Mechanistic studies on the roles of the oxidant and hydrogen bonding in determining the selectivity in alkene oxidation in the presence of molybdenum catalysts

Article abstract of DOI:10.1002/chem.201202597

When the molybdenum oxo(peroxo) acetylide complex [CpMo(O-O)(O)C≡CPh] is used as a catalyst for the oxidation of olefins, completely different product selectivity is obtained depending on the oxidant employed. When tert-butyl hydroperoxide (TBHP, 5.5 M) in dodecane is used as the oxidant for the oxidation of cyclohexene, cyclohexene oxide is formed with high selectivity. However, when H₂O₂ is used as the oxidant, the corresponding cis-1,2-diol is formed as the major product. Calculations performed by using density functional theory revealed the nature of the different competing mechanisms operating during the catalysis process and also provided an insight into the influence of the oxidant and hydrogen bonding on the catalysis process. The mechanistic investigations can therefore serve as a guide in the design of molybdenum-based catalysts for the oxidation of olefins. Copyright

Full text of DOI:10.1002/chem.201202597

FULL PAPER
DOI: 10.1002/chem.201202597
Mechanistic Studies on the Roles of the Oxidant and Hydrogen Bonding in
Determining the Selectivity in Alkene Oxidation in the Presence of
Molybdenum Catalysts
[
a]
[a]
[a]
Prakash Chandra, Swati L. Pandhare, Shubhangi B. Umbarkar,*
[
a]
[b]
Mohan K. Dongare,* and Kumar Vanka*
Abstract: When the molybdenum oxo-
(peroxo) acetylide complex [CpMo
selectivity. However, when H O is
vealed the nature of the different com-
peting mechanisms operating during
the catalysis process and also provided
an insight into the influence of the oxi-
dant and hydrogen bonding on the cat-
alysis process. The mechanistic investi-
gations can therefore serve as a guide
in the design of molybdenum-based
catalysts for the oxidation of olefins.
2
2
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
A
H
U
G
R
N
U
G
used as the oxidant, the corresponding
cis-1,2-diol is formed as the major
product. Calculations performed by
using density functional theory re-
O)(O)CꢀCPh] is used as a catalyst for
the oxidation of olefins, completely dif-
ferent product selectivity is obtained
depending on the oxidant employed.
When tert-butyl hydroperoxide (TBHP,
Keywords: epoxides · homogene-
ous catalysis · molybdenum · ole-
fins · oxidation
5.5 m) in dodecane is used as the oxi-
dant for the oxidation of cyclohexene,
cyclohexene oxide is formed with high
Introduction
What is of interest in the epoxidation catalysis process is
the mechanism of the reaction, which has been the subject
of much debate. Assuming the presence of a peroxido
The conversion of olefins into epoxides is an extremely im-
portant reaction because it allows the conversion of availa-
ble crude oil fractions into epoxides, which are important or-
ganic intermediates for the production of a wide variety of
chemicals, for example, drugs, agrochemicals and food
additives. They also play an important role in biological ac-
Although epoxidation can take place in the ab-
sence of a catalyst, considerable attention has been focused
in recent years on the metal-catalysed epoxidation process
due, in part, to the increased possibility of achieving higher
activity and selectivity by this route. Among the oxidants
employed in the epoxidation process, the one most studied
[17]
ligand at the metal centre, Mimoun et al. suggested a [2+
3] cycloaddition route for this process involving the reaction
between the olefin and the metal–peroxo species (see Fig-
[
1–4]
[18]
ure 1a). Sharpless et al. suggested the possibility of exoge-
nous attack of the olefin at the electrophilic peroxo oxygen
atom (see Figure 1b). Both these proposed mechanistic
routes assume that the peroxido ligand attached to the mo-
lybdenum centre is the source of the oxygen atom in the
[
2,5–7]
tivity.
[8]
AHCTUNGTRENNUNG
[15,19–29]
through the Sharpless and Mimoun pathways.
Of the
[9]
is H O2 because of its low cost and ready availability. The
transition-metal-based complexes that have been studied as
two mechanisms, the Mimoun pathway was found to have a
2
[23]
higher barrier than the Sharpless mechanism. Experimen-
[10,11]
[30]
epoxidation catalysts include complexes of iron,
manga-
Of these, the
compounds most studied are the high-oxidation-state com-
tal studies have confirmed this view.
However, recent
[
12,13]
[14]
[15,16]
[15]
nese,
tungsten
and molybdenum.
work by Galindo and co-workers has shown that a molyb-
denum diperoxo species is a stable intermediate during the
epoxidation of cis-cyclooctene with H O as the oxidant,
[16]
plexes of molybdenum.
2
2
which indicates that the Sharpless mechanism cannot be en-
tirely ruled out. However, it appears that a mechanism in
which the epoxide oxygen is derived from H O or tert-butyl
[
a] P. Chandra, S. L. Pandhare, Dr. S. B. Umbarkar, Dr. M. K. Dongare
CSIR-NCL Catalysis and Inorganic Chemistry Division
Dr. Homi Bhabha Road. Pashan, Pune, Maharastra (India)
E-mail: sb.umbarkar@ncl.res.in
2
2
hydroperoxide (TBHP) rather than from the oxido group on
molybdenum would be at least competitive, if not operative,
during the catalysis. Such a mechanism, shown in Figure 1c,
mk.dongare@ncl.res.in
[31,32]
[
b] Dr. K. Vanka
was proposed by Thiel and co-workers.
This pathway in-
Physical Chemistry Division CSIR-NCL
Dr. Homi Bhabha Road, Pashan, Pune, Maharashtra (India)
Fax : (+20)2590-2636
volves the coordination of the oxidant to the molybdenum
centre followed by the transfer of the oxygen from the oxi-
dant ligand to the olefin (see Figure 1c). Finally, a mecha-
nism that involves the insertion of the olefin into the MÀO
E-mail: k.vanka@ncl.res.in
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201202597.
[33]
bond was proposed by Calhorda and co-workers.
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

Figure 2. Experimentally observed change in the selectivity towards the
epoxide and cis-1,2-diol in the oxidation of cyclohexene with TBHP or
hydrogen peroxide as oxidant in the presence of a molybdenum catalyst.
[39]
calculations of Poli and co-workers in their investigations
of the epoxidation of cyclooctene with H O as oxidant.
2
2
The observed differences in selectivity when using differ-
ent oxidants suggest that there are competing mechanisms
operating during the catalysis process that lead to the for-
mation of either the cis-1,2-diol or the epoxide as the major
product. It is necessary to understand the nature of these
competing mechanisms because such an understanding can
help in the design of systems for specifically obtaining the
epoxide or the cis-1,2-diol as the major product, which was
the objective of this work. Detailed quantum mechanical
computational studies were performed on the following cat-
Figure 1. The different mechanisms proposed in the literature for the
olefin epoxidation reaction.
[34]
We have recently reported the experimental results of a
study of the oxidation of cyclohexene in the presence of the
alytic systems by using DFT: [CpMoO(O) CꢀCPh] with cy-
2
peroxo species [CpMo(O)
A
H
U
G
R
N
U
G
A
H
U
G
R
N
N
(CꢀCPh)] as catalyst and
clohexene as the substrate and H O or organic TBHP in
2
2
H O as the oxidant. This was the first reported case of the
dodecane as the oxidant. Three of the four mechanistic pos-
sibilities shown in Figure 1 were considered: the Mimoun
2
2
formation of a cis-1,2-diol by using a molybdenum-based
catalyst. Because cis-1,2-diols are important organic reagents
mechanism was not considered because it has been compu-
[
35–38]
[23]
in the pharmaceutical industry,
CpMo(O)
(OÀO)
the selectivity of the
tationally shown to be a high-barrier process.
Further-
[
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(CꢀCPh)] catalyst in forming cis-1,2-diols
more, a new mechanism was proposed to account for the
formation of the cis-1,2-diol product. It has been shown that
the presence of water has a significant influence on the se-
lectivity of the catalyst. It has also been shown that changing
the oxidant from H O to TBHP leads to a change in the
was an important result. However, as will be discussed in
the Results and Discussion section below, when, as an alter-
native to H O , organic TBHP in dodecane was used as the
2
2
oxidant, the following interesting results were obtained: the
oxidation of cyclohexene led to the formation of the cis-1,2-
diol rather than the epoxide when H O was used as the oxi-
2
2
preference for one mechanism over another and thus in the
eventual product (epoxide in the case of organic TBHP and
cis-1,2-diol in the case of H O as the oxidant). The results
2
2
dant, whereas the epoxide rather than the cis-1,2-diol was
formed as the major product when organic TBHP in dodec-
ane was used as the oxidant (Figure 2). It is clear that the
selectivity shown by the catalyst is significantly dependent
on the nature of the oxidant employed. Moreover, it is pos-
sible that the presence of water can also have an effect on
the energy barriers, as has been shown by the results of the
2
2
therefore shed light on the mechanistic possibilities of the
important epoxide and cis-1,2-diol forming processes, and
also provide an insight into how one could tune molybde-
num-based catalytic systems through the choice of oxidant
to obtain, as desired, the epoxide or the cis-1,2-diol as the
major product.
&
2
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Alkene Oxidation with Molybdenum Catalysts
FULL PAPER
In the first sub-section of the Results and Discussion sec-
tion, we discuss the experimental evidence for the formation
of the peroxo species upon addition of H O or TBHP
Information). The bands due to carbonyl stretching (1940
À1
and 2031 cm ) disappeared after the addition of H O . This
2
2
clearly indicates the loss of all the CO ligands after the addi-
tion of hydrogen peroxide with the formation of the higher
oxidation state molybdenum(VI) complex with retention of
the acetylide CꢀCPh group attached to the molybdenum
2
2
under the given reaction conditions, as well as the aforemen-
tioned observed selectivity for the epoxide when using
TBHP as the oxidant and the cis-1,2-diol when H O was
2
2
employed as the oxidant. In the following sub-sections we
discuss the computational studies of the different mecha-
nisms that are possible for the different systems and the re-
sults that corroborate the experimental findings.
centre.
The addition of TBHP (0.02 mol) to the catalyst results in
the formation of the oxo ACHUTNGRNENGU( peroxo) species, which is clearly in-
À1
dicated by the peaks in the range of 930–950 cm due to
the Mo=O bond and the broad peaks in the range of 840–
À1
8
60 cm due to the OÀO bond (Figure 4).
Results and Discussion
Experimental evidence for the formation of the peroxo spe-
cies of the acetylide-ligated molybdenum complexes: Exper-
imental evidence was found for the formation of the peroxo
molybdenum species [CpMo(O) OCꢀCPh] during the cata-
2
lytic oxidation of cyclohexene. [CpMo(O) OCꢀCPh] was ob-
2
tained from [CpMo(CO) CꢀCPh], which was prepared by
3
[40–42]
the previously reported procedure.
The FTIR spectrum
of the catalyst after the addition of H O confirmed the for-
2
2
mation of the (oxo)peroxo molybdenum species (Figure 3).
Figure 4. IR spectrum showing evidence for the formation of the oxo-
(peroxo)–molybdenum species after treatment of cyclohexene with
TBHP in dodecane in the presence of [CpMo(CO)
CꢀCPh] as catalyst.
AHCTUNGTRENNUNG
3
Cyclohexene oxidation experiments conducted with
[
CpMo(CO) CꢀCPh] and TBHP as oxidant and 1,2-di-
3
chloroethane as solvent led to the formation of the epoxide
with a selectivity as high as 79.2% within 12 h (Table 1). 1,2-
Figure 3. IR spectrum showing evidence for the formation of the oxo-
3
Table 1. Oxidation of cyclohexene with [CpMo(CO) AHCTUNGTERNNUNG
lyst.
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(peroxo) species after treatment of cyclohexene with H
2
O
2
in the pres-
[a]
ence of [CpMo(CO)
3
CꢀCPh] as catalyst.
Oxidant
Solvent
t
Conv. [%]
[
h]
Others
À1
[
b]
The peak at 953 cm shows the presence of a Mo=O termi-
TBHP
C H Cl
12 53
95
79.2
–
14.4
À91
6.4
9
2
4
2
À1
[c]
nal bond. The peak in the range of 930–950 cm corre-
H
2
O
2
(50%) tert-butyl
9
alcohol
sponds to the OÀO stretching vibration of the peroxo spe-
À1
cies. The weak bands at 668 and 573 cm can be assigned to
[a] Reagents and conditions: cyclohexene (0.02 mol), catalyst
0.01 mmol), oxidants H (0.04 mol) and TBHP (0.02 mol), 808C. [b]
trans-Diol. [c] cis-Diol.
(
2 2
O
MoÀO (peroxo) asymmetric and symmetric stretching vi-
2
brations, respectively. The IR peaks for the MoÀoxo and
Àperoxo bonds are in good agreement with the values re-
[34,41,42]
ported in the literature
for various molybdenum
Dichloroethane was selected as the solvent to eliminate hy-
drogen-bonding effects observed with protic solvents such as
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(oxo)peroxo complexes. The bands due to CÀH stretching
[34]
vibrations of the phenyl ring are observed in the range
tert-butanol. It has previously been shown
that the Cꢀ
À1
2
854–2955 cm and the C=C stretching vibrations of the
ring are observed at 1464 and 1377 cm , even after the ad-
dition of hydrogen peroxide (see Figure 1 in the Supporting
CPh-substituted molybdenum–peroxo catalyst gives cis-1,2-
diol with 91% conversion and 95% selectivity towards cis-
dihydroxylation when tert-butanol is employed as the sol-
À1
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

S. B. Umbarkar, M. K. Dongare, K. Vanka et al.
[34]
vent.
The following sections will focus attention on the
computational studies of the effect of changing the oxidant
on the selectivity of the catalysis process.
Importance of hydrogen bonding on the oxidation reactions:
Sharpless mechanism: As stated in the Introduction, the fact
that the use of [CpMo(CO) CꢀCPh] as the catalyst produces
3
the epoxide as the major product with TBHP as the oxidant,
[34]
whereas it produces the cis-1,2-diol as the major product
when 50% H O was used, indicates that there are compet-
2
2
ing mechanisms at work in the homogeneous catalysis sys-
tems. The possible mechanisms for the oxidation are illus-
trated in Figures 5–7. In the “Sharpless” mechanism, also
discussed in the Introduction (see Figure 1b), the catalyst is
approached directly by the cyclohexene (Figure 5). It is well
[
18]
established that [CpMo(CO) CꢀCPh] acts as the precur-
3
sor for [CpMo
Figure 5 b-1 can be converted into the corresponding dioxo
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
(OÀO)(O)(CꢀCPh)] (b-1). As indicated in
A
H
U
G
E
N
N
Figure 6. Potential energy surfaces for the first step of the Thiels mecha-
nism: approach of the oxidants TBHP and H
num catalyst. Energy values (DG) are given in kcalmol
2
O
2
towards the molybde-
À1
.
intermediate b-2 and in the process yield the epoxide, de-
noted as a-2. The activation barriers (DG) were determined
À1
to be 26.2 and 28.5 kcalmol for the reaction in tert-butanol
and 1,2-dichloroethane, respectively. For the case of aqueous
H O as the oxidant (with tert-butanol as the solvent), there
2
2
is the possibility that water molecules present in solution
during the reaction, as well as the H O oxidant itself, can
2
2
form a hydrogen bond with the [CpMo
A
H
U
G
R
N
U
G
A
H
U
G
R
N
U
catalyst molecule, and thereby affect the energy barrier for
this epoxidation reaction. These possibilities were consid-
ered in the epoxidation reaction with one and two water
molecules coordinated through hydrogen bonds to the
[
CpMo
A
H
U
G
R
N
U
G
A
H
U
G
R
N
U
G
water molecule was found to lower the energy barrier to
À1
2
2.3 kcalmol (see Figure 7 (ts-1/1) in the Supporting Infor-
mation), whereas the presence of two water molecules, as
shown in Figure 5 (t-2), was found to lead to a barrier of
À1
2
4.3 kcalmol for this epoxidation reaction. These results
therefore indicate that the presence of water molecules does
have an effect on the kinetics of the epoxidation process.
Similar calculations to show the involvement of hydrogen-
bonded H O in the “Sharpless mechanism” revealed a
lower energy barrier of 24.7 kcalmol (see Figure 9 (t-4/1)
in the Supporting Information). Hence, the results indicate
Figure 5. Potential energy surfaces for the first step of the “Sharpless
mechanism” for the oxidation of cyclohexene in the presence of the mo-
lybdenum catalyst [CpMo(O)(OO) ACHUTNRGNENUG( CCPh)]. The influence of hydrogen
2
2
À1
bonding by water and TBHP is considered. Energy values outside of pa-
that the interaction, through hydrogen bonding, of one H O2
rentheses are values of DG in tert-butanol and those in parentheses are
2
À1
values of DG in 1,2-dichloroethane. All values are given in kcalmol
.
or two H O molecules shows the same lowering of the
2
&
4
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Alkene Oxidation with Molybdenum Catalysts
FULL PAPER
of the catalyst, however, the results show that the two moi-
eties drift apart in the optimised structures. However,
TBHP itself, in addition to acting as an oxidant, can interact
with the catalyst through hydrogen bonding. Such an inter-
action also reduces the energy barrier for the epoxidation
À1
À1
process, from 28.5 kcalmol (see Figure 5) to 25.8 kcalmol
with one TBHP molecule hydrogen-bonded to the catalyst
(
see Figure 9e (t-3/1T) in the Supporting Information), and
À1
to 23.7 kcalmol with two TBHP molecules hydrogen-
bonded to the catalyst (see Figure 5 (t-3/2T)). These results
therefore underline the importance of the hydrogen bonding
of the oxidant and/or water molecules in the oxidation pro-
AHCTUNGTRENNUNGc esses occurring in the presence of molybdenum systems.
Thiel mechanism: The first step of the Thiel mechanism in-
volves the conversion of the [CpMo(O)(OO)CꢀCPh] cata-
lyst complex into [CpMoCꢀCPh(O)
ACTHUNGTRENUNNG( OOH) ACTHUNGTRENNU(GN OOR)] (R=H
H O ) and tert-butyl (TBHP)). In the absence of any hydro-
2 2
(
gen-bonding interactions, the energy barriers are 28.6 kcal
À1
À1
mol for H O as oxidant and 30.5 kcalmol for TBHP as
2
2
oxidant (see Figures 3 (ts-2) and 5 (t-4-H), respectively, in
the Supporting Information). These barriers are reduced
when two hydrogen-bonding water molecules are considered
in the mechanism for the H O oxidant and when two
2
2
TBHP molecules are considered in the mechanism for the
TBHP oxidant. For the case of H O , the barriers are re-
2
2
À1
Figure 7. Potential energy surfaces for the alternative reaction pathway
duced from 28.6 kcalmol (see Figure 3 (ts-2) in the Sup-
porting Information) to 24.7 kcalmol when one water mol-
for the reaction of the molybdenum catalyst [CpMo(O)(OO)
A
H
U
G
R
N
U
G
À1
with the two oxidants TBHP and H
2
O
2
ecule is hydrogen-bonded to the peroxo group of the cata-
À1
kcalmol
.
lyst (see Figure 7 (ts-2/1) in the Supporting Information)
À1
and to 22.7 kcalmol when two water molecules are consid-
ered (see Figure 6 (t-6)). For the case of TBHP, the corre-
À1
energy barrier for the “Sharpless mechanism”. Henceforth,
when considering aqueous H O as the oxidant, the model
sponding barriers are 30.5 kcalmol when no extra TBHP
molecules are employed (see Figure 5 (t-4-H) in the Sup-
2
2
À1
that will be employed that takes into account the effect of
hydrogen bonding will be that involving the presence of two
water molecules hydrogen-bonded to the molybdenum cata-
lyst. Note also that the solvent employed when aqueous
H O was used as the oxidant, tert-butanol, can also form
porting Information), 29.4 kcalmol with the addition of
one extra TBHP molecule (see Figure 9e (t-4/1T) in the
Supporting Information) and 26.5 kcalmol when two extra
TBHP molecules are considered (see Figure 6 (t-4/2T)). The
reason for the sharp decrease in the barrier, from 28.6 to
À1
2
2
À1
hydrogen bonds with the molybdenum catalyst and thus
could also influence the energetics of the oxidation process.
However, it is likely that water would displace tert-butanol
from the molybdenum catalyst because of its superior ability
to form hydrogen bonds with the catalyst. The calculations
indicate that given a choice of forming a hydrogen bond
with tert-butanol or water, the peroxo group of the catalyst
22.7 kcalmol for the case of H O is due to the fact that
2
2
one of the two water molecules that have been considered
performs the task of a “hydrogen atom relay” in the transi-
tion state (t-6): the water molecule accepts a hydrogen atom
from the coordinating H O oxidant species and donates
one of its own hydrogen atoms to the peroxo group of the
molybdenum catalyst (see Figure 6 (t-6)). This leads to the
conversion of the four-membered transition state for the hy-
drogen transfer in the absence of the water molecule to a
sterically more favoured six-membered transition state when
the water molecule is considered.
A similar hydrogen atom relay mechanism also occurs in
the case of TBHP. However, the effect with TBHP is less
pronounced, with the energy barrier decreasing by 6.1 kcal
mol , as compared with a decrease of 4.0 kcalmol in the
case of water.
2
2
À1
shows a preference, by 3.6 kcalmol , for bonding to the
water molecule (see Figure 6a in the Supporting Informa-
tion).
For the case of organic TBHP as the oxidant, then there
are no water molecules present that can contribute to hydro-
gen bonding with the catalyst. Moreover, the solvent that is
used in the oxidation reaction with TBHP, 1,2-dichloro-
ethane, also cannot form hydrogen bonds or participate in
other weak interactions with the catalyst. We attempted to
perform calculations in which the chlorine atoms of 1,2-di-
chloroethane are non-covalently bound to the oxygen atoms
À1
À1
The upshot of the calculations for the important first step
of the competing Sharpless and Thiel mechanisms is this:
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

S. B. Umbarkar, M. K. Dongare, K. Vanka et al.
for the case of aqueous H O , the first step of the Thiel
2
2
mechanism is more favoured than the first step of the
Sharpless mechanism, whereas for the case of TBHP, it is
the other way around (see Figures 5 and 6). This indicates
that when TBHP is employed as the oxidant, the Sharpless
mechanism would be favoured, leading to the formation of
the epoxide as the major product. This corroborates the ex-
perimental findings reported in the previous section. For the
case of H O , the epoxide would not be formed in the first
2
2
step, instead, the species [CpMo(O)
A
T
N
T
N
U
G
A
H
U
G
R
N
U
G
(
b-11) would be preferred. As we will show in a later sec-
tion, this will lead to the formation of the cis-1,2-diol as the
preferred major product.
Figure 8. Potential energy surface for the conversion of the molybdenum
dioxo species [CpMo(O)(O)
ACTHUNGTRENNU(GN CCPh)] into [CpMo(O)(OH) ACHTUNGTRENNUNG( OOtBu)-
Activation of the Mo=O bond: In addition to the two mech-
anisms already considered, the Sharpless and Thiel mecha-
nisms, a third possibility, proposed earlier by Poli and co-
À1
A
H
U
G
E
N
N
[23,39]
workers,
has also been considered. In this mechanism,
Figure 9. This intermolecular rearrangement reaction (see
Figure 9) is made more facile by the aid of a hydrogen-
bonded TBHP species. The barriers to the two steps shown
in Figures 8 and 9 are 24.5 and 23.0 kcalmol , respectively.
Note that the species [CpMo
(b-16), formed as shown in Figure 8, can also undergo attack
by a cyclohexene species leading to further epoxidation, as
has been speculated in the past.
(ts-4) in the Supporting Information, such an attack will be
as in the Thiel mechanism, the oxidant, H O or TBHP, ap-
2
2
proaches the molybdenum catalyst, but unlike in the Thiel
mechanism, the oxidant transfers a hydrogen atom to the
oxygen atom coordinated through a double bond to the mo-
lybdenum centre. As shown in Figures 3 (ts-3) and 5 (t-7-H)
in the Supporting Information, such a reaction has an
À1
A
H
U
G
R
N
N
(CꢀCPh)(O)(OH)
ACHTUNGTRENNUNG( OOtBu)
À1
[33,43]
energy barrier of 29.6 kcalmol for the H O oxidant and
3
As shown in Figure 13
2
2
À1
5.5 kcalmol for the TBHP oxidant. The presence of two
À1
H O molecules (for the case of aqueous H O as oxidant)
less favoured (energy barrier of 33.8 kcalmol ) than the in-
2
2
2
and two extra TBHP molecules (for the case of TBHP as
tramolecular pathway shown in Figure 9. Hence, the most
likely outcome of the oxidation process when using TBHP
as the oxidant will be the Sharpless mechanism, which yields
the epoxide as the major product.
À1
oxidant) reduces the barrier to 23.6 and 27.6 kcalmol , re-
À1
spectively. For the case of H O , the barrier (23.6 kcalmol )
2
2
is higher than for the first step of the Thiel Mechanism,
À1
whereas for the case of TBHP, the barrier (27.6 kcalmol ;
see Figure 7 (t-5/2T)) is higher than the first step of both the
Sharpless and the Thiel mechanisms. This indicates that the
activation of the Mo=O bond would not be a favoured path-
way during the oxidation reaction for either of the H O
2
2
and TBHP oxidants.
Completion of the epoxidation cycle: In the previous sections
we have established that the first step of the Sharpless
mechanism is preferred when organic TBHP is considered
as the oxidant, and that the first step of the Thiel mecha-
nism is preferred when aqueous H O is employed as the ox-
2
2
idant. In this section we discuss the completion of the cata-
lytic cycle for the Sharpless mechanism. Only the TBHP oxi-
dant case is considered here because the catalyst will prefer
the Thiel pathway when aqueous H O is used. The free-
2
2
energy pathway for the conversion of the molybdenum
dioxo species, formed at the end of the first step of the
Figure 9. Potential energy surface for the regeneration of the catalyst
Sharpless mechanism, into the [CpMo
A
H
U
G
R
N
U
G
from the intermediate [CpMo(O)(OH)
A
H
U
G
R
N
U
(OOtBu)
A
T
G
R
N
U
G
À1
(
DG) are given in kcalmol
.
ACHTUNGTRENNUNG( OOtBu)] species is shown in Figure 8. The aid of an addi-
tional tert-butyl species as a hydrogen relay agent has also
been considered for this conversion. The next step of the
cycle, the step that would convert the [CpMo
CPh)(O)(OH)(OOtBu)] species into the original catalyst,
CpMo(O)
(OÀO)(CꢀCPh)], and also form tert-butanol (one
of the observed products of the reaction) is shown in
A
H
U
G
R
N
N
(Cꢀ
Further reactions of the [CpMo
termediate: The previous sections have shown that the mo-
A
H
U
G
E
N
N
(CꢀCPh)
ACTHUNGTNERNUNG( OOH) CAHTUNGTRENUNNG( OOH)] in-
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
[
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
lybdenum-based catalyst, in conjunction with the H O oxi-
dant, will preferentially produce the [CpMo
2
2
A
H
U
G
E
N
N
(CꢀCPh)-
&
6
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Alkene Oxidation with Molybdenum Catalysts
FULL PAPER
ACHTUNGTRENNUNG( OOH) ACHTUNGTRENNUNG( OOH) ACHTUNGTRENNUNG( 2H O)] (b-11) species. Subsequent calcula-
2
tions (shown in Figure 10) indicate that this species can be
À1
converted, with a low energy barrier of 3.6 kcalmol , into
Figure 11. Interconversions between different isomers of the molybde-
num catalyst after its coordination.
Figure 10. Facile interconversion between the possible different conform-
ers of the intermediates formed from the reaction of hydrogen peroxide
with the acetylide-coordinated molybdenum catalyst. Energy values
À1
(
DG) are given in kcalmol
.
the conformer [CpMo
A
H
U
G
R
N
U
G
ACHTUNGTNRENUNG( OOH)(OO) ACHTUNGTNERNUNG( 2H O)]
2
(
À1
mol
into
the
conformer
[CpMo
A
H
U
G
E
N
N
(CꢀCPh)(OH)-
ACHTUNGTRENNUNG( OOH)(OO) ACHTUNGTRENNUNG( 2H O)] (b-14). Hence, there are three con-
2
formers of the intermediate that can exist in solution during
the catalysis process, as illustrated in Figure 11. Of the three,
the most stable intermediate is b-20, with a slightly lower
energy than the other two conformers. This species, with
two hydrogen-bonding water molecules, can undergo further
conversion into the structure referred to as b-20(i) in
Figure 12: in b-20(i), the hydrogen-bonding water molecules
are at the same end of the complex (the OOH end), instead
of the water molecules separately binding to the OOH and
the OH ends, as in b-20. Such a rearrangement of the hydro-
gen-bonding water molecules is quite a facile process, as evi-
denced by the small difference in the energies of the struc-
Figure 12. Potential energy surface for the formation of the cis-1,2-diol
from the reaction of cyclohexene with b-20(i), the most stable of the
three conformers of the intermediate of the reaction of hydrogen perox-
ide with the acetylide-coordinated molybdenum catalyst. Energy values
À1
(
DG) are given in kcalmol
.
facile process, with the barriers for the two reactions being
À1
15.9 and 10.5 kcalmol , respectively. Hence, for the case of
À1
tures b-20 and b-20(i), calculated to be 1.1 kcalmol , with
the molybdenum catalyst with aqueous H O as the oxidant,
2
2
b-20(i) being slightly higher in energy (see Figure 6b in the
Supporting Information). Such a rearrangement of the water
molecules has been considered in order to study the reaction
shown in Figure 12: a cyclohexene approaches the peroxo
group of the species b-20(i) and is oxidised to the corre-
sponding cis-1,2-diol through a two-step process. This is a
the first step, the approach and attack of the H O moiety, is
2 2
the slowest step of the reaction, having a barrier of 22.7 kcal
À1
mol . Note also that the cis-1,2-diol can also be formed by
the approach and attack of the cyclohexene species on the
intermediate conformers b-14 and b-11. As the potential
energy surfaces shown in Figures 13 and 14 indicate, the re-
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!

S. B. Umbarkar, M. K. Dongare, K. Vanka et al.
higher than the corresponding energy barriers for the forma-
tion of the cis-1,2-diol by approach and attack of the cyclo-
hexene on b-11(i), b-14(i) and b-20(i) (see the energy barri-
ers in the Figures 14, 13 and 12, respectively).
Thus, overall, for the molybdenum catalyst complex with
aqueous H O as the oxidant, the calculations predict that
2
2
the preferred product would be the cis-1,2-diol, thus corrob-
orating the experimental findings discussed earlier.
We also note here that the other mechanism possible
after the formation of the [CpMo
ACHTUNGTRENNU(GN CCPh) ACHTUNRTEGNNUNG( OOH) ACHTUNGTRENNUNG( OOH)-
AHCTUNGTRENNUNG
(2H O)] species, the “Calhorda” mechanism (see Figure 15),
2
which would lead to the formation of the epoxide, was also
studied and was found to have high barriers of 36.2, 43.5,
À1
and 45.1 kcalmol for b-14, b-11, and b-20, respectively.
Hence the calculations indicate that the two competing
mechanisms during the cyclohexene oxidation processes
would be the Sharpless mechanism and our proposed mech-
anism, which can be considered a modified Thiel-type mech-
Figure 13. Potential energy surface for the formation of the cis-1,2-diol
from the reaction of cyclohexene with b-14(i), one of the three conform-
ers of the intermediate of the reaction of hydrogen peroxide with the
acetylide-coordinated molybdenum catalyst. Energy values (DG) are
À1
given in kcalmol
.
Figure 14. Potential energy surface for the formation of the cis-1,2-diol
from the reaction of cyclohexene with b-11(i), one of the three conform-
ers of the intermediate of the reaction of hydrogen peroxide with the
acetylide-coordinated molybdenum catalyst. Energy values (DG) are
À1
given in kcalmol
.
actions in each of these cases, leading to the eventual forma-
tion of the cis-1,2-diol, were also found to have low barriers,
lower than the first step of the catalytic cycle, the approach
and attack of the hydrogen peroxide species.
It may also be possible for the intermediates b-11, b-14
and b-20 to form the epoxide by the approach and attack of
the cyclohexene. This has also been computationally investi-
gated and the results for the reactions of b-11, b-14 and b-20
with cyclohexene to yield the epoxide are shown in
Figure 21 in the Supporting Information. For each of the
cases b-11, b-14 and b-20, the barriers are 19.7, 18.1 and
Figure 15. Potential energy surfaces for the Calhorda mechanism for ep-
À1
24.0 kcalmol , respectively. All the energy barriers dis-
oxide formation from b-14, b-11 and b-20. Energy values (DG) are given
À1
cussed for the formation of the epoxide are found to be
in kcalmol
.
&
8
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Alkene Oxidation with Molybdenum Catalysts
FULL PAPER
anism, with the Sharpless mechanism being favoured when
TBHP is used as the oxidant and the mechanism with fea-
tures common to the Thiel mechanism being preferred when
H O is employed as the oxidant.
the catalyst centre determines the difference in the selectivi-
ty of the two cases.
2
2
Summary of the mechanistic studies: The calculations dis-
cussed in the preceding sections show the complete reaction
cycle for the formation of the epoxide in the presence of the
molybdenum catalyst [CpMo(O)(OO)CꢀCPh] when organic
Conclusion
It has been experimentally observed that the oxidation of
cyclohexene in the presence of molybdenum peroxo acety-
lide as catalyst and TBHP in dodecane or 50% H O as oxi-
TBHP is employed as the oxidant as well as the formation
2
2
of the cis-1,2-diol when aqueous H O is used as the oxidant.
Figure 16 encapsulates the essential features of the two dif-
dant can lead to different oxidation products. For the oxida-
tion of cyclohexene, when TBHP in dodecane is used as the
oxidant, the epoxide is formed as the major product, and
when 50% H O is used as the oxidant, the cis-1,2-diol is
2
2
2
2
formed as the major product. This indicates that there are
competing mechanisms at work during the catalytic oxida-
tion process in these systems. Calculations with DFT were
performed to elucidate the nature of these competing mech-
anisms, with a new mechanism being proposed to explain
the formation of the cis-1,2-diol when aqueous H O is used
2
2
as the oxidant. The DFT studies showed that the order in
which the two substrates, the olefin and oxidant, attack the
molybdenum centre is crucial for determining the nature of
the oxidation product that is formed. For the case of TBHP
in dodecane, it was found that the preferred order of attack
is that of cyclohexene followed by TBHP, which leads to the
preferential formation of the epoxide product. In contrast,
with 50% H O , the preferred order of attack is that of
2
2
H O followed by cyclohexene, which leads to the preferen-
2
2
tial formation of the cis-1,2-diol product. What is of crucial
importance in determining the selectivity of the catalyst
with the different oxidants is the hydrogen bonding that
occurs between the water or oxidant molecules and the
oxygen atoms of the catalyst.
The calculations have thus provided an insight into the
nature of the olefin oxidation processes at the catalytic
centre, and may serve as a guide in choosing the appropriate
oxidant to obtain the desired product in the oxidation of
olefins.
Experimental Section
Preparation of [CpMo(CO)
Aldrich, SD Fine) were used as received unless stated otherwise. The
3
CꢀCPh]: All reagents of commercial grade
Figure 16. Trend in reactivity for the competitive reactions between the
catalyst and the other species present in solution. When TBHP is the oxi-
dant, cyclohexene is seen to approach first followed by the TBHP mole-
(
hydrogen peroxide used was 50% w/w in water. THF was dried accord-
ing to the standard method and freshly distilled prior to use. Cyclopenta-
diene was prepared from dicyclopentadiene (Aldrich) by distillation
2 2
cule, but when H O is used as the oxidant, the order of approach is re-
versed.
[40]
prior to use and was prepared according to the literature procedure.
1
Yield of [CpMo(CO)
3
A
H
U
G
R
N
U
G
1
3
7
1
2
.15–7.38 ppm (C-HAr); C NMR: d=87.85 (CCp), 126.01, 127.0, 129.39,
30.8, 130.87 (CPh), 222.2–238.8 ppm (C=O); IR (KBr): n˜ =1940, 1980,
ferent catalytic cycles. The principal difference between the
two cycles lies in the sequence in which the two species, cy-
clohexene and the oxidant (TBHP or H O ), approach the
À1
038 (CO), 2109 (C=C), 2958 cm (CCp); elemental analysis calcd (%)
for [CpMo(CO)3
A
H
U
G
R
N
U
G
2
2
Catalytic reaction: The liquid phase catalytic oxidation reactions were
carried out in a two-necked round-bottomed flask equipped with a mag-
netic stirrer and immersed in a thermostatted oil bath. The reactions
were carried out at 808C in tert-butanol (10 g) for the reaction in which
O was used as the oxidant and in 1,2-dichloroethane for the re-
2 2
action in which TBHP in dodecane was used as the oxidant. The flask
molybdenum catalyst centre. For the case in which the oxi-
dant is TBHP, the preferred order is cyclohexene followed
by TBHP, whereas for the case in which the oxidant is
5
0% H
H O , the preferred order is H O followed by cyclohexene.
2
2
2
2
This difference in the order in which the substrates attack
was charged with cyclohexene (0.02 mol) and either 50%
H
2
O
2
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
9
&
ÞÞ
These are not the final page numbers!

S. B. Umbarkar, M. K. Dongare, K. Vanka et al.
(
1
6
0.04 mol) or TBHP (0.02 mol) and the reactions were performed for
2 h. Samples were withdrawn periodically and analysed on an Agilent
890 gas chromatograph equipped with a HP-5 dimethylpolysiloxane
and 2) the calculations were performed with three different functionals
and have yielded the same conclusions in each case, it is believed that
the results obtained, even for cases in which the competing mechanisms
have small differences in energy barriers, are reliable.
column (60 m length, 0.25 mm diameter and 0.25 mm film thickness) with
a flame ionisation detector. The products were characterised by GCMS
on an 6890N instrument and by GC-IR with a Perkin–Elmer Spectrum
2
001 spectrometer.
Computational details: The computational procedure adopted is as fol-
lows: all the structures reported in the manuscript were optimised by
Acknowledgements
[44]
[45]
DFT with Turbomole Version 6.0 by using the BP-86 functional. The
electronic configurations of the atoms were described by using a triple-
zeta basis set augmented by a polarisation function (TURBOMOLE
The authors are grateful to the Centre of Excellence in Scientific Com-
puting (COESC), NCL, Pune, for providing computational facilities. Mr.
Prakash Chandra acknowledges the Council of Scientific and Industrial
Research for a fellowship.
[
46]
basis set TZVP). Because it is possible that the geometry optimisation
procedure performed by DFT may be sensitive to the nature of the func-
tional, to ensure the reliability of the obtained geometry-optimised struc-
tures, all the structures were also optimised with the Perdew, Burke, and
[
47]
Erzenhof density functional (PBE) as well as with the B3LYP function-
[1] G. Sienel, R. Rieth, K. T. Rowbottom, Epoxides, Wiley-VCH, Wein-
heim, 1999.
[2] K. Bauer, D. Garbe, H. Surburg, Common Fragrance and Flavor
Materials, Wiley-VCH, Weinheim, 1985.
[3] K. Kamata, K. Yonehara, Y. Sumida, K. Yamaguchi, S. Hikichi, N.
Mizuno, Science 2003, 300, 964.
[4] X. Zuwei, Z. Ning, S. Yu, L. Kunlan, Science 2001, 292, 1139.
[5] B. Van Duuren, B. Goldschmidt, J. Med. Chem. 1966, 9, 77–79.
[6] A. Y. H. Lu, G. T. Miwa, Annu. Rev. Pharmacol. Toxicol. 1980, 20,
513–531.
[7] A. W. Wood, R. L. Chang, W. Levin, R. E. Lehr, M. Schaefer-
Ridder, J. M. Karle, D. M. Jerina, A. H. Conney, Proc. Natl. Acad.
Sci. USA 1977, 74, 2746.
[8] J. Schwesinger, T. Bauer, Stereoselective Synth. 1995, 21, 4599.
[9] B. S. Lane, K. Burgess, Chem. Rev. 2003, 103, 2457–2474.
[10] W. Nam, R. Ho, J. S. Valentine, J. Am. Chem. Soc. 1991, 113, 7052–
7054.
[
48,49]
[50]
al
and the TZVP basis set. The resolution of identity (RI) and the
[51]
multipole-accelerated resolution of identity (marij)
approximations
were employed for an accurate and efficient treatment of the electronic
Coulomb term for both sets of the density functional calculations. A com-
parison was then made between the corresponding structures obtained
with the BP-86 and the PBE and B3LYP functionals. The comparison
showed very little difference in bond lengths, bond angles or dihedral
angles between the corresponding structures for all cases. Further corrob-
oration of the smallness of the difference between the structures obtained
with the two functionals came from a comparison of the potential energy
surfaces for the different reactions discussed in the manuscript. The
values of DE for the insertion and termination barriers for the oligomeri-
sation mechanism, obtained from the three separate sets of calculations,
were compared and the results, shown in the Supporting Information, in-
dicate that there is only a small difference between the corresponding
values for almost all the cases for the energies obtained with the BP-86
and the PBE functionals. The energies obtained with the B3LYP func-
[11] Y. Naruta, F. Tani, N. Ishihara, K. Maruyama, J. Am. Chem. Soc.
1991, 113, 6865–6872.
À1
tionals were found to be different by 8–10 kcalmol from the corre-
sponding BP-86 and PBE values, but the trends discussed in the manu-
script with regard to the preference for the Sharpless and the Thiel mech-
anism were found to remain unchanged for the set of calculations per-
formed with the B3LYP functional. This suggests that changing the func-
tional may result in a change in the absolute values of the barriers in the
different mechanisms, but also that the essential conclusions reached in
the manuscript will remain unchanged. With regard to the transition
states obtained in the three sets of calculations, care was taken to ensure
that the transition-state structures obtained possessed only one imaginary
frequency corresponding to the correct normal mode.
[12] J. F. Hull, D. Balcells, E. L. O. Sauer, C. Raynaud, G. W. Brudvig,
R. H. Crabtree, O. Eisenstein, J. Am. Chem. Soc. 2010, 132, 7605–
7616.
[
[
[
[
[
[
[
[
[
[
13] D. E. De Vos, J. L. Meinershagen, T. Bein, Angew. Chem. 1996, 108,
2
355–2357; Angew. Chem. Int. Ed. Engl. 1996, 35, 2211–2213.
14] A. Raith, P. Altmann, M. Cokoja, W. A. Herrmann, F. E. Kꢁhn,
Coord. Chem. Rev. 2010, 254, 608–634.
15] M. Herbert, E. ꢂlvarez, D. J. Cole-Hamilton, F. Montilla, A. Galin-
do, Chem. Commun. 2010, 46, 5933–5935.
16] D. V. Deubel, G. Frenking, P. Gisdakis, W. A. Herrmann, N. Rçsch,
J. Sundermeyer, Acc. Chem. Res. 2004, 37, 645–652.
17] H. Mimoun, I. Seree de Roch, L. Sajus, Tetrahedron 1970, 26, 37–
The contributions of internal energy and entropy were further obtained
from frequency calculations performed on the DFT structures at
5
0.
18] K. Sharpless, J. Townsend, D. Williams, J. Am. Chem. Soc. 1972, 94,
95–296.
19] C. Di Valentin, P. Gisdakis, I. V. Yudanov, N. Rçsch, J. Org. Chem.
000, 65, 2996–3004.
2
98.15 K. Thus, the energies reported in the figures of the paper are the
DG values. To account for the fact that all the species are in solution, the
2
translational entropy term in all the calculated structures was corrected
[
52]
through a free volume correction introduced by Mammen et al.
and
2
was used for all the atoms. Solvent effects were incorporated by using
20] I. V. Yudanov, C. Di Valentin, P. Gisdakis, N. Rosch, J. Mol. Catal. A
Chem. 2000, 158, 189–197.
21] D. V. Deubel, J. Sundermeyer, G. Frenking, J. Am. Chem. Soc. 2000,
[
53]
the COSMO model, with tert-butanol (e=12.4) as the solvent for the
cases in which H was used as the oxidant and 1,2-dichloroethane (e=
0.4) as the solvent for the cases in which TBHP was used as the oxidant.
2
O
2
1
1
22, 10101–10108.
Note that different conformational possibilities were considered for the
structures reported for the different reactants, intermediates and products
described in the paper. The structures that have been reported and em-
ployed for evaluating the free energies and enthalpies for the different
reactions were the ones that were the most stable conformers. It is possi-
ble that a very exhaustive conformational search, by using search algo-
rithms to look at hundreds or thousands of conformations, could have
been employed to locate unequivocally the lowest energy conformer, the
22] D. V. Deubel, J. Sundermeyer, G. Frenking, Inorg. Chem. 2000, 39,
2314–2320.
[23] A. Comas-Vives, A. Lledꢃs, R. Poli, Chem. Eur. J. 2010, 16, 2147–
2158.
[24] D. V. Deubel, J. Phys. Chem. A 2001, 105, 4765–4772.
[25] D. V. Deubel, J. Sundermeyer, G. Frenking, Eur. J. Inorg. Chem.
2001, 2001, 1819–1827.
[26] D. V. Deubel, J. Sundermeyer, G. Frenking, Org. Lett. 2001, 3, 329–
“
global minima” in each case, but such an approach is currently beyond
332.
the scope of full quantum chemical methods. Note also that the differ-
ence in energy between the competing pathways is sometimes quite
small, about 2.0 kcalmol in some cases, but, given that 1) the same rela-
[27] M. Bꢁhl, R. Schurhammer, P. Imhof, J. Am. Chem. Soc. 2004, 126,
3310–3320.
À1
[28] P. Gisdakis, I. V. Yudanov, N. Rçsch, Inorg. Chem. 2001, 40, 3755–
tive differences are obtained at the same level of theory in these cases
3765.
&
10
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Alkene Oxidation with Molybdenum Catalysts
FULL PAPER
[
[
[
29] L. Salles, J. Y. Piquemal, R. Thouvenot, C. Minot, J. M. Brꢄgeault, J.
Mol. Catal. A Chem. 1997, 117, 375–387.
[43] A. M. Al-Ajlouni, D. Veljanovski, A. Capapꢄ, J. Zhao, E. Herdt-
weck, M. J. Calhorda, F. E. Kꢁhn, Organometallics 2009, 28, 639–
645.
[44] R. Ahlrichs, M. Bar, M. Haser, H. Horn, C. Kolmel, Chem. Phys.
Lett. 1989, 162, 165–169.
30] D. Chakraborty, M. Bhattacharjee, R. Krꢅtzner, R. Siefken, H. W.
Roesky, I. Usꢃn, H. G. Schmidt, Organometallics 1999, 18, 106–108.
31] a) W. R. Thiel, T. Priermeier, Angew. Chem. 1995, 107, 1870–1871;
Angew. Chem. Int. Ed. Engl. 1995, 34, 1737–1738; b) W. R. Thiel, J.
Eppinger, Chem. Eur. J. 1997, 3, 696–705.
[45] a) A. D. Becke, Phys. Rev. A 1988, 38, 3098–3100; b) J. P. Perdew,
Phys. Rev. B 1986, 33, 8822–8824.
[
46] a) F. Weigend, Phys. Chem. Chem. Phys. 2002, 4, 4285–4291; b) F.
Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297–3305;
c) A. Schaefer, H. Horn, R. Ahlrichs, J. Chem. Phys. 1992, 97, 2571–
[
[
32] W. R. Thiel, J. Mol. Catal. A Chem. 1997, 117, 449–454.
33] P. J. Costa, M. J. Calhorda, F. E. Kꢁhn, Organometallics 2010, 29,
3
03–311.
2
1
577; d) A. Schaefer, C. Huber, R. Ahlrichs, J. Chem. Phys. 1994,
00, 5829–5835.
[
[
[
34] A. V. Biradar, B. R. Sathe, S. B. Umbarkar, M. K. Dongare, J. Mol.
Catal. A Chem. 2008, 285, 111–119.
35] C. Bolm, J. P. Hildebrand, K. MuÇiz, N. Hermanns, Angew. Chem.
[
47] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77,
865–3868.
3
2
001, 113, 3382–3407; Angew. Chem. Int. Ed. 2001, 40, 3284–3308.
36] H. C. Kolb, M. S. Van Nieuwenhze, K. B. Sharpless, Chem. Rev.
994, 94, 2483–2547.
[
48] A. D. Becke, J. Chem. Phys. 1993, 98, 5648.
[
49] P. Stephens, F. Devlin, C. Chabalowski, M. J. Frisch, J. Phys. Chem.
1
1
994, 98, 11623–11627.
[
[
[
37] M. Schroeder, Chem. Rev. 1980, 80, 187–213.
38] J. Albert, A. Hamilton, &&& John Wiley & Sons, New York, 1995.
39] C. Dinoi, M. Ciclosi, E. Manoury, L. Maron, L. Perrin, R. Poli,
Chem. Eur. J. 2010, 16, 9572–9584.
40] M. Bruce, M. Humphrey, J. Matisons, S. Roy, A. Swincer, Austr. J.
Chem. 1984, 37, 1955–1961.
41] T. Fujihara, K. Hoshiba, Y. Sasaki, T. Imamura, Bull. Chem. Soc.
Jpn 2000, 73, 383–390.
42] C. Bianchi, F. Porta, Vacuum 1996, 47, 179–182.
[
[
[
[
50] K. Eichkorn, O. Treutler, H. Oehm, M. Haeser, R. Ahlrichs, Chem.
Phys. Lett. 1995, 240, 283–290.
51] M. Sierka, A. Hogekamp, R. Ahlrichs, J. Chem. Phys. 2003, 118,
9
136–9148.
52] M. Mammen, E. I. Shakhnovich, G. M. Whitesides, J. Org. Chem.
998, 63, 3168–3175.
53] A. Klamt, Andreas, J. Phys. Chem. 1995, 99, 2224–2235.
[
[
[
1
Received: July 23, 2012
Published online: && &&, 0000
Chem. Eur. J. 2013, 00, 0 – 0
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
11
&
ÞÞ
These are not the final page numbers!

S. B. Umbarkar, M. K. Dongare, K. Vanka et al.
Molybdenum-catalysed oxidation: Cal-
culations performed by using density
functional theory indicate that differ-
ent mechanisms compete during the
homogeneous catalysis of olefin oxida-
tion carried out in the presence of
molybdenum-based catalysts (see
scheme).
Homogeneous Catalysis
P. Chandra, S. L. Pandhare,
S. B. Umbarkar,* M. K. Dongare,*
K. Vanka* ........................ &&& — &&&
Mechanistic Studies on the Roles of
the Oxidant and Hydrogen Bonding in
Determining the Selectivity in Alkene
Oxidation in the Presence of Molyb-
denum Catalysts
&
12
&
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!