Dioxomolybdenum(VI) complexes with pyrazole based aryloxide ligands: Synthesis, characterization and application in epoxidation of olefins

Article abstract of DOI:10.1021/ic300648p

Synthesis, characterization, and epoxidation chemistry of four new dioxomolybdenum(VI) complexes [MoO₂(L)₂] (1-4) with aryloxide-pyrazole ligands L = L1-L4 is described. Catalysts 1-4 are air and moisture stable and easy to synthesize in only three steps in good yields. All four complexes are coordinated by the two bidentate ligands in an asymmetric fashion with one phenoxide and one pyrazole being trans to oxo atoms, respectively. This is in contrast to the structure found for the related aryloxide-oxazoline coordinated Mo(VI) dioxo complex 5. This was confirmed by the determination of the molecular structures of complexes 1-3 by X-ray diffraction analyses. Compounds 1-4 show high catalytic activities in the epoxidation of various olefins. Cyclooctene (S1) is converted to its epoxide with high activity, whereas the epoxidation of styrene (S2) is unselective. Internal olefins (S3 and S4) are also acceptable substrates, as well as the very challenging olefin 1-octene (S5). Catalyst loading can be reduced to 0.02 mol % and the catalyst can be recycled up to ten times without significant loss of activity. Supportive DFT calculations have been carried out in order to obtain deeper insights into the electronic situation around the Mo atom.

Full text of DOI:10.1021/ic300648p

Article
pubs.acs.org/IC
Dioxomolybdenum(VI) Complexes with Pyrazole Based Aryloxide
Ligands: Synthesis, Characterization and Application in Epoxidation
of Olefins
†
†
†
†
†
†
‡
Jo
̈
rg A. Schachner, Pedro Traar, Chris Sala, Michaela Melcher, Bastian N. Harum, Alexander F. Sax,
†
,†
Manuel Volpe, Ferdinand Belaj, and Nadia C. Mo
̈
sch-Zanetti*
†
‡
Institute of Chemistry, Karl-Franzens-University Graz, Stremayrgasse 16, 8010 Graz, Austria
Institute of Chemistry/Physical Chemistry, Karl-Franzens-University Graz, Heinrichstraße 28, 8010 Graz, Austria
*
S Supporting Information
ABSTRACT: Synthesis, characterization, and epoxidation chemistry of four new
dioxomolybdenum(VI) complexes [MoO (L) ] (1−4) with aryloxide-pyrazole ligands
2
2
L = L1−L4 is described. Catalysts 1−4 are air and moisture stable and easy to
synthesize in only three steps in good yields. All four complexes are coordinated by the
two bidentate ligands in an asymmetric fashion with one phenoxide and one pyrazole
being trans to oxo atoms, respectively. This is in contrast to the structure found for the
related aryloxide-oxazoline coordinated Mo(VI) dioxo complex 5. This was confirmed
by the determination of the molecular structures of complexes 1−3 by X-ray diffraction
analyses. Compounds 1−4 show high catalytic activities in the epoxidation of various
olefins. Cyclooctene (S1) is converted to its epoxide with high activity, whereas the
epoxidation of styrene (S2) is unselective. Internal olefins (S3 and S4) are also
acceptable substrates, as well as the very challenging olefin 1-octene (S5). Catalyst loading can be reduced to 0.02 mol % and the
catalyst can be recycled up to ten times without significant loss of activity. Supportive DFT calculations have been carried out in
order to obtain deeper insights into the electronic situation around the Mo atom.
INTRODUCTION
prevent the acidic ring-opening but usually makes the adduct
complexes more sensitive toward moisture rendering their
■
Molybdenum catalyzed oxygen atom transfer (OAT) chemistry
belongs to the very important chemical reaction both in
biological systems as well as on industrial scale. In biology,
molybdenum is found in a large family of enzymes that catalyze
important OAT reactions in the living cell. Mechanistic details
still remain unclear despite the determination of many crystal
25−27
handling more challenging.
Thus, it is still worthwhile to
develop further homogeneous catalysts for epoxidation of
olefins. A possible approach is the preparation of rhenium
complexes in lower oxidation states, such as oxorhenium(V)
compounds, or oxomolybdenum(VI) compounds as they tend
to be more stable toward protic conditions. In this endeavor,
we have developed and investigated as epoxidation catalysts
1
−4
structures. For this reason, several research groups, including
ours, developed functional and structural model molybdenum
complexes and investigated their OAT reactivity for a better
28−30
several oxorhenium(V) complexes.
With one of them we
5
−10
understanding of this important reaction.
reactions include the epoxidation of propylene with an
estimated scale of 5 million metric tons annually,
was patented more than 40 years ago and is now known as the
Halcon/ARCO process.
contains mainly molybdenum in the oxidation state VI, formed
in situ from the catalyst precursor [Mo(CO) ] and alkyl-
hydroperoxides as oxygen source. Subsequently, many homo-
Industrial OAT
were able to get an insight into the mechanism and found that
the rhenium(V) precursor is stepwise oxidized to rhenium(VII)
prior to the catalytic reaction which is again sensitive toward
11−13
which
30
decomposition. However, we have recently prepared
1
4,15
The active heterogeneous catalyst
oxorhenium(V) complexes equipped with aryloxide-pyrazole
29
ligands and investigated their epoxidation chemistry.
6
Although we found them to be only moderately active catalysts
in the epoxidation of cyclooctene, with yields ranging between
44 to 64% cyclooctane epoxide, their ease of preparation and
handling prompted us to investigate the ligand system in
molybdenum based epoxidation catalysis.
Here, we report air stable and easy to synthesize
dioxomolybdenum(VI) complexes 1−4 equipped with the
geneous molybdenum catalysts for epoxidation of olefins were
1
6−22
developed.
However, significantly more attention received
rhenium catalyzed homogeneous olefin epoxidation triggered
by the discovery of the easy entry into the chemistry of the
highly active methyltrioxorhenium (MTO) by Herrmann and
2
3,24
co-workers.
Nevertheless the high oxidation state rhenium
catalysts are hampered by their high Lewis acidity which results
25
in destructive ring-opening of the formed epoxide. The
addition of various mono- and bidentate Lewis bases can
Received: March 28, 2012
Published: June 29, 2012
©
2012 American Chemical Society
7642
dx.doi.org/10.1021/ic300648p | Inorg. Chem. 2012, 51, 7642−7649

Inorganic Chemistry
Article
aryloxide-pyrazole ligand family together with their interesting
catalytic behavior in epoxidation of various epoxides.
(vide infra). Because of their low solubility in common organic
1
solvents all H NMR experiments were performed in d -
4
methanol at 55 °C. For the same reason we could not obtain
13
RESULTS AND DISCUSSION
good quality C NMR spectra due to the crystallization of the
sample inside the NMR tube over the time of the experiment.
Because of the bidentate nature of ligands L1-L4 a total of
three geometric isomers are possible for bis substituted cis
dioxo molybdenum(VI) complexes, depending on the coordi-
nating atoms in trans position to the oxo groups: both N-
donors trans to the terminal oxo ligands (N,N-isomer), both
phenoxy groups trans to the terminal oxo ligands (O,O-isomer)
or one N-donor and one phenoxy group trans to the terminal
oxo ligands (N,O-isomer). The occurrence of two sets of ligand
resonances in the proton spectrum evidence the exclusive
formation of the N,O-isomer. Complexes 1−4 should be
compared to the related aryloxide-oxazoline complexes
■
Complex Synthesis. The pyrazole ligands were synthesized
according to published literature procedures.
2
0,29,31
Synthesis
involves a two step procedure by treating the respective methyl
aryl ketone with sodium hydride and ethyl formate to give the
aryl 3-oxopropanal intermediate product. The subsequent ring
closure procedure of the aldehyde intermediates with
methylhydrazine in refluxing ethanol yields the desired ligands
HL1−HL4 (Figure 1).
[
MoO (hoz) ] (Figure 4) featuring symmetric N,N-coordina-
2 2
tion. Several structurally related oxazoline containing Mo
complexes have been previously published and tested in olefin
1
8,19,21,22,33
epoxidation (vide infra).
For complexes 1−3 the molecular structures were
determined by single-crystal X-ray diffraction analysis. A
molecular view of 1 is given in Figure 2. Molecular views of
Figure 1. Aryloxide ligands used in this study with a pyrazol-3-yl
bonded moiety.
The Mo complexes [MoO (L) ] (L = L1−L4) were
2
2
synthesized by addition of the respective ligands HL1−HL4
to [MoO (acac) ] dissolved in hot MeOH (Scheme 1). After
2
2
Scheme 1. Synthesis of Pyrazole Containing Mo(VI)
Complexes 1−4
Figure 2. ORTEP plot of 1 at 50% probability with numbering
scheme. H atoms drawn with arbitrary radii.
refluxing for 4−6 h complexes 1−4 precipitated and were
isolated by filtration. After washing with cold MeOH the
complexes were analytically pure in 65−76% isolated yield. To
the best of our knowledge, complexes 1−4 represent the first
examples of aryloxide-pyrazol-3-yl containing
dioxomolybdenum(VI) complexes. A related complex, an
isomeric pyrazol-1-yl-ligated complex, was published by Sorrell
2 and 3 as well as selected bond lengths and angles and other
crystallographic information of all three complexes can be
found in the Supporting Information. All three complexes 1−3
display distorted octahedral coordination geometries around
the Mo atom in the N,O-isomeric form consistent to the
structure in solution.
32
and co-workers, but was not tested in olefin epoxidation.
Complexes 1−4 were isolated as yellow to orange powders.
They are stable for month in air in the solid state, and in
chloroform solution for approximately two weeks. They are
insoluble in apolar (hexanes etc.) as well as medium polar
As displayed in Figure 2, the nitrogen atom (N32) of one
pyrazole moiety trans to terminal oxygen atom O1 is only
weakly coordinated in the solid state, with a very long Mo1−
N32 distance of 2.4060(17) Å. The second pyrazole moiety
from the other ligand is bonded to the Mo atom with a Mo1−
N12 distance of 2.1980(15) Å. Trans to the second terminal
oxo ligand (O2) is a phenoxy moiety of one ligand (Mo1−O21
2.0178(13) Å) which is slightly longer than the corresponding
molybdenum phenoxide distance trans to pyrazole (Mo1−O41
1.9302(11) Å). This asymmetric bonding situation around the
Mo center is not reflected in the bond distances of the two
solvents (Et O, toluene) and only sparingly soluble in strong
2
polar solvents (MeOH, acetonitrile, CHCl ). X-ray quality
3
crystals of 1−3 were obtained by slow evaporation from
1
solutions in MeOH at 8 °C. The H NMR spectra of complexes
1
−4 display two sets of proton signals for two inequivalent
ligands. This is due to an unusual asymmetric coordination of
the ligands, which was also revealed by X-ray crystallography
7
643
dx.doi.org/10.1021/ic300648p | Inorg. Chem. 2012, 51, 7642−7649

Inorganic Chemistry
Article
terminal oxo ligands to the Mo center in the solid state. They
are virtually the same as Mo−O1 is 1.7084(15) Å and Mo−O2
is 1.7084(14) Å and fall within the expected range of other
oxazoline Mo(VI) complexes [MoO (hoz*) ] (Figure 4) which
reached 41% conversion after 22 h in toluene under conditions
2
2
34
published dioxomolybdenum complexes. A similar bonding
situation can also be observed in complexes 2 and 3 (see
Supporting Information). As mentioned above the related
aryloxide-oxazoline Mo(VI) complex 5 was found to exhibit the
19
symmetric N,N-isomeic structure. Similar to 1−3 the Mo
atom in 5 is coordinated in a distorted octahedral fashion by
two terminal oxo ligands and two bidentate aryloxide-oxazoline
ligands. Both terminal oxo ligands in 5 are at a distance of
1
.7054(13) Å to the Mo atom and therefore reside at a very
similar range as in complexes 1−3. Trans to both oxo ligands of
the nitrogen atoms of the oxazoline moieties are located,
5
resulting in the symmetric N,N-isomer. The nitrogen atoms are
at a distance of 2.3024(14) Å to the Mo atom, which lies right
between the distances for the two asymmetrically coordinated
pyrazole ligands in 1 (vide supra).
Catalytic Epoxidation. Complexes 1−4 were screened in
the catalytic epoxidation of five different olefins using tert-butyl
hydrogen peroxide (TBHP) as oxygen source (see Figure 3).
Figure 4. Selected literature examples of known Mo(VI) epoxidation
catalysts.
21
as described in Table 1, entry 5. When the peroxo complex
[
(
MoO(O )(hoz*) ] was used, the yield increased to 83%
2 2
Table 1, entry 6).
Figure 3. Substrates S1−S5 used in epoxidation experiments with
complexes 1−4.
Under standard conditions experiments were performed in 3
mL of CHCl with 2 mol % of catalyst and 3 equivalents of
3
TBHP at 25 or 55 °C. Different conditions for 1 were also
tested with substrate S5 to allow for better comparison with
previously published systems (vide infra). The five olefin
substrates used were cyclooctene S1, styrene S2, 3-buten-1-
ylbenzene S3, (E,Z)-1-methyl-2-(3-penten-1-yl)benzene S4 and
Figure 5. Epoxidation of styrene S2 (⧫) with complex 2 under
formation of epoxide (■) and side product 6 (▲).
1
-octene S5.
At 55 °C, the epoxidation of cyclooctene S1 was complete in
less than two minutes with complexes 1−4. At 25 °C the
reaction reached completion (based on the detection limit of
the GC-MS for S1) in less than 60 min (Table 1, Entries 1−4).
This is in contrast to previously published, structurally related
Bagherzadeh and co-workers published a comprehensive
reactivity study on the parent oxazoline complex
[MoO (hoz) ] 5 where they reported a very high conversion
2
2
Table 1. Epoxidation of Cyclooctene S1
a
entry
catalyst
mol %
equiv Ox.
solv.
T [°C]
t [min]
yield [%]
ref
1
2
3
4
5
6
7
8
9
1
2
3
4
2
3
CHCl3
CHCl3
CHCl3
CHCl3
toluene
toluene
C H Cl
25
25
25
25
25
25
80
80
55
55
80
40
30
>99
>99
>99
>99
41
this work
this work
this work
this work
21
2
3
2
3
30
2
3
40
[MoO (hoz*) ]
[MoO(O )(hoz*) ]
[MoO (hoz) ] (5)
2.5
2.5
0.02
0.02
1
1.5
1.5
1
1320
1320
60
2
2
83
21
2
2
96
18
2
2
2
4
2
2
1
1
C H Cl
60
44
this work
17
2
4
[MoO Cl (bipy*)]
[MoO Cl(Cp′)]
1.5
2
neat
5
>99
>99
>99
2
2
10
1
1
neat
240
45
16
2
1
[MoO (O,N,O)]
1
2
C H Cl
35
2
2
4
2
a
See reference section.
7
644
dx.doi.org/10.1021/ic300648p | Inorg. Chem. 2012, 51, 7642−7649

Inorganic Chemistry
Article
Table 2. Epoxidation of Styrene S2
a
entry
cat.
mol %
equiv Ox.
solv.
T [°C]
t [h]
yield [%]
ref
1
2
3
4
5
6
7
8
1
2
3
4
2
3
CHCl3
CHCl3
CHCl3
CHCl3
toluene
toluene
C H Cl
55
55
55
55
35
35
80
55
5
5
26
this work
this work
this work
this work
33
2
3
22
2
3
5
28
2
3
5
18
[MoO (acac) ]
2.5
2.5
0.02
1
1.5
1.5
1
18
18
5
2.8
12.8
45
2
2
[MoO (hoz*) ]
[MoO (hoz) ] (5)
33
2
2
18
2
2
2
4
2
[MoO Cl(Cp′)]
2
neat
24
84
16
2
a
See reference section.
Scheme 2. Formation of Side Products Observed for Catalysts 1−4 during Epoxidation of Styrene S2
Table 3. Comparison of Epoxidation Yields of S3, (E,Z)-S4, and S5
a
entry
catalyst
substrate
equiv Ox
solv.
T [°C]
t [h]
24
yield [%]
ref
1
2
3
4
5
6
7
8
9
1
2
3
4
1
2
3
4
5
1
S3
3
3
3
3
3
3
3
3
1
1
2
2
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
CHCl3
C H Cl
55
55
55
55
25
25
25
25
80
80
50
50
86
76
84
83
this work
this work
this work
this work
this work
this work
this work
this work
18
S3
24
24
S3
S3
24
b
c
(E,Z)-S4
(E,Z)-S4
(E,Z)-S4
(E,Z)-S4
S5
1(3)
99(85)
b
b
b
c
0.75(3)
2(3)
99(89)
c
98(74)
c
1(3)
98(88)
3
3
82
2
4
2
2
d
10
11
12
S5
C H Cl
0
this work
16
2
4
[MoO Cl(Cp′)]
S5
neat
24
24
15
25
2
1
S5
neat
this work
a
b
c
See Reference section. Numbers in brackets correspond to yield of second diastereomer. Number in brackets corresponds to time for slower
d
diastereomer. After 24 h 8% conversion was observed.
of cyclooctene S1 of 96% (Table 1, entry 7) within 1 h under
optimized conditions (0.02 mol % 5, 1 equiv TBHP, 80 °C,
example, a tridentate O,N,O-Schiff base complex (Table 1,
entry 11).
35
1
8
C H Cl ). Under these conditions 1 only yields 44% of
With the more challenging substrate styrene S2 the
selectivity of complexes 1−4 dropped significantly and is
inferior to some previously published systems. At 55 °C and 5 h
reaction time, the yield for styrene epoxide varied between 18
and 28% for complexes 1−4 (Table 2, entries 1−4). The low
yield is caused by concomitant side reactions. One side product
could be identified by GC-MS as benzaldehyd (<5%), the other
one as 2-(tert-butoxy)-2-phenylethanol 6 (vide infra).
2
4
2
epoxide after 1 h, and complete epoxidation is reached after 19
h (Table 1, entry 8). A similar activity to 1−4 was recently
published for a series of bipyridine containing Mo(VI)
complexes, the most active one reaching full conversion
under neat conditions at 55 °C in less than 5 min (Table 1,
1
7
entry 9). A series of half-sandwich dioxo Mo(VI) complexes
containing various cyclopentadienyl ligands (Cp) published by
Romao
̃
, Ku
̈
hn, and co-workers were all very active in
The problematic further reactivity of styrene oxide has also
epoxidation reactions (Table 1, entry 10), even surpassing
the activity of MTO in one case. Complexes 1−4 as well as
the oxazoline complexes
laboratory conditions, whereas the bipyridine and Cp
complexes are synthesized under exclusion of air and moisture.
Furthermore, not only mono- and bidentate active ligand
systems were published, but also tridentate systems, for
been observed in other published systems, with the main side
16
18,33
product being identified as benzaldehyde.
In our case, we
1
8,21,33
can be obtained under standard
also observe a small amount of benzaldehyde by GC-MS
(<5%). The other cleavage product from the styrene epoxide
36
oxidation, formaldehyde, was not observed. By in situ IR
spectroscopy we could observe the formation of CO , which we
2
attribute to a full oxidation of the initially formed formaldehyde.
7
645
dx.doi.org/10.1021/ic300648p | Inorg. Chem. 2012, 51, 7642−7649

Inorganic Chemistry
Article
The major side product (20−40%) was identified as tert-butyl
benzoate 6 (Scheme 2), the product of tert-butanol (the end
product from TBHP after oxygen transfer) and benzoic acid.
Table 4. Epoxidation Results of Catalyst 1 with Substrate S1
at Different Catalyst Loadings
a
37
b
TOF [h−1
]
c
mol %
2
yield [%]
t [h]
TON
Benzoic acid is the final oxidation product from either styrene
epoxide or styrene itself in a reaction series that starts with a
C−C bond cleavage.
For S3 good activities were observed for all four complexes,
with conversions of 76−86% at 55 °C after 24 h (Table 3,
entries 1−4) with high selectivity, since no side products were
detected by GC-MS. We were also interested to test an
unstrained internal olefin and screened the isomeric mixture of
>99
>99
>99
0.03
0.5
4
50
500
1666
3125
5000
0
0
.2
.02
5000
a
b
Conditions: CHCl , 55 °C, 3 equiv TBHP. mol product/mol
3
c
catalyst. TON/time; at the time interval of highest conversion.
epoxidation of cyclooctene S1 with catalyst 1 is slower than
with 5 and does not reach completion within 5 h. However
(E,Z)-S4. In this case complexes 1−4 also displayed good
activity, converting all of (E,Z)-S4 within 3 h at 55 °C (Table 3,
Entries 5−8). For substrate (E,Z)-S4 the formation of two
different products corresponding to both diastereomers was
observed (Figure 6). As displayed in the reaction profile in
under our optimal conditions (CHCl , 55 °C, 3 equiv TBHP)
3
S1 gets fully converted by 1 within 4 h.
The catalyst stability of 1 was tested by ten repeated
additions of aliquots of S1 and TBHP (2 mol % 1, CHCl , 55
3
°
C, 1.5 equiv TBHP). In contrast to our above-described
standard conditions, the repeated addition experiments were
performed with 1.5 equiv of TBHP as initial tests with repeated
additions of 3 equiv of TBHP increased the amount of side
products, mainly the ring-opened diole. This was probably due
to the increasing excess of TBHP that remained in the reaction
after each new addition of substrate and TBHP. Therefore the
amount of TBHP was reduced to 1.5 equiv of TBHP which
resulted in higher selectivity. To counter the concomitant loss
in catalyst activity with 1.5 equiv TBHP the temperature was
raised from 25 to 55 °C in these experiments. Under these
conditions conversion of cyclooctene S1 took ∼90 min in the
first six cycles and ∼120 min for the last 4 cycles to reach
complete conversion. Cycles 9 and 10 were done after the
reaction vial had been standing over the weekend at room
temperature with no apparent loss in activity compared to
cycles 7 and 8, which further demonstrates the stability of 1. It
seems that catalysts 1−4 have their optimal activity at 55 °C
and 3 equiv of TBHP. Reducing the amount of TBHP to 1.5
equiv significantly increases the reaction times for all substrates
Figure 6. Epoxidation of (E,Z)-S4 with complex 1.
Figure 6, one isomer of S4 reacts faster (■) than the other
isomer (⧫). Although not proven, we believe, from a steric
point of view it is feasible that the Z-isomer of S4 can approach
the active Mo catalyst more easily than the E-isomer and is
therefore converted faster. Comparing both isomers of S4 to S3
shows the internal, more electron-rich, olefin S4 to be more
reactive.
under investigation. It also seems that CHCl is the optimal
3
solvent for epoxidation, also compared to the higher boiling
solvent C H Cl . We also investigated the use of H O as
2
4
2
2
2
oxidant in different solvents but all attempts were unsuccessful.
Theoretical Calculations. For a better understanding of
the influence of the two found isomeric forms in complexes 1
and 5 on the catalytic properties the different bonding
situations of the two isolated and two hypothetical isomers
shown in Figure 7 were investigated by DFT calculations
employing the popular hybrid density functional B3-LYP.
Aryloxide-pyrazole ligands, as demonstrated in this manu-
script, coordinate in an asymmetric way to form the N,O-
isomer (Figure 7, 1(N,O)) whereas the oxazoline derivative was
Lastly, we chose the very challenging substrate 1-octene S5,
since terminal olefins are good model substrates for the
1
1
important industrial epoxidation of propylene. Epoxidation of
S5 with catalyst 1 leads to 25% selective conversion under neat
conditions at 50 °C (Table 3, entry 12). In comparison,
complex [MoO Cl(Cp′)] reaches 15% yield of epoxide after 24
2
1
6
h in the same neat conditions (Table 3, entry 11).
Significantly better is complex [MoO (hoz) ] (5) which
2
2
converts 82% of S5 to the epoxide within 3 h in C H Cl at
found to form the symmetric N,N-isomer (Figure 7,
2
4
2
18
18,19
8
0 °C (Table 3, entry 9). Complex 1 shows no conversion
5(N,N)).
The results of these calculations showed that
under the latter conditions (Table 3, entry 10).
the hypothetical symmetric complex 1(N,N) and the isolated
complex 5(N,N) feature two almost identical oxo groups in
trans position to the nitrogen atom whereas the asymmetric
isolated complex 1(N,O) and the hypothetical complex 5(N,O)
exhibit two electronically different oxo groups.
After the initial round of screening and testing it was
apparent that out of the four complexes 1−4 catalyst 1 was
displaying the best performance. Therefore we used catalyst 1
to further explore the catalytic performance by decreasing the
catalyst loading from 2 mol % to 0.02 mol % (Table 4) in the
epoxidation of model substrate cyclooctene S1. Experiments
were performed at 55 °C in CHCl or 80 °C in C H Cl to
Table 5 summarizes the calculated Shared Electron Numbers
(SENs) for molybdenum-oxo functionalities O and O and the
1
2
coordinating aryloxide ligands L as well as the natural charges
of all three of these coordinating moieties around Mo in both
systems 1 and 5, respectively. The calculations reveal that the
synthetically accessed oxazoline based complex 5(N,N) features
two electronically equivalent oxo groups whereas its asym-
3
2
4
2
study the influence of temperature. Interestingly, we found that
our catalyst 1 does not perform as well as under the optimal
1
8
conditions for [MoO (hoz) ] (5). At 80 °C in C H Cl with
2
2
2
4
2
0
.02 mol % catalyst loading and 1 equiv of TBHP the
7
646
dx.doi.org/10.1021/ic300648p | Inorg. Chem. 2012, 51, 7642−7649

Inorganic Chemistry
Article
Figure 7. N,N- and N,O- isomers of cis-dioxo molybdenum(VI) complexes 1 and 5. Hypothetical isomer labels are in italic. Terminal oxygen ligands
for isolated complexes of 1 and 5 are labeled (see Table 5).
Table 5. Electronic Properties of N,N- and N,O-Isomers of cis-Dioxomolybdenum(VI) Complexes 1 and 5 at the B3-LYP def2-
TZVP Level of Theory
a
SEN
natural charge
O1
b
b
c
c
complex
Mo•••O2
Mo•••O1
Mo
O2
L2
L1
1(N,N)
1(N,O)
5(N,N)
5(N,O)
1.085
0.858
1.061
0.931
1.084
1.029
1.061
1.072
1.834
1.815
1.805
1.780
−0.523
−0.632
−0.522
−0.589
−0.524
−0.543
−0.522
−0.511
−0.394
−0.303
−0.380
−0.335
−0.397
−0.337
−0.380
−0.346
a
Two center shared electron number (SEN) and atomic charges with multicenter corrections as calculated by population analysis based on
b
molecular orbitals. For the asymmetric complexes 1(N,O) and 5(N,O) O2 represents the oxo group in trans position to a phenolate oxygen.
c
Natural charges summed up over one ligand. For the asymmetric complexes 1(N,O) and 5(N,O) L2 represents the ligand with the phenolate
oxygen in trans position to the oxo group.
metric pyrazole analog 1(N,O) exhibits two different oxo
groups. The SEN in both oxo groups of 5(N,N) is with 1.061
significantly higher in comparison to the oxo groups in
asymmetric complex 1(N,O) (O1 SEN = 1.029; O2 SEN =
(S4) olefins are cleanly converted to the respective epoxides.
Challenging substrates like 1-octene do also show some
conversion under neat conditions, although an improved
activity would be desirable. With catalyst 1 and the benchmark
substrate cyclooctene S1 high activity reflected by a TOF of
0.858) pointing to generally weaker bonds in the latter. This is
−1
remarkable since this is not reflected in the experimentally
determined Mo−O bond lengths in 1(N,O) as they are
identical. Natural population analysis further indicates a shift of
electron density from the MoO1 bond to the O2 atom
leading to a more nucleophilic oxo group. Thereby,
molybdenum becomes slightly more positive and, hence,
more electrophilic. Furthermore, in the hypothetic asymmetric
oxazoline complex 5(N,O) the MoO1 bond is significantly
stronger in comparison to the pyrazole complex 1(N,O) (SEN
∼5000 h can be achieved. Only with styrene S2 all four
catalysts displayed poor selectivity similar to many other
published systems. When compared to the closely related
aryloxide-oxazoline complex 5, complexes 1−4 display lower
reactivity in olefin epoxidation. We attribute this to the higher
electrophilic nature of molybdenum in complexes 1−4 as
demonstrated by DFT calculations for 1 which is shown by the
analysis of partial charges. Possibly, the lower electrophilic
character of the molybdenum atom in 5(N,N) is beneficial for
the oxygen atom transfer to the olefin. This suggests that
asymmetric coordination of ligands like in 1(N,O) could
increase the overall activity if the electrophilicity on
molybdenum is lower as calculated for 5(N,O). However
these initial DFT results do not explain which isomer is formed
preferably.
1
.084 vs 1.029). In summary, compound 5(N,N) exhibits less
polar and stronger molybdenum oxo bonds compared to those
in 1(N,O). The natural charge on molybdenum is lower in
5
(N,N) compared to 1(N,O) and is even lower in the
hypothetical 5(N,O) consistent with a decreasing electrophilic
nature of the molybdenum atom. The catalytic activity under
similar conditions is higher with 5(N,N) compared 1(N,O).
Possibly, the lower electrophilic character of the molybdenum
atom in 5(N,N) is beneficial for the oxygen atom transfer to the
olefin. Thus, the hypothetical 5(N,O) compound would
potentially be more reactive.
EXPERIMENTAL SECTION
■
General. The metal precursor [MoO (acac) ] was purchased from
2
2
Sigma Aldrich and used as received. Ligands HL1−HL4 were prepared
2
0,31
according to literature.
Chemicals were purchased from
Consequently, DFT calculations suggest that the activity of
the investigated catalysts depends not only on the electronic
properties of the ligands, but also on the isomeric form of the
complexes. Therefore, in designing new ligands for catalysts the
rationale should also include the control of the appropriate
isomer. The preference for one isomer over the other with
regard to the ligand system used can not be answered by these
calculations and remains a topic for future investigations.
commercial sources and were used without further purification.
NMR spectra were recorded with a Bruker (300 MHz) instrument.
Chemical shifts are given in ppm and are referenced to residual
protons in the solvent. Signals are described as s (singlet), bs (broad
singlet), d (doublet), dd (doublet of doublet), t (triplet), m
(multiplet) and coupling constants (J) are given in Hertz (Hz).
Elemental analyses were carried out using a Heraeus Vario Elementar
automatic analyzer. Mass spectra were recorded with an Agilent 5973
MSD − Direct Probe using the EI ionization technique. Samples for
infrared spectroscopy were directly measured on a Bruker Optics
ALPHA FT-IR Spectrometer equipped with an ATR diamond probe
head. GC-MS measurements were performed on an Agilent 7890 A
with an Agilent 19091J−433 column coupled to a mass spectrometer
type Agilent 5975 C. Crystallographic data (excluding structure
factors) for 1−3 were deposited with the Cambridge Crystallographic
CONCLUSIONS
■
Within this paper we present aryloxide-pyrazole ligated dioxo
Mo(VI) complexes 1−4. Complexes 1−4 are easy to
synthesize, air- and moisture stable, and display good activity
in the epoxidation of olefins. Terminal (S3) as well as internal
7
647
dx.doi.org/10.1021/ic300648p | Inorg. Chem. 2012, 51, 7642−7649

Inorganic Chemistry
Article
Data Center as supplementary publication no. CCDC-851222 (1),
CCDC-851223 (2), CCDC-850224 (3).
ASSOCIATED CONTENT
Supporting Information
■
*
S
Synthesis of complex 1: [MoO (acac) ] (486 mg, 1.43 mmol) was
2
2
Selected bond distances and angles for 1, 2, and 3, crystal data
for 1, 2, and 3, and ORTEP plots of 1, 2, and 3. This material is
available free of charge via the Internet at http://pubs.acs.org.
dissolved in a minimum amount of methanol with gentle heating. To
this, a solution of 500 mg (2.87 mmol) HL1 in methanol (10 mL) was
added at room temperature and stirred for 16 h. The yellow precipitate
was filtered and washed with cold methanol to give 440 mg (0.927
1
AUTHOR INFORMATION
mmol, 65%) of analytical pure complex 1. H NMR (d -MeOH, 300
4
■
*
MHz, 55 °C) δ: 7.62 (m, 4H), 7.16 (m, 2H), 6.88 (m, 4H), 6.71 (m,
Corresponding Author
E-mail: nadia.moesch@uni-graz.at.
−
1
2
7
H), 4.03 (bs, 6H). IR (ATR, cm ): 1238.54, 916.40, 886.60, 854.85,
+
87.94, 756.58, 412.08. EI-MS: m/z 476.04 [M ]. Anal. Calcd. for
Notes
C H MoN O (474.34): C 50.64, H 3.82, N 11.81; found: C 50.23,
20
18
4
4
The authors declare no competing financial interest.
H 3.73, N 11.73.
Synthesis of complex 2: [MoO (acac) ] (433 mg, 1.33 mmol) was
2
2
dissolved in a minimum amount of methanol with gentle heating. To
this a solution of 500 mg (2.66 mmol) HL2 in methanol (10 mL) was
added at room temperature and stirred for 16 h. The orange
REFERENCES
1) Hille, R. Eur. J. Inorg. Chem. 2006, 1913−1926.
2) Romao, M. J. Dalton Trans. 2009, 4053−4068.
(3) Romao, M. J.; Archer, M.; Moura, I.; Moura, J. J. G.; LeGall, J.;
Engh, R.; Schneider, M.; Hof, P.; Huber, R. Science 1995, 270, 1170−
1176.
■
(
(
precipitate was filtered and washed with cold methanol to give 517 mg
1
(
1.03 mmol, 72%) of analytical pure complex 2. H NMR (d -MeOH,
4
3
00 MHz, 55 °C) δ: 7.63 (m, 2H), 7.42 (m, 2H), 6.99 (m, 2H), 6.80
−1
(
(
4) Hille, R. Arch. Biochem. Biophys. 2005, 433, 107−116.
5) Enemark, J. H.; Cooney, J. J. A.; Wang, J.-J.; Holm, R. H. Chem.
(
1
m, 2H), 6.70 (m, 2H), 2.28 (s, 6H), 2.10 (s, 6H). IR (ATR, cm ):
510.08, 1245.48, 867.58, 809.06, 771.82, 549.67. EI-MS: m/z 504.07
+
Rev. 2004, 104, 1175−1200.
[
M ]. Anal. Calcd. for C H MoN O (502.39): C 52.60, H 4.41, N
22 22 4 4
(
(
(
2
(
6) Sugimoto, H.; Tsukube, H. Chem. Soc. Rev. 2008, 37, 2609−2619.
11.15; found: C 52.33, H 4.40, N 11.13.
7) Schulzke, C. Eur. J. Inorg. Chem. 2011, 1189−1199.
Synthesis of complex 3: [MoO (acac) ] (364 mg, 1.12 mmol) was
2
2
8) Hine, F. J.; Taylor, A. J.; Garner, C. D. Coord. Chem. Rev. 2010,
dissolved in a minimum amount of methanol with gentle heating. To
this a solution of 500 mg (2.23 mmol) HL3 in methanol (10 mL) was
added at room temperature and stirred for 16 h. The brownish
precipitate was filtered and washed with cold methanol to give 464 mg
54, 1570−1579.
9) Lyashenko, G.; Saischek, G.; Judmaier, M. E.; Volpe, M.;
Baumgartner, J.; Belaj, F.; Jancik, V.; Herbst-Irmer, R.; Mosch-Zanetti,
N. C. Dalton Trans. 2009, 5655−5665.
(10) Judmaier, M. E.; Wallner, A.; Stipicic, G. N.; Kirchner, K.;
Baumgartner, J.; Belaj, F.; Mosch-Zanetti, N. C. Inorg. Chem. 2009, 48,
̈
(
0.806 mmol, 72%) of analytical pure complex 3. 1H NMR (d4-
MeOH, 300 MHz, 55 °C) δ: 8.40 (m, 2H), 7.73 (m, 7H), 7.40 (m,
H), 7.08 (m, 2H), 6.79 (m, 2H), 4.38 (s, 3H), 3.97 (s, 3H). IR (ATR,
̈
1
0211−10221.
7
−1
(11) Bregeault, J.-M. Dalton Trans. 2003, 3289−3302.
cm ): 1372.29, 1341.59, 909.30, 886.30, 821.50, 771.43, 733.74,
+
(12) Deubel, D. V.; Frenking, G.; Gisdakis, P.; Herrmann, W. A.;
6
71.5, 403.13. EI-MS: m/z 576.07 [M ]. Anal. Calcd. for
Roesch, N.; Sundermeyer, J. Acc. Chem. Res. 2004, 37, 645−652.
13) Kuhn, F. E.; Santos, A. M.; Herrmann, W. A. Dalton Trans.
005, 2483−2491.
14) Sheng, M. N.; Zajacek, G. J. Method for production of epoxides.
GB Patent 1136923, 1968.
15) Kollar, J. Catalytic epoxidation of an olefinically unsaturated
C H MoN O (574.46): C 58.54, H 3.86, N 9.75; found: C 58.34,
28
22
4
4
(
2
(
̈
H 3.93, N 9.39.
Synthesis of complex 4: [MoO (acac) ] (364 mg, 1.12 mmol) was
2
2
dissolved in a minimum amount of methanol with gentle heating. To
this a solution of 500 mg (2.23 mmol) HL4 in methanol (10 mL) was
added at room temperature and stirred for 16 h. The red-brown
precipitate was filtered and washed with cold methanol to give 489 mg
(
compound using an organic hydroperoxide as an epoxidizing agent.
U.S. Patent 3,350,422, 1967.
(
0.851 mmol, 76%) of analytical pure complex 4. 1H NMR (d4-
(16) Abrantes, M.; Santos, A. M.; Mink, J.; Ku
Organometallics 2003, 22, 2112−2118.
17) Gunyar, A.; Betz, D.; Drees, M.; Herdtweck, E.; Ku
Mol. Catal. A: Chem. 2010, 331, 117−124.
̈
hn, F. E.; Romao, C. C.
MeOH, 300 MHz, 55 °C) δ: 8.09 (bs, 2H), 7.75 (m, 6H), 7.40 (t,
H), 7.28 (t, 2H), 7.18 (d, 2H), 6.68 (bs, 2H), 4.10 (s, 6H). IR (ATR,
(
̈
̈
hn, F. E. J.
2
−1
cm ): 1235.52, 913.93, 874.76, 815.14, 791.97, 742.46, 592.56,
+
(18) Bagherzadeh, M.; Tahsini, L.; Latifi, R.; Woo, L. K. Inorg. Chim.
Acta 2009, 362, 3698−3702.
19) Bagherzadeh, M.; Tahsini, L.; Latifi, R.; Ellern, A.; Woo, L. K.
Inorg. Chim. Acta 2008, 361, 2019−2024.
20) Catalan, J.; Fabero, F.; Claramunt, R. M.; Santa Maria, M. D.;
5
52.71. EI-MS: m/z 574.46 [M ]. Anal. Calcd. for C H MoN O
28 22 4 4
(574.46): C 58.54, H 3.86, N 9.75; found: C 58.25, H 3.85, N 9.72.
(
Epoxidation: In a typical epoxidation experiment the respective
catalyst was stirred in 3 mL of CHCl and substrate was added. After
3
(
the respective experiment temperature was reached TBHP was added
to start the reaction and samples for GC-MS were withdrawn. GC
samples were diluted with ethyl acetate and mesitylene was used as
internal standard.
Foces-Foces, M. d. l. C.; Hernandez Cano, F.; Martinez-Ripoll, M.;
Elguero, J.; Sastre, R. J. Am. Chem. Soc. 1992, 114, 5039−5048.
(21) Brito, Jose
Dun
́ ́
A.; Gomez, M.; Muller, G.; Teruel, H.; Clinet, J.-C.;
̃
ach, E.; Maestro, Miguel A. Eur. J. Inorg. Chem. 2004, 4278−4285.
Computational Details. The geometries of two isomers (N,N-
and N,O-isomer) of complexes 1 and 5 were optimized in the gas
(22) Kandasamy, K.; Singh, H. B.; Butcher, R. J.; Jasinski, J. P. Inorg.
Chem. 2004, 43, 5704−5713.
23) Herrmann, W. A.; Fischer, R. W.; Marz, D. W. Angew. Chem.,
Int. Ed. Engl. 1991, 30, 1638−1641.
(24) Herrmann, W. A.; Wang, M. Angew. Chem., Int. Ed. Engl. 1991,
0, 1641−1643.
25) Rudolph, J.; Reddy, K. L.; Chiang, J. P.; Sharpless, K. B. J. Am.
Chem. Soc. 1997, 119, 6189−6190.
26) Ferreira, P.; Xue, W.-M.; Bencze, E.; Herdtweck, E.; Ku
Inorg. Chem. 2001, 40, 5834−5841.
(27) Qiu, C.-J.; Zhang, Y.-C.; Gao, Y.; Zhao, J.-Q. J. Organomet.
Chem. 2009, 694, 3418−3424.
3
8−40
(
phase using the hybrid density functional B3LYP
in the TURBOMOLE program. First geometry optimizations were
as implemented
4
1,42
performed with the standard double-ζ quality basis def2-SVP.
geometries were then reoptimized with the larger def2-TZVP
The
3
(
4
2,43
basis.
Input geometries were obtained from crystal structures if
available. The stationary points were confirmed as minima by
(
̈
hn, F. E.
calculation of analytical harmonic frequencies. Natural Population
44
Analysis (NPA) as well as calculations of the shared electron number
45
(
SEN) were performed with the larger def2-TZVP basis.
7
648
dx.doi.org/10.1021/ic300648p | Inorg. Chem. 2012, 51, 7642−7649

Inorganic Chemistry
Article
(
28) Schro
F.; Mosch-Zanetti, N. C. Inorg. Chem. 2009, 48, 11608−11614.
29) Traar, P.; Schachner, J. A.; Steiner, L.; Sachse, A.; Volpe, M.;
Mosch-Zanetti, N. C. Inorg. Chem. 2011, 50, 1983−1990.
30) Grunwald, K. R.; Saischek, G.; Volpe, M.; Mosch-Zanetti, N. C.
Inorg. Chem. 2011, 50, 7162−7171.
31) Catalan, J.; Del Valle, J. C.; Claramunt, R. M.; Maria, M. D. S.;
Bobosik, V.; Mocelo, R.; Elguero, J. J. Org. Chem. 1995, 60, 3427−
439.
32) Sorrell, T. N.; Ellis, D. J.; Cheesman, E. H. Inorg. Chim. Acta
986, 113, 1−7.
33) Gomez, M.; Jansat, S.; Muller, G.; Noguera, G.; Teruel, H.;
Moliner, V.; Cerrada, E.; Hursthouse, M. Eur. J. Inorg. Chem. 2001,
071−1076.
34) Wong, Y.-L.; Tong, L. H.; Dilworth, J. R.; Ng, D. K. P.; Lee, H.
K. Dalton Trans. 2010, 39, 4602−4611.
35) Rezaeifard, A.; Sheikhshoaie, I.; Monadi, N.; Stoeckli-Evans, H.
̈
ckeneder, A.; Traar, P.; Raber, G.; Baumgartner, J.; Belaj,
̈
(
̈
(
̈
̈
(
3
(
1
(
́
1
(
(
Eur. J. Inorg. Chem. 2010, 799−806.
(
(
36) Spyroudis, S.; Varvoglis, A. J. Org. Chem. 1981, 46, 5231−5233.
37) Thiel, W. R.; Angstl, M.; Hansen, N. J. Mol. Catal. A: Chem.
1
(
(
(
(
(
995, 103, 5−10.
38) Becke, A. D. J. Chem. Phys. 1993, 98, 1372−1377.
39) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.
40) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785−789.
̈
41) Schafer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97, 2571.
42) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7,
3
(
297−3305.
43) Weigend, F.; Has
Lett. 1998, 294, 143−152.
44) Reed, A. E.; Weinstock, R. B.; Weinhold, F. J. Chem. Phys. 1985,
3, 735−746.
45) Ehrhardt, C.; Ahlrichs, R. Theor. Chim. Acta 1985, 68, 231−245.
̈
ler, M.; Patzelt, H.; Ahlrichs, R. Chem. Phys.
(
8
(
7
649
dx.doi.org/10.1021/ic300648p | Inorg. Chem. 2012, 51, 7642−7649