Introducing copper as catalyst for oxidative alkane dehydrogenation

Article abstract of DOI:10.1021/ja310866k

The dehydrogenation of n-hexane and cycloalkanes giving n-hexene and cycloalkenes has been observed in the reaction of such hydrocarbons with hydrogen peroxide, in the presence of copper complexes bearing trispyrazolylborate ligands. This catalytic transformation provides the typical oxidation products (alcohol and ketones) with small amounts of the alkenes, a novel feature in this kind of oxidative processes. Experimental data exclude the participation of hydroxyl radicals derived from Fenton-like reaction mechanisms. DFT studies support a copper-oxo active species, which initiates the reaction by H abstraction. Spin crossover from the triplet to the singlet state, which is required to recover the catalyst, yields the major hydroxylation and minor dehydrogenation products. Further calculations suggested that the superoxo and hydroperoxo species are less reactive than the oxo. A complete mechanistic proposal in agreement with all experimental and computational data is proposed.

Full text of DOI:10.1021/ja310866k

Article
pubs.acs.org/JACS
Introducing Copper as Catalyst for Oxidative Alkane
Dehydrogenation
Ana Conde, Laia Vilella, David Balcells,* M. Mar Díaz-Requejo,* Agustí Lledos
́
,*,†
‡
†
,†,§
,‡
,‡
́
and Pedro J. Perez*
‡
Laboratorio de Catal
́
isis Homogen
en Química Sostenible, Universidad de Huelva, Campus de El Carmen 21007 Huelva, Spain
Departament de Química, Universitat Autonoma de Barcelona, 08193 Bellaterra, Spain
Centre for Theoretical and Computational Chemistry, Department of Chemistry, University of Oslo, P.O. Box 1033, Blindern
N-0315, Oslo, Norway
́
ea, Departamento de Química y Ciencia de los Materiales, Unidad Asociada al CSIC,
CIQSO-Centro de Investigacion
́
†
̀
§
*
S Supporting Information
ABSTRACT: The dehydrogenation of n-hexane and cycloalkanes giving n-
hexene and cycloalkenes has been observed in the reaction of such hydrocarbons
with hydrogen peroxide, in the presence of copper complexes bearing
trispyrazolylborate ligands. This catalytic transformation provides the typical
oxidation products (alcohol and ketones) with small amounts of the alkenes, a
novel feature in this kind of oxidative processes. Experimental data exclude the
participation of hydroxyl radicals derived from Fenton-like reaction mechanisms.
DFT studies support a copper-oxo active species, which initiates the reaction by H
abstraction. Spin crossover from the triplet to the singlet state, which is required
to recover the catalyst, yields the major hydroxylation and minor dehydrogenation
products. Further calculations suggested that the superoxo and hydroperoxo
species are less reactive than the oxo. A complete mechanistic proposal in
agreement with all experimental and computational data is proposed.
INTRODUCTION
Scheme 1. The Desaturase and Hydroxylase Reactions with
Dioxygen
■
The catalytic dehydrogenation of alkanes is probably one of the
emerging areas in both organometallic chemistry and
homogeneous catalysis that might have a tremendous impact
in chemical industry in the next decades, due to the increasing
1
use of olefins as raw materials. The availability of alkanes in
terms of their abundance and low cost makes this reaction an
attractive route toward olefins. Usually this reaction (eq 1)
enzymes have been modeled at different degree, with only a few
examples of transition metal complexes mimicking the
desaturase function being described to date. Que and co-
4a
workers reported the stoichiometric reaction of a dinuclear
iron-complex with cumene to yield α-methylstyrene as the
result of the desaturation reaction (Scheme 2a), a work later
requires the presence of a sacrificial hydrogen acceptor, another
4
b,c
1
expanded to several hydrocarbons. Crabtree, Eisenstein and
olefin in most cases. At the industrial level, the dehydrogen-
5
co-workers first described the catalytic use of a manganese-
ation of alkanes is a known process that is based on the use of
based system for the desaturation of dihydrophenanthrene with
PhIO or oxone as the oxidant (Scheme 2b), in a transformation
driven by the aromatization of the final products. Yin and co-
heterogeneous catalysts operating at high temperature (500−
2
9
00 °C), with low control over selectivity. In contrast, the
homogeneous catalysts developed to date operate at a lower
temperature (100−150 °C), allowing catalyst design toward
control of the reaction outcome. Among the catalysts described
6
workers later reported the use of oxone to desaturate
dihydroanthracene with other manganese-based catalysts
8
1
(Scheme 2c). Shaik and Nam have studied the interaction of
to date, those based on iridium are the most active by far. In
iron(IV)-oxo species with such substrates, providing dehydro-
genated compounds in substoichiometric amounts. In the
nature, dehydrogenation is catalyzed by the desaturase
oxidoreductase, which converts a C−C bond into the
corresponding CC bond, upon formal loss of two hydro-
3
gens. Desaturases and hydroxylases reduce dioxygen to give
Received: November 4, 2012
Published: February 14, 2013
the alkene or the alcohol, respectively (Scheme 1). Both
©
2013 American Chemical Society
3887
dx.doi.org/10.1021/ja310866k | J. Am. Chem. Soc. 2013, 135, 3887−3896

Journal of the American Chemical Society
Article
x
a
Scheme 2. Model Systems for Desaturase
Table 1. Cyclohexane Oxidation Using Tp Cu as Catalysts
dehydrog.
products (%)
hydroxylated
products (%)
entry
catalyst
yield (%)
P1
P2
P3
P4
Br3
1
2
3
4
5
6
7
8
9
Tp Cu
22
11
22
21
13
24.5
20
11
9
4
3
4
3
3
2
2
2
2
7
45
27
43
43
39
38
56
54
27
48
65
52
53
55
57
41
44
65
,Br
Tp* Cu
Tp*Cu
4
2
Me3
Tp Cu
1
Ph
Tp Cu
4
Br3
b
Tp Cu
3
Br3
c
Tp Cu
<1
nd
1
Br3
d
e
Tp Cu
Br3
Tp Cu
a
Conditions: 0.01 mmol of catalyst, 3 mL MeCN, 1 mmol of
cyclohexane, 10 mmol of H O , rxn time = 1 h, temp = 60 °C. See
above catalytic cases, the reaction took place onto benzylic C−
2
2
H bonds, a somewhat activated reaction site (bond dissociation
Scheme 4 for products numbering. Average of at least two runs.
Conversions correspond to mmol of products/mmol of initial
hydrocarbon. rt, 7 h. rt, 12 h. rt, 24 h. 80 °C, 1 h.
7
energy of ca. 90 kcal/mol), and additionally, in both
b
c
d
e
manganese systems, desaturation led to aromatization. It is
worth mentioning that this dehydrogenation process has not
yet been described for nonactivated C−H bonds as those of
cycloalkanes or lineal alkanes, with bond dissociation energies
(
Scheme 4). Surprisingly we also observed the appearance of
two minor products that have been identified as cyclohexene
7
of 96−100 kcal/mol, respectively.
x
We are currently interested in the development of catalysts
for the direct oxidation of hydrocarbons, particularly methane
or benzene into methanol and phenol respectively. This area
constitutes a major goal in modern chemistry, since neither
methanol nor phenol can be directly prepared in an efficient
manner from the parent hydrocarbons. Methanol is prepared
Scheme 4. Products Observed in the Tp Cu-Catalyzed
Cyclohexane Oxidation Reaction
9
a
from syngas, whereas the manufacture of phenol takes place
9b
through the cumene process. We have recently described the
x
x
and cyclohexene oxide. Blank experiments revealed that
cyclohexene was formed during the reaction of cyclohexane
and hydrogen peroxide only in the presence of the copper
catalysts (no cyclohexene was found upon reacting cyclohexane
catalytic potential of complexes of composition Tp Cu (Tp =
10
hydrotrispyrazolylborate ligand, Scheme 3) for the direct
Scheme 3. Trispyrazolylborate Ligands
x
and H O in the absence of Tp Cu) and that cyclohexene
2
2
epoxidized under the reaction conditions. It is also worth
pointing out that no cyclohexene was detected as an impurity in
the cyclohexane employed as substrate.
Table 1 shows the results obtained with the array of catalysts
employed, after 1 h of stirring at 60 °C a 1:100:1000 mixture of
[
Cu]:[C H ]:[H O ]. All the complexes tested led to the
6 12 2 2
formation of the four products shown in Scheme 4, but in a
different ratio. The products derived from the dehydrogenation
of cyclohexane, i.e., cyclohexene and cyclohexene oxide were
minor (4−8% of the distribution of products). This behavior
was not exclusive of cyclohexane, since the same protocol
applied to cyclooctane led to the observance of cyclooctene,
cyclooctene oxide and cyclooctenone (Scheme 5).
1
1
oxidation of benzene into phenol using H O as oxidant. We
2
2
herein report the results obtained in the course of our
investigations on expanding the above catalytic system to
alkanes. We have discovered that these hydrocarbons may
undergo catalytic dehydrogenation and subsequent formation
of alkenes. The transformation takes place under mild
conditions, and the oxidant also operates as the sacrificial
hydrogen acceptor, providing water as byproduct. With hexane
as the alkane, the dehydrogenation seems to be preferred
toward 1-hexene. Although conversions into alkenes at this
stage of research are yet low, we believe that this is a proof of
concept that copper-based catalysts, much cheaper than
iridium, could be developed for this purpose.
Br3
From the results in Table 1, entries 1−5, we chose Tp Cu
as the catalyst of choice, since it provided the best conversions
as well as yields into the dehydrogenated products. A somewhat
intriguing trend is observed when increasing the reaction time
Scheme 5. Oxidation of Cyclooctane with the [Cu]/H O
2
2
Catalytic System
RESULTS AND DISCUSSION
■
Oxidation of Cycloalkanes with Hydrogen Peroxide in
x
the Presence of Tp Cu Complexes. In our first set of
experiments, we tested the catalytic capabilities of five copper
x
complexes bearing different Tp ligands (Table 1). In all cases,
we observed the formation of cyclohexanol as well as some
cyclohexanone derived from overoxidation of the former
3
888
dx.doi.org/10.1021/ja310866k | J. Am. Chem. Soc. 2013, 135, 3887−3896

Journal of the American Chemical Society
Article
from 7 to 12 and 24 h: the yield values (intended as mmol of
products/mmol of initial cyclohexane) decreased with time
Table 2. Oxidation of n-Hexane with H O and
2 2
a
Br3
Tp Cu(MeCN) as the Catalyst
(
Table 1, entries 6−8). In the same way, the reaction carried
out at 80 °C resulted in a lower yield than that observed at 60
C (entries 1 vs 9). We have learnt that this is due to
mmol
2
yield
(%)
entry
H O2
P2 P3 P4 P5 P7
P9 P11
°
1
2
5
10
10
10
10
6.5
7
5
−
−
9
−
−
−
<2
−
−
−
8
12
8
42
36
34
38
41
41
45
33
41
46
cyclohexane loss by diffusion into the vapor phase (see
Supporting Information)
10
9
b
3
8.5
6.5
5
7
The reaction of cycloalkanes with hydrogen peroxide in the
presence of transition metal complexes has been previously
c
4
>9
5
−
−
1
9
d
12
5
3
5
described by several groups, although in no case dehydrogen-
ation was reported. It has been proposed that those
transformations seem to occur under Fenton-like pathways,
i.e., with the intermediacy of free hydroxyl radicals. Later on
this manuscript, we will provide mechanistic studies demon-
strating that this statement does not apply in our case, with
metal-oxo species being responsible of the observed trans-
formations.
a
See Scheme 6 for numbering. 0.01 mmol of catalyst, 3 mL of MeCN,
1
mmol of hexane, rxn time = 1 h, temp = 60 °C. Average of at least
two runs. Yield values correspond to mmol of products/initial mmol of
b
c
d
hydrocarbon. Rxn time = 3.5 h, at rt. Rxn time = 5 h, at rt. Rxn time
=
12 h, at rt.
n-hexane. As mentioned in the Introduction, this dehydrogen-
ation process in which an oxidant is employed has only been
5
,6
Oxidation of n-Hexane. At this stage we wondered if a
reported with manganese in a catalytic manner, and using
substrates in which the dehydrogenation of activated (benzylic)
C−H bonds also induces the aromatization of the resulting
product. In our case, loss of hydrogen takes place in
nonactivated C−H bonds, where there is not such a driving
force similar to aromatization. The use of copper as the metal
center and hydrogen peroxide as the oxidant are also novel for
this transformation.
linear alkane such as n-hexane could undergo this trans-
1
3
formation. Initially, and on the basis of the reactions with
cycloalkanes, we could expect the products derived from (i) C−
H hydroxylation, (ii) oxidation of the alcohol to give aldehyde
or ketone, (iii) dehydrogenation, and (iv) oxidation of the
olefin to the oxide. Scheme 6 shows the 12 possible products,
Scheme 6. Oxidation of n-Hexane with the [Cu]/H O
Catalytic System Showing All Possible Products
2
2
Oxidation of Dihydroaromatic Substrates. For the sake
a
of completeness, we have evaluated our catalytic system with
5
6
the substrates previously described by Crabtree and Yin with
manganese-based catalysts (as shown in Scheme 2). In our case,
both dihydrophenanthrene and dihydroanthracene were
dehydrogenated to give phenanthrene and anthracene (Scheme
7
), respectively, along with oxidation products. Conversions of
Scheme 7. Distribution of Products of 9,10-
Dihydrophenanthrene (Top) and Dihydroanthracene
(
Bottom) Oxidation Reactions
a
Detected products are framed. See Table 2 for distribution of
products.
from which we have identified seven in variable amounts (Table
10 and 27% (based on initial hydrocarbon) were obtained using
hydrogen peroxide with this copper-based catalyst, at variance
2), by comparison with commercial samples using GC/GCMS
techniques. Hydroxylation products and their overoxidation
derivatives were the main outcome of the reaction, particularly
that yield by the functionalization of the internal C2−H and
C3−H bonds. Primary sites were also oxidized, and 1-hexanal
was detected as the result of the oxidation of 1-hexanol.
Regarding the dehydrogenation reaction, 1-hexene was formed
as the sole olefin in most experiments. Only in a single case, 2-
hexene was detected at very low concentration (Table 2, entry
with PhIO or KHSO that were employed in those studies.
5
Mechanistic Studies. Evidencing the Lack of Fenton-Like
Pathways. A well-known feature of many transition metals is
the capability to promote the decomposition of hydrogen
•
peroxide and the generation of hydroxyl radicals (HO ) that
14
further induces oxidation reactions. This is the base of the so-
15
called Fenton chemistry, that is mainly characterized by the
lack of selectivity. Since several copper compounds or salts have
been reported to generate such radical promoting alkane
4
). No epoxides were detected in the reaction mixture.
Therefore, this catalytic system induces the dehydrogenation
of not only cycloalkanes, but also of less reactive alkanes such as
1
2,16
oxidations,
we have first performed a series of experiments
to elucidate whether or not our transformation occurs in a
3
889
dx.doi.org/10.1021/ja310866k | J. Am. Chem. Soc. 2013, 135, 3887−3896

Journal of the American Chemical Society
Article
similar manner. However, we must emphasize at this point that
no alkene formation has ever been described in those copper-based
systems.
The previous copper-based catalytic systems for cycloalkane
hydroxylation are dominated by the proposal of the
participation of hydroxyl radicals (Scheme 8) that induce
Although this behavior has not been reported with copper,
Lyons et al described this effect with iron-based catalysts.
17
In our search for additional evidence that could provide
information about the nature of the mechanism governing this
transformation, we have taken advantage of the already
x
described catalytic capabilities of these Tp Cu complexes to
11
oxidize aromatic C−H bonds. In our previous report on this
reaction, we observed that the reaction with benzene led to the
exclusive formation of phenol and benzoquinone. It has been
established that the oxidation of phenol with the intermediacy
of the hydroxyl radical also produces biphenyl derivatives as the
result of the coupling of the phenyl radical, but we have not
detected this product in our system. To reinforce this data, we
have now studied the reaction of toluene and H O with
Tp* Cu(NCMe) as the catalyst. GC and GCMS studies have
revealed the formation of a mixture of compounds identified as
benzaldehyde, ortho- and para-cresols and 2-methyl-1,4-
benzoquinone (Scheme 9). The oxidation of toluene by the
Scheme 8. Previously Proposed Free-Radical Copper
Induced Oxidation of Cyclohexane
2
2
,Br
Scheme 9. Oxidation of Toluene with Hydrogen Peroxide
,Br
Using Tp* Cu(NCMe) as the Catalyst
12,16
hydrogen abstraction from the cycloalkane.
The cyclohexyl
radical is further trapped by molecular oxygen to give the
alkylperoxide radical, from which cyclohexanol and cyclo-
hexanone are formed after several steps. Under the conditions
employed, cyclohexylperoxide is also formed. The latter can be
easily converted into cyclohexanol in a selective manner upon
12a
addition of PPh (Scheme 8), in what is considered a probe
3
to demonstrate the involvement of a peroxidative process
triggered by hydroxyl radicals. A second probe is the use of an
16a
inert atmosphere,
which should decrease the yield into
oxidation products since molecular oxygen is required for
cycloalkylperoxide formation.
In our system, we have collected some experimental data
related to the above. First of all, we have not observed any
variation of the reaction outcome upon varying the atmosphere
from air to nitrogen. Also, following the established
1
2a,16
protocols,
we have carried out experiments in which
15
hydroxyl radical is known to produce biaryl derivatives, which
we have not detected in the reaction mixtures. In contrast, the
distribution of products obtained with our copper catalyst
twin reactions were run, and PPh was added at the end to one
of each couple. As shown in Table 3, the catalytic results did
3
(
Scheme 9) was similar to that found by Sawyer and co-
Table 3. Effect of the Addition of PPh in the Reaction of
18
3
Br3
workers, where they proposed that some iron-based catalysts
for these transformations did not operate throughout the HO
route. Other groups have already found evidence against the
involvement of hydroxyl radicals in this catalytic process.
Particularly, studies by Marusawa, Tezuka and co-workers
demonstrated that the distribution of cresols derived from
photolytically generated HO radicals and toluene was 71:9:20
for o:m:p-cresols, a set of values quite distinct from those in
Scheme 9 (ca. 61:0:39, counting the quinone as derived from
the ortho derivative).
Cyclohexane and H O Using Tp Cu(NCMe) as the
2
2
•
a
Catalyst
19
dehydrog.
products (%)
hydrox.
products (%)
19b
entry
1
catalyst
yield (%)
P1
P2
P3
P4
•
Br3
Tp Cu
22%
21%
3.5
3
3
2
45
50
48.5
45
b
Br3
2
Tp Cu
a
Conditions: 0.01 mmol of catalyst, 3 mL of MeCN, 1 mmol of
cyclohexane, 10 mmol H O , rxn time = 1 h, temp = 60 °C, see
2
2
Overall, data collected and presented above seem to disfavor
Scheme 4 for products numbering. Conversion values correspond to
b
•
mmol of products/initial mmol of hydrocarbon. Values determined
the proposal of the involvement of free HO radicals in this
by GC after addition of 0.5 mmol of PPh3.
process.
Evidencing the Intermediacy of Copper-Oxo Species. The
not vary when comparing the results in the presence or absence
use of oxone, which contains KHSO as the oxidant, has also
5
of PPh . From here we could conclude that no cyclo-
alkylperoxides derived from the action of HO radicals are
provided valuable information since it does not produce
hydroxyl radicals. The reaction of cyclohexane and oxone in
3
•
20
formed in our system. However, it could happen that our
copper catalyst would also induce the catalytic decomposition
of the cycloalkylhydroperoxide in an efficient manner in such a
the presence of catalytic amounts of these copper complexes
originates a mixture of cyclohexanol and cyclohexanone, as well
as cyclohexene as the minor product, with yields similar to
those observed with hydrogen peroxide as the oxidant (Scheme
way that the experiment with PPh would not be informative.
3
3
890
dx.doi.org/10.1021/ja310866k | J. Am. Chem. Soc. 2013, 135, 3887−3896

Journal of the American Chemical Society
0). These results suggest that both hydrogen peroxide and
oxone involve a similar reaction mechanism in the copper-
catalyzed oxidation of cyclohexane.
Article
1
Scheme 12. Previously Proposed Pathways for the C−H
Bond Hydroxylation/Desaturation Reactions
Scheme 10. Oxidation of Cyclohexane with Oxone in the
Presence of the Copper Catalyst
species, which yields hydroxylation/desaturation product
mixtures. In the hydroxylation pathway, the OH group couples
with that radical, following the rebound mechanism, whereas in
the desaturation pathway, the vicinal C−H bond undergoes a
second H abstraction. Both pathways are initiated by a
common metal-oxo intermediate. Late transition metal oxo
complexes are extremely reactive and thus difficult to isolate
and characterize. Nonetheless, these complexes have been
Additional as well as crucial information has been found
when radical inhibitors when employed. The use of commonly
employed radical inhibitors such as 2,6-ditertbutyl-4-methyl-
phenol (BHT) or TEMPO was precluded since under the
reaction conditions (copper source and hydrogen peroxide)
they were also oxidized, due to the presence of either aromatic
or aliphatic C−H bonds. Thus, we moved toward radical traps
22
proposed as active species in several catalytic systems and
characterized for a few metal−ligand combinations able to
23
not having C−H bonds such as CCl or CBrCl . When the
4
3
provide enough stability to the system.
reaction was carried out in the presence of added CCl , some
4
To collect some additional information about the mechanism
that governs this transformation, the reaction mechanism was
explored by means of DFT calculations on the oxidation of
cyclohexyl chloride was detected at the end of the reaction
(
Scheme 11). This is in good agreement with the generation of
Br3
cyclohexane by the Tp Cu catalyst (see Computational
Scheme 11. The Effect of Added CCl or CBrCl to the
Details). The full real structures of the substrate and the
catalyst were considered without any simplifications. Free
4
3
Br3
Tp Cu-Catalyzed Reaction of Cyclohexane and H O
2
2
energies in solution, G , were computed by including both the
sol
solvent effects of acetonitrile (at the CPCM level with a triple-ζ
basis set) and the thermochemistry corrections (zero-point,
thermal and entropy energies).
Scheme 13 shows the three possible species that can be
formed upon oxidation of the catalyst precursor with hydrogen
Scheme 13. Possible Copper Active Species Considered in
This Study
•
the C H radical, as demonstrated by Mansuy and co-workers
6
11
21
in the related nitrene transfer reaction. Moreover, when the
chlorinated additive was changed to CBrCl , then the main
3
product was bromocyclohexane, the oxidation/dehydrogen-
ation products being detected in very minor amounts (<1%
8
overall). A recent work by Shaik, Nam and co-workers with a
Fe(IV)O complex has shown that the alkyl radical formed
upon H-abstraction from the alkane is similarly trapped with
CBrCl . Therefore, our data seem to support the involvement
3
of a copper-oxo species that abstracts a hydrogen from
cyclohexane yielding cyclohexyl radicals that are trapped by
halogenated reagents. The validity of this proposal has been
checked by the theoretical studies presented in the following
section.
2
4
peroxide: the copper oxo, superoxo and hydroperoxo species.
Mechanistic Studies: DFT Calculations. The aforemen-
We have investigated them as potential active species,
calculating the stability and reactivity with cyclohexane.
We have started optimizing the singlet and triplet states of
the oxo species, S and T, respectively. The ground state is the
5b
tioned previous work by Crabtree, Eisenstein and co-workers
has established the existence of two competing pathways
Scheme 12) involving the formation of a carbon-radical
(
3
891
dx.doi.org/10.1021/ja310866k | J. Am. Chem. Soc. 2013, 135, 3887−3896

Journal of the American Chemical Society
Article
triplet, which lies 33.2 kcal mol−1 below the singlet. The
coordination geometry of the ground state is tetrahedral, with
the trispyrazolylborate ligand bound to copper in a κ fashion.
a
Table 4. Local Spin Densities (ρ) and Charges (q) of the
Stationary Points Involved in the H Abstraction Step
3
T-1
T-TS1
T-2
The formation of the oxo species from the catalyst and
ρ(Cu)
ρ(O)
0.80
1.11
0.00
0.00
0.00
0.09
0.81
0.68
0.84
0.10
hydrogen peroxide (Scheme 13) is exergonic, ΔG = −3.6 kcal
−1
mol . Cyclohexane interacts with T yielding intermediate T-1,
in which a C−H bond of the substrate is weakly H-bonded to
the oxygen ligand, d(H···O) = 2.56 Å. This species, which is
similar to the prereaction complex characterized by Costas et al.
b
ρ(C)
0.47
0.96
b
ρ(H)
−0.06
0.48
−0.01
0.97
ρ(C H )
6
11
ρ(TpBr3
)
0.09
0.10
25
for C−H oxidation by Mn(IV)O species, lies 7.0 kcal
−1
mol below reactants (Figure 1). The local spin densities on
q(Cu)
q(O)
1.13
−0.65
−0.38
0.21
1.17
−0.94
−0.23
0.29
1.18
−1.14
−0.09
0.47
b
q(C)
b
q(H)
q(C H )
−0.21
−0.48
−0.01
−0.51
0.00
6
11
q(TpBr3
)
−0.51
a
Positive and negative spin densities are alpha and beta, respectively.
C and H belong to the activated C−H bond of cyclohexane.
b
Figure 2. Spin densities of T-1 (left), T-TS1 (middle) and T-2 (right).
Spin density colors: blue (alpha), green (beta). Atom colors: orange
(
(
Cu), red (O), white (H), gray (C), blue (N), maroon (Br), purple
B).
−1
Figure 1. Gibbs energy (G ) profile in acetonitrile (kcal mol ), for
sol
the hydroxylation and dehydrogenation of cyclohexane in the triplet
Figure 3. Optimized geometries of T-TS1 (left), MECP1 (middle),
and MECP2 (right). Atom colors: orange (Cu), red (O), white (H),
gray (C), blue (N), maroon (Br), purple (B).
(
solid line) and singlet (dashed line) states for the copper-oxo species.
copper and oxygen, ρ(Cu) = 0.80 and ρ(O) = 1.11 (Table 4
II
•
and Figure 2), are more consistent with a radical Cu −O
associated with the C H organic fragment. This fragment
6
11
26
•
configuration, referred to as oxyl, than with a closed-shell
corresponds to the neutral cyclohexyl radical, C H , as shown
6 11
III
Cu O oxo.
by its local charge, q(C H ) = 0.00, and spin density, ρ(C H )
6 11 6 11
The oxyl character of T-1 may promote radical H abstraction
= 0.97, which is concentrated upon the C involved in the C−H
cleavage, ρ(C) = 0.96. The generation of this radical is
consistent with the formation of C H Cl in the presence of
2
7
reactions. The transition state of this reaction, T-TS1, was
−1
located at a relative energy of 3.0 kcal mol (Figure 1). In T-
TS1, the H-bound C−H of T-1 breaks, d(C···H) = 1.25 Å, and
transfers the H atom to the oxyl ligand, d(H···O) = 1.25 Å
6
11
CCl or of C H Br in the presence of CBrCl (Scheme 11)
4
6
11
3
and the preferential oxidation of the secondary carbons of n-
(
Figure 3). The oxyl spin density is reduced to ρ(O) = 0.68,
hexane (Scheme 6), which stabilize the radical more than the
Br3
due to its polarization toward the activated C−H bond, which
gains radical character on its C atom, ρ(C) = 0.47 (Table 4 and
Figure 2). This is consistent with the pairing of two electrons to
form the new O−H bond, one coming from the oxyl and the
other from the homolytic cleavage of the C−H bond. The
relaxation of T-TS1 toward products converged into
intermediate T-2, which is a copper-hydroxo complex weakly
primary. The metal center and the Tp ligand seem to play a
somewhat spectator role, since their local charges and spin
densities are almost constant throughout the reaction. Starting
−
1
from T-1, the H abstraction step is exergonic by 8.3 kcal mol
−1
and involves a low energy barrier of 10.0 kcal mol . The triplet
state of the oxo species thus initiates the oxidation of
cyclohexane by a radical H abstraction.
3
892
dx.doi.org/10.1021/ja310866k | J. Am. Chem. Soc. 2013, 135, 3887−3896

Journal of the American Chemical Society
Article
The reactivity of the CuII superoxo and hydroperoxo
complexes (Scheme 13), which may also act as active species,
60.3 kcal mol−1 less stable in the singlet state, whereas S-3 and
S-4 are 88.2 kcal mol and 83.7 kcal mol , respectively, less
24
−1
−1
was investigated for this H abstraction process. The formation
of these species from the catalyst and hydrogen peroxide is
more exergonic than that of the oxo, with ΔG = −56.5 kcal
mol (superoxo) and ΔG = −27.3 kcal mol (hydroperoxo).
Similar to the oxo species, the superoxo abstracts one hydrogen
from cyclohexane, yielding the cyclohexyl radical (see Figures
S11 and S12 in the Supporting Information). However, the
stable in the triplet state.
−1
The release of cyclohexanol, ΔG = −7.2 kcal mol from S-3,
−
1
and cyclohexene, ΔG = −12.3 kcal mol from S-4, is
−1
−1
exergonic. The overall hydroxylation and dehydrogenation
−
1
reactions are much more exergonic, ΔG = −65.1 kcal mol
−
1
and ΔG = −65.5 kcal mol , respectively. The energy landscape
depicted in Figure 1 suggests that the reaction becomes
irreversible after spin crossover, which is thus determining the
final outcome of the reaction. Given the rather small energy
⧧
−1
barrier associated with this reaction, ΔG = 32.4 kcal mol , is
⧧
−1
higher than that given by the oxo, ΔG = 10.0 kcal mol . The
hydroperoxo complex also involves a higher energy barrier,
ΔG = 27.3 kcal mol , and, in contrast with the oxo and the
superoxo, promotes the reaction through a concerted
mechanism, inconsistent with the formation of cyclohexyl
radicals shown by the experiments (Scheme 11).
−
1
difference between MECP1 and MECP2, 3.2 kcal mol , a
mixture of cyclohexanol and cyclohexene is expected, with the
former being the major product. Further refinement with the
dispersion-corrected DFT functionals MPWB1K and M06-2X
gave energy differences of 1.0 and 1.3 kcal mol , respectively.
These values are in good agreement with the hydroxylation:de-
⧧
−1
−1
These results suggested that the superoxo and hydroperoxo
complexes are more stable but less reactive than the oxo, as
saturation ratio observed in the experiments, 93:7 at 60 °C
24a,b
previously reported for similar systems.
In addition, our
(
entry 1 in Table 1), which would correspond to an energy
experiments suggest that the reaction follows the same
mechanism when either oxygen is excluded or hydrogen
peroxide is replaced by oxone (vide supra). Under these
conditions, the formation of peroxidic complexes is unlikely.
These complexes were thus excluded as active species and the
theoretical investigations focused on the oxidation of the
cyclohexyl radical to cyclohexanol from intermediate T-2 on
the oxo pathway (Figure 1). In the classical rebound
mechanism, the organic radical abstracts the OH ligand of
−1
difference of 1.7 kcal mol , thus suggesting that dispersion may
play a role in the selectivity of these reactions. A recent work by
32
Shaik and co-workers has shown that ancillary ligand
flexibility influences the hydroxylation/dehydrogenation ratio.
In our case, the trispyrazolylborate ligand is quite rigid along
the pathway and seems to exert no influence in the reaction
outcome.
The low energy profiles found for the oxidation of
cyclohexane (Figure 1) support the participation of the oxo
complex (Scheme 13) as active species. For the sake of
completeness, we also investigated the formation of this
28
the hydroxo intermediate. All attempts to optimize the
transition state of this reaction in the triplet state led to high-
energy nonconverged geometries. This is due to the need of
,Br
29
complex for the Tp* ligand (Figure S13). After the initial
switching the spin state from the triplet of T-2, which can be
coordination of hydrogen peroxide to the metal center, spin
crossover from the singlet to the triplet state causes the
cleavage of the O−O bond, leading to the formation of a bis-
hydroxo intermediate, similar to the water oxidation catalyst
recently reported by Mayer and Goldberg. This species
undergoes intramolecular proton transfer followed by water
decoordination, yielding the oxo complex. At 22.9 kcal mol−
above reactants, the MECP for spin crossover is the highest
energy point along this reaction pathway, which, overall, is
exergonic by 10.1 kcal mol . These energies suggest that the
formation of the postulated oxo active species is feasible under
the mild conditions used in our experiments (vide supra).
Mechanistic Proposal. On the basis of the experimental
and theoretical data, we have built the mechanistic proposal
shown in Scheme 14. The Tp Cu core reacts with hydrogen
peroxide to give a copper-oxo intermediate with strong oxyl
character. Interaction with cyclohexane induces a hydrogen
abstraction process yielding the cyclohexyl radical and Tp Cu−
OH. From here, two competitive pathways may occur. Through
the hydroxylation pathway (Scheme 14b), the cyclohexyl
Br3
•
•
formulated as [Tp Cu (OH)]C H
(Figure 1), to the
6
11
Br3
singlet of S-3, [Tp Cu(C H OH)], which is a closed-shell
6
11
I
10
Cu d species. We located the minimum energy crossing point
MECP) associated with this spin crossover, MECP1 (Figure
), which has a geometry very similar to that of the classical OH
(
3
3
3
3
0
rebound transition state. In MECP1, the formation of the C−
O bond, d(C···O) = 2.36 Å (3.46 Å in T-2), is accompanied by
the elongation of the Cu−O bond, d(Cu···O) = 1.82 Å (1.78 Å
in T-2). The relaxation of MECP1 in the triplet and singlet
states yielded the expected T-2 and S-3 intermediates,
respectively. In S-3, the cyclohexanol product is formed and
coordinated to the metal center, d(C−O) = 1.43 Å and
d(Cu···O) = 2.06 Å.
1
−
1
The spin density surface of T-2 revealed the presence of
some radical oxyl character in the OH ligand (Figure 2). The
value of ρ(O) is rather small, 0.10 (Table 4), but this has been
proven to be enough to promote dehydrogenation by double H
x
x
5
abstraction. In addition, Solomon and co-workers have
recently shown the active role of copper-hydroxo species in
3
1
C−H oxidation. This prompted the location of an alternative
MECP for the formation of cyclohexene, MECP2 (Figure 3).
The geometrical parameters of this MECP are consistent with
the cleavage of the C−H bond, d(C···H) = 1.19 Å (1.09 Å in T-
x
radical collapses with Tp Cu−OH leading to cyclohexanol
formation, whereas through the dehydrogenation pathway
(
Scheme 14a), a second hydrogen abstraction from the α C−
2
), accompanied by the transfer of H to O, d(H···O) = 1.54 Å
H of the cyclohexyl radical leads to cyclohexene formation. In
the latter case, the oxidant itself acts as hydrogen acceptor, with
the net reaction providing two molecules of water from one
(
2.65 Å in T-2). The Cu−O bond elongates to 1.86 Å (1.78 Å
in T-2) and the C−C bond shortens to 1.43 Å (1.49 Å in T-2).
The full optimization of MECP2 in the triplet and singlet states
yielded intermediates T-2 and S-4, respectively, with the latter
containing the cyclohexene product, d(CC) = 1.33 Å. The
spin stability reversal around MECP1 and MECP2 was
confirmed by single point calculations. Intermediate T-2 is
molecule of H O and cyclohexane. These pathways involve
2 2
spin crossover through MECPs of similar energy, which leads
to product mixtures. The lowest-energy MECP yields cyclo-
hexanol as major reaction product. None of these pathways
involve the formation of free hydroxyl radicals.
3
893
dx.doi.org/10.1021/ja310866k | J. Am. Chem. Soc. 2013, 135, 3887−3896

Journal of the American Chemical Society
Article
Scheme 14. Mechanistic Proposal for the Copper-Catalyzed Oxidation of Cyclohexane Showing the Two Competitive Pathways
CONCLUSIONS
mmol (108 μL) of cyclohexane and 10 mmol (1 mL) of an aqueous
commercial solution of hydrogen peroxide (30% v/v) were added in
one portion. The mixture was stirred for 1 h at 60 °C. After cooling at
■
(
x
In summary, we have discovered that the complexes [Tp Cu-
NCMe)] catalyze the oxidation of nonactivated alkane C−H
room temperature, PPh was added (0.5 mmol), and the reaction
3
bonds providing alcohols and/or ketones as the major products
but inducing the unprecedented dehydrogenation of these
substrates with copper under mild reaction conditions. With n-
hexane as the substrate, 1-hexene has been formed upon such
dehydrogenation process, with other olefins being undetected
in most cases. The use of the above hydrocarbons as substrates,
of copper as catalyst, and of hydrogen peroxide as oxidant is
novel in this kind of transformation. Experimental data as well
as DFT calculations suggest a competitive mechanism in which
the hydroxylation and dehydrogenation pathways are initiated
from a common oxo intermediate. Work aimed at designing
new catalysts for selective dehydrogenation is currently
underway in our laboratories.
mixture was stirred 20 min. Additional dichloromethane (2 × 2.5 mL)
was then added to extract the organic products. The reaction was then
investigated by GC as described above.
Catalytic Oxidation of Cyclohexane in the Presence of CCl
4
or CBrCl . The reaction was performed in a 25 mL round bottomed
3
flask equipped with a reflux condenser and a magnetic stirrer bar. In a
typical procedure, 0.01 mmol of catalyst was dissolved in 2.5 mL of
acetonitrile and 0.5 mL of CCl and 1 mmol (108 μL) of cyclohexane
4
and 10 mmol (1 mL) of an aqueous commercial solution of hydrogen
peroxide (30% v/v) were added in one portion. The mixture was
stirred for 1 h at 60 °C. After cooling at room temperature, additional
dichloromethane (2.5 mL) was added to extract the organic products.
Analysis of the final reaction mixture was done by GC with the above
protocol. The reaction with CBrCl was performed in a similar manner
3
using 3 mL of acetonitrile and 1 mmol of CBrCl (ca. 0.1 mL). See
3
Supporting Information for GC traces.
EXPERIMENTAL SECTION
General. All catalytic experiments were carried out under air. The
■
Catalytic Procedure of Toluene Oxidation. The reactions were
performed in a 25 mL round bottomed flask equipped with a reflux
condenser and a magnetic stirrer bar. In a typical experiment, 0.01
mmol of catalyst was dissolved in 1 mL of acetonitrile and 1 mmol
chemicals were purchased from Aldrich and were used without
x
10
x
previous purification. The Tp ligands and the complexes [Tp Cu-
34
(
NCMe)] were prepared according to the literature procedures.
(106 μL) of toluene and 10 mmol (1 mL) of an aqueous commercial
NMR data were recorded in a Varian Mercury 400 spectrometer. GC
and GCMS data were collected with a Varian GC-3900 with a FID
detector or a Varian Saturn 2100, respectively.
solution of hydrogen peroxide (30% v/v) were added in one portion.
The mixture was stirred for 8 h at 75 °C. Additional dichloromethane
(2 × 2.5 mL) was then added to extract the organic products. Styrene
General Catalytic Oxidation Procedure. The reactions were
performed in a 25 mL round bottomed flask equipped with a reflux
condenser and a magnetic stirrer bar. In a typical experiment (for other
conditions, see the footnotes of Tables 1 and 2), 0.01 mmol of catalyst
was dissolved in 3 mL of acetonitrile and 1 mmol (108 μL) of
cyclohexane and 10 mmol (1 mL) of an aqueous commercial solution
of hydrogen peroxide (30% v/v) were added in one portion. The
mixture was stirred for 1 h at 60 °C. After cooling at room
temperature, additional dichloromethane (2 × 2.5 mL) was added to
extract the organic products. Cycloheptanone was added as internal
standard and the mixture was directly analyzed by GC and GCMS to
determine the mass balance, the conversion and relative ratio of
products by comparison with commercial samples (see Supporting
Information).
was added as internal standard and a sample of the mixture was
directly analyzed by GC/GCMS as above.
Catalytic Procedure Using Oxone as Oxidant. The reactions
were performed in an ampule. In a typical experiment, 0.02 mmol of
catalyst was dissolved in 5 mL of acetonitrile and 30 mmol (3.24 mL)
of cyclohexane. A solution of 0.5 mmol of oxone in 5 mL of water was
added in one portion. The mixture was stirred for 72 h at 60 °C.
Additional dichloromethane (2 × 2.5 mL) was then added to extract
the organic products. Cycloheptanone was added as internal standard
and a sample of the mixture was directly analyzed by GC/GCMS.
Computational Details. The calculations were carried out with
3
5
36
Gaussian09 at the DFT(BHLYP) level. Two distinct basis sets, BS-
I and BS-II, were used. With BS-I, H, B, C, N and O were described
3
7
Catalytic Procedure Using PPh as Additive. The reactions
were performed in a 25 mL round bottomed flask equipped with a
reflux condenser and a magnetic stirrer bar. In a typical procedure,
with the all-electron double-ζ 6-31G** basis set, whereas Br and Cu
3
were described with the Stuttgart−Dresden basis set including scalar
3
8
relativistic ECP. With BS-II, H, B, C, N and O were described with
3
9
0
.01 mmol of catalyst was dissolved in 3 mL of acetonitrile and 1
the triple-ζ 6-311+G** basis set. All stationary points were fully
3
894
dx.doi.org/10.1021/ja310866k | J. Am. Chem. Soc. 2013, 135, 3887−3896

Journal of the American Chemical Society
Article
optimized in gas phase with BS-I without any geometry or symmetry
constraints. Harmonic frequencies were computed analytically with
BS-I in order to classify the stationary points as either minima
REFERENCES
■
(
1) (a) Findlater, M.; Choi, J.; Goldman, A. S.; Brookhart, M. In
Alkane C-H Activation by Single-Site Metal Catalysis; Perez, P. J., Ed.;
Springer: Dordrecht, 2012; Chapter 4. (b) Haibach, M. C.; Kundu, S.;
Brookhart, M.; Goldman, A. S. Acc. Chem. Res. 2012, 45, 947−958.
c) Choi, J.; MacArthur, A. H. R.; Brookhart, M.; Goldman, A. S.
Chem. Rev. 2011, 111, 1761−1779.
(2) Caspary1, K. J.; Gehrke, H.; Heinritz-Adrian, M.; Schwefer, M. In
Handbook of Heterogeneous Catalysis; Ertl, G., Kno
F., Weitkamp, F., Eds.; Wiley: Weinheim, 2008.
(3) (a) Feiters, M. C.; Rowan, A. E.; Nolte, R. J. M. Chem. Soc. Rev.
́
(reactants, intermediates and products) or saddle points (transition
states). These calculations were also used to determine the difference
between the Gibbs and potential energies, (G − E), which includes the
zero-point, thermal and entropy corrections. Hindered rotor
calculations showed that the error associated with the lowest
(
−1
frequencies is smaller than 0.1 kcal mol . The nature of the transition
4
0
̈
̈
zinger, H., Schutz,
states was further confirmed by means of IRC calculations with BS-I.
The effect of the solvent, acetonitrile (ε = 35.688), was estimated by
41
computing the CPCM energy, ECPCM, in single-point calculations
2
000, 29, 375−384. (b) Costas, M.; Mehn, M. P.; Jensen, M. P.; Que,
L., Jr. Chem. Rev. 2004, 104, 939−986. (c) Hirao, H.; Kumar, D.; Que,
L., Jr.; Shaik, S. J. Am. Chem. Soc. 2006, 128, 8590−8606. (d) Que, L.,
Jr. Acc. Chem. Res. 2007, 40, 493−500. (e) Nam, W. Acc. Chem. Res.
2007, 40, 522−531. (f) Murahashi, S. I.; Zhang, D. Chem. Soc. Rev.
with BS-II. All energies given in the text are Gibbs energies in solution,
G , which were calculated by adding the thermodynamic corrections
sol
42
to the CPCM energies (eq 2).
Gsol = ECPCM + (G − E)
(2)
2
008, 37, 1490−1501. (g) Bell, C. B.; Wong, S. D.; Xiao, Y. M.;
The use of the hybrid BHLYP functional is based on previous
Klinker, E. J.; Tenderholt, A. L.; Smith, M. C.; Rohde, J. U.; Que, L.,
Jr.; Cramer, S. P.; Solomon, E. I. Angew. Chem., Int. Ed. 2008, 47,
9071−9074. (h) Bell, S. R.; Groves, J. T. J. Am. Chem. Soc. 2009, 131,
9640−9641. (i) Pan, Z.; Wang, Q.; Sheng, X.; Horner, J. H.;
Newcomb, M. J. Am. Chem. Soc. 2009, 131, 2621−2628. (j) Martinho,
43
studies by Sodupe and Rodriguez-Santiago. In these studies, a series
́
of copper-aqua complexes were studied with several DFT functionals,
with BHLYP giving the most accurate results when compared to
4
4
reference CCSD(T) calculations. Two other functionals, the
45
46
MPWB1K and M06-2X were also used to refine the critical
energy difference between MECP1 and MECP2. These functionals,
which have an amount of HF exchange (44% in MPWB1K and 54% in
M06-2X) similar to that of BHLYP (50%), were designed for an
accurate description of dispersion forces.
̈
M.; Xue, G.; Fiedler, A. T.; Que, L., Jr.; Bominaar, E. L.; Munck, E. J.
Am. Chem. Soc. 2009, 131, 5823−5830.
(4) (a) Kim, C.; Dong, Y.; Que, L., Jr. J. Am. Chem. Soc. 1997, 119,
3635−3636. (b) Company, A.; Prat, I.; Frisch, J. R.; Mas Balleste, R.;
̈
́
̈
Guell, M.; Juhasz, G.; Ribas, X.; Munck, E.; Luis, J. M.; Que, L., Jr.;
The minimum energy crossing points (MECPs) were located with
Costas, M. Chem.Eur. J. 2011, 17, 1622−1634. (c) Mukherjee, A.;
29a
the program developed by J. N. Harvey. The nature of the MECPs
was confirmed by their full optimization in each of the two spin states
involved in the crossing. The thermodynamic and solvent corrections
associated with the MECPs were evaluated by averaging the values
found for each spin state. The local charges and spin densities were
Martinho, M.; Bominaar, E. L.; Munck, E.; Que, L., Jr. Angew. Chem.,
̈
Int. Ed. 2009, 48, 1780−1783.
5) (a) Balcells, D.; Raynaud, C.; Crabtree, R. H.; Eisenstein, O.
(
Chem. Commun. 2009, 1772−1774. (b) Hull, J. F.; Balcells, D.; Sauer,
E. L. O.; Raynaud, C.; Brudvig, G. W.; Crabtree, R. H.; Eisenstein, O. J.
Am. Chem. Soc. 2010, 132, 7605−7616.
47
obtained from NPA calculations.
(
6) Shi, S.; Wang, Y.; Xu, A.; Wang, H.; Zhu, D.; Roy, S. B.; Jackson,
ASSOCIATED CONTENT
Supporting Information
T. A.; Busch, D. H.; Yin, G. Angew. Chem., Int. Ed. 2011, 50, 7321−
324.
7) Luo, Y.-R. In Comprehensive Handbook of Chemical Bond Energies;
CRC Press: Boca Raton, FL, 2007.
(8) Cho, K.-B.; Wu, X.; Lee, Y.-M.; Kwon, Y.-H.; Shaik, S.; Nam, W.
J. Am. Chem. Soc. 2012, 133, 20222−20225.
■
7
(
*
S
Detailed experimental catalytic and mechanistic procedures,
including GC traces; computational data including Cartesian
coordinates and energies of all stationary points reported in the
text as well as further details on the reactivity of the superoxo
and hydroperoxo species and the formation of the oxo active
species; complete ref 35. This material is available free of charge
via the Internet at http://pubs.acs.org.
́
(9) (a) Wender, I. Fuel Process. Technol. 1996, 48, 189−297.
(
b) Molinari, R.; Poeria, T. Asia-Pac. J. Chem. Eng. 2010, 5, 191−206.
(
10) (a) Trofimenko, S. In Scorpionates: The Coordination Chemistry
of Polypyrazolylborate Ligands; Imperial College Press: London, 1999.
b) Pettinari, C. Scorpionates II: Chelating Borate Ligands; Imperial
College Press: River Edge, NJ, 2008.
11) Conde, A.; Díaz-Requejo, M. M.; Per
011, 47, 8154−8156.
12) (a) Barton, D. H. R.; Bev
(
AUTHOR INFORMATION
(
́
ez, P. J. Chem. Commun.
■
2
Corresponding Author
perez@dqcm.uhu.es; agusti@klingon.uab.es; mmdiaz@dqcm.
uhu.es; david.balcells@kjemi.uio.no
(
́ ́
iere, S. D.; Chavarisi, W.; Csuhai, E.;
Doller, D. Tetrahedron 1992, 48, 2895−2910. (b) Velusamy, S.;
Punniyamurthy, T. Tetrahedron Lett. 2003, 44, 8955−8957. (c) Kirillov,
A. M.; Kopylovich, M. N.; Kirillova, M. V.; Haukka, M.; Guedes da
Silva, M. F. C.; Pombeiro, A. J. L. Angew. Chem., Int. Ed. 2005, 44,
Notes
The authors declare no competing financial interest.
4
345−4349. (d) Kirillov, A. M.; Kopylovich, M. N.; Kirillova, M. V.;
Karabach, E. Y.; Haukka, M.; Guedes da Silva, M. F. C.; Pombeiro, A.
J. L. Adv. Synth. Catal. 2006, 348, 159−174. (e) Silva, T. F. S.; Alegria,
E. C. B. A.; Martins, L. M. D. R. S.; Pombeiro, A. J. L. Adv. Synth.
Catal. 2008, 350, 706−716. (f) Mirkhani, V.; Moghadam, M.;
Tangestaninejad, S.; Mohammadpoor-Blatork, I.; Rasouli, N. Catal.
Commun. 2008, 9, 2411−2416. (g) Detoni, C.; Carvalho, N. M. F.;
Aranda, D. A. G.; Louis, B.; Antunes, O. A. C. Appl. Catal., A 2009,
ACKNOWLEDGMENTS
■
Dedicated to Prof. Gregorio Asensio on occasion of his 65th
birthday. We thank MINECO (Proyectos CTQ2011-28942-
CO2-01, CTQ2011-23336, Consolider Ingenio 2010,
CSD2006-00003 and CSD2007-00006) and Fondos Feder for
funding. D.B. and L.V. also thank MINECO for a Juan de la
Cierva contract and a predoctoral fellowship, respectively. Junta
3
65, 281−286. (h) Mahmudov, K. T.; Kopylovich, M. N.; Guedes da
Silva, M. F. C.; Figiel, P. J.; Karabach, Y. Y.; Pombeiro, A. J. L. J. Mol.
Catal. A: Chem. 2010, 318, 44−50. (i) Kopylovich, M. N.; Mahmudov,
K. T.; Guedes da Silva, M. F. C.; Figiel, P. J.; Karabach, Y. Y.;
Kuznetsov, M. L.; Luzyanin, K. V.; Pombeiro, A. J. L. Inorg. Chem.
2011, 50, 918−931. (j) Fernandes, R. R.; Lasri, J.; Guedes da Silva, M.
́
de Andalucia (P10-FQM-06292) is also acknowledged for
funding. D.B. thanks the Norwegian Research Council for
financial support through the CoE Centre for Theoretical and
Computational Chemistry (CTCC) Grant No. 179568/V30.
We thank Dr. Uwe Pischel for helpful discussions.
́
F. C.; da Silva, J. A. L.; Frausto da Silva, J. J. R.; Pombeiro, A. J. L. Appl.
3
895
dx.doi.org/10.1021/ja310866k | J. Am. Chem. Soc. 2013, 135, 3887−3896

Journal of the American Chemical Society
Article
Catal., A 2011, 402, 110−120. (k) Goberna-Ferron
Galan-Mascaros, J. R. Catal. Commun. 2012, 23, 30−33.
13) For one example of hexane hydroxylation see: Barton, D. H. R.;
Csuhai, E.; Doller, D.; Geletii, Y. V. Tetrahedron 1991, 47, 6561−6570.
14) See for example: Atkins, P.; Overton, T.; Rourke, J.; Weller, M.;
Armstrong, F. In Inorganic Chemistry, 4th ed.; Oxford University Press:
Oxford, 2006; Chapter 15.
́
, S.; Lillo, V.;
(33) Barnett, S. M.; Goldberg, K. I.; Mayer, J. M. Nat. Chem. 2012, 4,
498−502.
́
́
(
(34) (a) Mealli, C.; Arcus, C. S.; Wilkinson, J. L.; Marks, T. J.; Ibers,
J. A. J. Am. Chem. Soc. 1976, 98, 711−718. (b) Schneider, J. L.; Carrier,
S. M. C.; Ruggiero, E.; Young, V. G., Jr.; Tolman, W. B. J. Am. Chem.
Soc. 1998, 120, 11408−11418. (c) Mairena, M. A.; Urbano, J.; Carbajo,
(
́
J.; Maraver, J. J.; Alvarez, E.; Díaz-Requejo, M. M.; Perez, P. J. Inorg.
(
3
(
15) Walling, C.; Johnson, R. A. J. Am. Chem. Soc. 1975, 97, 363−
67.
16) (a) Kirillova, M. V.; Kirillov, A. M.; Guedes da Silva, M. F. C.;
Chem. 2007, 46, 7428−7435.
(35) Frisch, M. J. et al. Gaussian 09, Revision A.02; Gaussian, Inc.:
Wallingford, CT, 2009.
̀
Pombeiro, A. J. L. Eur. J. Inorg. Chem. 2008, 3423−3427. (b) Shulpin,
(36) (a) Becke, A. D. Phys. Rev. A 1988, 38, 3098−3100. (b) Lee, C.;
Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785−789. (c) Miehlich, B.;
Savin, A.; Stoll, H.; Preuss, H. Chem. Phys. Lett. 1989, 157, 200−206.
(37) (a) Hehre, W. J.; Ditchfield, R.; Pople, J. A. J. Chem. Phys. 1972,
G. B. J. Mol. Catal. A: Chem. 2002, 189, 39−66.
(
17) Lyons, J. E.; Ellis, P. E.; Myers, H. K., Jr. J. Catal. 1995, 155, 59−
73.
(
18) (a) Sawyer, D. T.; Kang, C.; Llobet, A.; Redman, C. J. Am.
56, 2257−2261. (b) Ehlers, A. W.; Bo
A.; Hollwarth, A.; Jonas, V.; Kohler, K. F.; Stegmann, R.; Veldkamp,
A.; Frenking, G. Chem. Phys. Lett. 1993, 208, 111−114. (c) Hollwarth,
A.; Bohme, H.; Dapprich, S.; Ehlers, A. W.; Gobbi, A.; Jonas, V.;
Kohler, K. F.; Stegmann, R.; Veldkamp, A.; Frenking, G. Chem. Phys.
̈
hme, M.; Dapprich, S.; Gobbi,
Chem. Soc. 1993, 115, 5817−5817. (b) Hagem, J. P.; Llobet, A.;
Sawyer, D. T. Bioorg. Med. Chem. 1995, 3, 1383−1388.
̈
̈
̈
(
19) (a) Bossmann, S. H.; Oliveros, E.; Go
̈
b, S.; Siegwart, S.; Dahlen,
rner, M.; Braun, A. M. J. Phys.
̈
E. P.; Payawan, Jr.; Straub, M.; Wo
Chem. A 1998, 102, 5542−5550. (b) Marusawa, H.; Ichikawa, K.;
Narita, N.; Murakami, H.; Ito, K.; Tezuka, T. Bioorg. Med. Chem. 2002,
̈
̈
Lett. 1993, 208, 237−240. (d) Hariharan, P. C.; Pople, J. A. Theor.
Chim. Acta 1973, 28, 213−222.
̈
(38) (a) Andrae, D.; Haussermann, U.; Dolg, M.; Stoll, H.; Preuss, H.
1
(
0, 2283−2290.
Theor. Chim. Acta 1990, 77, 123−141. (b) Bergner, A.; Dolg, M.;
Kuchle, W.; Stoll, H.; Preuss, H. Mol. Phys. 1993, 80, 1431−1441.
39) (a) McLean, A. D.; Chandler, G. S. J. Chem. Phys. 1980, 72,
639−5648. (b) Raghavachari, K.; Binkley, J. S.; Seeger, R.; Pople, J. A.
J. Chem. Phys. 1980, 72, 650−654.
40) (a) Hratchian, H. P.; Schlegel, H. B. J. Chem. Phys. 2004, 120,
918−9924. (b) Hratchian, H. P.; Schlegel, H. B. J. Chem. Theory
20) For the generation of metal-oxo species using oxone see: Chow,
T. W.-S.; Liu, Y.; Che, C.-M. Chem Commun. 2011, 47, 11204−11206
and references cited therein.
̈
(
5
(
1
(
8
21) Mahy, J.-P.; Bedy, G.; Battioni, P.; Mansuy, D. New J. Chem.
989, 13, 651−657.
22) Balcells, D.; Clot, E.; Eisenstein, O. Chem. Rev. 2010, 110, 749−
23.
23) (a) Hay-Motherwell, R. S.; Wilkinson, G.; Hussain-Bates, B.;
(
9
Comput. 2005, 1, 61−69.
(
(
(
41) Barone, V.; Cossi, M. J. Phys. Chem. A 1998, 102, 1995−2001.
42) (a) Balcells, D.; Ujaque, G.; Fernandez, I.; Khiar, N.; Maseras, F.
Hursthouse, M. B. Polyhedron 1993, 12, 2009−2012. (b) Jin, N.;
Ibrahim, M.; Spiro, T. G.; Groves, J. T. J. Am. Chem. Soc. 2007, 129,
J. Org. Chem. 2006, 71, 6388−6396. (b) Braga, A. A. C.; Ujaque, G.;
1
2416−12417. (c) Li, F.; Meier, K. K.; Cranswick, M. A.; Chakrabarti,
M.; Van Heuvelen, K. M.; Munck, E.; Que, L. J. Am. Chem. Soc. 2011,
33, 7256−7259. (d) Prat, I.; Mathieson, J. S.; Guell, M.; Ribas, X.;
Luis, J. M.; Cronin, L.; Costas, M. Nat. Chem. 2011, 3, 788−793.
24) (a) Yoshizawa, K.; Kihara, N.; Kamachi, T.; Shiota, Y. Inorg.
Maseras, F. Organometallics 2006, 25, 3647−3658.
̈
(
̀
43) (a) Poater, J.; Sola, M.; Rimola, A.; Rodríguez-Santiago, L.;
1
̈
Sodupe, M. J. Phys. Chem. A 2004, 108, 6072−6078. (b) Rios-Font, R.;
Sodupe, M.; Rodríguez-Santiago, L.; Taylor, P. R. J. Phys. Chem. A
(
2
(
010, 114, 10857−10863.
44) (a) Purvis, G. D., III; Bartlett, R. J. J. Chem. Phys. 1982, 76,
Chem. 2006, 45, 3034−3041. (b) Crespo, A.; Marti, M. A.; Roitberg,
A. E.; Amzel, L. M.; Estrin, D. A. J. Am. Chem. Soc. 2006, 128, 12817−
1
910−1918. (b) Scuseria, G. E.; Janssen, C. L.; Schaefer, H. F., III. J.
12828. (c) Huber, S. M.; Ertem, M. Z.; Aquilante, F.; Gagliardi, L.;
Chem. Phys. 1988, 89, 7382−7387. (c) Pople, J. A.; Head-Gordon, M.;
Raghavachari, K. J. Chem. Phys. 1987, 87, 5968−5975.
(
(
(
Tolman, W. B.; Cramer, C. J. Chem.Eur. J. 2009, 15, 4886−4895.
d) Poater, A.; Ribas, X.; Llobet, A.; Cavallo, L.; Sola, M. J. Am. Chem.
Soc. 2008, 130, 17710−17717. (e) Guell, M.; Luis, J. M.; Siegbahn, P.
E. M.; Sola, M. J. Biol. Inorg. Chem. 2009, 14, 273−285. (f) Itoh, S.;
(
̀
45) Zhao, Y.; Truhlar, D. G. J. Phys. Chem. A 2004, 108, 6908−6918.
46) Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215−241.
47) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88,
̈
̀
Taki, M.; Nakao, H.; Holland, P. L.; Tolman, W. B.; Que, L.;
Fukuzumi, S. Angew. Chem., Int. Ed. 2000, 39, 398−400. (g) García-
Bosch, I.; Ribas, X.; Costas, M. Chem.Eur. J. 2012, 18, 2113−2122.
8
99−926.
(
25) García-Bosch, I.; Company, A.; Cady, C. W.; Styring, S.;
Browne, W. R.; Ribas, X.; Costas, M. Angew. Chem., Int. Ed. 2011, 50,
648−5653.
26) (a) Lundberg, M.; Blomberg, M. R. A.; Siegbahn, P. E. M. Inorg.
5
(
Chem. 2004, 43, 264−274. (b) Siegbahn, P. E. M.; Crabtree, R. H. J.
Am. Chem. Soc. 1999, 121, 117−127.
(
27) Balcells, D.; Raynaud, C.; Crabtree, R. H.; Eisenstein, O. Chem.
Commun. 2008, 744−746.
28) (a) Groves, J. T. J. Chem. Educ. 1985, 62, 928−931. (b) Groves,
J. T.; McClusky, G. A. J. Am. Chem. Soc. 1976, 98, 859−861.
29) (a) Harvey, J. N.; Aschi, M.; Schwarz, H.; Koch, W. Theor.
Chem. Acc. 1998, 99, 95−99. (b) Poli, R.; Harvey, J. N. Chem. Soc. Rev.
003, 32, 1−8. (c) Manner, V. W.; Lindsay, A. D.; Mader, E. A.;
Harvey, J. N.; Mayer, J. M. Chem. Sci. 2012, 3, 230−243.
30) (a) Meunier, B.; de Visser, S. P.; Shaik, S. Chem. Rev. 2004, 104,
947−3980. (b) Shaik, S.; Cohen, S.; Wang, Y.; Chen, H.; Kumar, D.;
Thiel, W. Chem. Rev. 2010, 110, 949−1017.
31) Donoghue, P. J.; Tehranchi, J.; Cramer, C. J.; Sarangi, R.;
Solomon, E. I.; Tolman, W. B. J. Am. Chem. Soc. 2011, 133, 17602−
7605.
32) Usharani, D.; Janardanan, D.; Shaik, S. J. Am. Chem. Soc. 2011,
33, 176−179.
(
(
2
(
3
(
1
(
1
3
896
dx.doi.org/10.1021/ja310866k | J. Am. Chem. Soc. 2013, 135, 3887−3896