Dichlorodioxomolybdenum(VI) complexes bearing oxygen-donor ligands as olefin epoxidation catalysts

Article abstract of DOI:10.1039/c5dt02165k

Treatment of the solvent adduct [MoO₂Cl₂(THF)₂] with either 2 equivalents of N,N-dimethylbenzamide (DMB) or 1 equivalent of N,N′-diethyloxamide (DEO) gave the dioxomolybdenum(vi) complexes [MoO₂Cl₂(DMB)₂] (1) and [MoO₂Cl₂(DEO)] (2). The molecular structures of 1 and 2 were determined by single-crystal X-ray diffraction. Both complexes present a distorted octahedral geometry and adopt the cis-oxo, trans-Cl, cis-L configuration typical of complexes of the type [MoO₂X₂(L)_n], with either the monodentate DMB or bidentate DEO oxygen-donor ligands occupying the equatorial positions trans to the oxo groups. The complexes were applied as homogeneous catalysts for the epoxidation of olefins, namely cis-cyclooctene (Cy), 1-octene, trans-2-octene, α-pinene and (R)-(+)-limonene, using tert-butylhydroperoxide (TBHP) as oxidant. In the epoxidation of Cy at 55°C, the desired epoxide was the only product and turnover frequencies in the range of ca. 3150-3200 mol mol_Mo^-1 h^-1 could be reached. The catalytic production of cyclooctene oxide was investigated in detail, varying either the reaction temperature or the cosolvent. Complexes 1 and 2 were also applied in liquid-liquid biphasic catalytic epoxidation reactions by using an ionic liquid of the type [C₄mim][X] (C₄mim = 1-n-butyl-3-methylimidazolium; X = NTf₂, BF₄ or PF₆] as a solvent to immobilise the metal catalysts. Recycling for multiple catalytic runs was achieved without loss of activity.

Full text of DOI:10.1039/c5dt02165k

Dalton
Transactions
View Article Online
View Journal
PAPER
Dichlorodioxomolybdenum(VI) complexes bearing
oxygen-donor ligands as olefin epoxidation
catalysts†
Cite this: DOI: 10.1039/c5dt02165k
a
b
a
a,c
Tânia S. M. Oliveira, Ana C. Gomes, André D. Lopes,* João P. Lourenço,
b
b
b
Filipe A. Almeida Paz, Martyn Pillinger and Isabel S. Gonçalves*
Treatment of the solvent adduct [MoO
2
Cl
equivalent of N,N’-diethyloxamide (DEO) gave the dioxomolybdenum(VI) complexes
(DMB) ] (1) and [MoO Cl (DEO)] (2). The molecular structures of 1 and 2 were determined by
2 2
(THF) ] with either 2 equivalents of N,N-dimethylbenzamide
(
DMB) or
MoO Cl
1
[
2
2
2
2
2
single-crystal X-ray diffraction. Both complexes present a distorted octahedral geometry and adopt the
cis-oxo, trans-Cl, cis-L configuration typical of complexes of the type [MoO (L) ], with either the
X
2 2
n
monodentate DMB or bidentate DEO oxygen-donor ligands occupying the equatorial positions trans to
the oxo groups. The complexes were applied as homogeneous catalysts for the epoxidation of olefins,
namely cis-cyclooctene (Cy), 1-octene, trans-2-octene, α-pinene and (R)-(+)-limonene, using tert-butyl-
hydroperoxide (TBHP) as oxidant. In the epoxidation of Cy at 55 °C, the desired epoxide was the only
product and turnover frequencies in the range of ca. 3150–3200 mol molMo h−1 could be reached. The
catalytic production of cyclooctene oxide was investigated in detail, varying either the reaction tempera-
ture or the cosolvent. Complexes 1 and 2 were also applied in liquid–liquid biphasic catalytic epoxidation
−1
Received 8th June 2015,
Accepted 29th June 2015
reactions by using an ionic liquid of the type [C
X = NTf , BF or PF ] as a solvent to immobilise the metal catalysts. Recycling for multiple catalytic runs
was achieved without loss of activity.
4 4
mim][X] (C mim = 1-n-butyl-3-methylimidazolium;
DOI: 10.1039/c5dt02165k
2
4
6
www.rsc.org/dalton
excellent catalytic behaviour in olefin epoxidation using tert-
Introduction
1
–3
butylhydroperoxide (TBHP) as oxidant. They have also been
4
Olefin epoxidation is an important reaction in organic syn- studied as oxo-transfer agents and as models for molybdo-
5
thesis due to the role of epoxides as intermediates in the man- enzymes. The ligand(s) L may comprise one bidentate ligand
3
,6
7
8
ufacture of various industrial products, for example in the with nitrogen, oxygen or sulfur donor atoms, or two mono-
1
9
10
11
12
12a,13
petroleum and plastics industries. Since the beginning of dentate ligands (e.g. DMF, THF, RCN,
studies on metal-based epoxidation catalysts, numerous types OPR3
R
2
SO,
2
H O,
7
a,14
). Complexes formed with neutral bidentate ligands
of molybdenum complexes have been synthesised and exam- containing oxygen donor atoms have not attracted much atten-
2
ined as catalysts or catalyst precursors. In particular, dioxo- tion, possibly due to the instability of the complexes. However,
molybdenum(VI) complexes of the type [MoO X (L) ] with a limited number of studies hint at a promising catalytic be-
2
2
n
different anionic X (Cl, Br, F, alkyl, OR, OSiR , SR) and neutral haviour for complexes containing O-ligands. For example,
3
L ligands have been widely investigated because they exhibit complexes of the type [MoO
2 2 2
Cl (L) ] bearing monodentate N,
N-dialkylamide ligands have been shown to give high turnover
frequencies in the range of 561–577 mol molMo−
1
h
−1
for the
a
Faculty of Science and Technology, CIQA, University of the Algarve, Campus de
epoxidation of cis-cyclooctene with TBHP, giving the epoxide
Gambelas, 8005-136 Faro, Portugal. E-mail: adlopes@ualg.pt
as the only product in at least 98% yield after 6 h reaction at
b
Department of Chemistry, CICECO – Aveiro Institute of Materials, University of
9
e
5
5 °C.
Herein, we describe two new complexes of the type
[MoO Cl (L) ] bearing O-donor ligands, one containing the
Aveiro, Campus Universitário de Santiago, 3810-193 Aveiro, Portugal.
E-mail: igoncalves@ua.pt
c
CQE – Centro de Química Estrutural, Instituto Superior Técnico, Av. Rovisco pais,
2
2
n
1049-001 Lisboa, Portugal
monodentate ligand N,N-dimethylbenzamide (DMB), and
another containing the bidentate ligand N,N′-diethyloxamide
†
Electronic supplementary information (ESI) available: Olefin conversion
kinetic profiles for complexes 2 and 3 (Fig. S1). CCDC 1057618 (1) and 1057619
2a). For ESI and crystallographic data in CIF or other electronic format see DOI:
0.1039/c5dt02165k
(
DEO). To the best of our knowledge, the latter compound is
(
1
the first example of a molybdenum(VI) complex bearing a
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
neutral α-dicarbonyl ligand. The crystal structures of both ping signals, 10H), 3.01 (s, 6H), 2.88 (s, 6H). Single crystals of
complexes have been determined and a comprehensive study 1 suitable for X-ray diffraction were obtained by layering a solu-
has been performed to assess their performance as catalysts tion of the complex in acetonitrile with a 1 : 1 mixture of
for the epoxidation of olefins.
hexane and diethyl ether, followed by storage of the mixture in
the fridge for 24 h.
[
MoO Cl (DEO)] (2)
2 2
Experimental section
Materials and methods
After dissolving MoO Cl (1.07 g, 5.38 mmol) in THF (15 mL),
2
2
N,N′-diethyloxamide (0.775 g, 5.38 mmol) was added and the
mixture was stirred at room temperature for 1 h. The resultant
pale yellow precipitate was isolated by filtration, washed with
diethyl ether (2 × 15 mL), and vacuum-dried. Yield: 0.62 g
(
8
3
(
1
1
Transmission FT-IR spectra were measured on a Mattson 7000
spectrometer with 128 scans and a resolution of 4.0 cm−
1
.
FT-Raman spectra were recorded on an RFS-100 Bruker
FT-Spectrometer equipped with a Nd:YAG laser with an exci-
34%). Anal. Calcd for C
6
H
12Cl
2
MoN
.17. Found: C, 20.81; H, 3.65; N, 8.24%. FT-IR (KBr, cm ): ν =
330 (vs, ν(N–H)), 3178 (m), 3107 (w), 3057 (w), 2983 (m), 2939
2 4
O : C, 21.01; H, 3.53; N,
1
−
1
tation wavelength of 1064 nm. H NMR spectra were measured
with a Bruker CXP 300 instrument. Chemical shifts are quoted
in parts per million and referenced to tetramethylsilane.
All preparations and manipulations were carried out under
w), 2881 (w), 1643 (vs), 1523 (s), 1462 (s), 1446 (s), 1388 (m),
379 (m), 1350 (m), 1288 (m), 1267 (m), 1232 (m), 1151 (s),
097 (m), 1041 (m), 955 (vs, sym(MovO)), 912 (vs,
2 2
nitrogen using standard Schlenk techniques. MoO Cl (Sigma-
ν
Aldrich), anhydrous tetrahydrofuran (THF, 99.7%, Panreac), N,
N-dimethylbenzamide (DMB, ≥98%, Alfa Aesar), N,N’-diethyl-
oxamide (DEO, 98%, Alfa Aesar), acetonitrile (99.9%, Lab-
Scan), n-hexane (≥98.5%, Carlo Erba), diethyl ether (99.8%,
Sigma-Aldrich), cis-cyclooctene (95%, Sigma-Aldrich), 1-octene
ν
asym(MovO)), 741 (m), 673 (m), 565 (s), 507 (s), 391 (s), 337
vs, ν(Mo–Cl)). FT-Raman (cm− ): ν = 3337 (w), 2981 (w), 2940
1
(
(
s), 2878 (m), 1678 (w), 1636 (w), 1599 (m), 1554 (w), 1522 (w),
1
447 (m), 1378 (s), 1332 (m), 1290 (w), 1267 (m), 1145 (w), 1058
(
(
(
(
w), 1040 (w), 959 (vs), 913 (s), 803 (w), 505 (m), 390 (s), 316
vs), 265 (s), 254 (s), 239 (vs), 192 (m), 140 (m). H NMR
300 MHz, 298 K, CD
br, 6H). Single crystals (2a) suitable for X-ray diffraction were
1
(98%, Sigma-Aldrich), trans-2-octene (97%, Sigma-Aldrich),
α-pinene (98%, Sigma-Aldrich), (R)-(+)-limonene (97%, Sigma-
Aldrich), n-octane (98%, Merck-Schuchardt), undecane (≥99%,
Sigma-Aldrich), tert-butylhydroperoxide (5–6 M in decane,
Sigma-Aldrich), benzotrifluoride (99%, Alfa Aesar), anhydrous
3
CN): δ = 8.55 (br, 2H), 3.54 (br, 4H), 1.26
obtained by slow diffusion of diethyl ether into a solution of
complex 2 in acetonitrile.
1
,2-dichloroethane (99.8%, Sigma-Aldrich), ethanol (99.9%,
X-ray crystallography
Scharlau), and nitromethane (>95%, Sigma-Aldrich) were pur-
chased from commercial sources and used as received. The
The single crystals obtained upon recrystallisation of 1 and 2
ionic liquids [C
mim][NTf ] were acquired from Solchemar and used as
received. [MoO Cl (DMF) ] (3) was synthesised according to
4 4 4 4 6
mim][Cl], [C mim][BF ], [C mim][PF ] and
(
see above) were collected and immersed in highly viscous
[C
4
2
FOMBLIN Y perfluoropolyether vacuum oil (LVAC 140/13,
Sigma-Aldrich) in order to prevent any possible degradation
caused by the evaporation of the solvent. Crystals were
mounted on a Hampton Research CryoLoop. Data were col-
lected at 180(2) K on a Bruker X8 Kappa APEX II CCD area-
2
2
2
9
f,13a
15
the published procedure.
[
MoO Cl (DMB) ] (1)
2
2
2
After dissolving MoO Cl (1.10 g, 5.53 mmol) in THF (15 mL), detector diffractometer (Mo-Kα graphite-monochromated radi-
2
2
N,N-dimethylbenzamide (1.65 g, 11.06 mmol) was added and ation, λ = 0.71073 Å) controlled by the APEX2 software
1
6
the mixture was stirred at room temperature for 1 h. The resul- package and equipped with an Oxford Cryosystems Series
1
7
tant pale yellow precipitate was isolated by filtration, washed 700 cryostream monitored remotely using Cryopad. Images
18
with diethyl ether (2 × 15 mL), and vacuum-dried. Yield: 2.51 g were processed using SAINT+, and data were corrected
91%). Anal. Calcd for C18 MoN : C, 43.48; H, 4.46; N, for absorption by the multiscan semi-empirical method
.63. Found: C, 43.29; H, 4.54; N, 5.68%. FT-IR (KBr, cm ): ν = implemented in SADABS. Structures were solved using the
(
5
3
H
22Cl
2
2 4
O
−
1
19
2
0
109 (w), 3057 (w), 3035 (w), 2966 (w), 2935 (w), 1587 (vs), 1566 algorithm implemented in SHELXT-2014, which allowed the
(vs), 1510 (s), 1475 (s), 1444 (s), 1404 (s), 1286 (w), 1265 (s), immediate location of almost all of the heaviest atoms com-
1
1
9
6
217 (w), 1180 (w), 1161 (w), 1147 (w), 1097 (m), 1078 (m), posing the molecular unit of the two compounds. The remain-
057 (m), 1024 (m), 1003 (w), 991 (w), 947 (vs, νsym(MovO)), ing missing non-hydrogen atoms were located from difference
10 (vs, νasym(MovO)), 852 (m), 783 (s), 741 (s), 719 (s), 700 (s), Fourier maps calculated from successive full-matrix least-
2
52 (s), 559 (m), 447 (m), 395 (s), 332 (vs, ν(Mo–Cl)). FT-Raman squares refinement cycles on F using the latest SHELXL from
−
1
21
(cm ): ν = 3067 (s), 2967 (w), 2939 (m), 2808 (w), 1601 (s), the 2014 release. All structural refinements were performed
2
2
1
1
573 (w), 1514 (m), 1447 (m), 1414 (m), 1263 (w), 1217 (w), using the graphical interface ShelXle and all non-hydrogen
181 (w), 1162 (w), 1099 (w), 1025 (w), 1001 (s), 949 (vs), 910 atoms were refined assuming anisotropic displacement
(
(
1
m), 782 (w), 741 (m), 725 (m), 651 (w), 615 (w), 561 (w), 447 parameters.
w), 397 (w), 347 (w), 322 (w), 302 (w), 259 (w), 207 (s), 144 (s), The crystalline product obtained upon recrystallisation of 2
07 (s). H NMR (300 MHz, 298 K, CD CN): δ = 7.36 (overlap- is referred to as 2a. Besides the presence of two identical
1
3
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
complex molecular units in the asymmetric unit (i.e., Z′ = 2), Table 1 Crystal and structure refinement data
the crystal structure determination for 2a revealed the presence
of an acetonitrile solvent molecule of crystallisation and one
1
2a
disordered, uncoordinated DEO molecule. One branch of this
Formula
18
C H22Cl
MoN O
2 2 4
20 4 2 7 10
C H39Cl Mo N O
uncoordinated molecule was found to be disordered over two Formula weight
497.22
180(2)
Orthorhombic
Fdd2
16.9586(11)
27.5327(19)
8.5225(5)
90
90
90
3979.3(4)
8
1.660
871.26
180(2)
Temperature/K
distinct crystallographic positions which were modelled into
Crystal system
Triclinic
the final structure with rates of occupancy of 0.582(10) and
ˉ
Space group
P1
9.6274(7)
0
.418(10), respectively.
Hydrogen atoms bound to carbon were placed at their ideal-
a/Å
b/Å
c/Å
α/°
12.3705(9)
16.4500(11)
96.854(3)
105.901(3)
107.175(4)
1756.9(2)
2
1.647
1.073
880
0.16 × 0.14 × 0.11
Yellow plates
3.52 to 29.13
−13 ≤ h ≤ 13
−16 ≤ k ≤ 16
−22 ≤ l ≤ 22
43 689
ised positions using appropriate HFIX instructions in SHELXL:
4
3 (aromatic carbon atoms), 23 (alkylic carbon atoms) or 137 β/°
γ/°
Volume/Å
(
for the terminal methyl groups). These hydrogen atoms were
3
included in subsequent refinement cycles with isotropic
thermal displacement parameters (Uiso) fixed at 1.2 (for the
two former families of hydrogen atoms) or 1.5 × Ueq (solely for
those associated with the methyl group) of the parent carbon
atoms.
Hydrogen atoms associated with the amino groups of
DEO molecules were directly located from difference Fourier
maps. All these hydrogen atoms were included in the final
structural model with the N–H distances restrained to 0.95(1)
Å. The isotropic thermal displacement parameters (Uiso) of
Z
D
/g cm−
3
c
1
μ (Mo-K
F (000)
Crystal size/mm
Crystal type
θ range
α
)/mm−
0.954
2016
0.20 × 0.10 × 0.08
Colourless blocks
3.81 to 29.11
Index ranges
−23 ≤ h ≤ 23
−
37 ≤ k ≤ 23
−11 ≤ l ≤ 11
Reflections collected
Independent reflections
8189
2605 (Rint
.0513)
99.6%
=
9428 (Rint = 0.0358)
0
these hydrogen atoms were fixed at 1.5 × Ueq of the parent Data completeness to θ =
99.7%
2
5.24°
nitrogen atoms.
The last difference Fourier map synthesis showed: for 1, the
highest peak (0.378 e Å ) and deepest hole (−0.528 e Å
Data/restraints/parameters 2605/1/125
Final R indices [I > 2σ(I)]
9428/19/442
a,b
R
1
= 0.0331
wR = 0.0705
= 0.0378
wR = 0.0736
R
1
= 0.0328
wR = 0.0702
= 0.0427
wR = 0.0754
−
3
−3
)
2
2
a,b
Final R indices (all data)
R
1
R
1
located at 1.53 and 0.79 Å from H5 and Mo1, respectively; for
2
2
−
3
c
2
Å
a, the highest peak (0.870 e Å ) and deepest hole (−0.739 e
) located at 0.49 and 0.79 Å from H13A and Mo2, respect-
Weighting scheme
m = 0.0227
n = 2.3684
m = 0.0270
n = 1.3078
0.870, −0.739 e Å
−
3
−
3
Largest diff. peak and hole 0.378, −0.528 e
ively. Structural drawings were created using Crystal Impact
−3
Å
2
3
Diamond. Information concerning crystallographic data col-
lection and structure refinement details is summarised in
Table 1.
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
P
P
a
c
b
2
2
2
2
2
R
1
= ∑||F
o
| − |F
c
|/∑|F
o
|. wR
2
=
½wðF ꢀ F Þ ꢁ= ½wðF Þ ꢁ.
o
c
o
2
2
2
2
o
2
w = 1/[σ (F
o
) + (mP) + nP] where P = (F
+ 2F
c
)/3.
Catalytic tests
The liquid-phase catalytic epoxidation of olefins was carried
out with magnetic stirring (1100 rpm), under air, in closed as those used in the tests without cosolvent, but with the
borosilicate micro-reactors (5 mL) equipped with a valve to addition of 0.5 mL of IL. The ILs studied were [C mim]Cl,
allow sampling. The reaction mixtures had a catalyst : [C mim]BF , [C mim]PF and [C mim]NTf . For recycling,
4
4
4
4
6
4
2
substrate : TBHP ratio of 1 : 100 : 200. In typical runs, the fol- after the catalytic reaction the upper phase was removed from
lowing quantities were used: For 1, 0.025 mmol of molyb- the reaction vessel with a Pasteur pipette. The IL phase was
denum complex, 2.5 mmol of olefin and 5.0 mmol of oxidant; washed with n-hexane (with stirring). The upper phase was
for 2 and 3, 0.035 mmol of molybdenum complex, 3.5 mmol of again removed with a Pasteur pipette and this operation was
olefin and 7.0 mmol of oxidant.
repeated 4 times. The organic phase was analysed by GC to
The epoxidation of cyclooctene was performed without check whether any TBHP, n-octane or cyclooctene oxide were
adding a cosolvent or by using 2 mL of benzotrifluoride (BTF), still present. Additionally, oil pump vacuum for 30 min
n-hexane, 1,2-dichloroethane (DCE), ethanol, acetonitrile or allowed the removal of tBuOH from the remaining IL phase
nitromethane. The mixture of olefin, TBHP and cosolvent before the next run. In the next run, the reaction time starts at
(
when used) was pre-heated to the reaction temperature in a the instant TBHP was added.
temperature-controlled oil bath for 10 min, before adding the
The catalytic reactions were monitored by gas chromato-
catalyst. The instant the catalyst was added was considered as graphy (GC Bruker SCION 456-GC with a 30 m Macherey-Nagel
time zero. Additional substrates (1-octene, trans-2-octene, OPTIMA delta-3 capillary column and a flame ionisation detec-
α-pinene and (R)-(+)-limonene) were also studied, without the tor (FID)) using n-octane or undecane as internal standards.
presence of a cosolvent.
Samples were analysed without any further treatment. Samples
Catalytic olefin epoxidation tests in the presence of ionic that came from the catalytic experiments with ILs were taken
liquids (ILs) were carried out with exactly the same quantities after phase separation (stop stirring first).
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
corresponding signals for the free ligand are shifted slightly
Results and discussion
Synthesis and characterisation of the dioxomolybdenum(VI)
downfield, appearing at 2.93 and 3.04 ppm. For complex 2, the
1
H NMR spectrum shows broad signals at δ = 1.26 (CH
3
), 3.54
complexes
(
CH
2
) and 8.55 (NH).
Complexes 1 and 2 precipitated as pale yellow solids upon
addition of DMB (2 equivalents) or DEO (1 equivalent) to a
solution of the adduct [MoO
The compounds may be handled for short periods of time in
air. Both complexes are soluble in ethanol, CH CN and
CH NO , and insoluble in BTF, n-hexane, DCE and diethyl
ether.
The IR spectrum of 1 contains a pair of bands at 910 and
2 2 2
Cl (THF) ] in THF (Scheme 1).
Crystal structure description
2 2 2
The complexes [MoO Cl (DMB) ] and [MoO Cl (DEO)] present
2
2
3
in the crystalline compounds 1 and 2a (Table 1) exhibit the
expected structural features discussed above. While the crystal
structure of 1 is solely composed of the individual discrete
complexes, in 2a the complex molecular units co-crystallised
alongside solvent (acetonitrile) molecules and uncoordinated
organic ligands.
3
2
47 cm− (cf. 910 and 949 cm in the Raman) which are
1
−1
9
assigned to the asymmetric and symmetric stretching
2
+
vibrations of the cis-[MoO
2
]
group, respectively. The corres-
Cl (THF) ] are
VI
In both compounds the Mo centres are coordinated to
ponding frequencies for the THF adduct [MoO
higher (ca. 960 and 920 cm ), consistent with the stronger
coordination (trans to the oxo groups) of the amide ligand
2
2
2
two chlorine atoms, two terminal oxygen atoms, and two
oxygen atoms from the organic ligand(s), which in 2a appear
as an O,O′-chelated molecule (Fig. 1 and 2). In both complexes
the organic ligands form, alongside the terminal oxo groups,
−
1
9
a,e
in 1.
evidently lower than that for DMF in the complex
MoO Cl (DMF) ] (3), which has the νsym(MovO) band at
about 940 cm
cies (compared with 1) for the ν(MovO) modes (912 and
The donor capacity of the DMB ligand in 1 is, however,
2 4
the equatorial plane of the distorted {MoCl O } coordination
[
2
2
2
polyhedra, with the chlorine atoms being positioned in the
remaining apical positions. The terminal oxo groups in both
complexes markedly exert the well-known trans effect with the
MovO bond lengths being found in the 1.681(3)–1.6901(18) Å
range, which contrast with the longer Mo–O bonds to the
organic ligands, 2.2157(15)–2.2394(16) Å (Tables 2 and 3). The
longer Mo–Cl bond distances, allied with the large dispersion
in the cis and trans internal octahedral angles [70.76(5)–105.14
−
1 9a
.
Complex 2 exhibits slightly higher frequen-
1
−1
9
55 cm− in IR, 913 and 959 cm in the Raman), possibly
indicating an even lower donor capacity for the organic ligand
and consequently a higher Lewis acidity for the metal centre.
The IR spectra of 1 and 2 point to coordination of the organic
ligands through the oxygen atom since the carbonyl stretching
−
1
frequency shifts from 1626 and 1657 cm for the free ligands
DMB and DEO, respectively, to 1587 for 1 and 1643 cm− for 2.
This result is explained by the decrease in the double bond
character of CvO and the subsequent increase in the C–N
(
10)° and 159.76(3)–168.15(16)°, respectively], further demon-
1
strate the highly distorted coordination environment for all
VI
Mo centres.
9
e
double bond character. The smaller shift observed for 2
seems to be consistent with the weaker coordination of the
ligand when compared with DMB in 1.
1
The H NMR spectrum for complex 1 shows several overlap-
ping signals centred at 7.36 ppm due to the hydrogen atoms of
the phenyl ring, and two singlets at 2.88 and 3.01 ppm, one
due to the methyl group cis to the carbonyl oxygen atom and
the other due to the methyl group in the trans position. The
Fig. 1 Schematic representation of the molecular unit composing the
crystal structure of 1. Non-hydrogen atoms are represented as thermal
ellipsoids drawn at the 50% probability level and hydrogen atoms as
small spheres with arbitrary radii. For simplicity, only the atoms compos-
ing the first coordination sphere of the crystallographically independent
molybdenum centre are labelled. Symmetry transformation used to
generate equivalent atoms: (i) 1 − x, 1 − y, z.
Scheme 1 Preparation of 1 and 2. Reaction conditions: THF, RT, 1 h.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
Table 3 Selected bond lengths (in Å) and angles (in degrees) for the
two crystallographically independent molybdenum centres present in 2a
Mo1–O1
Mo1–O2
Mo1–O5
1.6847(17)
1.6888(17)
2.2259(15)
104.89(9)
90.72(7)
161.32(8)
95.84(7)
95.26(7)
164.36(7)
93.68(7)
Mo1–O6
Mo1–Cl1
Mo1–Cl2
O2–Mo1–Cl2
O5–Mo1–Cl1
O5–Mo1–Cl2
O6–Mo1–O5
O6–Mo1–Cl1
O6–Mo1–Cl2
Cl1–Mo1–Cl2
2.2186(14)
2.3726(6)
2.3853(7)
95.77(7)
82.86(4)
83.47(4)
70.76(5)
84.57(4)
80.59(4)
162.47(2)
O1–Mo1–O2
O1–Mo1–O5
O1–Mo1–O6
O1–Mo1–Cl1
O1–Mo1–Cl2
O2–Mo1–O5
O2–Mo1–O6
O2–Mo1–Cl1
Mo2–O3
94.43(7)
1.6901(18)
1.6823(18)
2.2394(16)
91.42(8)
162.35(8)
97.33(8)
Mo2–O8
Mo2–Cl3
Mo2–Cl4
O4–Mo2–Cl4
O7–Mo2–Cl3
O7–Mo2–Cl4
O8–Mo2–O7
O8–Mo2–Cl3
O8–Mo2–Cl4
Cl3–Mo2–Cl4
2.2157(15)
2.3669(8)
2.3673(8)
95.37(7)
81.30(5)
82.98(5)
71.11(6)
82.89(5)
80.01(5)
159.76(3)
Mo2–O4
Mo2–O7
O3–Mo2–O7
O3–Mo2–O8
O3–Mo2–Cl3
O3–Mo2–Cl4
O4–Mo2–O3
O4–Mo2–O7
O4–Mo2–O8
O4–Mo2–Cl3
95.73(8)
105.14(10)
163.44(8)
92.35(8)
96.07(7)
than the decane present in the TBHP solution, yielded cyclooc-
tene oxide (CyO) as the only reaction product (Fig. 3 and
Table 4). Complex 3 has been previously studied in this reac-
Fig. 2 (a) Schematic representation of one independent molecular unit
composing the crystal structure of 2a. Please see the caption to Fig. 1
for general comments about the graphical representation. (b) Overlay of
2 2
the two crystallographically independent [MoO Cl
(DEO)] complexes tion (although under different conditions from those used
9
f
composing the asymmetric unit in 2a (hydrogen atoms have been here), and was used in this work as a reference.
omitted for clarity), emphasising the distinct conformations for the
terminal alkylic chains of the organic ligands.
Fig. 3 displays the conversion of cis-cyclooctene obtained at
selected reaction times for the complexes studied. The data
reveal the outstanding catalytic activity of 1 and 2 under the
reaction conditions used, with a nearly full conversion being
Table 2 Selected bond lengths (in Å) and angles (in degrees) for the
reached at 2 min reaction time, the first point evaluated.
crystallographically independent molybdenum centre present in com-
a
2 2 2
pound [MoO Cl (DMB) ] (1)
Mo1–O1
Mo1–O1i
Mo1–O2
O1–Mo1–O1i
O1–Mo1–Cl1
O1–Mo1–Cl1i
O1i–Mo1–Cl1
O1i–Mo1–Cl1i
O2–Mo1–O1
O2–Mo1–O1i
O2–Mo1–Cl1
2.219(3)
2.219(3)
1.681(3)
77.82(15)
81.98(8)
82.52(8)
82.52(8)
81.98(8)
90.45(14)
168.15(16)
97.69(14)
Mo1–O2i
Mo1–Cl1
Mo1–Cl1i
O2i–Mo1–O1
O2i–Mo1–O1i
O2i–Mo1–O2
O2i–Mo1–Cl1
O2–Mo1–Cl1i
O2i–Mo1–Cl1i
Cl1–Mo1–Cl1i
1.681(3)
2.3800(12)
2.3800(12)
168.15(16)
90.45(14)
101.3(2)
94.92(14)
94.93(14)
97.69(14)
160.04(6)
a
Symmetry transformation used to generate equivalent atoms: (i) 1 −
x, 1 − y, z.
In compound 2a, as mentioned above, the asymmetric unit
is composed of two identical [MoO Cl (DEO)] complexes
2
2
(Fig. 2b). The configuration of the complexes is, nevertheless,
identical, apart from conformational differences regarding the
terminal alkylic chains of the DEO ligand.
Catalytic epoxidation
Fig. 3 Kinetic profiles of the catalytic epoxidation of cis-cyclooctene
The epoxidation of cis-cyclooctene using complexes 1, 2 and 3
(
Cy) using 1 ( ), 2 (▲) and 3 (◆) as catalyst precursors, without
□
as catalyst precursors at 55 °C, with no solvent added other cosolvent.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
When compared with complex 3, which had already been mental conditions the catalytic performance of complex 2,
9
f
shown to be an excellent catalyst for this reaction, a much when compared with that of complex 1, is always higher at the
higher conversion is obtained with 1 and 2 in the early stages early stages of the reaction and independent of the type of
of the reaction, although a similar catalytic performance is olefin used as substrate. These differences in catalytic behav-
observed for longer reaction times. It is expected, based on iour point to the key role of the rate of formation of the active
6
c,k,9e,f
previous studies of [MoO
2
X
2
(L)
n
]-type complexes,
that species in the catalytic process, as suggested above. The com-
the initial complexes act as pre-catalysts, forming the active parison of the three complexes in terms of initial TOF
species in the reaction medium. Therefore, the data obtained numbers and kinetic profile shows that the relative catalytic
may suggest that the formation of the active species is faster behaviour is the same for cyclooctene, α-pinene and limonene,
for complexes 1 and 2 than for complex 3. On the other hand, but quite different for the linear olefins 1-octene and trans-2-
the FT-IR data discussed above suggest that the ligands DMB octene. For these cases, complex 3 shows higher initial
(complex 1) and DEO (complex 2) have less donating character activity than complex 1. The reason for this fact is not clear at
than DMF (complex 3), which may give rise to a more acidic this point but the influence of steric effects cannot be
metal centre in both complexes that could also contribute to a discarded.
higher activity of the actives species. The turnover frequency
TOF) values obtained at 2 min reaction time (Table 4) are olefins has shown that the three complexes behave in a similar
strikingly high for all three catalysts (in the range of way. The catalytic activity is in accordance with the increasing
The study of the epoxidation of the structurally different
(
−1
−1
2
500–3200 mol molMo
h ), which probably reflects the presence of alkyl groups that increase the electron density at
same type of active species and reaction mechanism. Table 5 the double bond, i.e., 1-octene < trans-2-octene < α-pinene <
compares the results obtained in the epoxidation of cis- limonene < cis-cyclooctene, as expected for catalytic systems
9
e,f
cyclooctene with complexes 1–3 with values found in the litera- involving these types of catalysts.
ture for complexes with similar characteristics, namely Concerning selectivity, for all three complexes the reaction
dichlorodioxomolybdenum(VI) complexes bearing monoden- of 1-octene and trans-2-octene gave, respectively, 1,2-epoxyoc-
tate ligands which coordinate to the metal through an oxygen tane and 2,3-epoxyoctane as the only reaction products. On the
atom.
other hand, and despite the significant conversions obtained
The epoxidation of other olefins, namely 1-octene, trans-2- for both α-pinene and (R)-(+)-limonene, the selectivities to the
octene, α-pinene and (R)-(+)-limonene, was also studied for corresponding epoxides were rather low. For α-pinene, the
complexes 1, 2 and 3, without cosolvent at 55 °C (Table 4, selectivity to the epoxide was below 5% for all the catalysts
Fig. 4 for complex 1 and Fig. S1 in the ESI† for complexes 2 under study, and several byproducts were obtained that
and 3). The data obtained clearly show that under these experi- included campholenic aldehyde. In the case of (R)-(-
+)-limonene, the reaction also originated a wide variety of com-
pounds that included limonene oxide and carvone as the main
products. Nevertheless, no products were identified that could
Table 4 Catalytic performance of complexes 1, 2 and 3 in the epoxi- indicate the epoxidation of the terminal double bond, which
dation of cis-cyclooctene, 1-octene, trans-2-octene, α-pinene, and (R)-
agrees with the results found for the linear olefins 1-octene
and trans-2-octene where the more electron-rich double bond
of trans-2-octene is more easily oxidised.
(
+)-limonene, using TBHP at 55 °C
a
TOF (mol
−
1
molMo
The influence of an additional solvent in the reaction
medium was investigated using cis-cyclooctene as model sub-
strate (Fig. 5 and Table 6). The catalytic tests were carried out
at 35 °C due to the high activity of the complexes at 55 °C that
might mask any solvent-dependent differences. Data in
Table 6 also give complementary information on the influence
of the temperature on the catalytic activity. For all the com-
plexes, the TOF values and the conversion of cis-cyclooctene
are, as expected, significantly lower than those obtained at
55 °C. Depending on the type of solvent used, benzotrifluoride
−
1
b
b
Olefin
Cat.
h
)
Conv. (%) Select. (%)
1
2
3
3199
3149
2499
99/100
100/100
96/100
100/100
100/100
100/100
1
2
3
1
2
3
1
2
3
41
182
107
91
208
239
1268
1527
581
32/62c
100/100
100/100
100/100
100/100
100/100
100/100
51/70c
33/64
c
70/91c
67/86
56/85c
d
55/70
55/81
4/2
c
d
5/3
(BTF), n-hexane, 1,2-dichloroethane (DCE), ethanol, CH
3
CN or
d
51/61
4/5
CH NO , the catalytic results changed significantly, although
3
2
e
e
1
2
3
1162
2414
557
93/97
96/99
96/98
33/19; 37/12
33/21; 22/16
20/32; 7/12
with no effect on the selectivity to the epoxide which was
always 100%.
Nitromethane is a polar solvent and in spite of the dilution
effect caused by its addition to the reaction medium, its use is
beneficial in all cases since the reaction rate increases signifi-
e
a
b
Turnover frequency calculated at 2 min reaction. Cyclooctene
conversion and selectivity to epoxide after 6/24 h reaction, unless
c
d
indicated otherwise. 3/24 h reaction. Other products were formed
e
which were not identified. Selectivity to carvone (other products were cantly when compared with the reaction carried out in the
formed which were not identified).
absence of an additional solvent or with other solvents tested.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
Table 5 Epoxidation of cis-cyclooctene at 55 °C with TBHP (in decane) and different molybdenum complexes as catalyst precursors
a
b
c
Complex
Mo : Cy : TBHP
Cosolvent
TOF
Ref.
1
2
3
1 : 100 : 200
1 : 100 : 200
1 : 100 : 200
1 : 100 : 160
1 : 100 : 150
1 : 100 : 150
1 : 100 : 150
1 : 100 : 150
ns
ns
ns
ns
ns
ns
ns
DCE
3199 (2)
3149 (2)
2499 (2)
1003 (10)
150 (30)
71 (30)
This work
This work
This work
9f
7a
7a
9e
9e
[MoO
[MoO
[MoO
[MoO
[MoO
2
Cl
2
Cl
2
Cl
2
Cl
2
Cl
2
2
2
2
2
(DMF)
(OPMePh
dppmO
2
]
2
)
2
]
2
]
(DPF)
(DPF)
2
]
]
575 (10)
605 (10)
2
a
b
OPMePh
2
= methyl(diphenyl)phosphine oxide; dppmO
2
= bis(diphenylphosphineoxide)methane; DPF = N,N-diphenylformamide. ns = no
c
−1 −1
cosolvent was added. TOF (mol molMo
conversion at 24 h was 100%.
h ). The reaction time used to calculate the TOF is given in parentheses. For all systems the Cy
Fig. 4 Oxidation of cis-cyclooctene (◆), 1-octene (□), trans-2-octene
▲), α-pinene (×) and (R)-(+)-limonene (—) with TBHP using complex 1
Fig. 5 Kinetic profiles of cis-cyclooctene epoxidation in the presence
of 2, using TBHP at 35 °C without a cosolvent (◆), or with BTF (□), n-
(
as catalyst precursor at 55 °C, without cosolvent.
hexane (▲), DCE (×), ethanol (—), CH
3 3 2
CN (●) or CH (○) as
NO
cosolvents.
The low catalytic activity obtained in presence of n-hexane,
non-polar and aprotic, may be explained by the low solubility
With the aim of studying the catalytic behaviour of these
of the complexes in this solvent that will give rise to a small dioxomolybdenum(VI) complexes in a biphasic system, epoxi-
amount of active species in the reaction medium. Data in dation of cis-cyclooctene was carried out in the presence of
Table 6 are therefore significantly influenced by the dielectric different ionic liquids (IL). Such biphasic systems may have
properties of the solvent, probably due to changes in the solu- implications on the stability of the active species since the
bilities of the complexes.
CH NO has similar characteristics in terms of dielectric simplify the recycling procedure.
constant to CH CN but gives rise to higher catalytic activity.
Table 7 shows the catalytic data obtained with the com-
This behaviour is probably related with the coordinating pro- plexes 1 and 2 at 55 °C. Data for [C mim][Cl] are not shown
complex is dissolved in the IL phase, and additionally may
3
2
3
4
3
perties of CH CN that allow the participation of solvent mole- since this IL did not allow a phase separation and, addition-
cules in the catalytic process, causing a decrease in activity. In ally, no catalytic activity could be observed. For the remaining
fact, the use of ethanol, another polar and coordinating ILs a complete phase separation was achieved and both com-
solvent, give rise to an even lower catalytic activity when com- plexes showed significant activity with 100% selectivity to the
pared with polar and non-coordinating solvents such as epoxide. When compared with the system in the absence of
CH NO , BTF and DCE.
any solvent (Table 4), both complexes show a decrease in
3
2
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Table 6 Catalytic performance of complexes 1, 2 and 3 in the epoxi-
dation of cis-cyclooctene
The behaviour of the catalytic system is clearly influenced
both by the nature of the anionic part of the IL and by the
complex. Comparing complex 1 with complex 2, the data show
that the catalytic activity of complex 1 is higher at the early
stages of the reaction and does not seem to change with the
nature of the IL. A similar behaviour in terms of activity of
complexes 1 and 2 was already encountered in the presence of
cosolvents at 35 °C. Thus, decreasing the reaction rate, either
by addition of a solvent, lowering the temperature or addition
of an IL, emphasises different rates of formation of active
species between the catalysts. The formation of the active
species may involve the loss or substitution of ligands and
further work is in progress in order to clarify this important
issue. Concerning the conversion it is also clear that complex 1
is more active than complex 2, in line with the TOF values
obtained at the beginning of the reaction.
a
b
TOF (mol molMo h−1
c
−1
d
Catalyst
Cosolvent
None
)
Conv. (%)
1
1680
114
1597
2601
887
2885
1274
1496
3
1148
2111
457
2152
1790
1666
161
1320
2013
83
68/84
47/52
83/90
94/98
42/66
97/99
82/85
65/76
41/54
77/92
88/94
50/80
93/97
89/93
77/86
47/55
79/91
93/96
45/78
97/98
85/91
Hexane (1.89)
BTF (9.18)
DCE (10.4)
EtOH (24.6)
CH
CH
3
NO
3
2
(35.9)
CN (36.6)
2
3
None
Hexane
BTF
DCE
EtOH
CH
CH
3
NO
CN
2
3
None
Hexane
BTF
DCE
EtOH
Comparing the different ILs, higher conversions at 24 h are
−
obtained with the IL comprising NTf
2
as anionic part fol-
−
−
lowed by the one with BF
4
and finally the IL with PF
6
CH
CH
3
NO
CN
2
2692
885
3
counter-ions. This behaviour is observed for both complexes
but is more evident for complex 2, probably due to the lower
reaction rate. The data suggest that the anionic part of the IL
a
Reaction conditions: 35 °C, catalyst/substrate/TBHP molar ratio =
b
c
1
: 100 : 200. Dielectric constants are given in parentheses. Turnover
d
frequency calculated at 2 min reaction. Cyclooctene conversion after may have an influence on the catalytically active species. In
6/24 h reaction.
−
fact, the interaction of the NTf
2
anion has already been pro-
2
4
posed for a palladium(II) complex. Illner et al. even referred
that the small interaction observed for one of their complexes
may raise the question of whether the coordination to the
metal centre might be possible at a very large excess of this
anion (for example, on dissolving the complex in an ionic
liquid) in spite of its low nucleophilicity.
The catalytic systems containing the ILs allow a complete
and fast phase separation once the stirring is stopped, and
therefore offer the opportunity to devise a simple and effective
recycling procedure. In order to evaluate the stability of these
catalysts during consecutive runs, several recycling experi-
ments were undertaken. Catalysts 1 and 2 were tested with
Table 7 Epoxidation of cis-cyclooctene in the presence of ionic liquids
a
−1
b
Catalyst
Ionic liquid
TOF (mol molMo h−1
)
Conv. (%)
1
[C
4
4
4
4
4
4
mim][NTf
mim][BF
mim][PF
mim][NTf
mim][BF
mim][PF
2
4
]
]
]
2914
2572
2899
1913
1849
1967
100/100
98/100
96/96
87/97
93/93
[
[
C
C
6
2
[C
2
]
[
[
C
C
4
]
]
6
70/71
a
b
Turnover frequency calculated at 2 min reaction. Cyclooctene
conversion after 6/24 h reaction.
[C
4
mim][NTf
2 4
], and catalyst 2 was also tested with [C mim]
[
BF ]. The results in Table 8 for the IL [C mim][NTf ] indicate
4
4
2
that the active species are stable in the reaction media and
activity at the early stages of the reaction, reflected in the lower that the catalysts are dissolved in the IL phase since no loss of
values of TOFs obtained. This was expected due to the biphasic activity is observed in consecutive runs.
nature of the system, with the catalyst and cis-cyclooctene in
different phases.
Surprisingly, the activity of the catalysts seems to increase
with the consecutive runs when [C mim][NTf ] is used. This is
4
2
Table 8 Recycling runs with cis-cyclooctene
Run 1
Run 2
Run 3
Run 4
a
b
a
b
a
b
a
b
Catalyst/IL
TOF
Conv.
TOF
Conv.
TOF
Conv.
TOF
Conv.
1/[C
2/[C
2/[C
4
4
4
mim][NTf
mim][NTf
mim][BF ]
4
2
]
]
2914
1913
1849
99
87
93
2041
1011
846
99
99
72
2217
3031
594
99
100
95
2464
2252
690
100
100
92
2
a
Turnover frequency (mol molMo h−1) calculated at 2 min reaction. Cyclooctene conversion after 2 h reaction for complex 1 and 4 h reaction
−1
b
for complex 2.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
The reaction products (cyclooctene oxide and tBuOH) could be
efficiently removed by extraction and evaporation steps, and
the resultant IL–catalyst mixture could be recycled for multiple
catalytic runs without loss of activity, suggesting that the dis-
solved catalyst was stable and resistant to leaching. Moreover,
we have shown that a regenerated IL–catalyst mixture can be
successfully reused with a different substrate from that used
initially. Of the four different ILs examined in this study (chan-
ging only the nature of the counter-anion), the best results
−
were obtained with [C
4
mim][NTf
2
]. The NTf
2
anion could play
VI
a non-innocent role, interacting with Mo centres. To gain
Fig. 6 Olefin conversion with complex 2 and [C
4
mim][NTf
2
]. Runs 1 further insight into the chemistry of these systems and to opti-
and 2 correspond to the epoxidation of cis-cyclooctene, and runs 3 and
correspond to the epoxidation of (R)-(+)-limonene.
mise the catalytic process, future work should focus on a sys-
tematic investigation of a wider range of imidazolium-based
ILs, varying not only the anion but also the cation.
4
more evident with catalyst 2, probably due to its lower catalytic
activity. For this catalyst the conversion obtained after 4 h in
the first run is only 87%. However, in the following runs, the
conversions are 99–100%, indicating a progressive increase in
activity. Although the present data do not allow a full compre-
Acknowledgements
We acknowledge funding by FEDER (Fundo Europeu de Desen-
volvimento Regional) through COMPETE (Programa Operacio-
nal Factores de Competitividade). National funding through
the FCT (Fundação para a Ciência e a Tecnologia) within the
projects FCOMP-01-0124-FEDER-029779 (FCT ref. PTDC/
QEQ-SUP/1906/2012, including research grants to A.C.G. (ref.
BPD/UI89/4864/2014) and T.S.M.O.) and FCOMP-01-0124-
FEDER-020658 (FCT ref. PTDC/EQU-EQU/121677/2010) is
thanked. This work was developed in the scope of the project
CICECO – Aveiro Institute of Materials (FCT ref. UID/CTM/
hension of the reasons behind this behaviour, as mentioned
−
before, the participation of the NTf
2
ion in the stabilisation
of the active species cannot be discarded at this point. It
should be noted that the initial reaction rates are necessarily
lower in the second and following runs when compared with
the first run due to the fact that the reaction is initiated with
the addition of TBHP, which is at room temperature and there-
fore colder than the reaction media.
5
0011/2013), financed by national funds through the FCT/MEC
Due to the exceptional stability of the catalytic species and
the simplicity of the recycling procedure, the possibility of
changing the olefin between recycling cycles was also tested.
The system comprising catalyst 2 and [C mim][NTf ] was used
to convert cis-cyclooctene for 2 cycles and then the olefin was
changed and the system used to convert limonene for 2 more
cycles. The results in Fig. 6 clearly show that the system is
highly robust and flexible, opening up a wide range of
applications.
and co-financed by FEDER under the PT2020 Partnership
Agreement. The FCT and CICECO are acknowledged for finan-
cial support towards the purchase of the single-crystal
diffractometer.
4
2
Notes and references
1
(a) M. G. Clerici, M. Ricci and G. Strukul, in Metal-catalysis
in Industrial Organic Processes, ed. G. P. Chiusoli and
P. M. Maitlis, Royal Society of Chemistry Publishing, Cam-
bridge, 2008; (b) K. Jeyakumar and D. K. Chand, J. Chem.
Sci., 2009, 121, 111.
Conclusions
2 2 n
Two new complexes of the type [MoO Cl (L) ] bearing oxygen-
donor ligands have been prepared and characterised. To the
best of our knowledge, complex 2 containing bidentate DEO is
the first example of a dioxomolybdenum(VI) complex with a
neutral α-dicarbonyl ligand. These complexes are extremely
active and selective catalysts for the epoxidation of standard
substrates such as cis-cyclooctene, using TBHP as oxidant. The
enhanced Lewis acidity of the complexes probably contributes
in a major way to their high catalytic activity. On the other
hand, this high acidity may explain the lower selectivities
measured for more sensitive epoxides such as α-pinene oxide
and limonene oxide. Complexes 1 and 2 were also applied in
liquid–liquid biphasic catalytic epoxidation reactions by using
an ionic liquid as a solvent to immobilise the metal catalysts.
2 R. Sanz and M. R. Pedrosa, Curr. Org. Synth., 2009, 6, 239.
3 (a) F. E. Kühn, A. M. Santos, I. S. Gonçalves, C. C. Romão
and A. D. Lopes, Appl. Organomet. Chem., 2001, 15, 43;
(b) F. E. Kühn, A. M. Santos and M. Abrantes, Chem. Rev.,
2006, 106, 2455; (c) K. R. Jain, W. A. Herrmann and
F. E. Kühn, Coord. Chem. Rev., 2008, 252, 556.
4 F. J. Arnáiz, R. Aguado and J. M. Martinez de llarduya, Poly-
hedron, 1994, 13, 3257.
5 J. H. Enemark, J. J. A. Cooney, J.-J. Wang and R. H. Holm,
Chem. Rev., 2004, 104, 1175.
6 (a) F. E. Kühn, A. D. Lopes, A. M. Santos, E. Herdtweck,
J. J. Haider, C. C. Romão and A. G. Santos, J. Mol. Catal. A:
Chem., 2000, 151, 147; (b) F. E. Kühn, M. Groarke,
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
É. Bencze, E. Herdtweck, A. Prazeres, A. M. Santos,
M. J. Calhorda, C. C. Romão, I. S. Gonçalves, A. D. Lopes
A. A. Valente, M. Pillinger, J. P. Lourenço and
I. S. Gonçalves, J. Mol. Catal. A: Chem., 2009, 297, 110.
and M. Pillinger, Chem. – Eur. J., 2002, 8, 2370; (c) A. Al- 10 C. D. Nunes, A. A. Valente, M. Pillinger, J. Rocha and
Ajlouni, A. A. Valente, C. D. Nunes, M. Pillinger, I. S. Gonçalves, Chem. – Eur. J., 2003, 9, 4380.
A. M. Santos, J. Zhao, C. C. Romão, I. S. Gonçalves and 11 F. E. Kühn, E. Herdtweck, J. J. Haider, W. A. Herrmann,
F. E. Kühn, Eur. J. Inorg. Chem., 2005, 1716; (d) S. M. Bruno,
B. Monteiro, M. S. Balula, C. Lourenço, A. A. Valente,
I. S. Gonçalves, A. D. Lopes and C. C. Romão, J. Organomet.
Chem., 1999, 583, 3.
M. Pillinger, P. Ribeiro-Claro and I. S. Gonçalves, Molecules, 12 (a) F. J. Arnáiz, R. Aguado, M. R. Pedrosa, J. Mahía and
2
006, 11, 298; (e) A. Günyar, M.-D. Zhou, M. Drees,
M. A. Maestro, Polyhedron, 2002, 21, 1635; (b) F. J. Arnáiz,
R. Aguado, M. R. Pedrosa and A. D. Cian, Inorg. Chim. Acta,
2003, 347, 33; (c) N. Manwani, M. C. Gupta, R. Ratnani,
J. E. Drake, M. B. Hursthouse and M. E. Light, Inorg. Chim.
Acta, 2004, 357, 939; (d) A. L. Bingham, J. E. Drake,
M. B. Hursthouse, M. E. Light, R. Kumar and R. Ratnani,
Polyhedron, 2006, 25, 3238.
P. N. W. Baxter, G. Bassioni, E. Herdtweck and F. E. Kühn,
Dalton Trans., 2009, 8746; (f) A. Günyar, D. Betz, M. Drees,
E. Herdtweck and F. E. Kühn, J. Mol. Catal. A: Chem., 2010,
3
31, 117; (g) A. Günyar and F. E. Kühn, J. Mol. Catal. A:
Chem., 2010, 319, 108; (h) S. M. Bruno, C. C. L. Pereira,
M. S. Balula, M. Nolasco, A. A. Valente, A. Hazell,
M. Pillinger, P. Ribeiro-Claro and I. S. Gonçalves, J. Mol. 13 (a) F. J. Arnáiz, R. Aguado, J. Sanz-Aparicio and
Catal. A: Chem., 2007, 261, 79; (i) A. C. Coelho, M. Nolasco,
S. S. Balula, M. M. Antunes, C. C. L. Pereira, F. A. A. Paz,
A. A. Valente, M. Pillinger, P. Ribeiro-Claro, J. Klinowski
and I. S. Gonçalves, Inorg. Chem., 2011, 50, 525;
M. Martinez-Ripoll, Polyhedron, 1994, 13, 2745;
(b) F. J. Arnáiz, R. Aguado, M. R. Pedrosa, J. Mahía and
M. A. Maestro, Polyhedron, 2001, 20, 2781; (c) F. Li and
G. Yuan, Angew. Chem., Int. Ed., 2006, 45, 6541; (d) Y. Luan,
G. Wang, R. L. Luck and M. Yang, Eur. J. Inorg. Chem.,
2007, 1215.
(
j) T. R. Amarante, P. Neves, F. A. A. Paz, A. A. Valente,
M. Pillinger and I. S. Gonçalves, Dalton Trans., 2014, 43,
6
059; (k) A. A. Valente, J. Moreira, A. D. Lopes, M. Pillinger, 14 (a) G. Wang, G. Chen, R. L. Luck, Z. Wang, Z. Mu,
C. D. Nunes, C. C. Romão, F. E. Kühn and I. S. Gonçalves,
New J. Chem., 2004, 28, 308; (l) Z. Petrovski, M. Pillinger,
A. A. Valente, I. S. Gonçalves, A. Hazell and C. C. Romão,
J. Mol. Catal. A: Chem., 2005, 227, 67; (m) S. Gago,
J. E. Rodríguez-Borges, C. Teixeira, A. M. Santos, J. Zhao,
D. G. Evans and X. Duan, Inorg. Chim. Acta, 2004, 357,
3223; (b) M. B. Hursthouse, W. Levason, R. Ratnani and
G. Reid, Polyhedron, 2004, 23, 1915; (c) M. F. Davis,
W. Levason, R. Ratnani, G. Reid, T. Rose and M. Webster,
Eur. J. Inorg. Chem., 2007, 306.
M. Pillinger, C. D. Nunes, Z. Petrovski, T. M. Santos, 15 T. Kottke and D. Stalke, J. Appl. Crystallogr., 1993, 26, 615.
F. E. Kühn, C. C. Romão and I. S. Gonçalves, J. Mol. Catal. 16 APEX2 Data Collection Software, Version 2.1-RC13, Bruker
A: Chem., 2005, 236, 1.
AXS, Delft, The Netherlands, 2006.
7
8
(a) A. Jimtaisong and R. L. Luck, Inorg. Chem., 2006, 45, 17 Cryopad, Remote monitoring and control, Version 1.451,
1
0391; (b) K. Dreisch, C. Andersson, M. Hakansson and
Oxford Cryosystems, Oxford, United Kingdom, 2006.
18 SAINT+, Data Integration Engine, version 8.27b, Bruker AXS,
Madison, Wisconsin, USA, 1997–2012.
S. Jagner, J. Chem. Soc., Dalton Trans., 1993, 1045.
(a) M. D. Brown, M. B. Hursthouse, W. Levason, W. Ratnani
and G. Reid, Dalton Trans., 2004, 2487; (b) M. F. Davis, 19 G. M. Sheldrick, SADABS, version 2012/1, Bruker AXS Area
W. Levason, M. E. Light, R. Ratnani, G. Reid, K. Saraswat
and M. Webster, Eur. J. Inorg. Chem., 2007, 1903.
Detector Scaling and Absorption Correction, Bruker AXS,
Madison, Wisconsin, USA, 2012.
9
(a) R. J. Butcher, H. P. Gunz, R. G. A. R. Maclagan, 20 G. M. Sheldrick, SHELXT-2014, Program for Crystal Structure
H. K. J. Powell, C. J. Wilkins and Y. S. Hian, J. Chem. Soc., Solution, University of Göttingen, 2014.
Dalton Trans., 1975, 1223; (b) P. Chaumette, H. Mimoun, 21 (a) G. M. Sheldrick, Acta Crystallogr., Sect. C: Cryst. Struct.
L. Saussine, J. Fischer and A. Mitschler, J. Organomet.
Chem., 1983, 250, 291; (c) R. Sanz, J. Escribano,
M. R. Pedrosa, R. Aguado and F. J. Arnáiz, Adv. Synth.
Catal., 2007, 349, 713; (d) J. E. Drake, M. B. Hursthouse,
Commun., 2015, 71, 3; (b) G. M. Sheldrick, SHELXL, version
2014, Program for Crystal Structure Refinement, University of
Göttingen, 2014; (c) G. M. Sheldrick, Acta Crystallogr., Sect.
A: Found. Crystallogr., 2008, 64, 112.
M. E. Light, R. Kumar and R. Ratnani, J. Chem. Crystallogr., 22 C. B. Hübschle, G. M. Sheldrick and B. Dittric, J. Appl. Crys-
007, 37, 421; (e) S. Gago, P. Neves, B. Monteiro, tallogr., 2011, 44, 1281.
M. Pessêgo, A. D. Lopes, A. A. Valente, F. A. A. Paz, 23 K. Brandenburg, DIAMOND, Version 3.2f, Crystal Impact
M. Pillinger, J. Moreira, C. M. Silva and I. S. Gonçalves, GbR, Bonn, Germany, 1997–2010.
Eur. J. Inorg. Chem., 2009, 4528; (f) B. Monteiro, 24 P. Illner, R. Puchta, F. W. Heinemann and R. van Eldik,
S. S. Balula, S. Gago, C. Grosso, S. Figueiredo, A. D. Lopes, Dalton Trans., 2009, 2795.
2
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015