atoms were
included in subsequent refinement cycles in riding-motion
approximation with isotropic thermal displacements parameters
Reagents, solvent, yield: LiCpMe (0.55 g, 6.44 mmol), (
h
3-C3H5)
Mo(CO)2(NCMe)2Cl (2.00 g, 6.44 mmol), THF, 88% (1.54 g). Anal.
Calcd for C11H12MoO2 (272.15): C, 48.55; H, 4.44. Found: C, 48.56; H,
Chart 1.

P. Neves et al. / Journal of Organometallic Chemistry 695 (2010) 2311e2319
2313
(Uiso) fixed at 1.2 or 1.5 (only for the Me moieties) times Ueq of the
attached carbon atom.
To recover and reuse the ionic liquid-standing catalysts, the
reaction mixtures after the first run were cooled to room temper-
ature, n-hexane (1 mL) was added, and the mixtures stirred for ca.
5 min, followed by centrifugation to enhance the separation of the
two liquid phases. The colourless, transparent, upper phase con-
sisting of hexane/olefin/products was separated from the
IL þ catalyst phase. The IL phase was extracted twice more with
hexane, and the complete removal of reagent/products was
confirmed by GC analysis. A second run was initiated by adding
substrate and oxidant in equimolar amounts to those charged in the
first run, and monitoring the reaction carried out at 55 ꢀC for 24 h.
The last difference Fourier map synthesis showed, for 2, the
ꢀꢁ3
ꢀꢁ3
highest peak (0.263 eA ) and deepest hole (ꢁ0.272 eA ) located
ꢀ
ꢀ
at 0.50 A from H(8) and 0.32 A from H(8) (both associated with the
ꢀꢁ3
h3-allyl ligand), respectively; for 3, the highest peak (0.362 eA
)
ꢀꢁ3
ꢀ
and deepest hole (ꢁ0.295 eA ) located at 0.66 A from C(6) and
ꢀ
1.26 A from H(9C) (both associated with the Cp* ligand and located
on the crystallographic mirror plane of space group Pnma),
respectively. Information concerning crystallographic data collec-
tion and structure refinement details is summarised in Table 1.
3. Results and discussion
2.4. Catalytic olefin epoxidation
3.1. Synthesis and characterisation of the
complexes
h
3-allyl dicarbonyl
The liquid phase epoxidation of cis-cyclooctene was carried out
at 55 ꢀC under air (atmospheric pressure) in a micro reaction vessel
equipped with a magnetic stirrer and immersed in a thermostated
oil bath. The micro vessel was loaded with the complex (36 mmol),
olefin (3.6 mmol), oxidant (5.5 mmol of either 5.5 M TBHP in
The reaction of (
salt of cyclopentadiene (Cp) or methylcyclopentadiene (CpMe) in
THF gives the complexes CpMo(CO)2(
3-C3H5) (1) and (CpMe)Mo
(CO)2(
3-C3H5) (2) in 80e90% yield. When the same reaction is
carried out using the lithium salt of pentamethylcyclopentadiene
(Cp*), the yield of the desired
3-allyl dicarbonyl complex (3) is
h
3-C3H5)Mo(CO)2(NCMe)2Cl with the lithium
h
decane, 70% aq. TBHP or 30% aq. H2O2), and, optionally, 1,2-
dichloroethane (2 mL) or a room temperature ionic liquid (100 mL).
h
The TBHP/decane solution was used as received from Sigma-
eAldrich and may contain up to 4% water. In the experiments
with different olefins, 1,2-dichloroethane (2 mL) was used as
solvent. Samples were withdrawn periodically and analysed using
a Varian 3800 GC equipped with a capillary column (SPB-5,
h
much lower, but can be improved to 58% by using toluene as the
solvent instead of THF. Complexes 1e3 (Chart 1) are bright-yellow
solids, which are moderately air stable and may be stored
unchanged for long periods under nitrogen. They are very soluble in
polar solvents such as CH2Cl2 and ethanol, and also in non-polar
solvents such as n-hexane and diethyl ether. As expected, the IR
spectra of the complexes show a pair of carbonyl stretching bands
between ca. 1845 and 1950 cmꢁ1. The stretching frequencies for
dichloromethane solutions follow the order 1 (1859,1946 cmꢁ1) z 2
20 m ꢂ 0.25 mm ꢂ 0.25
mm) and a flame ionisation detector.
Undecane was used as an internal standard added after the
reaction.
Table 1
(1858, 1944 cmꢁ1) > 3 (1846, 1933 cmꢁ1). The lower
n(CO) for
complex 3 can be explained by the higher donor capacity of the Cp*
ligand, i.e., the substitution of the hydrogen atoms in Cp and CpMe
by methyl groups (Cp*) increases the electron density at the metal,
Crystal and structure refinement data for (CpMe)Mo(CO)2(
h
3-C3H5) (2) and Cp*Mo
(CO)2(
h
3-C3H5) (3).
2
3
Formula
C
11H12MoO2
C15H20MoO2
328.25
180(2)
Orthorhombic
Pnma
9.5295(7)
13.4922(9)
11.4888(7)
e
which results in greater
p back-bonding to the carbonyl groups and
Formula weight
Temperature (K)
Crystal system
Space group
272.15
150(2)
Monoclinic
P21/n
8.2649(5)
11.4169(7)
11.1609(6)
103.381(2)
1024.55(10)
4
hence a lower CeO bond order. At room temperature the solution 1H
NMR spectra reveal only one set of signals for the allyl ligands,
presumably due to rapid exoeendo interconversion on the NMR
ꢀ
time scale. The complex (h h
5-C5H4SiMe3)Mo(CO)2( 3-C3H5) exhibits
a (A)
ꢀ
b (A)
similar temperature-dependent dynamic behaviour in solution [25].
The structures of 2 and 3 were determined by X-ray crystallog-
raphy (Fig. 1 and Table 2). While complex 3 is rigorously symmetric,
having a crystallographic mirror plane bisecting the O^CeMoeC^O
ꢀ
c (A)
b
(ꢀ)
3
ꢀ
Volume (A )
1477.16(17)
4
1.476
Z
Dc (g cmꢁ3
)
1.764
1.249
angle (and consequently the
this is not the case for 2. Indeed, even though the complex (CpMe)Mo
(CO)2(
3-C3H5) could, in theory, exhibit the same kind of crystallo-
graphic symmetry as that registered for 3, the
5-CpMe ligand is
h h
5-Cp* and 3-allyl ligands, see Fig. 1b),
m
(Mo-K
a
) (mmꢁ1
)
0.880
F(000)
544
672
h
Crystal size (mm)
Crystal type
0.26 ꢂ 0.08 ꢂ 0.08
Yellow blocks
3.57e25.35
ꢁ9 ꢃ h ꢃ 9
ꢁ13 ꢃ k ꢃ 13
ꢁ13 ꢃ l ꢃ 13
29069
0.30 ꢂ 0.06 ꢂ 0.01
Yellow needles
3.55e25.34
ꢁ10 ꢃ h ꢃ 11
ꢁ16 ꢃ k ꢃ 16
ꢁ13 ꢃ l ꢃ 13
23397
h
q
range
rotated by approximately 27ꢀ along the axis containing its centre of
gravity (Cg,Cp) and the molybdenum centre (Fig.1a), ultimatelyleading
to an overall reduction in both local and crystallographic symmetries.
The isolation of this conformational isomer can be rationalised by
taking into account the supramolecular interactions mediating the
crystal packing of 2 (see Discussion below).
Index ranges
Reflections collected
Independent reflections
Data completeness
Data/parameters
1861 (Rint ¼ 0.0308)
1397 (Rint ¼ 0.0499)
99.3%
99.4%
1861/128
R1 ¼ 0.0139,
wR2 ¼ 0.0329
R1 ¼ 0.0157,
wR2 ¼ 0.0338
m ¼ 0.0128,
n ¼ 0.7356
1397/91
,
b
Final R indices [I > 2
s
(I)]a
R1 ¼ 0.0263,
wR2 ¼ 0.0549
R1 ¼ 0.0380,
wR2 ¼ 0.0593
m ¼ 0.0199,
n ¼ 2.1687
The molybdenum centres in complexes 2 and 3 are coordinated
to two carbonyl (C^O) groups, one methylcyclopentadienyl ring
(for 2) orone pentamethylcyclopentadienyl ring (for 3), and one allyl
ligand (Fig.1). Even though the structures of a considerable number
of molybdenum dicarbonyleallyl complexes containing a Cp0
aromatic ring have been reported [7,26], those which most resemble
Final R indices (all data)a
,
b
Weighting schemec
ꢀꢁ3
ꢀꢁ3
Largest diff. peak and hole
0.263 and ꢁ0.272 eA
0.362 and ꢁ0.295 eA
2 and 3 were described in the 1980s: CpMo(CO)2(h
3-C3H5) by
P
P
a
R1 ¼
jjFoj ꢁ jFcjj= jFoj.
Faller et al. [27], and (AcCp)Mo(CO)2(
h
3-C3H5) (Ac ¼ acetyl) by
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
P
P
b
c
wR2 ¼
½wðFo2 ꢁ Fc2Þ2ꢄ= ½wðFo2Þ2ꢄ.
Vanarsdale and Kochi [28]. Besides the striking similarities con-
cerning the molybdenum coordination spheres, the allyl ligands in
w ¼ 1=½s2ðFo2Þ þ ðmPÞ2 þ nPꢄ where P ¼ ðFo2 þ 2Fc2Þ=3.

2314
P. Neves et al. / Journal of Organometallic Chemistry 695 (2010) 2311e2319
Fig. 1. Schematic representation of the molecular units present in the crystal structures of (a) (CpMe)Mo(CO)2(
scheme for all non-hydrogen atoms. Thermal ellipsoids are drawn at the 30% probability level and hydrogen atoms are represented as small spheres with arbitrary radii (those
belonging to the Cp* ligand in complex 3 have been omitted). MoeC bonds to the
5-CpMe, 5-Cp* and 3-allyl ligands have been replaced by black-filled dashed bonds to the
h h
3-C3H5) (2) and (b) Cp*Mo(CO)2( 3-C3H5) (3) showing the labelling
h
h
h
corresponding centres of gravity (Cg,Cp and Cg,allyl, respectively). For selected bond lengths and angles see Table 2. Symmetry transformation used to generate equivalent atoms: (i) x,
1.5 ꢁ y, z.
all four crystal structures are coordinated via a typical exo confor-
mation, further substantiating the assumptions of Faller and co-
workers that this isomer is the most energetically favourable in the
solid state [17,27]. The average dihedral angles between the mean
rings the MoeC bond lengths were found in the ranges of 2.2982
(16)e2.4077(16) A for 2 and 2.296(4)e2.379(2) A for 3 (Table 2),
which are in agreement with the values reported for related
ꢀ
ꢀ
ꢀ
materials (data from the CSD: CpMe, 152 entries, range 2.15e2.53 A
planes containing the
h
3-allyl groups and the
h
5-Cp-sustituted rings
with median of 2.38 A; Cp*, 394 entries, range 2.03e2.53 A with
ꢀ
ꢀ
are approximately identical and calculated as ca. 34.2ꢀ for 2 and ca.
35.0ꢀ for 3.
median 2.36 A). The MoeC bond lengths with the
h
3-allyl groups
ꢀ
ꢀ
were found in the ranges of 2.2261(16)e2.3427(17) A for 2 and
2.202(5)e2.321(3) A for 3, which, once again, are well within the
The observed MoeC distances for the coordinated Cp0 and allyl
ꢀ
groups are typical of
h
5 and
h
3 coordination modes, respectively, as
expected interval (141 entries in the CSD with values between 2.13
ꢀ
ꢀ
revealed by systematic searches in the Cambridge Structural
Database (CSD, Version 5.31, November 2009 [29]) for geometrical
data on similar structural arrangements. For the substituted Cp
and 2.39 A, median of 2.22 A). As generally encountered for
complexes with a geometry resembling a three-legged piano-stool,
the MoeC distances with the CpMe group clearly indicate a slight
“slip” from the expected h h
5-coordination mode toward a 3-type,
with the MoeC(2,3) bond lengths of 2 being statistically longer
(Fig. 1 and Table 2). This structural feature registered for 2 (but not
3) can be rationalised by taking into account that, on the one hand,
the Cp* moiety has local pseudo-C5 symmetry and, consequently,
the effect of steric repulsion is evenly distributed across the five
carbon atoms of the ring and, on the other hand, the methyl
substituent of CpMe is closer to C(2) and C(3). The MoeC bond
lengths with the C^O groups were found in the range of 1.928(3)e
1.9554(17) (Table 2), and are consistent with the values reported for
related molybdenum-containing structures (3709 entries in the
Table 2
Selected bond lengths (A) and angles (ꢀ) for the molybdenum coordination
ꢀ
environments present in compounds 2 and 3.a,b,c
(CpMe)Mo(CO)2(
Mo(1)eC(2)
Mo(1)eC(3)
Mo(1)eC(4)
Mo(1)eC(5)
Mo(1)eC(6)
Mo(1)eC(7)
Mo(1)eC(8)
Mo(1)eC(9)
Mo(1)eC(10)
Mo(1)eC(11)
Mo(1)/Cg,Cp
Mo(1)/Cg,allyl
h
3-C3H5) (2)
2.4077(16)
C
C
C
g,CpeMo(1)eC(10)
g,CpeMo(1)eC(11)
g,Cp/Mo(1)/Cg,allyl
120.65(6)
122.54(6)
126.33(6)
79.68(7)
99.72(6)
96.60(6)
2.3812(16)
2.3263(16)
2.2982(16)
2.3364(16)
2.3427(17)
2.2261(16)
2.3293(17)
1.9554(17)
1.9423(18)
2.0185(1)
C(10)eMo(1)eC(11)
C(10)eMo(1)eCg,allyl
C(11)eMo(1)eCg,allyl
ꢀ
ꢀ
CSD, range 1.48e2.37 A with median of 1.98 A). Taking into account
the absence of structurally significant interactions directly
involving these coordinated carbonyl groups (see the Discussion
below about the intermolecular interactions mediating the crystal
packing of individual complexes), it is noteworthy that the geom-
etries of the MoeC^O bonds are clearly different for the two
complexes: while in 2 the :[Mo(1)eC(10,11)eO(1,2)] interaction
angles are almost linear [178.1(1)ꢀ and 179.7(1)ꢀ], in 3 the :[Mo
(1)eC(1)eO(1)] of the crystallographically independent connection
deviates from linearity [176.8(1)ꢀ]. These structural differences
2.0488(1)
Cp*Mo(CO)2(
Mo(1)eC(1)
Mo(1)eC(2)
Mo(1)eC(3)
Mo(1)eC(4)
Mo(1)eC(5)
Mo(1)eC(6)
Mo(1)/Cg,Cp
h
3-C3H5) (3)
1.928(3)
C
C
g,CpeMo(1)eC(1)
g,Cp/Mo(1)/Cg,allyl
120.13(2)
127.05(2)
82.2(2)
2.321(3)
2.202(5)
2.379(2)
2.327(3)
2.296(4)
2.008(1)
2.035(1)
C(1)eMo(1)eC(1)i
C(1)eMo(1)eCg,allyl
98.41(2)
arise from the steric hindrance created by the proximity of the h3
allyl group which is, on average, closer to the C^O groups in 3.
-
Mo(1)/Cg,allyl
By replacing the aforementioned MoeC bonds to each organic
ligand by a single connection to the corresponding centres of
gravity (Cg,Cp and Cg,allyl), the overall coordination geometries of
molybdenum can be envisaged as highly distorted tetrahedrons,
typical of the class of three-legged piano-stool complexes to which
a
Cg,Cp e centroid of the coordinated
h
5-CpMe and
h
5-Cp* aromatic rings [C(2)-C
(6) and C(4)-C(6) for 2 and 3, respectively].
b
Cg,allyl e centroid of the coordinated
h
3-allyl ligands [C(7)-C(9) and C(2)-C(3) for
2 and 3, respectively].
c
Symmetry transformation used to generate equivalent atoms: (i) x, 1.5 ꢁ y, z.

P. Neves et al. / Journal of Organometallic Chemistry 695 (2010) 2311e2319
2315
these compounds belong. Even though the bond distances are, for
each compound, all approximately identical [found in the ranges of
and <(CHCg,Cp) of ca. 149ꢀ] but cooperative CeH/
p
interactions
involving the methyl substituent of one complex and the aromatic
ring of another (Fig. 2a). These interactions lead to the formation of
a supramolecular centrosymmetric dimer which close packs in
a parallel fashion in the ab plane of the unit cell (Fig. 2b). In the
crystal structure of 3 no significant supramolecular interactions
ꢀ
ꢀ
1.9423(18)e2.049(1) A for 2 and 1.928(3)e2.035(1) A for 3], the
internal tetrahedral angles clearly reflect the different chemical
moieties coordinated to these metal centres, ranging from 79.68
(7)ꢀ to 126.33(6)ꢀ for 2, and from 82.2(2)ꢀ to 127.05(2)ꢀ for 3 (Table
2). Nevertheless, the CeMoe(C,Cg,allyl) basal angles for the two
complexes are approximately identical, ranging from 79.68(7)ꢀ to
99.72(6)ꢀ for 2, and from 82.2(2)ꢀ to 98.41(2)ꢀ for 3 (Fig. 1 and Table
2). Notably, the different internal tetrahedral angles mentioned
above are also intimately related with the steric strain imposed in
each individual complex by the coordination of the Cp0 ligands.
Since Cp* is clearly sterically more hindered than CpMe, the average
distance in 3 between the former moiety and the coordinated allyl
could be found between neighbouring Cp*Mo(CO)2(h
3-C3H5)
complexes with the packing being essentially driven by the need to
effectively fill the space. Also, in 3 individual complexes self-orga-
nise into layers but in this case their relative orientation is typical of
a herringbone tiling (Fig. 3a); layers close pack in the unit cell in
a typical ABAB/ fashion along the [001] direction (Fig. 3b).
3.2. Olefin epoxidation
group tends to be slightly longer, leading to
a
larger
C
g,Cp/Mo/Cg,allyl angle (Fig.1 and Table 2). As a result, and in order
Complexes 1, 2 and 3 were investigated as catalyst precursors
for the epoxidation of cis-cyclooctene, used as a model substrate for
unfunctionalised olefins, with TBHP as oxidant and without co-
solvent (other than the decane present in the oxidant solution),
under atmospheric air at 55 ꢀC. Whereas in the absence of a metal
complex (but with TBHP) or of TBHP (but with metal complex)
olefin conversion was negligible, in the presence of complexes 1e3
to minimise the repulsion between the moieties composing the
coordination sphere of molybdenum, the coordinated C^O groups
in
3 tend to be farther apart, which leads to a larger
O^CeMoeC^O angle.
The complexes (CpMe)Mo(CO)2(
h
3-C3H5) interact in the crystal
ꢀ
structure of 2 via two moderately strong [d(C/Cg,Cp) of 3.563(1) A
Fig. 2. (a) Schematic representation of the cooperative CeH/
p
interactions (dashed white-filled bonds) between coordinated
h
5-CpMe moieties belonging to adjacent (CpMe)Mo
3-C3H5) complexes in 2: C(1)eH(1A)/Cg,Cp with d(C/Cg,Cp) of 3.563(1) A and <(CHCg,Cp) of ca. 149ꢀ. (b) Crystal packing of compound 2 viewed along the [100] direction of
i
ꢀ
(CO)2(
the unit cell. MoeC bonds to the
Symmetry transformation used to generate equivalent atoms: (i) ꢁx, 1 ꢁ y, 2 ꢁ z.
h
h
5-CpMe and 3-allyl ligands have been replaced by black-filled dashed bonds to the corresponding centres of gravity (Cg,Cp and Cg,allyl, respectively).
h

2316
P. Neves et al. / Journal of Organometallic Chemistry 695 (2010) 2311e2319
Fig. 3. (a) Herringbone-type parallel packing of individual Cp*Mo(CO)2(
h
3-C3H5) complexes, which leads to the formation of a layer placed in the ac plane of the unit cell of 3.
5-Cp* and
(b) Crystal packing of compound 3 viewed in perspective along the [001] direction of the unit cell. Hydrogen atoms have been omitted for clarity and MoeC bonds to the
h
h
3-allyl ligands have been replaced by black-filled dashed bonds to the corresponding centres of gravity (Cg,Cp and Cg,allyl, respectively).
and TBHP, conversion reached 84e100% at 24 h (Table 3). For 1 and
2 the epoxide was the only product, while 3 gave 1,2-cyclo-
octanediol as a minor by-product (possibly indicating the forma-
tion of metal species with enhanced Lewis acidity, which may
favour epoxide ring opening). These results indicate that the metal
complex is required for the activation of the oxidant and that the
oxidant is required for the formation of active species responsible
for oxygen atom transfer to the olefin. For comparison, the epoxi-
dation of cis-cyclooctene was carried out using CpMo(CO)3Cl (Table
Table 3
cis-Cyclooctene epoxidation with TBHP (in decane) at 55 ꢀC using molybdenum
carbonyl complexes as catalyst precursors, without a co-solvent (unless otherwise
specified).
Catalyst precursor TOF a
Conversion b (%) Selectivity b (%)
Run 1 Run 2 Run 1 Run 2
99/100 84/99
(mol mol hꢁ1
)
ꢁ1
Mo
1
2
3
310
307
32
100/100 100/100
100/100 100/100
94/98
50/84
67/89
53/86
93/96
98/98
CpMo(CO)3Cl
CpMo(CO)3Mec
CpMo(CO)3Me/
499
99/100 85/100 100/100 100/100
3), under similar reaction conditions, which gave a somewhat
248
96/100 82/97
86/98 44/89
100/100 100/100
100/100 100/100
ꢁ1
higher turnover frequency (TOF ¼ 499 mol mol
h
ꢁ1) than that
121 (14)c
Mo
ꢁ1
h
ꢁ1), and the epoxide yield was
c
observed for 1 (310 mol mol
Mo
[BMIM]BF4
nearly quantitative within 20 min (Fig. 4). The TOFs for 1 and 2 are
comparable to those found for CpMo(CO)3Me [8j] (Table 3) and
a
Turnover frequency calculated at ca. 10 min.
cis-Cyclooctene conversion and epoxide selectivity at 6/24 h for runs 1 and 2.
Published in ref. [8j]; values in brackets refer to run 2.
b
c

P. Neves et al. / Journal of Organometallic Chemistry 695 (2010) 2311e2319
2317
Fig. 5. Kinetic profiles for the conversion of olefins with TBHP (in decane) at 55 ꢀC
Fig. 4. Kinetic profiles of cis-cyclooctene epoxidation with TBHP (in decane) at 55 ꢀC
using molybdenum carbonyl complexes as catalyst precursors and, unless otherwise
stated, no co-solvent: 1 (,); 2 (:); 3 (B); CpMo(CO)3Cl (þ); 1 þ 1,2-dichloroethane
(ꢂ); 1 þ [BMIM]BF4 (*); 1 þ [BMPy]BF4 (-).
using complex 1 as the catalyst precursor: cis-cyclooctene (ꢂ), (R)-(þ)-limonene (
6),
cyclododecene (,), trans-2-octene (þ) and 1-octene (B).
(CO)3Me (Cp0 ¼ Cp or Cp*) used as catalyst precursors for the same
reaction revealed a similar activity difference [8d], which suggested
that the cyclopentadienyl ligands were retained during the in situ
oxidative decarbonylation to give the MoVI catalysts, i.e., that steric
and/or electronic effects involving the Cp0 ring strongly influenced
the overall performance of the catalysts [8f]. Modifying Cp0 in the
precursor complexes 1e3 may change the nature of the oxomo-
lybdenum species formed in situ by oxidation. In an effort to get
some insight into the nature of such species, the catalytic reaction
using cis-cyclooctene, TBHP and 1 was scaled up bya factor of 10 and
allowed to run for 6 h at 55 ꢀC. After cooling to room temperature,
a small amount of manganese dioxide was added to destroy the
excess of TBHP, and the mixture was filtered. Evaporation of the
filtrate under reduced pressure gave an intractable yellow oil which
we were unable to purify and characterise. Based on the previous
work with the complexes Cp0Mo(CO)3X (see the Introduction,
some of the most active MoO2X2L complexes used as catalysts for
the same reaction under similar conditions [30].
The kinetic profiles for 1e3 are similar in that there is a rapid rise
in conversion during the first hour or so, with no induction period
being observed, followed by a more gradual increase as the reaction
proceeds (Fig. 4). This behaviour, which is also typical of the tri-
carbonyl complexes Cp0Mo(CO)3X [8], suggests that the active oxi-
dising species are formed within the first few minutes of reaction,
and, possibly, that the initial rate of olefin epoxidation is not
significantly limited by the oxidative decarbonylation of the
precursor complex. The TOF for 3 is about one order of magnitude
lower than that for 1 and 2, and this activity difference is maintained
throughout the 24 h of reaction. Previous studies with Cp0Mo
Table 4
section 1) and the well-known chemistry of Cp0Mo(CO)2( 3-C3H5),
h
Olefin epoxidation at 55 ꢀC using CpMo(CO)2(
h
3-C3H5) (1) as the catalyst precursor.
we may expect that the dicarbonyl complexes 1e3 will lose the allyl
a
Olefin
Oxidant Solvent
TOFꢁ1(mol Conv.b
Select.b
(%)
molMo hꢁ1
)
(%)
and CO groups upon reaction with TBHP. The products may include
bimetallic systems like [Cp0MoO2]2(
m-O), as found for the oxidation
cis-Cyclooctene
TBHP aq. None
H2O2 aq. None
97
80/99
11/27
99/100
100/100
100/100
100/100
of (h
5-C5Bz5)Mo(CO)3Me [8f], as well as their corresponding peroxo
<1
310
TBHP
None
species or the anionic complex [Cp0MoO3]ꢁ.
(decane)
To assess the stability of the catalytic systems derived from
complexes 1e3 and, for comparison, CpMo(CO)3Cl, two consecutive
24 h runs were carried out by recharging the micro reactor after the
first 24 h run with cis-cyclooctene and TBHP in the same amounts
as those used initially, and monitoring the reaction for a further
24 h, at 55 ꢀC. For 1 and 3 the catalytic results at 6/24 h are roughly
the same for runs 1 and 2 (Table 3). The results for 1 are comparable
to those observed for CpMo(CO)3Cl, as well as to those found for
CpMo(CO)3Me used as a catalyst precursor under similar reaction
conditions [8j]. In the case of 2 the reaction is slower in the second
run, which may be due to the formation of different metal species.
In an effort to facilitate catalyst recycling with complex 1 as the
precursor, the ionic liquids (ILs) 1-butyl-3-methylimidazolium
tetrafluoroborate ([BMIM]BF4) and 1-butyl-4-methylpyridinium
tetrafluoroborate ([BMPy]BF4) were used, which are both readily
available and inexpensive. Under the reaction conditions used
DCE
361
100/100 100/100
[BMIM]BF4 156
90/94
(64/87)
76/93
(50/70)
35/51
100/100
(100/100)
100/100
(100/95)
100/100
(57)c
c
c
[BMPy]BF4 108
(36)c
c
c
1-Octene
TBHP
(decane)
TBHP
(decane)
TBHP
(decane)
DCE
DCE
DCE
DCE
6
trans-2-Octene
Cyclododecene
81
74/89
86/92
98/100
100/100
100/100
188
289
d
(R)-(þ)-Limonene TBHP
79/70
(decane)
a
b
c
Turnover frequency calculated at ca. 10 min.
Olefin conversion and epoxide selectivity at 6/24 h.
Information relative to the second run.
d
Sum of selectivities to 1,2-epoxy-p-menth-8-ene and l,2-8,9-diepoxy-p-men-
thane. At 24 h, the molar ratio epoxide/diepoxide was 1.2.

2318
P. Neves et al. / Journal of Organometallic Chemistry 695 (2010) 2311e2319
Scheme 1. Limonene and its oxidation products.
liquideliquid biphasic mixtures were obtained comprising a col-
ourless organic phase containing cis-cyclooctene and decane, and
a denser, clear yellow IL phase containing the oxidant and the fully
dissolved metal species. Control experiments performed with IL but
without metal complex gave 36% and 27% conversion at 24 h for
[BMIM]BF4 and [BMPy]BF4, with epoxide yields of 31% and 24%,
respectively. For the catalytic systems 1 þ IL, the epoxide was
obtained as the only reaction product in ca. 94% yield at 24 h
(Table 4). The kinetic profiles are quite similar (Fig. 4); the TOF was
slightly higher for [BMIM]BF4 than for [BMPy]BF4 (156 and
which might have been expected to give a lower TOF, was appar-
ently offset by the favourable polarity of the reaction medium
(e.g., readily dissolving and/or stabilising metal species). The cata-
lytic performance of 1 was further investigated for the epoxidation
of other unfunctionalised olefins, namely 1-octene, trans-2-octene,
cyclododecene and (R)-(þ)-limonene, using DCE as solvent, at 55 ꢀC
(Table 4, Fig. 5). With the exception of limonene, the corresponding
epoxide was always the only observed product. In the reaction of
limonene, 1,2-epoxy-p-menth-8-ene and l,2-8,9-diepoxy-p-men-
thane were the main products (70% epoxide yield at 24 h), and the
corresponding 1,2-diol was a by-product (Scheme 1). The catalytic
epoxidation is regioselective, occurring preferentially at the endo-
cyclic double bond rather than at the exocyclic one. When the
reaction was carried out at 35 ꢀC, the total epoxide yield at 24 h
increased to 76%. The reaction rates for the studied olefins
increased in the following order (at 55 ꢀC): 1-octene < trans-2-
octene < cyclododecene < (R)-(þ)-limonene < cis-cyclooctene.
Hence, electron-rich olefins such as cis-cyclooctene display
a greater reactivity than electron deficient olefins such as 1-octene.
This probably reflects the electrophilic nature of the oxygen atom
transfer from the active intermediate to the olefinic double bond.
Comparing the two cyclic olefins, the lower reactivity of cyclo-
dodecene may be attributed to steric hindrance being more
significant in the formation of the transition state of the epoxida-
tion, leading to a slower epoxidation reaction.
ꢁ1
108 mol mol
h
ꢁ1, respectively, Table 4). The results for complex
Mo
1 þ [BMIM]BF4 are somewhat comparable to those reported for the
[BMIM]BF4estanding CpMo(CO)3Me (Tables 3 and 4) [8j]. The
reactions performed with the ILs are significantly slower than those
involving a single liquid phase, which may be due to mass transfer
limitations [8j].
When the catalyst þ IL systems were recycled (see Section 2.4
for details about the hexane extraction procedure used to recover
the catalyst þ IL phase), the reaction of the olefin in the second run
was somewhat slower than that in the first run, and epoxide
selectivity decreased slightly in the case of [BMPy]BF4. To check
whether active metal species were extracted during the work-up
procedure, the three n-hexane fractions resulting from the extrac-
tion of the catalyst þ [BMPy]BF4 system were combined, and the
volatiles were evaporated under reduced pressure, giving
a yellowish oily residue. cis-Cyclooctene and TBHP were added to
the residue in the same amounts as those charged for the first run,
without co-solvent, and the mixture was stirred during 6 h at 55 ꢀC.
Conversion at 6 h was 59%, which is quite significant compared to
the 99% conversion observed for complex 1 without solvent (Table
3). Hence, the n-hexane fractions contained active metal species,
explaining the lower reaction rates for the recycled catalyst þ IL
systems. This drawback has been reported for several related
molybdenum complex/IL catalytic systems, occurring within the
first couple of batch runs [8j, 31]. Somewhat more efficient
molybdenum complex/IL catalytic systems (stable during at least
two batch runs) include [MoO2Cl(HC(bim)3)]Cl/([BMIM]PF6 or
[BMPy]PF6) [32], where HC(bim)3 ¼ tris(benzimidazolyl)methane,
and [MoO2L]Na/[BMIM]NTf2 [33], where L is a Schiff base and
NTf2 ¼ anionic bis{(trifluoromethyl)sulfonyl}amide. It is desirable
to choose an immiscible IL-extraction solvent pair which favours
high partition ratios of the catalyst in the IL and of the target
product in the extraction solvent; otherwise further work-up
procedures for purification of the target products may be required.
On the other hand, distillation of the target product may be an
attractive alternative to solvent extraction but its feasibility will
depend on the thermal stability and vapour pressures of the
different compounds of the reaction mixture.
To study the activity of these systems in aqueous solution, TBHP
or H2O2 were used as oxidants in cis-cyclooctene epoxidation with
complex 1 as the catalyst precursor (Table 4). The reaction mixtures
consisted of two immiscible liquid phases, i.e., an organic phase
containing cis-cyclooctene and cyclooctene oxide, and an aqueous
phase containing the oxidant and molybdenum species. With 70%
ꢁ1
aq. TBHP the initial reaction rate (97 mol mol
h
ꢁ1) was lower than
Mo
ꢁ1
that measured for the decane solution of TBHP (310 mol mol
h
ꢁ1),
Mo
but after 24 h practically the same conversion was reached, giving
99% epoxide yield. Under similar conditꢁio1ns, the catalyst precursor
CpMo(CO)3Cl gave a TOF of 108 mol molMo h
ꢁ1 and an epoxide yield
of 80% at 24 h [16]. The reaction of cis-cyclooctene with 1 as the
precursor and 30% aq. H2O2 as the oxidant was very sluggish (27%
conversion at 24 h), although the epoxide was the only product.
Nevertheless, these results are an improvement over those obtained
with the complexes CpMo(CO)3Cl, Cp*MoO2Cl and [Cp*MoO2]2(m-
O), which are unable to activate aqueous H2O2 for cis-cyclooctene
epoxidation [8d,8f,16].
4. Conclusions
The present work has demonstrated that complexes of the type
Cp0Mo(CO)2( 3-C3H5) are potentially interesting as a new class of
h
Compared with the results obtained using 1 as the catalyst
precursor and no additional solvent (other than the decane present
molybdenum carbonyl precursors to olefin epoxidation catalysts.
The stability of these catalytic systems seems fair, based on the
catalyst reuse tests. The poorer performance (slower reaction) of
the Cp* analogue indicates some dependence of catalytic perfor-
mance on the ring substituents. In the future, in situ spectroscopic
measurements of the reaction of the precursors with the oxidant
in the oxidant solution), the catalytic performance improved in
ꢁ1
terms of initial reaction rate (TOF ¼ 361 mol mol
h
ꢁ1) by addition
Mo
of 1,2-dichloroethane (DCE) as solvent, with no decrease in epoxide
selectivity (Table 4, Fig. 4). The dilution of the reaction mixture,

P. Neves et al. / Journal of Organometallic Chemistry 695 (2010) 2311e2319
1826e1833;
2319
may help to elucidate the nature of the molybdenum species
formed.
(j) M. Abrantes, P. Neves, M.M. Antunes, S. Gago, F.A.A. Paz, A.E. Rodrigues,
M. Pillinger, I.S. Gonçalves, C.M. Silva, A.A. Valente, J. Mol. Catal. A: Chem. 320
(2010) 19e26;
(k) A. Capapé, A. Raith, E. Herdtweck, M. Cokoja, F.E. Kühn, Adv. Synth. Catal.
351 (2010) 547e556;
Acknowledgements
(l) P.J. Costa, M.J. Calhorda, F.E. Kühn, Organometallics 29 (2010) 303e311.
[9] J.C. Alonso, P. Neves, M.J.P. da Silva, S. Quintal, P.D. Vaz, C. Silva, A.A. Valente,
P. Ferreira, M.J. Calhorda, V. Félix, M.G.B. Drew, Organometallics 26 (2007)
5548e5556.
[10] V.V.K.M. Kandepi, A.P. da Costa, E. Peris, B. Royo, Organometallics 28 (2009)
4544e4549.
[11] T.R. Amarante, P. Neves, A.C. Coelho, S. Gago, A.A. Valente, F.A.A. Paz,
M. Pillinger, I.S. Gonçalves, Organometallics 29 (2010) 883e892.
[12] C.A. Gamelas, T. Lourenço, A.P. da Costa, A.L. Simplício, B. Royo, C.C. Romão,
Tetrahedron Lett. 49 (2008) 4708e4712.
We are grateful to the Fundação para a Ciência e a Tecnologia
(FCT, Portugal), POCI 2010, OE and FEDER for funding through the
project with ref. no. PTDC/QUI/71198/2006. CICECO is acknowl-
edged for financial support through the project entitled “Metal
carbonyl intercalated anion exchangers as drug delivery systems”.
The FCT is thanked for grants to PN, CCLP and SG, and for financial
support toward the purchase of the single-crystal diffractometer.
[13] A.V. Biradar, M.K. Dongare, S.B. Umbarkar, Tetrahedron Lett. 50 (2009)
2885e2888.
Appendix A. Supplementary material
[14] M.K. Trost, R.G. Bergman, Organometallics 10 (1991) 1172e1178.
[15] A. Comas-Vives, A. Lledós, R. Poli, Chem. Eur. J. 16 (2010) 2147e2158.
[16] S.S. Balula, A.C. Coelho, S.S. Braga, A. Hazell, A.A. Valente, M. Pillinger,
J.D. Seixas, C.C. Romão, I.S. Gonçalves, Organometallics 26 (2007) 6857e6863.
[17] J.W. Faller, C.-C. Chen, M.J. Mattina, A. Jakubowski, J. Organomet. Chem. 52
(1973) 361e386.
[18] T. Kottke, D. Stalke, J. Appl. Crystallogr. 26 (1993) 615.
[19] APEX2, Data Collection Software (Version 2.1-RC13). Bruker AXS, Delft, The
Netherlands, 2006.
CCDC-777390 (2) and CCDC-777391 (3) contain the supple-
mentary crystallographic data (excluding structure factors) for this
paper. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
datarequest/cif.
[20] Cryopad, Remote Monitoring and Control (Version 1.451). Oxford Cry-
osystems, Oxford, UK, 2006.
References
[21] SAINTþ, Data Integration Engine (Version 7.23a). Bruker AXS, Madison, WI,
[1] (a) R.B. King, Inorg. Chem. 5 (1966) 2242e2243;
USA, 2005.
(b) A. Davison, W.C. Rode, Inorg. Chem. 6 (1967) 2124e2125;
(c) J.W. Faller, M.J. Incorvia, Inorg. Chem. 7 (1968) 840e842;
(d) M. Cousins, M.L.H. Green, J. Chem. Soc. (1963) 889e894.
[2] D.W. Norman, M.J. Ferguson, J.M. Stryker, Organometallics 23 (2004) 2015e2019.
[3] (a) C. Limberg, A.J. Downs, T.M. Greene, T. Wistuba, Eur. J. Inorg. Chem. (2001)
2613e2618;
[22] G.M. Sheldrick, SADABS Version 2.01, Bruker/Siemens Area Detector Absorp-
tion Correction Program. Bruker AXS, Madison, Wisconsin, USA, 1998.
[23] G.M. Sheldrick, SHELXS-97, Program for Crystal Structure Solution. University
of Göttingen, Göttingen, Germany, 1997.
[24] G.M. Sheldrick, SHELXL-97, Program for Crystal Structure Refinement.
University of Göttingen, Göttingen, Germany, 1997.
(b) T.E. Bitterwolf, J.T. Bays, B. Scallorn, C.A. Weiss, M.W. George, I.G. Virrels,
J.C. Linehan, C.R. Yonker, Eur. J. Inorg. Chem. (2001) 2619e2624.
[4] A. Ariafard, S. Bi, Z. Lin, Organometallics 24 (2005) 2241e2244.
[5] (a) J.R. Ascenso, C.G. de Azevedo, I.S. Gonçalves, E. Herdtweck, D.S. Moreno,
C.C. Romão, J. Zühlke, Organometallics 13 (1994) 429e431;
(b) J.R. Ascenso, C.G. de Azevedo, I.S. Gonçalves, E. Herdtweck, D.S. Moreno,
M. Pessanha, C.C. Romão, Organometallics 14 (1995) 3901e3919;
(c) I.S. Gonçalves, C.C. Romão, J. Organomet. Chem. 486 (1995) 155e161;
(d) I.S. Gonçalves, P. Ribeiro-Claro, C.C. Romão, B. Royo, Z.M. Tavares,
J. Organomet. Chem. 648 (2002) 270e279;
[25] S.S. Braga, I.S. Gonçalves, A.D. Lopes, M. Pillinger, J. Rocha, C.C. Romão,
J.J.C. Teixeira-Dias, J. Chem. Soc. Dalton Trans. (2000) 2964e2968.
[26] (a) C.J. Beddows, M.R. Box, C. Butters, N. Carr, M. Green, M. Kursawe,
M.F. Mahon, J. Organomet. Chem. 550 (1998) 267e282;
(b) M. Green, T.D. McGrath, R.L. Thomas, A.P. Walker, J. Organomet. Chem. 532
(1997) 61e70;
(c) L.C.F. Chao, H.K. Gupta, D.W. Hughes, J.F. Britten, S.S. Rigby, A.D. Bain,
M.J. McGlinchey, Organometallics 14 (1995) 1139e1151;
(d) S. Wan, M.J. Begley, P. Mountford, J. Organomet. Chem. 489 (1995)
C28eC31;
(e) M.-F. Liao, G.-H. Lee, S.-M. Peng, R.-S. Liu, Organometallics 13 (1994)
4973e4977;
(f) J.W. Faller, Y. Ma, Organometallics 12 (1993) 1927e1930;
(g) W.-J. Vong, S.-M. Peng, S.-H. Lin, W.-J. Lin, R.-S. Liu, J. Am. Chem. Soc. 113
(1991) 573e582;
(h) N.W. Murrall, A.J. Welch, Acta Crystallogr. Sect. C: Cryst. Struct. Commun.
40 (1984) 401e403.
ꢁ
(e) J. Honzícek, A. Mukhopadhyay, T. Santos-Silva, M.J. Romão, C.C. Romão,
Organometallics 28 (2009) 2871e2879;
ꢁ
(f) J. Honzícek, A. Mukhopadhyay, C. Bonifacio, C.C. Romão, J. Organomet.
Chem. 695 (2010) 680e686.
[6] (a) C.-L. Li, R.-S. Liu, Chem. Rev. 100 (2000) 3127e3161 (and references cited
therein);
(b) A. Alcudia, R.G. Arrayás, L.S. Liebeskind, J. Org. Chem. 67 (2002) 5773e5778
(and references cited therein).
[7] D.R. van Staveren, T. Weyhermüller, N. Metzler-Nolte, Organometallics 19
(2000) 3730e3735.
[27] J.W. Faller, D.F. Chodosh, D. Katahira, J. Organomet. Chem. 187 (1980)
227e231.
[28] W.E. Vanarsdale, J.K. Kochi, J. Organomet. Chem. 317 (1986) 215e232.
[8] (a) M. Abrantes, A.M. Santos, J. Mink, F.E. Kühn, C.C. Romão, Organometallics
22 (2003) 2112e2118;
(a) F.H. Allen, Acta Crystallogr. Sect. B. Struct. Sci. 58 (2002) 380e388;
(b) F.H. Allen, W.D.S. Motherwell, Acta Crystallogr. Sect. B. Struct. Sci. 58
(2002) 407e422.
[29]
(b) C. Freund, M. Abrantes, F.E. Kühn, J. Organomet. Chem. 691 (2006)
3718e3729 (and references cited therein);
(c) F.E. Kühn, A.M. Santos, M. Abrantes, Chem. Rev. 106 (2006) 2455e2475
(references cited therein);
(d) J. Zhao, A.M. Santos, E. Herdtweck, F.E. Kühn, J. Mol. Catal. A: Chem. 222
(2004) 265e271;
[30] (a) S.M. Bruno, J.A. Fernandes, L.S. Martins, I.S. Gonçalves, M. Pillinger,
P. Ribeiro-Claro, J. Rocha, A.A. Valente, Catal. Today 114 (2006) 263e271;
(b) S.M. Bruno, C.C.L. Pereira, M.S. Balula, M. Nolasco, A.A. Valente, A. Hazell,
M. Pillinger, P. Ribeiro-Claro, I.S. Gonçalves, J. Mol. Catal. A: Chem. 261 (2007)
79e87.
(e) A.A. Valente, J.D. Seixas, I.S. Gonçalves, M. Abrantes, M. Pillinger,
C.C. Romão, Catal. Lett. 101 (2005) 127e130;
(f) A.M. Martins, C.C. Romão, M. Abrantes, M.C. Azevedo, J. Cui, A.R. Dias,
M.T. Duarte, M.A. Lemos, T. Lourenço, R. Poli, Organometallics 24 (2005)
2582e2589;
(g) A. Capapé, A. Raith, F.E. Kühn, Adv. Synth. Catal. 351 (2009) 66e70;
(h) A.M. Al-Ajlouni, D. Veljanovski, A. Capapé, J. Zhao, E. Herdtweck,
M.J. Calhorda, F.E. Kühn, Organometallics 28 (2009) 639e645;
(i) M. Abrantes, F.A.A. Paz, A.A. Valente, C.C.L. Pereira, S. Gago, A.E. Rodrigues,
J. Klinowski, M. Pillinger, I.S. Gonçalves, J. Organomet. Chem. 694 (2009)
[31] (a) A.A. Valente, Z. Petrovski, L.C. Branco, C.A.M. Afonso, M. Pillinger,
A.D. Lopes, C.C. Romão, C.D. Nunes, I.S. Gonçalves, J. Mol. Catal. A: Chem. 218
(2004) 5e11;
(b) F.E. Kühn, J. Zhao, M. Abrantes, W. Sun, C.A.M. Afonso, L.C. Branco,
I.S. Gonçalves, M. Pillinger, C.C. Romão, Tetrahedron Lett. 46 (2005) 47e52.
[32] S. Gago, S.S. Balula, S. Figueiredo, A.D. Lopes, A.A. Valente, M. Pillinger,
I.S. Gonçalves, Appl. Catal. A. Gen. 372 (2010) 67e72.
[33] C. Bibal, J.-C. Daran, S. Deroover, R. Poli, Polyhedron 29 (2010) 639e647.