Dimeric μ-oxo bridged molybdenum(vi) dioxo complexes as catalysts in the epoxidation of internal and terminal alkenes

Article abstract of DOI:10.1039/c3nj00115f

The preparation of the tridentate phenol based amine ligands HL1-HL4 is achieved via a convenient one-pot synthesis by reductive amination in quantitative yield in an autoclave under 7 bar H₂ gas. Reaction of [MoO₂(acac)₂] and the corresponding ligand HLX (X = 1, 2 and 4) in methanol-H₂O results in the formation of orange to red dimeric μ-oxo bridged [{MoO₂(LX)}₂(μ-O)] (X = 1, 2 and 4) complexes 1-3 in high yield and high purity. Complexes 1-3 are stable towards air and water. Both ligands coordinate via the phenolic O atom, the amine N atom and the third donor atom in the side chain (OMe for 1 and NMe ₂ for 2 and 3) in a fac mode to the metal center. The molybdenum atoms are linked by a bridging μ-oxo moiety to each other as confirmed by X-ray diffraction analyses of complexes 2 and 3. All complexes have been tested in the epoxidation of several internal and terminal alkenes using TBHP as an oxidant. Depending on the nature of the substrate, the epoxides are obtained in moderate to good yields and high selectivities. In the epoxidation of cyclooctene a TOF = 467 h^-1 with complex 1 has been observed, significantly higher compared to other dimeric complexes reported in the literature. In the more challenging epoxidation of styrene, complexes 1 and 2 have proven to be highly selective as only the formation of styrene oxide is observed. The OMe based complex 1 has also proven to be more active than the NMe₂ based counterparts 2 and 3. The basic conditions induced by the NMe₂ groups in complexes 2 and 3 lower their catalytic activity.

Full text of DOI:10.1039/c3nj00115f

NJC
View Article Online
View Journal
PAPER
Dimeric l-oxo bridged molybdenum(VI) dioxo
complexes as catalysts in the epoxidation of internal
Cite this: DOI: 10.1039/c3nj00115f
and terminal alkenes†
Martina E. Judmaier, Chris H. Sala, Ferdinand Belaj, Manuel Volpe and
Nadia C. Mosch-Zanetti*
¨
The preparation of the tridentate phenol based amine ligands HL1–HL4 is achieved via a convenient
one-pot synthesis by reductive amination in quantitative yield in an autoclave under 7 bar H2 gas.
Reaction of [MoO2(acac)2] and the corresponding ligand HLX (X = 1, 2 and 4) in methanol–H2O results
in the formation of orange to red dimeric m-oxo bridged [{MoO2(LX)}2(m-O)] (X = 1, 2 and 4) complexes
1–3 in high yield and high purity. Complexes 1–3 are stable towards air and water. Both ligands
coordinate via the phenolic O atom, the amine N atom and the third donor atom in the side chain
(OMe for 1 and NMe2 for 2 and 3) in a fac mode to the metal center. The molybdenum atoms are
linked by a bridging m-oxo moiety to each other as confirmed by X-ray diffraction analyses of complexes
2 and 3. All complexes have been tested in the epoxidation of several internal and terminal alkenes
using TBHP as an oxidant. Depending on the nature of the substrate, the epoxides are obtained in
moderate to good yields and high selectivities. In the epoxidation of cyclooctene a TOF = 467 hꢀ1 with
complex 1 has been observed, significantly higher compared to other dimeric complexes reported in
the literature. In the more challenging epoxidation of styrene, complexes 1 and 2 have proven to be
highly selective as only the formation of styrene oxide is observed. The OMe based complex 1 has also
proven to be more active than the NMe2 based counterparts 2 and 3. The basic conditions induced by
the NMe2 groups in complexes 2 and 3 lower their catalytic activity.
Received (in Montpellier, France)
29th January 2013,
Accepted 4th April 2013
DOI: 10.1039/c3nj00115f
www.rsc.org/njc
several monomeric3–11 and dimeric m-oxo bridged6,7,10,12–17
molybdenum(VI) dioxo complexes coordinated by various types
Introduction
Epoxides are important organic intermediates as they allow for
a wide range of further applications in synthetic organic
chemistry and chemical technology.1 The catalytic epoxidation
of alkenes is an essential synthetic method and high valent oxo
metal species have been demonstrated to be versatile catalysts
in the presence of soft oxidants such as H2O2, alkyl hydro
peroxides or air,2 thus effectively overcoming the limitations
connected to the use of stoichiometric amounts of peracids.
Since the late 1960’s molybdenum and tungsten complexes
have been playing an important role in the industrial propylene
oxide production in the Arco/Halcon process using alkyl
hydroperoxides as oxygen sources.1 During the last few years,
of ligands have been investigated as catalysts for liquid phase
alkene epoxidation, usually employing tert-butyl hydroperoxide
(TBHP) or H2O2 as an oxygen source. Among them, the mono-
meric [CpMoO2Cl] complex3 as well as the air and water stable
dimeric m-oxo bridged [{MoO2Cl(pzH)2}2(m-O)] complex15 sur-
pass the high catalytic activity of the well-known [CH3ReO3]
(MTO) epoxidation catalyst.18,19
In general, molybdenum(VI) dioxo species have proven to be
highly active and selective in the epoxidation of internal
aliphatic alkenes (e.g. cyclooctene or cyclohexene). The epoxi-
dation of terminal alkenes (e.g. styrene) remains challenging,
as in most cases the formation of ring-opening products is
preferred.11,20 Only a limited number of highly selective
molybdenum(VI) dioxo complexes in the epoxidation of styrene
can be found in the literature.3,5,8,21 Styrene oxide is often
used for the manufacture of important commercial products
(e.g. epoxy resins, cosmetics, surface coatings, sweeteners,
perfumes, drugs, etc.).22
¨
¨
Institut fur Chemie, Karl-Franzens-Universitat Graz, Stremayrgasse 16, 8010 Graz,
Austria. E-mail: nadia.moesch@uni-graz.at; Fax: +43 (0)316 380 9835;
Tel: +43 (0)316 380 5286
† CCDC 909614 (2) and 909615 (3). For crystallographic data in CIF or other
electronic format see DOI: 10.1039/c3nj00115f
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem.

View Article Online
Paper
NJC
Thus, the search for highly active and selective homogenous reduction with M[BH4] (M = Na or K) or Na[BH3(CN)].12,13,24
molybdenum catalysts is still ongoing. For easy catalyst handling, However in our hands, the synthesis via this pathway led to a
the formation of air and water stable molybdenum complexes is mixture of products and the desired ligands HL1–HL4 could
highly preferred.
only be obtained in pure form via column chromatography in
Our research group has focused in recent years on the synthesis significant lower yields (o50%).
of rhenium(V) monooxo23 and molybdenum(VI) dioxo11,21 com-
All ligands were characterized by common spectroscopic
plexes as catalysts in epoxidation reactions. Among them is a set methods such as NMR and IR spectroscopy and mass spectro-
of highly active and selective monomeric [MoO2L2] complexes metry. The diagnostic protons of the methylene group at the
ligated by bidentate Schiff base ligands with pendant donor arms aromatic ring (Ar-CH2N) reveal a single resonance in the region
1
(D = OMe, NMe2), influencing both the reactivity as well as the between 3.93 and 4.00 ppm in H NMR spectra, measured in
selectivity.21 Complexes with OMe donor atoms have proven to be chloroform-d, which is in good accordance with the literature.12
more reactive than complexes with an NMe2 group despite the fact The methylene C (Ar-CH2N) atom shows a single resonance
that they are not coordinated in the catalyst precursors. Further- between 52.43 and 53.55 ppm in the 13C NMR spectra. Due to a
more, in the epoxidation of styrene, complexes with donor atoms fast exchange, the protons of the NH group as well as of the OH
1
(OMe, NMe2) have proven to be more selective than a related moiety are not visible in H NMR spectra. IR spectra confirm
complex without an additional donor. Herein, we present the the formation of the desired ligands, as the absorption of
synthesis and characterization of molybdenum(VI) complexes the NH vibration is visible as a weak band between 1605 and
ligated by tridentate phenol based amine ligands in which the 1612 cmꢀ1
imine functionality is reduced to an amine. This led to dinuclear
.
Synthesis of the complexes 1–3
m-oxo bridged complexes that were found to be air and water
stable. Their interesting catalytic behaviour in the epoxidation of
various internal and terminal alkenes is described.
The dimeric m-oxo bridged [{MoO2(LX)}2(m-O)] (X = 1, 2 and 4)
complexes (1–3) are accessible by the reaction of [MoO2(acac)2]
and the respective ligand HLX (X = 1, 2 and 4) in methanol in
the presence of water at room temperature (see Scheme 2). The
complexes precipitate as solid material after stirring overnight
at room temperature. After workup, complexes 1–3 are isolated
Results and discussion
Synthesis of the ligands HL1–HL4
Ligands HL1–HL4 have been prepared in a convenient one-pot as yellow (1) to red (2 and 3) solids in excellent yields (79–96%).
synthesis by reductive amination. Inexpensive benzaldehyde Reaction conditions were evaluated using the example com-
derivatives are reacted with the appropriate primary amines in pound 3. Equimolar reaction of [MoO2(acac)2] and the ligand
an autoclave under 7 bar H2 gas (see Scheme 1). The reaction is HL4 in the presence of 2 equiv. of water in methanol at room
catalyzed by Pd/C (10 wt%). After standard workup, the ligands temperature (see Table 1, entry 4) gave complex 3 in highest
are obtained as viscous oils in quantitative yields.
yields. Elevated reaction temperatures have a negligible effect
Similar ligands were previously prepared via the intermediate on the isolated amount of complex 3 (entry 5). Lower amounts
formation of the corresponding Schiff base and subsequent of water result in lower yields. Dry solvents (methanol or
acetonitrile) often still contain residual traces of water, con-
sequently low amounts of complex 3 are observed in these
reactions, even though no water was added (entries 1 and 6).
Scheme 2 Synthetic procedure for the preparation of dimeric m-oxo bridged
Scheme 1 Synthetic procedure for the preparation of the ligands HL1–HL4.
molybdenum(VI) dioxo complexes 1–3.
c
New J. Chem.
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
NJC
Paper
Table 1 Optimization of reaction conditions for the synthesis of 3
Table 2 Selected 1H NMR resonances and IR vibrations of the ligands HL1, HL2,
HL4 and the dimeric molybdenum complexes 1–3
Entrya
Solvent
H2O (equiv.)
T (1C)
Yield (%)
Ar-CH2Na
Ligand [ppm]
Ar-CH2Na NHa
Complex [ppm]
[ppm] [cmꢀ1
nMoQO
nMo-m-O
b
b
1
2
3
4
5
6
7
MeOH
MeOH
MeOH
MeOH
MeOH
CH3CN
CH3CN
0.0
0.5
1.5
2.0
2.0
0.0
2.0
25
25
25
25
50
25
25
18
72
80
88
86
11
74
]
[cmꢀ1
739
]
HL1
HL2
HL4
4.00
3.97
3.93
1
2
3
3.93
4.77
3.95
4.75
3.85
4.75
4.77
4.75
4.83
897
885
918
880
912
878
711
710
a
An equimolar ratio (1 : 1) of [MoO2(acac)2] and HL4 has been used for
the synthesis of complex 3 in entries 1–7.
a
NMR spectra are measured in chloroform-d. The shifts are reported in
b
ppm vs. the solvent residual peak. IR spectra are measured in the
solid state.
Also reactions performed in acetonitrile in the presence of
2 equiv. of water result in the formation of complex 3 in good
yield (74%). Therefore it seems obvious that the bridging
oxygen atom derives from water, as it has been previously
described in the literature.12
shifted upon coordination to the metal center. The diagnostic
resonance of the methylene moiety (Ar-CH2N) at the aromatic
ring splits into two multiplets (for 1 and 2) or doublets (for 3)
with the integration of one between 3.85 and 4.77 ppm (see
Table 2) upon coordination of the N atom to the metal center.
The proton of the NH moiety is visible as a broad signal between
4.75 and 4.83 ppm. The corresponding C atom of the methylene
moiety (Ar-CH2N) occurs between 53.46 and 54.29 ppm in
13C NMR spectra.
IR spectra of complexes show two strong nMoQO absorptions at
878–885 cmꢀ1 and 897–918 cmꢀ1, characteristic of the symmetric
and asymmetric stretching mode of the cis-[MoO2]2+ fragment.
The absorption of the nMo–(m-O) vibration is visible as a broad band
between 710 and 739 cmꢀ1. All absorptions are in good accor-
dance with the literature.12,17,27 Elemental analyses confirmed
the existence of m-oxo bridged dimeric molybdenum complexes
of the type [{MoO2(LX)}2(m-O)] (X = 1, 2 and 4).
Surprisingly enough, analogous reaction conditions employing
ligand HL3 and [MoO2(acac)2] did not result in any complex
1
formation. Monitoring the reaction via H NMR spectroscopy
after 24 hours of reaction time showed only the existence of the
free ligand and the metal precursor. Furthermore, the for-
mation of monomeric complexes of the type [MoO2L2] has
not proven to be successful with this type of ligand. On the
other hand, we have previously published a set of such mono-
nuclear complexes ligated by corresponding Schiff base ligands
featuring an imine (Ar-CHQN) moiety.21 With these ligands no
dimeric complexes could be obtained, since in the presence
of water they tend to decompose into unidentified polyoxo
materials. Monomeric compounds are also formed with a
ligand where the proton attached to the amine nitrogen atom
is substituted by a methyl group. Reaction of this type of
ligands with [MoO2Cl(Z2-tBu2Pz)]25 or [MoO2Cl2] led to com-
plexes [MoO2ClL] in which the ligand is coordinated in a
tridentate fashion.26 Any attempt to substitute the remaining
chlorine atom in the latter compounds proved to be futile. Thus,
the protons at nitrogen in the here described ligands are crucial
for the stability of the complexes 1–3. This fact is supported by
the molecular structure determined by X-ray diffraction analysis
(vide infra) where strong MoQO–H bonds are apparent.
Dimeric [{MoO2Ln}2(m-O)] (n = 1 or 2) complexes are wide-
spread in the literature.6,7,9,10,12–15,17,27 Mitchell and Finney
published in 2001 a very similar complex with a pyridine moiety
in the side chain.12 This complex has proven to be highly active
in the epoxidation of aromatic alkenes in the presence of tert-
butyl hydroperoxide (TBHP).
Molecular structure in the solid state
Single crystals suitable for X-ray diffraction analyses of complex 2
Complexes 1–3 are well soluble in common organic solvents were obtained from dilute acetonitrile solutions at room tempera-
such as dichloromethane, chloroform, THF and methanol at ture and those for complex 3 were obtained from concentrated
room temperature, but less so in acetonitrile and water. All dichloromethane solutions at ꢀ30 1C, in both cases in the form of
complexes are insoluble in aliphatic and aromatic hydro- orange blocks. Complex 1 also crystallized from an acetonitrile
carbons such as pentane, heptane, benzene and toluene. solution, however the crystals were found to be unsuitable for
Complexes 1–3 are stable in air and can be stored without X-ray diffraction analysis. A molecular view of each of the two
decomposition for several months.
All complexes have been characterized by common spectro- are given in Table 3 and crystallographic data in Table 7.
scopic methods such as NMR and IR spectroscopy and ele- The molecular structures of complexes 2 and 3 are quite
complexes is shown in Fig. 1. Selected bond lengths and angles
mental analyses. MS measurements – even with very mild similar and both complexes reveal a m-oxo bridging moiety.
ionization techniques – failed to show the molecular peak. The two molybdenum atoms are six coordinated in a distorted
Only free ligand could be detected. However the molecular octahedral environment. Each molybdenum atom is ligated by
structures of complexes 2 and 3 were determined by single a tridentate ligand, two terminal oxygen atoms and a bridging
crystal X-ray diffraction analyses (vide infra).
m-oxo atom.
The tridentate ligand coordinates via its phenolic O atom
1H and 13C NMR spectroscopy confirmed the formation of
the desired complexes as the resonances of the free ligand are and both N atoms (NH and NMe2) to the metal center.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem.

View Article Online
Paper
NJC
Table 3 Selected bond lengths [Å] and angles [1] of complexes 2 and 3
2
3
Mo1–O1
Mo1–O2
Mo1–O3
Mo1–O11
Mo1–N21
Mo1–N22
Mo2–O1
Mo2–O4
Mo2–O5
Mo2–O41
Mo2–N51
Mo2–N52
Mo1ꢁ ꢁ ꢁMo2
O1–Mo1–O11
O2–Mo1–N21
O3–Mo1–N22
O1–Mo2–O41
O4–Mo2–N52
O5–Mo2–N51
Mo1–O1–Mo2
O3–Mo1–O1
O3–Mo1–O2
O3–Mo1–O11
O3–Mo1–N21
O4–Mo2–O1
O4–Mo2–O5
O4–Mo1–O41
O4–Mo2–N51
1.9148(10)
1.7113(11)
1.7258(11)
1.9515(11)
2.3276(13)
2.4168(13)
1.9104(10)
1.7203(12)
1.7122(11)
1.9537(11)
2.3350(13)
2.4429(14)
3.7110
155.96(5)
158.13(5)
166.71(5)
155.21(5)
167.50(5)
157.81(6)
151.93(6)
96.39(5)
1.9104(12)
1.7178(13)
1.7206(13)
1.9548(12)
2.3298(15)
2.4138(15)
1.9055(12)
1.7256(14)
1.7091(14)
1.9577(13)
2.3233(14)
2.4221(16)
3.6954
154.87(5)
160.08(6)
165.45(6)
155.63(6)
166.28(6)
158.91(6)
151.12(8)
96.64(6)
108.36(6)
94.64(5)
92.93(5)
96.86(5)
107.63(6)
94.25(5)
108.13(7)
95.08(6)
91.42(6)
96.32(6)
108.20(7)
95.82(6)
94.15(5)
92.54(6)
D–Hꢁ ꢁ ꢁA
d(D–H)
d(Hꢁ ꢁ ꢁA)
d(Dꢁ ꢁ ꢁA)
o(DHA)
Hydrogen bonds for complex 2
N21–H21ꢁ ꢁ ꢁO4
0.93
0.93
1.99
1.99
2.9093(17)
2.9126(17)
167.7
168.8
N51–H51ꢁ ꢁ ꢁO3
Fig. 1 Molecular structure and atom labeling scheme for complexes 2 (top) and 3
(bottom). Thermal ellipsoids were drawn at the 50% probability level. Hydrogen
atoms (except those involved in hydrogen bonding) are omitted for clarity.
Hydrogen bonds for complex 3
N21–H21ꢁ ꢁ ꢁO4
0.93
0.93
1.99
1.98
2.899
2.900
166.09
167.92
N51–H51ꢁ ꢁ ꢁO3
The molybdenum oxo groups show the expected mutual cis
configuration and are located trans to the NH and NMe2
moieties. The phenolic O atom and the m-oxo bridged O atom
are found to be trans to each other. The conformation of
complexes 2 and 3 is locked in position by two intra-molecular
hydrogen bonds, as protons of the NH group weakly coordinate
to a terminal oxygen atom close by.
(e.g. Mo1ꢁ ꢁ ꢁMo2 3.7110 Å for 2 and 3.6954 Å for 3). Both, the
angle around the m-oxo bridge as well as the intermetallic
distances are in good accordance with the literature.9,12,15
Epoxidation of alkenes
All dimeric molybdenum(VI) dioxo complexes 1–3 have been
All Mo–O bond lengths (ranging from 1.9515(11) to tested as catalysts in the epoxidation of several internal and
1.9577(13) Å) as well as all MoQO bond lengths (ranging from terminal alkenes using tert-butyl hydroperoxide (TBHP, 5.5 M in
1.7091(14) to 1.7258(11) Å) are in the expected range of cis- decane) as an oxidant. Optimal reaction conditions regarding
[MoO2]2+ complexes. The Mo–(m-O) bond lengths (ranging from temperature, oxidant loading and catalyst loading were evaluated
1.9104(10) to 1.9148(10) Å) are slightly longer than those using the substrate cyclooctene and complex 1 as a catalyst.
reported in the literature.9,12,27 The Mo–NH bond lengths
Table 4 summarizes the results of temperature and solvent
(ranging from 2.3233(14) to 2.3350(13) Å) as well as the screening. In accordance with the literature, chlorinated
Mo–NMe2 bond lengths (ranging from 2.4138(15) to 2.4429(14) Å) solvents proved to be the best cosolvent. Full conversion to
are somewhat longer due to the influence of the trans coordi- the corresponding epoxide was obtained in chloroform at 50 1C
nated MoQO ligands.9,12,27
within one hour of reaction time. With heptane and toluene as
Bond angles around the metal centers deviate in both com- co-solvents, acceptable conversions at higher temperature
plexes considerably from those of an ideal octahedron (see (75 1C) were obtained. Lower conversions are observed in
Table 3). The distorted octahedral geometry is also evident by methanol and tert-butyl alcohol, whereas no conversion of
the obtuse angles of the apical ligands (e.g. O3–Mo1–O2 108.36(6)1 epoxide is reached in THF and diethyl ether. Again, these
for 2 and O1–Mo1–O2 108.13(7)1 for 3). The angle around the m-O results are in good accordance with the literature, where higher
bridge (e.g. Mo1–O1–Mo2 151.93(6)1 for 2 and Mo1–O7–Mo2 temperatures and chlorinated solvents are preferred. All further
151.12(8)1 for 3) promotes a short intermetallic distance experiments were performed in chloroform at 50 1C (Scheme 3).
c
New J. Chem.
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
NJC
Paper
Table 4 Epoxidation of cyclooctene catalyzed by 1; the effect of cosolvent and
temperature
Cosolventa
35 1C
50 1C
75 1C
C2H4Cl2 87
Chlorinated
Heptane
Toluene
Diethyl ether
THF
CH2Cl2
82
CHCl3
98
63
72
69
81
o10
o10
27
MeOH
tBuOH
12
a
Reaction conditions: 0.5 mol% complex 1, 1.41 mmol cyclooctene,
2.82 mmol TBHP (2 equiv.). Conversions were determined by GC-MS
measurements after 60 minutes. Mesitylene was used as internal
standard. Complete selectivity towards the epoxide was observed.
Fig. 2 Activity test of complex 1 via the epoxidation of cyclooctene in four
consecutive runs. After 120 minutes, 24 and 48 hours 1.41 mmol cyclooctene and
2.82 mmol TBHP are added to the initial reaction mixture. Reactions are
performed in chloroform (5 ml) at 50 1C and 0.5 mol% catalyst loading.
Scheme 3 Epoxidation of cyclooctene.
most probably due to the increasing concentration of tert-butyl
alcohol in solution, which may act as a competitive inhibitor
for the attack of TBHP at the molybdenum(VI) center.28 After
48 hours, the catalytic activity of complex 1 is significantly
lower and only 40% of epoxide is obtained after 60 minutes of
reaction time.
Complexes 1–3 have been used as precatalysts in the epoxi-
dation of different aliphatic and aromatic alkenes. All reactions
were performed in chloroform at 50 1C using 0.5 mol% catalyst,
1.41 mmol substrate and 2.82 mmol TBHP. In all cases control
experiments confirmed a low conversion of the substrate
(o10%) in the absence of a catalyst.
The different alkenes are generally oxidized in moderate to
good yields with high selectivities (see Table 6). Complete
conversion to the corresponding epoxide is observed for both
cyclic alkenes (cyclohexene and cyclooctene) after one hour of
reaction time for the OMe based complex 1. Lower conversions
are observed after 24 hours for both NMe2 based complexes 2
and 3. These complexes possess similar catalytic activities.
Sterically more demanding substituents (tBu in 2 vs. Me in 3)
in the ligand backbone have a negligible effect.
Further epoxidation reactions were performed with com-
plexes 1 and 2. Both complexes have sterically demanding tBu
substituents at ortho and para positions on the phenyl ring, but
differ in the donor atoms in the side chain (OMe for 1 and
NMe2 for 2).
Complex 1 allows for a catalyst loading down to 0.25 mol%
in the presence of two equivalents of TBHP without loss of
activity (TOF = 467 hꢀ1). After one hour of reaction time, full
conversion to the corresponding epoxide is observed. A further
catalyst lowering to 0.1 mol% with a similar TBHP amount
results in lower epoxide yield (87%). Reactions performed with
1 equiv. of TBHP are slower. Only 50% conversion is obtained
with a catalyst : substrate ratio of 1 : 200. After 3 hours of
reaction time, 65% of epoxide yield is observed. For subsequent
epoxidation reactions we chose 0.5 mol% of catalyst loading
and 2 equiv. of TBHP at 50 1C in chloroform as the standard
conditions (Table 5).
The stability of complex 1 under catalysis conditions was
tested in four consecutive runs in the epoxidation of cyclo-
octene using TBHP (5.5 M in decane) as an oxidant (see Fig. 2).
After 90 minutes, 24 hours and 48 hours, catalysis was restarted by
the addition of cyclooctene and TBHP (ratio 1 : 2). Full conversion
to the corresponding epoxide was obtained in the first two runs
after 60 minutes of reaction time. After 24 hours, complex 1 is still
active, but the conversion rate is lower than in the former
two runs (e.g. after 60 minutes 58% of epoxide are obtained),
Table 5 Epoxidation of cyclooctene catalyzed by 1; the effect of catalyst and
oxidant (TBHP) loading
Conversions for linear alkenes are in general somewhat
lower, but the best catalyst 1 reaches 70% of 1,2-epoxyoctane
and 83% of 2,3-epoxy-1-propanol after 24 hours. More challen-
ging are aromatic alkenes, such as styrene, cis-stilbene and
trans-stilbene, as these substrates are less electron rich.
Moderate conversions, but always with excellent selectivities,
are obtained in the epoxidation of styrene. The formation of
ring opening products is negligible. Again, complex 1 is some-
Conversiona (%)
TBHPb (1 equiv.)
TBHPb (2 equiv.)
mol% 1
[Mo] : alkene ratio
0.5
0.25
0.1
0.05
0.01
1 : 200
1 : 400
1 : 1000
1 : 2000
1 : 10 000
50
48
33
22
18
98
96
87
80
21
a
Reactions were carried out at 50 1C in chloroform (5 ml) using what more active (e.g. after 24 hours 44% for 1 vs. 36% for 2). In
1.41 mmol cyclooctene, 1.41 mmol mesitylene (internal standard) and
1.41 (1 equiv.) or 2.82 (2 equiv.) TBHP. Conversions were determined by
GC-MS measurements after 60 minutes. Mesitylene was used as internal
standard. Complete selectivity towards the epoxide was observed.
the epoxidation of cis-stilbene and trans-stilbene significantly
higher conversions are obtained with complex 1. Complex 2 is
less active and only in the case of cis-stilbene, epoxide formation
b
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem.

View Article Online
Paper
NJC
Table 6 Epoxidation of aliphatic and aromatic alkenes using complexes 1–3 as
as the formation of 1,2-epoxy-limonene as a single product is
preferred.
precatalysts
Recently we published a set of monomeric [MoO2L2] com-
plexes ligated by bidentate Schiff base ligands (Ar-CHQN
moiety) including the same donor atoms in the side chain
(OMe and NMe2).21 For both complex types (monomeric vs.
dimeric), the OMe based compounds have proven to be more
active than their NMe2 based counterparts. The more basic
NMe2 donor group is obviously more prone towards protona-
tion leading to the formation of a less active species and hence
slowing down the catalytic activity.21 The higher catalytic
activity of OMe based complexes may be explained by the
influence of the OMe group in the formation of hydrogen
bonds. Such bonds may be stable enough to lower the activa-
tion barriers during the catalytic cycle.30,31 Moreover, the
catalytic activity of both, the monomeric and dimeric com-
plexes, is influenced by the nature of the substrate. In the
epoxidation of internal aliphatic alkenes the dimeric complex 1
has proven to be more active than the monomeric [MoO2L2]
complexes (e.g. in the epoxidation of cyclooctene TOF = 467 hꢀ1
for 1 vs. TOF = 359 hꢀ1 for the best [MoO2L2] complex21). In the
epoxidation of terminal alkenes (e.g. styrene) higher conver-
sions are observed in reactions catalyzed by the monomeric
complexes.
Substrate
Catalysta Epoxide yield (%) Time (h) Selectivityb (%)
Cyclohexenec
1
2
3
1
2
3
1
2
1
2
1
2
1
2
1
2
1
2
98
90
95
98
9/88
7/90
34/70
26
66/83
21
35/44
36
36/62
17
32/55
—
55/100
36/86
53/98/53
37/77
1
24
24
1
1/24
1/24
4/24
24
8/24
24
5/24
24
4/24
24
4/24
24
0.25/1
4/24
0.25/2/4 53
4/24 99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
99
—
99
99
Cyclooctenec
1-Octened
2-Propenold
Styrenec
cis-Stilbenee
trans-Stilbenee
a-Terpineolc
R-(+)-Limonenee 1f
2
a
Reactions were carried out at 50 1C in chloroform using 1.41 mmol
(1 equiv.) alkene, 1.41 mmol internal standard, 2.82 mmol (2 equiv.)
b
c
TBHP. Selectivity after 24 hours. Reaction yields were determined by
integration of GC-MS chromatograms; mesitylene was used as internal
standard. Isolated yields: determined by integration of 1H NMR spec-
d
e
tra in CDCl3 vs. dichloroethane as internal standard. Reaction yields
were determined by integration of HPLC chromatograms; mesitylene
was used as internal standard. Dipentene dioxide (limonene dioxide)
as a side product.
The reaction mechanism in the epoxidation of alkenes is
still under debate.17,30–32 However, the more likely mechanism
includes the addition of TBHP across a terminal MoQO group,
resulting in the formation of Mo–OH and Mo–O–O–tBu moieties.
f
is observed (e.g. after 24 hours 17%). In the epoxidation of The a-O atom of the latter moiety is then transferred to the
a-terpineol the oxidizable OH group is not affected and after alkene, producing the epoxide under concomitant elimination
one hour of reaction time full conversion is obtained with of tert-butyl alcohol, yielding the initial complex. Theoretical
complex 1, whereas 86% of epoxide is obtained after 24 hours studies evidence the importance of H bond formations at
with 2. The epoxidation of limonene is more challenging as the various steps during the catalytic cycle.30,31 Both the oxygen
substrate includes two different alkene moieties – one internal atom transfer as well as the H atom transfer represent a barrier.
and one terminal double bond. Full conversion within two Single crystal X-ray diffraction analyses of complexes 2 and 3
hours of reaction is observed for the OMe based complex 1. indicate the existence of MoQOꢁ ꢁ ꢁH bonds.
Nevertheless, the selectivity after 24 hours is low as the for-
In the case of dimeric complexes, a splitting across the m-oxo
mation of the double epoxide (dipentene dioxide) is preferred. bridge prior to the coordination of TBHP seems to be necessary.
Reactions catalyzed by 2 are slower, but more selective. After It has previously been described in the literature that in
24 hours, only 1,2-epoxy-limonene (77%) as a single product is comparative studies related monomeric and dimeric complexes
obtained.
exhibit similar epoxidation activities after 24 hours of reaction
Complex 1 could definitely not surpass the high catalytic time pointing to a common monomeric catalytically active
activity of the dimeric [{MoO2Cl(pzH)2}2(m-O)] complex pub- species.7,12,16,17 ATR-IR spectroscopic investigations of the
lished by Gonçalves and coworkers in 2007.15 This complex epoxidation of cyclooctene employing complex 1 were performed
reaches under the best reaction conditions a TOF = 32 000 hꢀ1 by taking aliquots of the reaction solution. The absorption of
in the epoxidation of cyclooctene, which is even higher than the the Mo–(m-O) vibration in the IR spectrum at 739 cmꢀ1 dis-
TOFs recorded for the [ReO3CH3]–H2O2 catalytic epoxidation appears during the catalytic cycle indicating the transformation
system.18,29 Compared to other dimeric [{MoO2Ln}2(m-O)] com- of the catalyst precursor. Both, the symmetric and the asym-
plexes (n = 1, 2) in the literature,6,7,12,14,17 the catalytic activity metric absorption of the MoQO vibration are shifted and
of complex 1 is higher. Under the best reaction conditions a appear now as broad overlapping bands at 907 cmꢀ1. The
TOF = 467 hꢀ1 in the epoxidation of cyclooctene is observed. formation of a monomeric active species is likely, but the real
For example, the [{MoO2Cl(dmf)2}2(m-O)] complex7 representing nature of the compound is as yet unclear. Furthermore, it is
one of the more active dimeric catalysts reaches a TOF = unknown whether both metal atoms of the dinuclear com-
102 hꢀ1 in the epoxidation of cyclooctene. In the epoxidation pounds represent active centers. Comparison of the activity
of R-(+)-limonene similar conversions to the literature are of compounds 1–3 to related mononuclear compounds would
obtained,6,14 but complex 2 has proven to be more selective be helpful. However, such compounds are as yet elusive
c
New J. Chem.
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
NJC
Paper
Table 7 Crystal data and structure refinement of complexes 2 and 3
2
3
Empirical formula
Formula weight
Mo2O7N4C38H66ꢁ2 C2H3N
Mo2O7N4C26H42ꢁ2 CH2Cl2
884.37
964.94
Crystal description
Crystal size (mm)
Crystal system, space group
Unit cell dimensions, a (Å)
Block, orange
0.29 ꢂ 0.22 ꢂ 0.20
Monoclinic, P21/n
13.8832(5)
Block, orange
0.46 ꢂ 0.34 ꢂ 0.29
%
Triclinic, P1
7.4666(6)
b (Å)
c (Å)
a (1)
11.6611(4)
30.2445(9)
90
14.8332(11)
16.7476(12)
98.171(2)
b (1)
94.1130(10)
90
4883.8(3)
4, 1.312
2024
90.647(2)
92.937(2)
1833.3(2)
2, 1.602
900
1.022
Multi-scan
100
g (1)
Volume (Å3)
Z, calculated density (g cmꢀ3
F(000)
)
Linear absorption coefficient m (mmꢀ1
Absorption correction
)
0.563
Semi-empirical from equivalents
100
Temperature (K)
Wavelength (MoKa) (Å)
Theta range for data collection (1)
Limiting indices
0.71073
0.71073
1.87 to 30.00
ꢀ19 r h r 11
ꢀ16 r k r 14
ꢀ42 r l r 42
44 653/14 241
12 714
2.46 to 34.35
ꢀ11 r h r 11
ꢀ21 r k r 23
ꢀ24 r l r 25
360 456/12 236
11 100
Reflections collected/unique
Reflections with I 4 2s(I)
R(int), R(sigma)
Completeness to theta max.
Refinement method
0.0216, 0.0229
0.999
0.0425, 0.0270
0.983
Full matrix least squares on F2
14 241/585/1
1.060
Full matrix least squares on F2
12 236/5/441
1.071
Data/restraints/parameters
Goodness-of-fit on F2
a
b
Final R1 wR2 [I 4 2s(I)]
R1 = 0.0268
wR2 = 0.0657
R1 = 0.0322
wR2 = 0.0686
1.200 and ꢀ0.691
909614
R1 = 0.0297
wR2 = 0.0754
R2 = 0.0340
wR2 = 0.0790
1.364 and ꢀ1.380
909615
R indices (all data)
Largest diff. peak and hole (e Åꢀ3
CCDC deposition no.
)
ꢂ
ꢃ
1=2
h
i
h
i
2
ꢀ
ꢁ
ꢀ
ꢁ
P
P
P
P
2
2
a
b
2
2
R1 ¼ kFojꢀjFck= jFoj. wR2 ¼
w wFo ꢀ Fc2
=
w Fo
.
preventing the elucidation of this question. Attempts to isolate yields with high selectivities. Complex 1 has proven to be less
any intermediate molybdenum(VI) tert-butyl hydroperoxide selective in the epoxidation of R-(+)-limonene as the formation of
complex have not been successful as yet. Unfortunately, the use the double epoxide (dipentene dioxide) is observed after 24 hours
of H2O2 as a terminal oxidant in the epoxidation of cyclooctene of reaction time. These results revealed complex 1 to be a highly
did not yield significant amounts of the corresponding epoxide active catalyst in comparison to other published molybdenum
(17% of cyclooctene oxide for 1).
complexes but could not surpass the high catalytic activity of the
dimeric [{MoO2Cl(pzH)2}2(m-O)] complex published by Gonçalves
and coworkers in 2007.15 Furthermore, compounds 1 and 2 proved
to be highly selective in the epoxidation of more challenging
substrates such as 1-octene and styrene as no further oxidation
products were observed. The nature of the real catalyst is as yet
unclear, but IR measurements of complex 1 indicate a splitting of
the dimeric m-oxo bridged compound during the catalytic cycle.
Conclusions
Here we presented a set of new dimeric m-oxo bridged
molybdenum(VI) dioxo complexes ligated by tridentate phenol
based ligands including an NH moiety. Complexes 1–3 are
obtained by a simple procedure in high yield and high purity.
The stability of complexes 1–3 towards moisture and air simplifies
the handling. X-ray diffraction analyses of complexes 2 and 3 prove
the existence of a bridging O atom, which most probably occurs
from water. Complexes 1–3 act as precatalysts in the epoxidation of
Experimental
General remarks
aliphatic and aromatic alkenes. The OMe based complex 1 has 3,5-Di-methyl-2-hydroxybenzaldehyde was prepared according
proven to be more active than the NMe2 based counterparts 2 and to the literature.33 [MoO2(acac)2] was purchased from Aldrich
3. However, consecutive catalytic runs with 1 showed after two and used as received. All other chemicals were obtained from
cycles reduced activity. Terminal as well as internal alkenes includ- different suppliers and used without further purification. All
ing challenging substrates such as 1-octene and styrene are cleanly 1H and 13C NMR spectra were recorded on a Bruker Avance III
converted to the corresponding epoxides in moderate to good Spectrometer (300 MHz for 1H and 75 MHz for 13C NMR).
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem.

View Article Online
Paper
NJC
The 1H NMR spectroscopic data are reported as s = singlet, Anal. calcd for C18H31NO2 (293.4): C 73.17, H 10.65, N 4.77%;
d = doublet, t = triplet, m = multiplet, coupling constants are found: C 72.80, H 9.77, N 4.90%.
reported in Hz and chemical shifts are given in ppm relative to
Ligand HL2. 3,5-Di-tert-butyl-2-hydroxybenzaldehyde (4.69 g,
the solvent residual peak. All deuterated solvents were pur- 20 mmol) and N,N-dimethylethylene-diamine (2.29 g, 26 mmol)
chased from Deutero GmbH and dried over molecular sieves. IR were dissolved in methanol (80 ml). Pd/C (0.47 g) was added
spectra were measured as solid samples on a Bruker Alpha P and the mixture was exposed to H2 gas (7 bar). After purifica-
Diamond FTIR-ATR spectrometer. Mass spectra have been tion, the ligand was obtained as a colorless viscous oil. Yield:
measured on an Agilent 5973 MSD-Direct Probe using the EI 6.00 g (98%).
ionisation technique. GC-MS measurements were performed
on an Agilent 7890A with an Agilent 19091J-433 column C(CH3)3), 1.44 (s, 9H, C(CH3)3), 2.24 (s, 6H, N(CH3)2), 2.46
1H NMR (300 MHz, chloroform-d, 298 K, ppm): d 1.30 (s, 9H,
3
3
coupled to a mass spectrometer type Agilent 5975C. HPLC (t, JH-H = 5.9 Hz, 2H, N(CH2)2N), 2.75 (t, JH-H = 5.9 Hz, 2H,
4
measurements were performed on an Agilent 1200 Series with N(CH2)2N), 3.97 (s, 2H, Ar-CH2N), 6.88 (d, JH-H = 2.4 Hz, 1H,
a Zorbax eclipse XDB-C18 column and a UV/Vis detector. Ar-H), 7.23 (d, JH-H = 2.5 Hz, 1H, Ar-H). 13C NMR (75 MHz,
4
Elemental analyses were carried out using a Heraeus Vario chloroform-d, 298 K, ppm): d 29.77 (C(CH3)3), 31.83 (C(CH3)3),
Elementar automatic analyzer at the Institute of Inorganic 34.25 (C(CH3)3), 35.01 (C(CH3)3), 45.59 (2ꢂ N(CH3)2), 46.09
Chemistry at the University of Technology in Graz.
(N(CH2)2N), 53.55 (Ar-CH2N), 58.41 (N(CH2)2N), 122.22 (Ar),
122.90 (Ar-H), 123.21 (Ar-H), 135.86 (Ar), 140.35 (Ar), 154.90
(Ar-OH). The protons of the NH and OH group were not
General procedure for the synthesis of the ligands
The aromatic benzaldehyde (12.8 mmol, 1 equiv.) and the detected in benzene-d6 or in chloroform-d. IR (ATR, cmꢀ1):
appropriate primary amine (16.6 mmol, 1.3 equiv.) were dis- n 2950 (s), 1606 (w, NH), 1458 (s), 1438 (s), 1360 (m), 1236 (s),
solved in methanol (80 ml) and placed in a stainless steel 1202 (m), 1040 (m), 876 (m), 821 (m), 761 (m). MS (EI) (70 eV)
250 ml autoclave. Pd/C (10 %wt) as a catalyst was added. The m/z (%): 306.3 (23) [M]+, 247.3 (17) [M-(CH2N(CH3)2)-H]+, 219.3
mixture was degassed several times and saturated with H2. The (100) [M-NH(CH2)2N(CH3)2]+. Anal. calcd for C19H34N2O (306.5):
hydrogenation was carried out at 7 bar H2 with stirring at 50 1C C 74.46, H 11.18, N 9.14%; found: C 74.09, H 11.02, N 9.08%.
for 24 hours. The hydrogen gas was then carefully released and
Ligand HL3. 3,5-Di-methyl-2-hydroxybenz-aldehyde33 (3.00 g,
the Pd/C catalyst was filtered off. The solution of the product 20 mmol) and 2-methoxyethylamine (1.96 g, 26 mmol) were
was dried over Na2SO4 and again filtered. The solvent was dissolved in methanol (80 ml). Pd/C (0.30 g) was added and
removed under vacuum affording the corresponding ligand as the mixture was exposed to H2 gas (7 bar). After purifica-
colorless oils in quantitative yield. The ligands were used tion, the ligand was obtained as a colorless viscous oil. Yield:
without further purification.
4.06 g (97%).
Reduction of the respective Schiff base ligands with M[BH4]
1H NMR (300 MHz, chloroform-d, 298 K, ppm) d 2.21 (s, 3H,
3
(M = Na or K) or Na[BH3(CN)] according to the literature results Ar-CH3), 2.22 (s, 3H, Ar-CH3), 2.82 (t, JH-H = 5.1 Hz, 2H,
as well in the formation of the desired amine based ligands.12 N(CH2)2O), 3.37 (s, 3H, OCH3), 3.52 (t, JH-H = 5.1 Hz, 2H,
3
Nevertheless, as purification via column chromatography N(CH2)2O), 3.95 (s, 2H, Ar-CH2N), 6.66 (s, 1H, Ar-H), 6.87
on aluminium oxide was needed, yields were significantly lower (s, 1H, Ar-H). 13C NMR (75 MHz, chloroform-d, 298 K, ppm)
(o50%).
d 15.73 (Ar-CH3), 20.50 (Ar-CH3), 47.96 (N(CH2)2N), 52.43
Ligand HL1. In a 250 ml stainless steel autoclave 3,5-di-tert- (Ar-CH2N), 58.95 (OCH3), 71.13 (N(CH2)2N), 121.46 (Ar),
butyl-2-hydroxybenzaldehyde (4.69 g, 20 mmol) and 2-methoxy- 124.93 (Ar), 126.57 (Ar-H), 127.57 (Ar), 130.55 (Ar-H), 153.97
ethylamine (1.96 g, 26 mmol) were dissolved in methanol (Ar-OH). The protons of the NH and OH group were not
(80 ml). Pd/C (0.47 g) was added and the mixture was exposed detected in benzene-d6 or in chloroform-d. IR (ATR, cmꢀ1):
to H2 gas (7 bar). After purification, the ligand was obtained as a n 2917 (m), 1612 (w, NH), 1481 (s), 1244 (s), 1192 (m), 1157 (m),
colorless viscous oil. Yield: 5.63 g (96%).
1124 (s), 1097 (s), 857 (s), 770 (s). MS (EI) (70 eV) m/z (%): 209.2 (43)
1H NMR (300 MHz, chloroform-d, 298 K, ppm) d 1.32 (s, 9H, [M]+, 164.2 (46) [M-CH2OCH3]+, 135.2 (100) [M-NH(CH2)2OCH3]+.
3
C(CH3)3), 1.46 (s, 9H, C(CH3)3), 2.88 (t, JH-H = 5.1 Hz, 2H, Anal. calcd for C12H19NO2 (209.3): C 68.87, H 9.15, N 6.69%; found:
N(CH2)2O), 3.40 (s, 3H, OCH3), 3.56 (t, JH-H = 4.8 Hz, 2H, C 68.68, H 8.78, N 6.32%.
3
4
N(CH2)2O), 4.00 (s, 2H, Ar-CH2N), 6.90 (d, JH-H = 2.4 Hz, 1H,
Ligand HL4. 3,5-Di-methyl-2-hydroxybenz-aldehyde33 (3.00 g,
4
Ar-H), 7.24 (d, JH-H = 2.4 Hz, 1H, Ar-H). 13C NMR (75 MHz, 20 mmol) and N,N-dimethylethylenediamine (2.29 g, 26 mmol)
chloroform-d, 298 K, ppm) d 29.77 (C(CH3)3), 31.83 (C(CH3)3), were dissolved in methanol (80 ml). Pd/C (0.30 g) was added
34.26 (C(CH3)3), 35.02 (C(CH3)3), 48.07 (N(CH2)2O), 53.33 and the mixture was exposed to H2 gas (7 bar). After purifica-
(Ar-CH2N), 58.96 (OCH3), 71.22 09 (N(CH2)2O), 122.04 (Ar), tion, the ligand was obtained as a colorless viscous oil. Yield:
123.02 (Ar-H), 123.31 (Ar-H), 135.93 (Ar), 140.49 (Ar), 154.83 4.06 g (97%).
(Ar-OH). The protons of the NH and OH group were not
detected in benzene-d6 or in chloroform-d. IR (ATR, cmꢀ1): Ar-CH3), 2.22 (s, 3H, Ar-CH3), 2.23 (s, 6H, N(CH3)2), 2.44
n 2952 (s), 1605 (w, NH), 1441 (s), 1360 (m), 1237 (s), 1128 (m), (t, JH-H = 5.7 Hz, 2H, N(CH2)2N), 2.70 (t, JH-H = 5.7 Hz, 2H,
1098 (m), 877 (m), 724 (m). MS (EI) (70 eV) m/z (%): 293.3 (50) N(CH2)2N), 3.93 (s, 2H, Ar-CH2N), 6.65 (s, 1H, Ar-H), 6.86 (s, 1H,
[M]+, 248.2 (30) [M-CH2OCH3]+, 219.3 (100) [M-NH(CH2)2OCH3]+. Ar-H). 13C NMR (75 MHz, chloroform-d, 298 K, ppm) d 15.71 (Ar-CH3),
1H NMR (300 MHz, chloroform-d, 298 K, ppm) d 2.21 (s, 3H,
3
3
c
New J. Chem.
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
NJC
Paper
20.48 (Ar-CH3), 45.49 (2ꢂ N(CH3)2), 45.87 (N(CH2)2N), 52.64 45.96 (N(CH2)2N), 47.93 (N(CH3)3), 50.76 (N(CH3)3), 54.29
(Ar-CH2N), 58.31 (N(CH2)2N), 121.70 (Ar), 124.82 (Ar), 126.42 (Ar-CH2N), 58.77 (N(CH2)2N), 120.43 (Ar), 122.97 (Ar-H), 123.92
(Ar-H), 127.37 (Ar), 130.39 (Ar-H), 154.07 (Ar-OH). The protons of (Ar-H), 137.80 (Ar), 142.70 (Ar), 157.73 (Ar-O). IR (ATR, cmꢀ1):
the NH and OH group were not detected in benzene-d6 or in n 1439 (w), 1241 (m), 918 (MoQO, m), 880 (MoQO, s),
chloroform-d. IR (ATR, cmꢀ1): n 2856 (m), 1612 (w, NH), 1482 (s), 842 (s), 711 (Mo-(m-O), s, br), 547 (s). Found: C, 50.02; H, 6.34;
1245 (s), 1156 (m), 1039 (m), 856 (s), 774 (s), 735 (m). MS (EI) N, 7.32%; calcd for Mo2O7N4C38H66ꢁ0.49 CH2Cl2: C, 50.00; H,
(70 eV) m/z (%): 222.2 (27) [M]+, 163.1 (100) [M-(CH2N(CH3)2)-H]+, 6.06; N, 7.30%.
135.2 (38) [M-NH(CH2)2N(CH3)2]+. Anal. calcd for C13H22N2O
(222.3): C 70.23, H 9.97, N 12.60%; found: C 70.50, H 10.52, general procedure described above. A solution of the ligand
[{MoO2(L4)}2(l-O)] (3). Synthesis of complex 3 followed the
N 12.36%.
HL3 (0.22 g, 1 mmol) in methanol (1 ml) was added to a solu-
tion of [MoO2(acac)2] (0.33 g, 1 mmol) in methanol (3 ml). After
purification, complex 3 was obtained as orange solid. Yield:
General procedure for the synthesis of the complexes
The appropriate ligand (1 mmol, 1 equiv.) was dissolved in 0.39 mg (88%). Crystals suitable for X-ray diffraction analyses
methanol (3 ml) and slowly added to a solution of 326 mg were obtained from concentrated dichloromethane solutions at
[MoO2(acac)2] (1 mmol, 1 equiv.) in methanol (3 ml). Distilled ꢀ30 1C.
water (0.36 ml, 2 mmol, 2 equiv.) was added via syringe. The
1H NMR (300 MHz, chloroform-d, 298 K, ppm) d 2.22 (s, 3H,
solution was stirred over-night at 25 1C. The resulting precipitate Ar-CH3), 2.23 (s, 3H, Ar-CH3), 2.32 (dd, J = 13.0, 2.6 Hz, 1H,
was filtered and washed twice with cold methanol (2 ꢂ 3 ml). N(CH2)2N), 2.47 (s, 3H, N(CH3)2), 2.63 (dt, J = 13.6, 10.9, 3.0 Hz,
The complex was dissolved in CH2Cl2 (5 ml) and filtered over a 1H, N(CH2)N), 2.76 (m, 1H, N(CH2)2N), 2.84 (s, 3H, N(CH3)2),
pad of Celite. The pure complex was dried under vacuum.
3.11 (dt, J = 13.0, 3.5 Hz, 1H, N(CH2)N), 3.85 (dd, J = 14.7, 1.9 Hz,
[{MoO2(L1)}2(l-O)] (1). Synthesis of complex 1 followed the 1H, Ar-CH2N), 4.75 (dd, J = 14.6, 3.0 Hz, 1H, Ar-CH2N), 4.83
general procedure described above. A solution of the ligand (m, 1H, Ar-CH2NH), 6.67 (s, 1H, Ar-H), 6.91 (s, 1H, Ar-H).
HL1 (0.29 g, 1 mmol) in methanol (1 ml) was added to a solu- 13C NMR (75 MHz, chloroform-d, 298 K, ppm) d 16.78
tion of [MoO2(acac)2] (0.33 g, 1 mmol) in methanol (3 ml). After (Ar-CH3), 20.62 (Ar-CH3), 44.58 (N(CH2)2N), 46.66 (N(CH3)2),
purification, complex 1 was obtained as yellow solid. Yield: 50.85 (N(CH3)2), 53.46 (Ar-CH2N), 58.54 (N(CH2)2N), 120.63
0.68 g (79%).
(Ar), 127.26 (Ar), 127.63 (Ar-H), 129.62 (Ar), 130.63 (Ar-H),
1H NMR (300 MHz, chloroform-d, 298 K, ppm) d 1.29 (s, 3H, 155.69 (Ar-O). IR (ATR, cmꢀ1): n 1475 (m), 1231 (m), 1161 (m),
C(CH3)3), 1.44 (s, 3H, C(CH3)3), 2.77–2.98 (m, 2H, N(CH2)2O), 912 (MoQO, s), 878 (MoQO, s), 828 (s), 710 (Mo-(m-O), s, br),
3.58 (dd, J = 9.5, 2.9 Hz, 1H, N(CH2)2O), 3.88 (s, 1H, OCH3), 3.93 589 (m), 536 (s). Found: C, 43.26; H, 5.78; N, 7.68%; calcd for
(m, 1H, Ar-CH2N), 4.06 (dt, J = 12.4, 9.5, 3.0 Hz, 1H, N(CH2)2O), Mo2O7N4C26H42ꢁ0.21 CH2Cl2: C, 43.00; H, 5.84; N, 7.65%.
4.77 (m, 2H, Ar-CH2NH), 6.87 (d, J = 2.5 Hz, 1H, Ar-H), 7.28
(d, J = 2.4 Hz, 1H, Ar-H). 13C NMR (75 MHz, chloroform-d,
Epoxidation of alkenes
298 K, ppm) d 29.79 (Ar-C(CH3)3), 31.75 (Ar-C(CH3)3), 34.47 In a typical epoxidation reaction, the catalyst (7.05 ꢂ 10ꢀ3 mmol,
(Ar-C(CH3)3), 35.28 (Ar-C(CH3)3), 45.98 (N(CH2)2O), 54.09 0.5 mol%), the corresponding alkene (1.41 mmol, 1 equiv.) and
(Ar-CH2N), 62.42 (OCH3), 70.74 (N(CH2)2O), 120.31 (Ar), 123.47 internal standard (1.41 mmol, 1 equiv.) were combined in
(Ar-H), 124.19 (Ar-H), 138.21 (Ar), 142.98 (Ar), 155.65 (Ar-O). IR chloroform (5 ml). After stirring the mixture for 5 minutes, the
(ATR, cmꢀ1): n 1468 (w), 1442 (w), 1244 (m), 1055 (m), 933 (m), epoxide reaction was started with the addition of TBHP (0.5 ml of
919 (m), 897 (MoQO, s), 885 (MoQO, s), 854 (s), 739 (Mo-(m-O), a 5.5 M solution in decane, 2.82 mmol, 2 equiv.). The reactions
s, br), 554 (m). Found: C, 48.23; H, 6.78; N, 3.39%; calcd for were monitored quantitatively by GC-MS (cyclooctene, cyclo-
Mo2O9N2C36H60ꢁ0.74 CH2Cl2: C, 48.00; H, 6.74; N, 3.05%.
hexene, styrene, a-terpineol, R-(+)-limonene), HPLC (cis-stilbene,
[{MoO2(L2)}2(l-O)] (2). Synthesis of complex 2 followed the trans-stilbene) or 1H NMR (1-octene and 2-propenol) analyses.
general procedure described above. A solution of the ligand At fixed intervals samples were taken and residual TBHP traces
HL2 (0.31 g, 1 mmol) in methanol (1 ml) was added to a solu- were quenched with MnO2. After centrifugation, sample aliquots
tion of [MoO2(acac)2] (0.33 g, 1 mmol) in methanol (3 ml). After were diluted with ethyl acetate (for GC-MS) or acetonitrile (for
purification, complex 2 was obtained as light red solid. Yield: HPLC). Mesitylene was used as internal standard for GC-MS
1
0.85 g (96%). Crystals suitable for X-ray diffraction analyses and HPLC measurements. H NMR spectra were measured in
were obtained from diluted acetonitrile solutions at room chloroform-d using dichloroethane as internal standard.
temperature.
1H NMR (300 MHz, chloroform-d, 298 K, ppm) d 1.27 (s, 3H,
X-ray structure determination
C(CH3)3), 1.47 (s, 3H, C(CH3)3), 2.39 (d, J = 13.0 Hz, 1H, For X-ray structure analyses the crystals were mounted on the
N(CH2)2N), 2.59 (s, 3H, N(CH3)2), 2.80 (m, 2H, N(CH2)2N), 2.86 tip of glass fibres and data collection was performed at 100 K
(s, 3H, N(CH3)2), 3.12 (m, 1H, N(CH2)2N), 3.95 (d, J = 14.8 Hz,1H, using graphite monochromated MoKa radiation (l = 0.71073 Å)
Ar-CH2N), 4.75 (d, J = 14.5 Hz, overlapping signals, 2H, with a BRUKER-AXS SMART APEX II diffractometer equipped
Ar-CH2NH), 6.82 (s, 1H, Ar-H), 7.25 (s, 1H, Ar-H). 13C NMR with a CCD detector. The structures were solved by direct
(75 MHz, chloroform-d, 298 K, ppm) d 30.44 (Ar-C(CH3)3), methods (SHELXS-97 (ref. 34) or SIR92 (ref. 35)) and refined by
31.73 (Ar-C(CH3)3), 34.40 (Ar-C(CH3)3), 35.38 (Ar-C(CH3)3), full-matrix least-squares techniques against F2 (SHELXL-97).36
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem.

View Article Online
Paper
NJC
(m) Y. Sui, X. Zeng, X. Fang, X. Fu, Y. Xiao, L. Chen, M. Li
and S. Cheng, J. Mol. Catal. A: Chem., 2007, 270, 61–67;
(n) A. Gu¨nyar, M.-D. Zhou, M. Drees, P. N. W. Baxter,
G. Bassioni, E. Herdtweck and F. E. Ku¨hn, Dalton Trans.,
2009, 8746–8754; (o) M. Bagherzadeh, L. Tahsini, R. Latifi
and L. K. Woo, Inorg. Chim. Acta, 2009, 362, 3698–3702.
5 S. M. Bruno, S. S. Balula, A. A. Valente, F. A. Almeida Paz,
M. Pillinger, C. Sousa, J. Klinowski, C. Freire, P. Ribeiro-
Claro and I. S. Gonçalves, J. Mol. Catal. A: Chem., 2007, 270,
185–194.
If not otherwise stated, all non-hydrogen atoms were refined
with anisotropic displacement parameters without any con-
straints. All hydrogen atoms were fixed in calculated positions
to correspond to standard bond lengths and angles. All diagrams
were drawn with 50% probability thermal ellipsoids, and all
H atoms, except for the H atom of the NH moiety, were omitted
for clarity. Crystallographic data (excluding structure factors)
for the structures of compounds 2 and 3 are reported.†
Acknowledgements
´
6 J. A. Brito, G. Muller, S. Massou and M. Gomez, Inorg. Chim.
Financial support from the Austrian Science Foundation FWF
(P19309-N19) is gratefully acknowledged.
Acta, 2008, 361, 2740–2746.
7 S. Gago, P. Neves, B. Monteiro, M. Pessego, A. D. Lopes,
ˆ
A. A. Valente, F. A. Almeida Paz, M. Pillinger, J. Moreira,
C. M. Silva and I. S. Gonçalves, Eur. J. Inorg. Chem., 2009,
4528–4537.
References
´
8 P. Neves, S. Gago, C. Pereira, S. Figueiredo, A. Lemos,
A. Lopes, I. Gonçalves, M. Pillinger, C. Silva and
A. Valente, Catal. Lett., 2009, 132, 94–103.
9 A. C. Coelho, M. Nolasco, S. S. Balula, M. M. Antunes,
C. C. L. Pereira, F. A. Almeida Paz, A. A. Valente, M. Pillinger,
P. Ribeiro-Claro, J. Klinowski and I. S. Gonçalves, Inorg. Chem.,
2011, 50, 525–538.
10 C. A. Gamelas, A. C. Gomes, S. M. Bruno, F. A. Almeida Paz,
A. A. Valente, M. Pillinger, C. C. Romao and I. S. Gonçalves,
Dalton Trans., 2012, 41, 3474–3484.
1 J. M. Bregeault, J. Chem. Soc., Dalton Trans., 2003,
3289–3302.
2 (a) K. C. Gupta and A. K. Sutar, Coord. Chem. Rev., 2008,
252, 1420–1450; (b) Q.-H. Xia, H.-Q. Ge, C.-P. Ye, Z.-M. Liu
and K.-X. Su, Chem. Rev., 2005, 105, 1603–1662;
(c) D. Sherrington, Catal. Today, 2000, 57, 87–104;
(d) K. C. Gupta, A. K. Sutar and C. C. Lin, Coord. Chem.
Rev., 2009, 253, 1926–1946; (e) B. S. Lane and K. Burgess,
Chem. Rev., 2003, 103, 2457–2474; ( f ) K. A. Jørgensen, Chem.
Rev., 1989, 89, 431–458; (g) C.-M. Che and J.-S. Huang,
Coord. Chem. Rev., 2003, 242, 97–113; (h) S. Liu and
J. Xiao, J. Mol. Catal. A: Chem., 2007, 270, 1–43.
11 J. A. Schachner, P. Traar, C. Sala, M. Melcher, B. N. Harum,
¨
A. F. Sax, M. Volpe, F. Belaj and N. C. Mosch-Zanetti, Inorg.
Chem., 2012, 51, 7642–7649.
3 M. Abrantes, A. M. Santos, J. Mink, F. E. Ku¨hn and
˜
12 J. M. Mitchell and N. S. Finney, J. Am. Chem. Soc., 2001, 123,
C. C. Romao, Organometallics, 2003, 22, 2112–2118.
862–869.
4 (a) A. Gu¨nyar, D. Betz, M. Drees, E. Herdtweck and
F. E. Ku¨hn, J. Mol. Catal. A: Chem., 2010, 331, 117–124;
(b) Y. L. Wong, L. H. Tong, J. R. Dilworth, D. K. P. Ng and
H. K. Lee, Dalton Trans., 2010, 39, 4602–4611;
(c) A. Rezaeifard, I. Sheikhshoaie, N. Monadi and
13 Y.-L. Wong, D. K. P. Ng and H. K. Lee, Inorg. Chem., 2002, 41,
5276–5285.
´
14 J. A. Brito, M. Gomez, G. Muller, H. Teruel, J.-C. Clinet,
D. Elisabet and M. A. Maestro, Eur. J. Inorg. Chem., 2004,
M.
Alipour,
Polyhedron,
2010,
29,
2703–2709;
4278–4285.
(d) A. Rezaeifard, I. Sheikhshoaie, N. Monadi and 15 C. C. L. Pereira, S. S. Balula, F. A. Almeida Paz, A. A. Valente,
H. Stoeckli-Evans, Eur. J. Inorg. Chem., 2010, 799–806;
(e) P. Neves, S. Gago, S. S. Balula, A. D. Lopes,
M. Pillinger, J. Klinowski and I. S. Gonçalves, Inorg. Chem.,
2007, 46, 8508–8510.
´
A. A. Valente, L. Cunha-Silva, F. A. A. Paz, M. Pillinger, 16 J. A. Brito, N. Saffon, M. Gomez and B. Royo, Curr. Inorg.
J. Rocha, C. M. Silva and I. S. Gonçalves, Inorg. Chem., 2011,
Chem., 2011, 1, 131–139.
50, 3490–3500; ( f ) F. Madeira, S. Barroso, S. Namorado, 17 J. Morlot, N. Uyttebroeck, D. Agustin and R. Poli,
P. M. Reis, B. Royo and A. M. Martins, Inorg. Chim. Acta,
ChemCatChem, 2013, 5, 601–611.
2012, 383, 152–156; (g) G. Chahboun, J. A. Brito, B. Royo, 18 W. A. Herrmann, R. W. Fischer and D. W. Marz, Angew.
´
M. A. El Amrani, E. Gomez-Bengoa, M. E. G. Mosquera,
T. Cuenca and E. Royo, Eur. J. Inorg. Chem., 2012, 19 W. A. Herrmann, R. W. Fischer, M. U. Rauch and
2940–2949; (h) J. Pisk, D. Agustin, V. Vrdoljak and R. Poli,
W. Scherer, J. Mol. Catal., 1994, 86, 243–266.
Adv. Synth. Catal., 2011, 353, 2910–2914; (i) P. M. Reis, 20 K. Jeyakumar and D. K. Chand, Synthesis, 2008, 807–819.
Chem., Int. Ed. Engl., 1991, 30, 1638–1641.
´
¨
C. A. Gamelas, J. A. Brito, N. Saffon, M. Gomez and 21 M. E. Judmaier, C. Holzer, M. Volpe and N. C. Mosch-
B. Royo, Eur. J. Inorg. Chem., 2011, 666–673;
Zanetti, Inorg. Chem., 2012, 51, 9956–9966.
( j) F. Romano, A. Linden, M. Mba, C. Zonta and G. Licini, 22 L. Shechter, J. Wynstra and R. Kurkjy, Ind. Eng. Chem., 1957,
Adv. Synth. Catal., 2010, 352, 2937–2942; (k) S. M. Bruno, 49, 1107–1109.
C. C. L. Pereira, M. S. Balula, M. Nolasco, A. A. Valente, 23 (a) B. Terfassa, P. Traar, M. Volpe, N. C. Mosch-Zanetti,
¨
A. Hazell, M. Pillinger, P. Ribeiro-Claro and I. S. Gonçalves,
J. Mol. Catal. A: Chem., 2007, 261, 79–87; (l) F. E. Ku¨hn, A. M.
Santos and M. Abrantes, Chem. Rev., 2006, 106, 2455–2475;
V. J. T. Raju, N. Megersa and N. Retta, Eur. J. Inorg. Chem.,
2011, 4434–4440; (b) P. Traar, J. A. Schachner, L. Steiner,
¨
A. Sachse, M. Volpe and N. C. Mosch-Zanetti, Inorg. Chem.,
c
New J. Chem.
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
NJC
Paper
2011, 50, 1983–1990; (c) B. Machura, M. Wolff, D. Tabak, 29 W. A. Herrmann and M. Wang, Angew. Chem., Int. Ed. Engl.,
¨
J. A. Schachner and N. C. Mosch-Zanetti, Eur. J. Inorg. Chem.,
1991, 30, 1641–1643.
´
2012, 3764–3773.
30 A. Comas-Vives, A. Lledos and R. Poli, Chem.–Eur. J., 2010,
24 (a) Y. L. Wong, E. S. H. Chan, Q. Yang, T. C. Mak and
16, 2147–2158.
ˆ
˜
D. K. P. Ng, J. Chem. Soc., Dalton Trans., 1998, 3057–3064; 31 L. F. Veiros, A. Prazeres, P. J. Costa, C. C. Romao, F. E. Ku¨hn
´
(b) C. J. Whiteoak, G. J. P. Britovsek, V. C. Gibson and
A. J. P. White, Dalton Trans., 2009, 2337–2344.
and M. Jose Calhorda, Dalton Trans., 2006, 1383.
32 P. J. Costa, M. J. Calhorda and F. E. Ku¨hn, Organometallics,
2010, 29, 303–311.
¨
¨
25 K. Most, S. Kopke, F. Dall’Antonia and N. C. Mosch-Zanetti,
Chem. Commun., 2002, 1676–1677.
26 M. E. Judmaier, A. Wallner, G. N. Stipicic, K. Kirchner,
33 P. D. Knight, P. N. O’Shaughnessy, I. J. Munslow,
B. S. Kimberley and P. Scott, J. Organomet. Chem., 2003,
683, 103–113.
¨
J. Baumgartner, F. Belaj and N. C. Mosch-Zanetti, Inorg.
Chem., 2009, 48, 10211–10221.
34 G. M. Sheldrick, SHELXS, 97, Program or Structure Solution,
ˇ
ˇ
¨
27 N. Kitanovski, A. Golobic and B. Ceh, Inorg. Chem. Commun.,
2006, 9, 296–299.
University of Gottingen, Germany, 1997.
35 A. Altomare, G. Cascarano, C. Giacovazzo and A. Gualardi,
28 A. Al-Ajlouni, A. A. Valente, C. D. Nunes, M. Pillinger,
J. Appl. Crystallogr., 1993, 26, 343–350.
˜
A. M. Santos, J. Zhao, C. C. Romao, I. S. Gonçalves and 36 G. M. Sheldrick, SHELXL-97, Program for Crystal Structure
¨
Refinement, University of Gottingen, Germany, 1997.
F. E. Ku¨hn, Eur. J. Inorg. Chem., 2005, 1716–1723.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem.