Lewis base adducts of bis-(halogeno)dioxomolybdenum(VI): Syntheses, structures, and catalytic applications

Article abstract of DOI:10.1016/S1381-1169(99)00262-9

Reaction of solvent substituted MoO₂X₂(Solv)₂ complexes ((Solv) = THF, CH₃CN) with mono- and bidentate nitrogen and oxygen donor ligands leads to complexes of the type MoO₂X₂L₂ in nearly quantitative yields at room temperature within a few minutes. The ⁹⁵Mo and ¹⁷O NMR data of selected complexes as well as the Mo=O IR vibrations were used to probe the influence of the ligands on the electronic properties of the metal and the Mo=O bond. Two complexes have additionally been examined by single crystal X-ray analysis. The activity of the MoO₂X₂L₂ complexes as catalysts in olefin epoxidation with t-butylhydroperoxide as oxidizing agent depends on both the nature of the organic ligand L and the halogeno ligand X. The difference in activity observed between Cl and Br substituted complexes is not very pronounced. In general, the Cl derivatives are more active than their Br analogues. The organic ligands L display a significant influence on the catalytic performance. Complexes with ligands bearing aromatic substituents at N are in all cases, much more active than those bearing aliphatic substituents. The less active complexes can be activated by raising the temperature and extending the reaction time. In all observed cases, these changes produce a significant increase of the product yield. (C) 2000 Elsevier Science B.V.

Full text of DOI:10.1016/S1381-1169(99)00262-9

Ž
.
Journal of Molecular Catalysis A: Chemical 151 2000 147–160
www.elsevier.comrlocatermolcata
ž
/
ž /
Lewis base adducts of bis- halogeno dioxomolybdenum VI :
syntheses, structures, and catalytic applications
b
a
a
Fritz E. Kuhn a,), Andre D. Lopes , Ana M. Santos , Eberhardt Herdtweck ,
¨
´
a
b
c
Joachim J. Haider , Carlos C. Romao , A. Gil Santos
Anorganisch-chemisches Institut der Technischen UniÕersitat Munchen, Lichtenbergstraße 4, D-85747 Garching bei Munchen, Germany
˜
a
¨
¨
¨
b
Instituto de Tecnologia Quımica e Biologica da UniÕersidade NoÕa de Lisboa, Quinta do Marques, EAN, Apartado 127, P-2781-901
´
´
ˆ
Oeiras, Portugal
c
Departamento de Quımica, Centro de Quımica Fina e Biotecnologia, Faculdade de Ciencias e Tecnologia, UniÕersidade NoÕa de Lisboa,
´
´
ˆ
P-2825 Monte de Caparica, Portugal
Received 15 April 1999; received in revised form 25 May 1999; accepted 22 June 1999
Abstract
Ž
.
ŽŽ
.
.
Reaction of solvent substituted MoO2 X2 Solv complexes Solv sTHF, CH3CN with mono- and bidentate nitrogen
2
and oxygen donor ligands leads to complexes of the type MoO2 X2L2 in nearly quantitative yields at room temperature
within a few minutes. The 95Mo and 17O NMR data of selected complexes as well as the Mo5O IR vibrations were used to
probe the influence of the ligands on the electronic properties of the metal and the Mo5O bond. Two complexes have
additionally been examined by single crystal X-ray analysis. The activity of the MoO2 X2L2 complexes as catalysts in olefin
epoxidation with t-butylhydroperoxide as oxidizing agent depends on both the nature of the organic ligand L and the
halogeno ligand X. The difference in activity observed between Cl and Br substituted complexes is not very pronounced. In
general, the Cl derivatives are more active than their Br analogues. The organic ligands L display a significant influence on
the catalytic performance. Complexes with ligands bearing aromatic substituents at N are in all cases, much more active than
those bearing aliphatic substituents. The less active complexes can be activated by raising the temperature and extending the
reaction time. In all observed cases, these changes produce a significant increase of the product yield. q 2000 Elsevier
Science B.V. All rights reserved.
Keywords: Catalysis; Chelating ligands; Molybdenum; Olefins; Oxidation; X-ray structures
1. Introduction
regard to the catalyst precursors, two types of
complex are particularly important, namely the
q
bisperoxo complexes A and the cis-MoO22
Ž
.
Several molybdenum VI complexes have
proven to be useful catalysts for the epoxidation
of olefins with hydroperoxides like H2O2 and
alkylperoxides and the mechanism of these reac-
derivatives B.
w
x
tions has been recently reexamined 1–3 . With
)
Corresponding author.
1381-1169r00r$ - see front matter q 2000 Elsevier Science B.V. All rights reserved.
Ž
.
PII: S1381-1169 99 00262-9

(
)
148
F.E. Kuhn et al.rJournal of Molecular Catalysis A: Chemical 151 2000 147–160
¨
.
Peroxomolybdenum compounds have been
recently examined in great detail by the groups
w x w x
hexyl, o-tolyl, p-tolyl ligands and the related
N,N - dicyclohexyl-2,3- dimethyl-1,4-diazabuta-
diene. These ligands were chosen considering
their good coordination capabilities. For com-
parison purposes, complexes with non-chelating
organic oxygen donor ligands were also pre-
pared with picoline N-oxide, acac acac s
acetylacetonate related ligands, 4-anilino-pent-
of Thiel and Priermeier 2 , Thiel 3 , Thiel and
w x
w x
Eppinger 4 , Ballisteri et al. 5 and Campestrini
w x
and di Furia 6 . In these complexes, the control
of the activity can be achieved exclusively by
varying the ligand L.
y
Ž
q
In the case of the cis-MoO22 complexes, the
.
X
Ž
.
Ž . Ž Ž . .
general formula is MoO2 L4 or MoO2 L2 L2 .
Depending on the nature of the ligands, three
main complex types have been examined in
more detail for catalytic purposes, despite the
fact that a much broader variety of compounds
3-ene-2-one, CH3C O CHC N H Ph CH3, and
w
Ž .
4-methylamino-pent-3-ene-2-one CH3C O -
Ž Ž .
x
x
CHC N H CH3 CH3 . These complexes were
characterized and their behavior as olefin epoxi-
dation catalysts with t-butylhydroperoxide as
oxidant was studied.
Ž
.
is known. The first dioxomolybdenum VI com-
plex examined in epoxidation catalysis was the
Ž
.
Ž
.
bis acetylacetonato dioxomolybdenum VI , de-
w x
scribed by Indictor and Brill 7 . Aminoalcoho-
late complexes followed, leading to patents of
2. Results and discussion
w
x
the so-called ARCO-process 8–11 . Finally,
Schiff-bases have been used as ligands of the
2.1. Synthesis and spectroscopic examinations
of compounds 1–20
2
q
Ž
Ž
.
.
w x
Mo VI O2
moiety 12 . The acetylacetonato
complexes and derivatives have been examined
more often in epoxidation catalysis, due to their
Organic ligands with donor functionalities
such as nitrogen or oxygen atoms react readily
w
x
easy and straightforward synthesis 13,14 .
Few catalytic studies of MoO2 X2 L2 com-
plexes have been carried out to date. The most
important ones regard the complexes with non-
chelating O donor ligands L and those with
Ž
.
with complexes of the type MoO2 X2 Solv
2
Ž
Ž
.
.
XsCl, Br; Solv sTHF, NCCH3 forming
the octahedrally coordinated complexes 1–20
Ž
.
Eq. 1
w
x
solvento complexes 15–19 . It has been shown
that these systems form monoperoxo complexes
in stoichiometric reactions with certain
w
x
organoperoxides 18 , but the reaction with ex-
cess peroxide has never been examined in de-
tail. MoO2 X2 L2 systems with chelating nitro-
gen ligands display two important advantages:
first, the two different ligand sets X and L can
be easily varied in order to fine tune the ligand
Ž
.
surrounding of the Mo VI center. Secondly,
bidentate nitrogen donor ligands offer a great
versatility in modifying the electron donor and
acceptor properties of the ligands L.
Ž
.
at room temperature within minutes Fig. 1 .
The product complexes are in general signifi-
cantly less soluble than the starting materials
and precipitate from the reaction mixture. They
Ž
In the present study, MoO2 X2 L2-type Xs
.
Cl, Br complexes were prepared from the sol-
Ž
.
vent adducts of the halogenated molybdenum
oxides and a variety of substituted 1,4-diazabu-
can be purified e.g., from excess of ligand by
washing with diethyl ether and pentanes or hex-
anes.
Ž
. Ž
tadienes R-DAB Rsi-propyl, t-butyl, cyclo-

(
)
F.E. Kuhn et al.rJournal of Molecular Catalysis A: Chemical 151 2000 147–160
149
¨
Žt
. Ž
Ž
.
Ž
..
O2 X2 Bubipy XsCl 11 , Br 12 and
Žt
. Ž
.
MoO2 X2- Bu-DAB 17 , have been already
mentioned in the literature 20–22 .
Most of the complexes 1–20 are soluble
enough in polar organic solvents to produce H
w
x
1
NMR spectra of sufficient quality. However, the
chemical shifts of the protons of these com-
pounds do not differ significantly from those of
the free organic ligands.
95Mo NMR spectra of the more soluble com-
plexes are shown in Table 1. Unfortunately,
several of the synthesized compounds 1–20 are
not soluble enough to give 95Mo NMR spectra
of sufficient quality. The 95Mo NMR spectra
display their signals in the region between ca.
190 and 280 ppm, the same found for the
Ž
.
starting complex MoO2 Br2 NCCH3 2, which
95
Ž95
.
displays its Mo signal at d Mo s278 ppm.
Solvent effects have not been observed: in all
cases examined, the chemical shift of a given
Ž
complex in different solvents e.g., methylene
.
chloride or toluene vs. THF or CH3CN differs
less then 5 ppm. This observation indicates that
Ž
.
the ligand surrounding of the Mo VI center is
not significantly influenced by the solvent. The
95Mo NMR spectra of the product complexes
show a higher shielding of this nucleus as com-
pared to the solvent coordinated starting materi-
als. Despite the fact that the generally low
solubility of the synthesized complexes has
hampered a more thorough study of the ligand
influence on the chemical shift of this system by
Fig. 1. Formulae of the complexes synthesized in this work
according to Eq. 1.
The complexes 1–6, bearing non-chelating
oxygen organic ligands decompose within a few
hours when exposed to air. The chelating nitro-
gen ligand complexes 7–20 are in general more
stable. Several of them are fully stable under
laboratory atmosphere and all of them can be
handled in air. This represents a remarkable
difference to the acetonitrile or THF substituted
starting materials, which are both air- and mois-
ture sensitive and can only be handled and
stored under moisture-free inert gas atmosphere.
The better donor capabilities and perhaps also,
the greater steric bulk of the chelating product
complexes 7–20, stabilize these molecules sig-
nificantly. The mixed N,O ligands of the b-ke-
toimine type act as monodentate and coordinate
via the O atom, as demonstrated crystallograph-
Table 1
17
95 Mo and O NMR terminal oxoligands data of selected com-
plexes. CH2Cl2 , CH3CN and toluene, respectively, were used as
solvents
Ž
.
Compound Chemical shift
Compound Chemical shift
Ž
.
Ž
.
ppm
ppm
Ž95
.
Ž17
.
.
.
.
3
4
5
d
d
d
d
d
d
d
d
d
Mo s160
1
2
3
4
d
d
d
d
O s961
Ž95
Ž95
Ž95
Ž95
Ž95
Ž95
Ž95
Ž95
.
Ž17
Ž17
Ž17
Mo s192
O s936
.
Mo s189
O s970
.
6
Mo s220
O s927
.
13
17
18
19
20
Mo s221
.
Mo s216
.
Mo s277
.
Mo s207
.
Mo s265
Ž
.
ically for 3 see below . The complexes Mo-

(
)
150
F.E. Kuhn et al.rJournal of Molecular Catalysis A: Chemical 151 2000 147–160
¨
95Mo NMR, some conclusions may still be
drawn.
transition metal oxo complexes e.g., OsO4 26 .
It is also known that donor adducts of
w
x
Ž
.
There is a clear inverse halogen dependence
organorhenium VII oxides, RReO3Ln, readily
exchange Lewis base ligands, especially in donor
on the chemical shift of the 95Mo NMR spectra
Ž
w
x
since the signals of the chloro complexes 5, 17,
solvents at room temperature and above 26,27 .
However, in our experiments, we did not ob-
serve any significant ligand exchange between
complexes substituted with nitrogen donor lig-
.
19 are observed at higher field than those of
Ž
. Ž
.
the bromo derivatives 6, 18, 20 Table 1 . The
Ž95
.
d
Mo shift difference between the Cl and the
Ž .
Br complexes lies between 30 and 60 ppm in
the examined cases. Such an inverse halogen
dependence has been observed in other few
ands such as 7 and methyltrioxorhenium VII
Scheme 1 even after several hours in toluene
Ž
.
at 908C. In the case of oxygen donor derivatives
such as 1–6 oxygen exchange with 17O labelled
CH3Re17O3 occurred after about 2 h reaction
time in toluene at 908C to a minor but de-
tectable degree. Unlabelled complexes mea-
sured under comparable conditions did not give
rise to detectable 17O NMR signals. However,
we did not find any indication for the exchange
of the organic donor ligands between the Mo
and the Re complexes. In the Mo case, both
oxygen and nitrogen atoms of the organic lig-
ands seem to be bound more strongly. An oxy-
gen ligand exchange does also not occur easily.
This could be interpreted in terms of the coordi-
Ž
.
cases, which include other Mo VI derivatives
q
of the MoO22 core 23 .
w
x
The 95Mo NMR signals of the complexes
with monodentate oxygen ligands 3–6 are
Ž
.
shifted to higher field compared to the com-
Ž
.
plexes with N donor ligands 13, 17–20 regard-
less of their denticity. Given the high sensitivity
of the 95Mo chemical shift, it is interesting to
note the similarity of the chemical shifts among
complexes with N-substituted 1,3-DAB ligands
Ž
.
13, 17, 19 . It demonstrates that different R
groups in the DAB ligand do not strongly influ-
ence the electron density at the metal center of
these complexes. The half widths of all exam-
ined complexes are below 100 Hz and therefore,
comparatively narrow, not indicating special
ligand exchange phenomena or equilibria be-
tween different species. Temperature changes
do not lead to significant changes of the half
widths of the signals.
Ž
.
native saturation of the Mo VI center by the
strongly coordinated organic ligands. The for-
mation of oxygen bridges, which are suspected
Ž
The precursor compound MoO2 Br2 NCC-
.
17
Ž17
.
H3 2 displays its O NMR signal at d O s
1024 ppm. The chemical shifts of the terminal
oxygen atoms measured for the sufficiently sol-
Ž
.
uble complexes see Table 1 are in the ex-
pected range for transition metal oxo complexes
w
x
in high oxidation states 24,25 , namely between
900 and 1000 ppm. The half width of the
observed signals is ca. 50 Hz. It seems that the
O atoms of the Br containing compounds are
more deshielded than those of the Cl substituted
derivatives. We examined both the oxygen and
the organic ligand exchange with 17O marked
Scheme 1. Neither exchange of terminal oxygens nor that of the
N-donor ligands between MoO2 X2 L2 and CH3ReO3 can be
observed.
Ž
.
methyltrioxorhenium VII , which is known to
exchange easily its oxo functionalities with other

(
)
F.E. Kuhn et al.rJournal of Molecular Catalysis A: Chemical 151 2000 147–160
151
¨
Ž .
to be responsible for the exchange mechanism is
therefore, seemingly hindered.
position to the organic donor ligand s in the
equatorial plane and the halogeno atoms in trans
position to each other in the apical positions.
The complexes 1–20 display their Mo5O
w
x
stretching vibrations in the expected range 28 .
The symmetric and unsymmetric vibrations are
observed at ca. 935 and 905 cmy1, varying ca.
"10 cmy1. Within the experimental error of
2.2. X-ray crystallography
The molecular structure of the compounds 3
and 5PNCCH3 was determined by X-ray crys-
tallography. Details of the single crystal X-ray
experiments are given in Table 2. Key bond
distances are listed in Table 3. Both compounds
y
1
Ž
.
the measurement itself "4 cm , the vibra-
tions are all very similar. The IR data also
support the fact that the complexes 1–20 dis-
Ž
play a distorted octahedral C2v-geometry see
.
Ž
.
X-ray section with the oxygen atoms in trans
Figs. 2 and 3 are monomeric. The core geome-
Table 2
w
Ž
.
Ž
Ž
.
.
x2 Ž . .
Ž
.
Ž
Crystallographic data for MoO2Cl2 CH3C O CHC N H CH3 CH3 3 , and MoO2Cl2 pic.N-oxide 2 PNCCH3 5PNCCH3
3
5
Chemical formula
fw
C12 H22Cl2 MoN2O4
425.16
C12 H14Cl2 MoN2O4 PNCCH3
458.15
Colorrshape
yellowrfragment
0.38P0.25P0.20
monoclinic
yellowrfragment
0.66P0.56P0.46
monoclinic
Ž
.
Cryst. size mm
Cryst. system
Space group
C2rc
P21rn
Ž
.
Ž .
Ž .
1113.1 1
a pm
838.1 1
Ž
.
.
Ž .
Ž .
Ž .
1197.7 1
b pm
1657.0 2
1400.3 1
Ž
Ž .
c pm
1258.0 2
Ž .
a
b
g
8
90
Ž .
93.20 1
90
1744.3 4
4
90
95.59 1
90
Ž .
1857.9 3
Ž .
Ž .
8
Ž .
8
V 10 pm3
6
Ž
.
Ž .
Z
4
Ž
.
Ž
T
K
193
1.619
1.07
193
1.638
1.02
y3
.
rcalcd. g cm
y1
Ž
.
m mm
F000
864
920
Ž
.
l pm
Devicerscan method
Ž .
71.073
71.073
IPDSrf-rotation
2.73 to 24.81
"9, "19, "14
10,624
IPDSrf-rotation
3.68 to 30.38
"15, "19, "16
39,186
Q-range 8
Data collected h,k,l
Number of reflections collected
Ž
.
Number of independent reflections
1473
4448
Ž
.
Ž
.
Number of observed reflections
1473 all data
4448 all data
Number of parameters refined
140
291
Rint
0.0209
0.0257
0.0700
1.153
0.0408r3.03
q0.27ry0.43
0.0399
0.0313
0.0743
1.069
0.0369r1.39
q0.45ry0.50
R1a
wR2b
GOFc
Weights arbd
y3
˚
Ž
.
Drmaxrmin e A
a
Ž5
<
Ž
Ž
<
5.
<
<
R1sS Fo y Fc rS Fo .
b
c
d
2
2
2
1r2
w
.
Ž
2 .2 x
wR2s Sw Fo yFc rSw Fo
.
2
2
2
1r2
w
.
Ž
.x
GOFs Sw Fo yFc r NOyNV
.
2
2
2
2
w
2Ž
w'1r s Fo q aPP qbPP with P: max 0 or Fo q2Fc r3.
.
Ž
.
x
w
Ž
.
x

(
)
152
F.E. Kuhn et al.rJournal of Molecular Catalysis A: Chemical 151 2000 147–160
¨
try around the molybdenum atom is best de-
scribed as a distorted octahedron with two chlo-
ride atoms trans to each other, two oxygen
atoms and two NrO-ligands in a cis position
The Mo–Cl, Mo5O, and the Mo–OrN bond
distances are in good accordance with data re-
ported in the literature for analogous com-
w
x
pounds of the MoCl2O2 L2 type 29 .
Table 3
Ž
.
Ž .
Selected interatomic distances pm and angles 8 for MoO2-
Ž .
w
Ž
.
Ž
Ž .
5
Ž
.
.
x2 Ž .
Ž
Fig. 2. ORTEP style plot of 3 with the atomic labelling scheme.
Cl2 CH3C O CHC N H CH3 CH3
3 , and MoO2Cl2 pic.N-
.
Thermal ellipsoids are drawn at the 50% probability level. Hydro-
gen atoms are omitted for clarity. The molecule has C2-symmetry
with the two-fold axis bisecting the O1–Mo–O1a angle.
oxide 2PNCCH3
3
5PNCCH3
Ž .
238.81 8
Ž .
Ž .
240.33 6
238.54 6
Mo–Cl
Mo–Cl 1
Ž .
Ž .
Mo–Cl 2
Ž .
Mo–O 1
Ž
.
.
Ž .
Ž .
Mo–O 2
Ž
.
.
.
.
218.29 19
Mo–O 1
218.87 15
Ž
216.55 14
Ž .
Mo–O 2
Ž
Ž .
Ž
168.78 19
Mo–O 3
169.86 15
2.3. MoO2 X2 L2 complexes in oxidation cataly-
Ž .
Mo–O 4
Ž
168.72 19
sis
Ž . Ž .
O 1 –C 2
Ž .
N–C 4
Ž .
131.7 4
Ž .
O 2 –N 21
Ž
Ž
.
.
Ž .
134.9 2
128.3 3
Ž .
O 1 –N 11
Ž .
134.1 2
Ž .
We examined most of the complexes de-
scribed above in the catalytic oxidation of cy-
clooctene with t-butylhydroperoxide. The broad
variety of ligands enabled us to do a systematic
study on the influence of the ligands on the
catalytic performance of the catalysts. We addi-
tionally wanted to compare the catalytic behav-
ior of chloride and bromide substituents in oth-
erwise identical molecules. The experimental
Ž .
N–C 6
Ž .
149.6 5
Ž
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Ž .
144.5 5
Ž .
N 11 –C 12
134.6 3
Ž .
134.4 3
Ž . Ž .
Ž
C 1 –C 2
N 11 –C 16
Ž . Ž .
C 2 –C 3
Ž .
Ž
Ž .
134.0 3
Ž .
133.8 3
138.7 5
Ž .
N 21 –C 22
Ž . Ž .
Ž
C 3 –C 4
139.7 4
N 21 –C 26
Ž . Ž .
C 4 –C 5
Ž .
149.9 5
Ž
Ž .
C 12 –C 13
136.8 3
Ž
Ž .
138.9 3
Ž .
138.6 3
C 13 –C 14
Ž
C 14 –C 15
Ž
Ž .
C 14 –C 17
149.9 3
Ž
Ž .
137.0 4
Ž .
136.8 3
C 15 –C 16
Ž
C 22 –C 23
Ž
Ž .
C 23 –C 24
138.8 3
Ž
Ž .
138.1 3
Ž .
150.6 3
Ž .
138.1 3
C 24 –C 25
Ž
C 24 –C 27
Ž
C 25 –C 26
Ž .
Cl–Mo–O 1
Ž .
82.38 5
Ž .
Ž .
Ž .
Cl 1 –Mo–O 2
Ž . Ž .
81.99 4
Ž .
95.93 5
Ž .
Ž .
Cl–Mo–O 2
95.46 6
Cl 1 –Mo–O 3
Ž . Ž .
Ž .
161.76 3
Ž .
Cl–Mo–Cl_a
Cl 1 –Mo–Cl 2
Ž . Ž .
164.13 3
Ž .
Ž .
Ž .
Cl–Mo–O 1 a
83.02 5
Cl 1 –Mo–O 1
Ž . Ž .
84.77 4
Ž .
92.12 8
Ž .
Cl–Mo–O 2 a
Ž .
95.96 6
Cl 1 –Mo–O 4
Ž . Ž .
Ž .
Ž .
Ž .
Ž .
O 1 –Mo–O 2
Ž .
92.15 8
O 2 –Mo–O 3
Ž . Ž .
87.61 6
Ž .
Ž .
83.02 5
Cl_a–Mo–O 1
Ž . Ž .
Cl 2 –Mo–O 2
Ž . Ž .
86.08 4
Ž .
Ž .
O 1 –Mo–O 1 a 73.45 7
O 1 –Mo–O 2
Ž . Ž .
78.57 5
Ž . Ž . Ž .
Ž .
167.76 7
O 1 –Mo–O 2 a 165.60 8
Ž . Ž .
O 2 –Mo–O 4
Ž . Ž .
Ž .
Cl_a–Mo–O 2 95.96 6
Cl 2 –Mo–O 3
Ž . Ž .
93.99 5
Ž . Ž . Ž .
Ž .
165.94 7
Ž .
103.72 8
O 1 a–Mo–O 2 165.60 8
Ž . Ž . Ž .
O 1 –Mo–O 3
Ž . Ž .
O 2 –Mo–O 2 a 102.25 9
Ž . Ž .
O 3 –Mo–O 4
Ž . Ž .
Ž .
Cl_a–Mo–O 1 a 82.38 5
Ž . Ž .
Cl 2 –Mo–O 1
Ž . Ž .
82.63 4
Ž .
97.54 8
Ž .
90.26 7
Cl_a–Mo–O 2 a 95.46 6
Ž . Ž . Ž .
Cl 2 –Mo–O 4
Ž . Ž .
O 1 a–Mo–O 2 a 92.15 8
Ž . Ž .
O 1 –Mo–O 4
Ž
.
Ž .
Ž
.
.
Ž
.
.
Mo–O 1 –C 2 142.63 19
Mo–O 1 –N 11
Ž .
123.13 11
Ž
.
Fig. 3. ORTEP style plot of 5PNCCH3 with the atomic labelling
scheme. Thermal ellipsoids are drawn at the 50% probability
level. Hydrogen atoms are omitted for clarity.
Ž
Ž
124.73 11
Mo–O 2 –N 21
Ž
.
Symmetry operation O O_a: y x, y, 0.5y z .

(
)
F.E. Kuhn et al.rJournal of Molecular Catalysis A: Chemical 151 2000 147–160
153
¨
details are given in the experimental part. Blank
reactions have been performed and, as expected,
it turned out that without catalyst, no significant
epoxide formation was observed under any of
the applied conditions. Significant byproduct
formation e.g., diol instead of epoxide, was
never observed either in the uncatalysed or
catalysed runs. The oxidizing agent was only
consumed during the epoxidation process of the
olefin. A typical catalytic run is shown in Fig. 4
Fig. 5. Yield of cyclooctene epoxide in the presence of selected
Ž
.
for the efficient catalyst MoO2Cl2 o-tolyl-DAB
Ž
epoxidation catalysts after 4 and 24 h, respectively Compound
. .
Ž .
Ž
.
Ž
IsMo2O2 acac , IIsMo CO
.
7 . The reaction reaches a conversion of ca.
6
90% after about 4 h. Running the reaction 24 h
Ž
.
does not significantly improve the yield Fig. 5 .
Compounds showing a worse catalytic perfor-
mance than compound 7 within 4 h usually
display a significant increase of product forma-
tion within 24 h as can also be seen in Fig. 5.
This indicates that the low yield after 4 h is not
mainly due to decomposition of the particular
catalyst under the applied reaction conditions
but to a sluggish reaction. This activity is of the
same order of magnitude of the activity of more
A considerable influence of the coordinating
ligands on the product yield can be observed. A
first general observation clearly seen in Fig. 6 is
the higher catalytic activity of the chloro com-
plexes in comparison to the bromo analogues
e.g., 5 vs. 6 or 7 vs. 8. It is also obvious that the
complexes containing ligands with aromatic
substituents are more active than those with
ligands containing aliphatic substituents in iden-
tical positions e.g., 7 vs. 17. However, the yield
difference between chlorinated and brominated
complexes is in general considerably smaller
than the differences caused by the changes of
the organic ligand. This shows that activity
growth correlates with electron density at the
metal: electron attracting ligands improve activ-
ity. Furthermore, this fact suggests that the or-
ganic ligands remain attached to the metal in the
catalytic active center and determine its activity.
Ž
.
common Mo catalysts like Mo CO
and
6
Ž
. Ž
.
MoO2 acac Fig. 5 .
2
The influence of the ratio of oxidant to olefin
was examined for compound 11. The conver-
sion raises with this ratio and after 4 h reaction
time 558C 43%, 53% and 60% yields are
achieved for a 1:1, 1:1.5 and 1:2 ratio, respec-
tively. After 24 h, these yields increase to 67%,
70% and 72%, respectively.
Ž
.
Ž
. Ž .
Fig. 4. Catalytic activity of MoCl2O2 o-phenyl-DAB
7
in the
Fig. 6. Yield of cyclooctene epoxide in the presence of com-
pounds 1–20 as catalysts after 4 h reaction time. See text and
Section 4 for reaction details.
presence of TBHP in the catalytic epoxidation of cycloctene
epoxidation. See text and Section 4 for reaction details.

(
)
154
F.E. Kuhn et al.rJournal of Molecular Catalysis A: Chemical 151 2000 147–160
¨
Of course, we cannot rule out that the true
catalytic activity is due to another species de-
rived from our precursor complexes. In order to
establish the number and eventually the nature
of the Mo complexes present in solution under
catalytic conditions, we tried to follow the course
1
of the reaction by 95Mo and H NMR. Unfortu-
nately, the rather limited solubility of these Mo
catalysts or catalyst precursors prevented the
observation of any 95Mo resonance at 558C or
even higher temperatures and no significant
changes were detected in the resonances of the
ligand protons. This at least shows that no
significant oxidation or liberation of the ligands
took place under the reaction conditions ap-
plied. Steric hindrance does not seem to play an
important role as the activity of complexes 7–10
are similar as well as the activity of 13–20.
Another point worth mentioning is the behav-
ior of the complexes at higher and lower tem-
peratures as depicted in Fig. 7 with the medium
active complex 11 as an example. The higher
the temperature, at least within the examined
range between 25 and 908C, the higher the
product yield. This statement seems to be trivial
on the first glance but several examples are
known where the catalytic activity of oxidation
catalysts decreases with increasing temperature
above a certain level e.g., in the case of
Fig. 8. Time-dependent yield of cyclooctene epoxide in the pres-
ence of selected oxidation catalysts at 558C. A run without
catalyst at 558C is shown for comparison.
All catalytic reactions show the same time-
Ž
.
dependent curve form Fig. 8 . After a quick
increase of the yield within the first hour the
reaction velocity slows down. The appearance
of these curves does not indicate the transforma-
tion of the original catalyst in another species
during the reaction time and does not show an
observable induction period. According to the
related examples in the literature, the active
w
x
species may be an alkylperoxo complex 2,3 or
a peroxo complex in which case, it may be
either a monoperoxo or a bisperoxo complex
w
x
18 . In the present case, all attempts to identify
the nature of the active peroxo species were
thwarted by the low solubility of the complexes.
Further work designed to overcome this limita-
tion and help clarifying the nature of the active
species of these reactions is currently under way
in our laboratories.
w
x
Ž
.
w x
CH3ReO3 30 and CH3 2 MoO2 L2 31 . It is
also important to notice that the conversion
without catalyst at 908C is considerably lower
than the conversion with catalyst at 258C.
3. Conclusions
A range of 1,3-diazabutadiene derivatives of
Ž
.
MoO2 X2 XsCl, Br has been prepared and
characterized by IR as well as 95Mo and 17O
NMR spectroscopy. In spite of the similar spec-
troscopic properties of all these complexes, con-
siderable activity differences have been found in
the catalytic epoxidation of cyclooctene with
t-butylhydroperoxide. The stability of the com-
plexes and the ligand dependence observed sug-
gest that these chelating ligands remain bound
Fig. 7. Temperature dependence of the catalytic activity of com-
pound 11. A run without catalyst at 908C reaction temperature is
shown for comparison.

(
)
F.E. Kuhn et al.rJournal of Molecular Catalysis A: Chemical 151 2000 147–160
155
¨
to the active center throughout the reaction.
This suggestion raises the possibility of achiev-
ing enantioselective oxidations by means of the
appropriate chiral ligands. This possibility to-
gether with the study of the nature of the active
species is currently under investigation in our
laboratories.
dryness to yield a yellow powder, which was
washed with CH2Cl2 and diethyl ether. Yield:
87%.
Ž
.
C12 H22 N2O4Cl2 Mo 425.17 : calcd. C 48.10,
H 5.10, N 4.77; found C 48.11, H 5.02, N 4.97.
y
Ž
Ž
1..
IR KBr, y cm
: 3165 vs, 3094 vs, 3013 vs,
2971 vs, 2851 vs, 1611 vs, 1547 vs, 1492 vs,
1427 vs, 1364 vs, 1335 vs, 1316 vs, 1273 vs,
Ž
.
1209 vs, 1080 s, 936, 897, vs, y Mo5O , 835 s,
1
Ž
4. Experimental section
783 vs, 756 vs. H NMR CHCl3-d1, 300 MHz,
r.t., d ppm : 12.22 b, 2H , 7.35–6.93 m,
. Ž . Ž . Ž .
10H , 3.75 s, 2H , 2.41 s, 6H , 1.96 s, 6H .
Ž
..
Ž
.
Ž
All preparations and manipulations were done
with standard Schlenk techniques under an at-
mosphere of nitrogen. Solvents were dried by
[
( )
( ( )
)
]
4.2. MoO2 Br2 2CH3C O CHC N H Ph CH3
2
Ž
( )
2
standard procedures THF, n-hexane and Et2O
over Narbenzophenone ketyl; CH2Cl2 and
.
Ž
.
Ž
2
NCMe over CaH2 , distilled under argon and
A solution of MoO2 Br2 NCMe
0.47 g,
˚
˚
Ž
.
.
Ž
.
.
kept over 4 A molecular sieves 3 A for NCMe .
Microanalyses were performed at the ITQB.
NMR spectra were measured on a Bruker CXP
1.28 mmol in THF 15 ml was treated with
Ž .
Ž
Ž
.
CH3C O CHC HNPh CH3 0.45 g, 2.56 mmol .
The colour of the solution changed to orange
and an orange precipitate was formed. The mix-
ture reaction was stirred for 30 min and the
suspension was then taken to dryness to yield an
orange powder which was washed with CH2Cl2
and diethyl ether. Yield: 80%.
C12 H22 N2O4Br2 Mo 526.10 : calcd. C 41.40,
H 4.11, N 4.39; found C 41.15, H 4.13, N 4.24.
: 3127 m, 3061 m, 2978 m,
1597 vs, 1580 vs, 1543 vs, 1495 vs, 1433 s,
1
300 and Bruker Avance DPX-400. H NMR
spectra were recorded at 300 MHz, 17O NMR
spectra were recorded at 54.14 MHz, 95Mo NMR
spectra at 26.07 MHz. The IR spectra were
measured on a Unican Mattson Mod 7000 FTIR
spectrometer. 4-Anilino-pent-3-ene-2-one
Ž
.
w
Ž .
Ž
.
x w
x
CH3C O CHC HNPh CH3
32 , 4-methyl-
y
Ž
Ž
1..
w
Ž
.
amino-pent-3-ene-2-one CH 3C O CHC-
N H CH3 CH3 32 , Bubipy 33 , CYDAB
IR KBr, y cm
t
Ž Ž .
.
x w
x
w x
w
x
w
x
w x
34 , p-phen.-DAB 35 , o-phen.-DAB 36 ,
Me-CYDAB 37 , Bu-DAB 38 , iprop.-DAB
1360 vs, 1339 vs, 1202 s, 937, 899, vs,
y Mo5O , 756 vs, 694 s. H NMR CHCl3-d1,
300 MHz, r.t., d ppm : 7.41–7.17 m, 10H ,
5.15 b, 2H , 2.54 s, 6H , 2.01 s, 6H .
t
1
w
x
w x
Ž
.
Ž
w
x
w
x
Ž
.
.
w x
Ž
Ž
..
Ž
.
35 , MoO2Cl2 36 , MoO2Cl2 THF
MoO2 Br2 36 , MoO2 Br2 NC-Me
37 ,
w
x
Ž
w
2x
Ž
.
.
Ž
.
38 were
2
prepared as published or with minor changes.
Ž
.
Ž
.
[
( )
( ( )
)
]
4.3. MoO2Cl2 CH3C O CHC N H CH3 CH3
2
( )
3
MoO2 X2 P2 THF and MoO2 X2 P2 CH3CN
w
x
were prepared as described 19 .
[
( )
( ( )
)
]
4.1. MoO2Cl2 CH3C O CHC N H Ph CH3
2
( )
1
Ž
. Ž
2
A solution of MoO2Cl2 THF 1.10 g, 3.20
.
Ž
.
mmol in THF 15 ml , was treated with
CH3C O CHC HNCH3 CH3 0.72 g, 6.40
mmol . The initial light blue solution changed
immediately to bright yellow and the mixture
reaction was allowed to stir for further 30 min.
The yellow turbid solution was taken to dryness
to yield a yellow powder which was washed
with CH2Cl2 and diethyl ether. Yield: 80%.
Ž .
Ž
.
Ž
Ž
. Ž
2
.
A solution of MoO2Cl2 THF 1.10 g, 3.20
mmol in THF 15 ml , was treated with
CH3C O CHC HNCH3 CH3 0.72 g, 6.40
mmol . The colour of the solution changed to
yellow and a yellow precipitate was formed.
After 30 min, the suspension was brought to
.
Ž
.
Ž .
Ž
.
Ž
.

(
)
156
F.E. Kuhn et al.rJournal of Molecular Catalysis A: Chemical 151 2000 147–160
¨
Ž
.
(
) ( )
C12 H22 N2O4Cl2 Mo 425.17 : calcd. C 32.10,
H 5.17, N 5.99; found: C 31.69; H 4.94; N 6.07.
4.6. MoO2 Br2 pic.N-oxide 2
6
y
Ž
Ž
1..
IR KBr, y cm : 3165 vs, 3094 vs, 3013 vs,
2971 vs, 2851 vs, 1611 vs, 1547 vs, 1492 vs,
1427 vs, 1364 vs, 1335 vs, 1316 vs, 1273 vs,
1209 vs, 1080 s, 936, 897, vs, y Mo5O , 835 s,
783 vs, 756 vs. H NMR CHCl3-d1, 300 MHz,
r.t., d ppm : 5.03 b, 2H , 3.08 s, 6H , 2.32
Ž
.
.
Ž
2
A solution of MoO2 Br2 NCMe
2.11 mmol in THF 15 ml was treated with
0.78 g,
.
Ž
Ž
.
picoline N-oxide 0.46 g, 7.04 mmol . A yellow
solution immediately formed. After concentra-
tion to ca. 3 ml and addition of diethyl ether, a
crystalline yellow precipitate was obtained. The
solid was filtered off, washed with diethyl ether
and dried under vacuum. Yield: 98%.
Ž
.
1
Ž
Ž
..
Ž
.
Ž
.
Ž
.
Ž
.
s, 6H , 2.11 s, 6H .
Ž
.
C12 H14N2O4Br2 Mo 518.02 : calcd. C 28.50,
[
( ) ]
4.4. MoO2 Br2 2CH3C O CHC N H CH3 CH3
2
( )
4
( ( )
)
H 2.79, N 5.54; found C 28.55, H 2.79, N 5.46.
y
Ž
Ž
1..
IR KBr, y cm
: 3103 s, 3046 m, 1628 m,
1499 vs, 1458 s, 1213 vs, 939, 906, vs,
Ž
.
Ž
2
A solution of MoO2 Br2 NCMe
1.25 g,
1
Ž
.
y Mo5O , 858 vs, 828 s, 770 vs, 579 w. H
.
Ž
.
3.39 mmol in THF 15 ml was treated with
CH3C O CHC HNCH3 CH3 0.77 g, 6.78
mmol . The colour of the solution became yel-
Ž
Ž
..
NMR CHCl3-d1, 300 MHz, r.t., d ppm : 8.70
Ž .
Ž
.
Ž
Ž
.
Ž
.
Ž
.
s, 4H , 7.42 d, 4H , 2.48 s, 6H .
.
low and an orange precipitate appeared. After
30 min, the suspension was taken to dryness
and the orange powder was washed with CH2Cl2
and diethyl ether. Yield: 80%.
(
) ( )
4.7. MoO2Cl2 o-phenyl-DAB
7
Ž
. Ž
2
A solution of MoO2Cl2 THF 0.42 g, 1.23
mmol in THF 15 ml , was treated with o-
.
Ž
.
Ž
.
C12 H22 N2O4Br2 Mo 526.10 : calcd. C 26.07,
Ž
.
phenyl-DAB 0.29 g, 1.23 mmol . The slightly
blue solution immediately changed to bright
yellow and the reaction was stirred for further
30 min. The solution was taken to dryness and
the yellowish red product was washed with
diethyl ether and dried under vacuum. Yield:
94%.
H 4.04, N 4.68; found: C 26.17, H 4.12, N 5.00.
y
Ž
Ž
1..
IR KBr, y cm
: 3115 s, 2922 s, 1722 vs,
1613 vs, 1408 m, 1362 s, 1167 m, 1035 m, 957,
Ž
.
916, vs, y Mo5O , 866 s, 768 vs, 640 w, 571 s.
1
Ž
Ž
..
Ž
H NMR CHCl3-d1, 300 MHz, r.t., d ppm :
5.00 b, 2H , 3.03 s, 6H , 2.33 s, 6H , 2.12 s,
Ž
.
Ž
.
Ž
.
.
6H .
Ž
.
C16H16N2O2Cl2 Mo 435.16 : calcd. C 44.16,
H 3.71, N 6.44; found C43.95, H 3.94, N 6.74.
y
1..
(
) ( )
5
4.5. MoO2Cl2 pic.N-oxide 2
Ž
Ž
IR KBr, y cm
: 3023 w, 2972 m, 1485 s,
1456 s, 1368 s, 1185 s, 1115 s, 945, 914, vs,
y Mo5O , 772 s, 752 vs, 712 s. 1H NMR
Ž
. Ž
2
Ž
.
A solution of MoO2Cl2 THF 1.21 g, 3.52
.
Ž
Ž
.
Ž
Ž
..
.
Ž
mmol in THF 15 ml , was treated with picol-
CH3CN-d3, 400 MHz, r.t., d ppm : 8.42 s,
2H , 7.55–7.30 m, 8H , 234 s, 6H .
.
.
Ž
.
Ž
ine N-oxide 0.77 g, 7.04 mmol . The
yellowrgreenish turbid solution which formed
was concentrated and the pale green product
was precipitated by diethyl ether. Yield: 98%.
(
) ( )
8
4.8. MoO2 Br2 o-phenyl-DAB
Ž
.
Ž
.
Ž
1.22 g,
2
C12 H14N2O4Cl2 Mo 417.10 : calcd: C 36.70,
A solution of MoO2 Br2 NC-Me
.
Ž
.
H 3.72, N 9.17; found C 36.69, H 3.72, N 9.16.
: 3123 s, 1626 s, 1493 vs,
1460 vs, 1209 vs, 1179 vs, 1121 s, 1040 m,
3.29 mmol in THF 15 ml , was treated with
y
Ž
Ž
1..
Ž
.
IR KBr, y cm
o-phenyl-DAB 0.78 g, 3.29 mmol . The initial
greenish colour of the solution changed immedi-
ately to orange. The reaction was stirred for 15
min and the solution was then concentrated to
ca. 2 ml. An orange solid was precipitated by
Ž
.
935, 903, vs, y Mo5O , 866 vs, 826 vs, 788 vs,
1
Ž
694 s. H NMR CHCl3-d1, 300 MHz, r.t., d
ppm : 8.62 d, 4H , 7.40 d, 4H , 2.50 s, 6H .
Ž
..
Ž
.
Ž
.
Ž
.

(
)
F.E. Kuhn et al.rJournal of Molecular Catalysis A: Chemical 151 2000 147–160
157
¨
Ž .. Ž .
300 MHz, r.t., d ppm : 8.30 s, 2H , 7.62–7.31
m, 8H , 2.41 s, 6H .
addition of Et2O and the solid was dried under
vacuum. Yield: 90%.
Ž
.
Ž
.
Ž
.
C16H16N2O2 Br2 Mo 524.07 : calcd. C 36.67,
(
) (
)
H 3.08, N 5.35; found C 36.70, H 3.23, N 5.32.
4.11. MoO2Cl2 Me-CYDAB 13
y
Ž
Ž
1..
IR KBr, y cm
: 3057 w, 2958 w, 1485 m,
Ž
.
Ž
. Ž
2
13371 s, 11,220 m, 939, 909, vs, y Mo5O ,
A solution of MoO2Cl2 THF 1.20 g, 3.50
1
Ž
.
Ž
.
768 s, 712 m. H NMR CHCl3-d1, 300 MHz,
mmol in THF 15 ml , was treated with Me-
Ž
..
Ž
.
Ž
.
Ž
.
r.t., d ppm : 8.46 s, 2H , 7.82–7.34 m, 8H ,
CYDAB 0.87 g, 3.50 mmol . The colour of the
solution changed immediately to pinkish red
and the turbid solution was stirred for further 15
min. After concentration to ca. 3 ml. A pinkish
red solid was precipitated by addition of diethyl
ether. The product was washed with diethyl
ether and dried under vacuum. Yield: 90%.
Ž
.
2.51 s, 6H .
(
) ( )
4.9. MoO2Cl2 p-phenyl-DAB
9
Ž
. Ž
2
A solution of MoO2Cl2 THF 0.62 g, 1.81
.
Ž
.
mmol in THF 15 ml , was treated with o-
Ž
.
Ž
.
phenyl-DAB 0.43 g, 1.81 mmol . The colour of
the solution changed immediately to bright yel-
low and the reaction was stirred for further 15
min. The solution was then taken to dryness to
yield a yellowishrred product, which was
washed with diethyl ether and dried under vac-
uum. Yield: 97%.
C16H28N2O2Cl2 Mo 447.26 : calcd. C 42.97,
H 6.31, N 6.26; found C 43.13, H 6.35, N 6.34.
y
Ž
Ž
1..
IR KBr, y cm
: 2926 vs, 2857 s, 1586 m,
1454 m, 1366 m, 1173 m, 1069 w, 939, 910, vs,
1
Ž
.
Ž
y Mo5O , 839 w, 802 w. H NMR CHCl3-d1,
Ž
..
Ž
.
300 MHz, r.t., d ppm : 4.08 TT, 2H , 2.66–
Ž
.
Ž
.
1.32 m, 20H , 2.42 s, 6H .
Ž
.
C16H16N2O2Cl2 Mo 435.16 : calcd. C 44.16,
(
) (
)
H 3.71, N 6.44; found C 44.37, H 3.80, N 6.52.
4.12. MoO2 Br2 Me-CYDAB 14
y
Ž
Ž
1..
IR KBr, y cm
: 3017 m, 2976 m, 1597 s,
Ž
.
Ž
0.89 g,
2
1501 vs, 1371 vs, 1213 s, 1179 s, 945, 912, vs,
A solution of MoO2 Br2 NC-Me
y Mo5O , 856 s, 814 vs, 552 s, 519 s. 1H
Ž
.
.
Ž
.
2.40 mmol in THF 15 ml , was treated with
Ž
Ž
..
.
Ž
.
NMR CHCl3-d1, 300 MHz, r.t., d ppm : 8.28
s, 2H , 7.58–7.31 m, 8H , 2.40 s, 6H .
Me-CYDAB 0.60 g, 2.40 mmol . The greenish
solution changed to a brownish colour and the
reaction was stirred for further 20 min. The
turbid solution was then taken to dryness, the
remaining pale brown solid was washed with
cold CH2Cl2 and diethyl ether and the resulting
white powder was dried under vacuum. Yield:
80%.
Ž
.
Ž
.
Ž
(
) ( )
4.10. MoO2 Br2 p-phenyl-DAB 10
Ž
.
Ž
2
A solution of MoO2 Br2 NC-Me
0.71 g,
.
Ž
.
1.92 mmol in THF 15 ml , was treated with
Ž
.
p-phenyl-DAB 0.45 g, 1.92 mmol . The colour
of the solution changed to orange and the reac-
tion was allowed to stir for an additional 15
min. The solution was then concentrated to ca. 2
ml and a red solid was precipitated by addition
of diethyl ether. The product was filtered off,
washed with diethyl ether and dried under vac-
uum. Yield: 90%.
Ž
.
C16H28N2O2 Br2 Mo 536.16 : calcd. C 35.84,
H 5.2, N 5.22; found C 35.92, H 5.41, N 5.08.
y
Ž
Ž
1..
IR KBr, y cm
: 2930 vs, 2857 s, 1578 w,
1454 m, 1397 w, 1366 m, 1171 m, 937, 909,
1
Ž
.
Ž
y Mo5O , 802 w. H NMR CHCl3-d1, 300
Ž
..
Ž
.
MHz, r.t., d ppm : 4.20 TT, 2H , 2.63–1.31
m, 20H , 2.43 s, 6H .
Ž
.
Ž
.
Ž
.
C16H16N2O2 Br2 Mo 524.07 : calcd. C 36.67,
(
) (
)
H 3.08, N 5.35; found C 36.91, H 3.25, N 5.62.
4.13. MoO2Cl2 CYDAB 15
y
Ž
Ž
1..
IR KBr, y cm
: 3020 m, 2969 m, 2922 m,
Ž
. Ž
2
1501 s, 13770 s, 1211 m, 941, 909, vs,
A solution of MoO2Cl2 THF 1.21 g, 3.52
1
Ž
.
Ž
.
Ž
.
y Mo5O , 812 vs, 519 m. H NMR CHCl3-d1,
mmol in THF 15 ml , was treated with

(
)
158
F.E. Kuhn et al.rJournal of Molecular Catalysis A: Chemical 151 2000 147–160
¨
Ž
.
Ž
.
CYDAB 0.78 g, 3.52 mmol . A white precipi-
tate was immediately formed and the reaction
mixture was stirred for an additional 15 min.
After concentration to ca. 4 ml and addition of
diethyl ether, the solid was filtered off and the
white powder was washed with diethyl ether
and dried in oil pump vacuum. Yield: 98%.
C10H20N2O2Cl2 Mo 367.13 : calcd. C 32.72,
H 5.49, N 7.63; found C 32.90, H 5.71, N 7.52.
: 2980 s, 2936 m, 1478 m,
1454 m, 1389 s, 1366 s, 1238 m, 1186 s, 1038
y
Ž
Ž
1..
IR KBr, y cm
Ž
.
w, 968 s, 949, 912, vs, y Mo5O , 864 s, 758
m. 1H NMR CHCl3-d1, 300 MHz, r.t., d
Ž
Ž
..
Ž
.
Ž
.
ppm : 8.03 s, 2H , 1.58 s, 18H .
Ž
.
C14H24N2O2Cl2 Mo 416.21 : calcd. C 40.11,
H 5.77, N 6.68; found C 39.97, H 5.80, N 6.50.
t
(
) (
)
y
4.16. MoO2 Br2 Bu-DAB 18
Ž
Ž
1..
IR KBr, y cm
: 2941 vs, 2914 vs, 2854 vs,
1448 vs, 1404 vs, 1352 vs, 1076 vs, 931, 912,
vs, y Mo5O , 881 vs, 632 s. 1H NMR
A solution of MoO2 Br2 NC-Me
0.33 g,
Ž
.
Ž
.
Ž
2
Ž
Ž
Ž
..
Ž
.
Ž
Ž
.
CH3CN-d3, 400 MHz, r.t., d ppm : 8.22 s,
2H , 4.17 TT, 2H , 2.13–1.24 m, 20H .
0.90 mmol in THF 15 ml , was treated with
t
.
Ž
.
.
.
Bu-DAB 0.15 g, 0.90 mmol . The colour of
the solution changed immediately to yellowr
green and was stirred for further 15 min. The
solution was then concentrated to ca. 2 ml and
the addition of diethyl ether precipitated a pale
green product. The product was filtered off and
dried under vacuum. Yield: 90%.
(
) ( )
4.14. MoO2 Br2 CYDAB 16
Ž
.
Ž
2
A solution of MoO2 Br2 NC-Me
0.89 g,
.
Ž
.
2.40 mmol in THF 15 ml , was treated with
Ž
.
CYDAB 0.53 g, 2.40 mmol . A yellowr
greenish turbid solution was formed and the
mixture reaction was stirred for 30 min. After
concentration to ca. 3 ml, a pale green solid was
precipitated by addition of diethyl ether. The
solid was filtered off, washed with diethyl ether
and dried under vacuum. Yield: 98%.
Ž
.
C10H20N2O2 Br2 Mo 456.03 : calcd. C 26.34,
H 4.42, N 6.14; found C 25.97, H 4.37, N 5.89.
y
Ž
Ž
1..
IR KBr, y cm
: 2976 vs, 2934 s, 1478 s,
1389 vs, 1388 s, 1236 s, 1192 vs, 970 s, 943,
910, vs, y Mo5O , 866 s, 58 m. 1H NMR
Ž
.
Ž
Ž
..
Ž
CHCl3-d1, 300 MHz, r.t., d ppm : 8.05 s,
.
Ž
.
2H , 1.57 s, 18H .
Ž
.
C12 H24N2O2 Br2 Mo 508.11 : calcd. C 33.0,
H 4.76, N 5.51; found C 33.50, H 4.84, N 5.57.
y
1..
i
Ž
Ž
IR KBr, y cm
: 2942 vs, 2853 vs, 1449 vs,
(
) (
)
4.17. MoO2Cl2 prop-DAB 19
Ž
.
1404 s, 1352 s, 1072 s, 930, 909, vs, y Mo5O ,
880 s. H NMR CH3CN-d3, 400 MHz, r.t., d
ppm : 8.27 s, 2H , 4.28 TT, 2H , 2.22–1.25
m, 20H .
1
Ž
Ž
. Ž
2
A solution of MoO2Cl2 THF 0.58 g, 1.68
Ž
Ž
..
Ž
.
Ž
.
i
.
Ž
Ž
.
mmol in THF 15 ml , was treated with prop-
.
.
DAB 0.24 g, 1.68 mmol . The colourless solu-
tion was stirred for 15 min, and after concentra-
tion and addition of diethyl ether, a white prod-
uct was precipitated. The solid was filtered off,
washed with diethyl ether and dried under vac-
uum. Yield: 95%.
t
(
) ( )
4.15. MoO2Cl2 Bu-DAB 17
Ž
. Ž
2
A solution of MoO2Cl2 THF 1.35 g, 4.20
t
.
Ž
.
mmol in THF 15 ml , was treated with Bu-
Ž
.
Ž
.
DAB 0.71 g, 4.20 mmol . The colourless solu-
tion which formed was stirred for 15 min. The
solution was then concentrated and then a white
solid was precipitated by addition of diethyl
ether. The product was filtered off, washed with
diethyl ether and dried under vacuum. Yield:
90%.
C8H16N2O2Cl2Mo 339.07 : calcd. C 28.34,
H 4.76, N 8.24; found C 28.30, H 4.93, N 8.26.
y
Ž
Ž
1..
IR KBr, y cm
: 2940 vs, 2853 vs, 1447 vs,
Ž
.
1402 s, 1352 m, 1072 s, 932, 909, vs, n Mo5O ,
880 s, 630 w. H NMR CHCl3-d1, 300 MHz,
r.t., d ppm : 8.21 s, 2H , 4.70 m, 2H , 1.63
1
Ž
Ž
..
Ž
.
Ž
.
Ž
.
d, 12H .

(
)
F.E. Kuhn et al.rJournal of Molecular Catalysis A: Chemical 151 2000 147–160
159
¨
i
(
) (
)
Ž
4.18. MoO2 Br2 prop-DAB 20
exposure time of 150r60 s per image oscilla-
tion scan modus from fs0.08 to 3608 with
Dfs18 . A total number of 10,624r39,186
Ž
.
.
Ž
2
A solution of MoO2 Br2 NC-Me
0.90 mmol in THF 15 ml , was treated with
0.33 g,
.
.
Ž
reflections were collected. 216r850 systematic
absent reflections 237r1519 negative intensities
together with 0r6156 very weak intensities with
Q)28.008 were rejected from the original data
set. After merging a sum of 1473r4448, inde-
pendent reflections remained and were used for
all calculations. Data were corrected for Lorentz
and polarization effects. Corrections for absorp-
tion and decay effects were applied with the
i
Ž
.
prop-DAB 0.13 g, 0.90 mmol . The greenish
solution changed immediately to a light yellow
and was stirred for 15 min. The solution was
then concentrated to ca. 2 ml and the addition of
diethyl ether precipitated a yellow solid, which
was washed with diethyl ether and dried under
vacuum. Yield: 95%.
Ž
.
C8H16N2O2Br2Mo 427.98 : calcd. C 22.45,
H 3.77, N 6.55; found C 22.35, H 3.81, N 6.47.
w
x
program DECAY 44 . The unit cell parameters
were obtained by full-matrix least-squares re-
finements of 1882r1976 reflections with the
program CELL 44 . All ‘‘heavy atoms’’ of the
asymmetric unit were anisotropically refined.
Hydrogen atoms were found in the difference
maps and refined with individual isotropic ther-
mal displacement parameters. For compound
5PNCCH3 the hydrogen atoms of the disor-
dered solvent acetonitrile were placed in calcu-
y
Ž
Ž
1..
IR KBr, y cm
: 2982 s, 2930 s, 1458 s,
1406 vs, 1368 s, 1329 s, 1149 s, 1127 vs, 941,
907, vs, y Mo5O , 831 s. H NMR CHCl3-d1,
300 MHz, r.t., d ppm : 8.20 s, 2H , 4.74 m,
2H , 1.68 d, 12H .
1
Ž
.
Ž
w
x
Ž
..
Ž
.
Ž
.
Ž
.
4.19. X-ray crystallography
Suitable single crystals for the X-ray diffrac-
tion studies were grown by standard techniques
from saturated solutions. Both structures were
solved by a combination of direct methods,
difference-Fourier syntheses and full-matrix
least-squares refinements. Neutral atom scatter-
ing factors for all atoms and anomalous disper-
sion corrections for the non-hydrogen atoms
were taken from International Tables for X-Ray
lated positions. Full-matrix least-squares refine-
ments were carried out by minimizing Sw Fo
2
Ž
2
yFc2
with SHELXL-93 weighting scheme
.
and stopped at shiftrerr-0.006r0.001.
Ž
Crystallographic data excluding structure
.
factors for the structures reported in this paper
have been deposited with the Cambridge
Crystallographic data center as supplementary
publication no. CCDC-103338 and CCDC-
103339. Copies of the data can be obtained
free of charge on application to CCDC, 12
Union Road, Cambridge CB2 1EZ, UK Fax:
int. code q44 1223 336-033; E-mail: deposit
w
x
Crystallography 39 . All calculations were per-
formed on a DEC 3000 AXP workstation with
w
x
the STRUX-V 40 system, including the pro-
grams PLATON-92 41 , PLUTON-92 41 , Sir-
92 42 , and SHELXL-93 43 .
w
x
w x
w
w
x
w x
Ž
.
x
@ccd.cam.ac.uk .
4.20. Data collection, structure solution and
(
)
refinement for the complexes 3r5PNCCH3
4.21. Catalytic reactions with compounds 1–20
Preliminary examinations and data collec-
tions were carried out on an imaging plate
as catalysts
Ž
.
Ž .
800 mg 7.3 mmol cis-cyclooctene, 800 mg
n-dibutylether internal standard , 1 mol% 73
diffraction system IPDS; STOE&CIE equipped
Ž
Ž . Ž
with a rotating anode NONIUS FR591; 50r50
kV; 80r80 mA; 4.0r4.0 kW and graphite
.
.
Ž
.
mmol 1–20 as catalyst , and 2 ml 5.5 M
tert-butylhydroperoxide in decane were added
to a thermostated reaction vessel and stirred for
4 h at 558C.
monochromated MoKa radiation. Data collec-
tions were performed at 193 K within the Q-
range of 2.738 to 24.81r3.688 to 30.388 with an

(
)
160
F.E. Kuhn et al.rJournal of Molecular Catalysis A: Chemical 151 2000 147–160
¨
w
w
x
18 P. Chaumette, H. Mimoun, L. Saussine, J. Fischer, A.
The course of the reaction was monitored by
Ž
.
Mitschler, J. Organomet. Chem. 250 1983 291.
quantitative GC-analysis. Samples were taken
every 30 min, diluted with chloroform, and
chilled in an icebath. For the destruction of
hydroperoxide and removal of water, a catalytic
amount of manganese dioxide and magnesium
sulfate was added. After the gas evolution
ceased, the resulting slurry was filtered over a
filter equipped Pasteur pipette and the filtrate
injected in the GC column.
x
19 F.E. Kuhn, E. Herdtweck, J.J. Haider, W.A. Herrmann, I.S.
¨
Gonc¸alves, A.D. Lopes, C.C. Romao, J. Organomet. Chem.
˜
Ž
.
583 1999 3.
w
w
w
w
w
x
20 K. Dreisch, C. Andersson, C. Stalhandske, Polyhedron 12
Ž
.
1993 303.
x
21 H. Arzoumanian, G. Agrifoglio, H. Krentzien, M. Capparelli,
Ž
.
J. Chem. Soc., Chem. Commun. 1995 655.
x
22 D.M. Baird, F.L. Yang, D.J. Kavanaugh, G. Finess, K.R.
Ž
.
Dunbar, Polyhedron 15 1996 2597.
x
23 M. Minelli, J.H. Enemark, R.T.C. Brownlee, M.J. O’Connor,
Ž
.
A.G. Wedd, Coord. Chem. Rev. 68 1985 169.
x
24 S. Berger, S. Braun, H.C. Kalinowski, NMR-Spektroskopie
von Nichtmetallen, Vol. 1, Georg Thieme Verlag, Stuttgart,
1992.
The conversion of cyclooctene and formation
of cyclooctene oxide was calculated from a
2
Ž
.
calibration curve r s0.999 recorded prior to
w
w
w
x
25 W.A. Herrmann, F.E. Kuhn, P.W. Roesky, J. Organomet.
¨
Ž
.
Chem 485 1995 243.
the reaction course.
x
26 W.A. Herrmann, F.E. Kuhn, M.U. Rauch, J.D.G. Correia, G.
¨
Ž
.
Artus, Inorg. Chem. 34 1995 2914.
x
27 C.C. Romao, F.E. Kuhn, W.A. Herrmann, Chem. Rev. 97
˜
¨
Acknowledgements
Ž
.
1997 3197.
w
w
x
Ž
.
28 F.A. Cotton, R.M. Wing, Inorg. Chem. 4 1965 867.
x
This work was supported by PRAXIS XXI,
29 F.H. Allen, O. Kennard, 3D search and research using the
Ž
.
Project 2r2.1rQUIr419r94 . The authors
Cambridge structural database, Chemical Design Automation
Ž
.
Ž
News 8 1993 1 and 31.
want to thank the DAAD and CRUP INIDA
w
w
x
x
30 W.A. Herrmann et al., personal communication.
31 F.E. Kuhn, A.D. Lopes, I.S. Gonc¸alves, A.M. Santos, E.
.
and Acc¸oes Integradas Programme for gener-
˜
¨
ous support. A.D.L. thanks PRAXIS XXI,
A.M.S. thanks the Bayerische Forschungss-
tiftung for a grant. The authors greatly acknowl-
edge Prof. W.A. Herrmann for generous sup-
port.
Herdtweck, C.C. Romao, submitted for publication.
˜
w
w
w
x
32 H. Iida, Y. Yuasa, C. Kibayashi, Y. Iitaka, J. Chem. Soc.
Ž
.
Dalton Trans. 1956 640.
x
33 H.F. Holtzclaw, J.P. Coleman Jr., R.M. Alire, J. Am. Chem.
Ž
.
Soc. 80 1958 1100.
x
34 P. Kiprof, W.A. Herrmann, F.E. Kuhn, W. Scherer, M.
¨
Kleine, M. Elison, K. Rypdal, H.V. Volden, S. Gundersen,
Ž
.
A. Haaland, Bull. Soc. Chim. Fr. 129 1992 655.
w
x
35 G. van Koten, K. Vrieze, Advances in Organometallic Chem-
References
istry, Vol. 21, Academic Press, New York, 1982.
w x
Ž
.
1
w x
2
R.A. Sheldon, J. Mol. Catal. 7 1980 107.
w
w
w
x
Ž
.
36 R. Colton, I.B. Tomkins, Aust. J. Chem. 18 1965 447.
W.R. Thiel, T. Priermeier, Angew. Chem., Int. Ed. Engl. 34
x
Ž
.
37 H.L. Kraus, W. Huber, Chem. Ber. 94 1961 2864.
Ž
.
1995 1737.
x
38 W.M. Carmichael, D.A. Edwards, G.W.A. Fowles, P.R. Mar-
w x
Ž .
3
W.R. Thiel, J. Mol. Catal. 117 1997 449.
Ž
.
shall, Inorg. Chim. Acta 1 1967 93.
w x
Ž
.
4
W.R. Thiel, J. Eppinger, Chem. Eur. J. 3 1997 696.
F.P. Ballisteri, G.A. Tomaselli, R.M. Toscano, V. Conte, F.
di Furia, J. Mol. Catal. 89 1994 295.
S. Campestrini, F. di Furia, J. Mol. Catal. 79 1993 13.
N. Indictor, W.F. Brill, J. Org. Chem. 29 1965 2075.
U.S. Patent No. 3.350.422, 1967.
U.S. Patent No. 3.351.635, 1967.
w
x
39 International Tables for Crystallography, Vol. C, Tables
w x
5
Ž
.
Ž
.
6.1.1.4 pp. 500–502 , 4.2.6.8 pp. 219–222 , and 4.2.4.2
pp. 193–199 , in: A.J.C. Wilson Ed. , Kluwer Academic
Publishers, Dordrecht, The Netherlands, 1992.
Ž
.
Ž
.
Ž
.
w x
Ž .
.
6
w x
Ž
7
w
w
x
40 G.R.J. Artus, W. Scherer, T. Priermeier, E. Herdtweck, S T
R U X–V, A Program System to Handle X-ray Data, TU
w x
8
w x
9
Munchen, Germany, 1997.
¨
w
w
w
w
x
10 U.S. Patent No. 3.507.809, 1970.
x
41 A.L. Spek, PLATON-92–PLUTON-92, an integrated tool for
x
11 U.S. Patent No. 3.625.891, 1971.
the analysis of the results of a single crystal structure deter-
mination, Acta Crystallogr., Sect. A 46 1990 C34.
x
Ž
.
12 D.D. Agarwal, Inorg. Chem. 25 1986 65.
Ž
.
x
13 E.P. Talsi, K.V. Shalyaev, K.I. Zamaraev, J. Mol. Catal. 83
w
w
x
42 G. Cascarano, C. Giacovazzo, A. Guagliardi, M.C. Burla, G.
Ž
.
1993 347.
Polidori, M. Camalli, SIR-92, University Bari, Italy, 1992.
w
w
w
w
x
Ž
.
14 R.H. Holm, Chem. Rev. 87 1987 1401.
x
43 G.M. Sheldrick, SHELXL-93, in: G.M. Sheldrick, C. Kruger,
¨
x
Ž
.
15 L.R. Florian, E.G. Corey, Inorg. Chem. 7 1968 722.
Ž
.
R. Goddard Eds. , Crystallographic Computing 3, Oxford
Univ. Press, England, 1993, pp. 175–189.
x
Ž
.
16 B. Spivack, Z. Dori, Coord. Chem. Rev. 17 1975 99.
x
17 F. Trifiro, P. Forzatti, S. Preite, I. Pasquon, J. Less-Common
w
x
44 IPDS Operating System Version 2.8 STOE&CIE., Darmstadt,
Deutschland, 1997.
Ž
.
Met. 36 1974 319.