Cyclohexane oxidation with dioxygen catalyzed by supported pyrazole rhenium complexes

Article abstract of DOI:10.1016/j.molcata.2008.01.022

The pyrazole complexes [ReCl₂{N₂C(O)Ph}(Hpz)(PPh₃)₂ ] 2 (Hpz = pyrazole), [ReCl₂{N₂C(O)Ph}(Hpz)₂(PPh₃) ] 3 and [ReClF{N₂C(O)Ph}(Hpz)₂(PPh₃)] 4, and their precursor [ReOCl₃(PPh₃)₂] 1, immobilized on 3-aminopropyl functionalized silica, catalyze the cyclohexane oxidation with dioxygen, to cyclohexanol and cyclohexanone (the main product), in the absence of solvent and additives and under relatively mild conditions. Complex 4, whose synthesis and characterization are reported herein for the first time, provides the best activity (ca. 16% overall conversion towards the ketone and alcohol, at the O₂ pressure of 19 atm, at 150 °C, 8 h reaction time). The reaction is further promoted by pyrazinecarboxylic acid. TGA analysis shows that the supported complexes are stable up to ca. 200 °C. The use of radical traps supports the involvement of a free-radical mechanism via carbon- and oxygen-centred radicals. The effects of various factors were studied towards the optimization of the processes. Complex 4 also catalyzes the oxidation of other cycloalkanes to the corresponding cycloalkanols and cycloalkanones.

Full text of DOI:10.1016/j.molcata.2008.01.022

Available online at www.sciencedirect.com
Journal of Molecular Catalysis A: Chemical 285 (2008) 92–100
Cyclohexane oxidation with dioxygen catalyzed by
supported pyrazole rhenium complexes
Gopal S. Mishraa, Elisabete C.B. Alegriaa,b, Luısa M.D.R.S. Martins
,
a,b
´
a
a,∗
˜
´
Joao J.R. Frausto da Silva , Armando J.L. Pombeiro
a Centro de Qu´ımica Estrutural, Complexo I, Instituto Superior Te´cnico, TU Lisbon,
Av. Rovisco Pais, 1049-001 Lisboa, Portugal
b Departamento de Engenharia Qu´ımica, ISEL, R. Conselheiro Em´ıdio Navarro, 1950-062 Lisboa, Portugal
Received 25 October 2007; received in revised form 9 January 2008; accepted 13 January 2008
Available online 20 January 2008
Abstract
The pyrazole complexes [ReCl2{N2C(O)Ph}(Hpz)(PPh3)2]
2
(Hpz = pyrazole), [ReCl2{N2C(O)Ph}(Hpz)2(PPh3)]
3
and
[ReClF{N2C(O)Ph}(Hpz)2(PPh3)] 4, and their precursor [ReOCl3(PPh3)2] 1, immobilized on 3-aminopropyl functionalized silica, cat-
alyze the cyclohexane oxidation with dioxygen, to cyclohexanol and cyclohexanone (the main product), in the absence of solvent and additives
and under relatively mild conditions. Complex 4, whose synthesis and characterization are reported herein for the first time, provides the best
activity (ca. 16% overall conversion towards the ketone and alcohol, at the O2 pressure of 19 atm, at 150 ◦C, 8 h reaction time). The reaction is
further promoted by pyrazinecarboxylic acid. TGA analysis shows that the supported complexes are stable up to ca. 200 ◦C. The use of radical
traps supports the involvement of a free-radical mechanism via carbon- and oxygen-centred radicals. The effects of various factors were studied
towards the optimization of the processes. Complex 4 also catalyzes the oxidation of other cycloalkanes to the corresponding cycloalkanols and
cycloalkanones.
© 2008 Elsevier B.V. All rights reserved.
Keywords: Rhenium; Cyclohexane; Dioxygen; Carboxylic acids; Radical mechanism; Cyclohexanol; Cyclohexanone
1. Introduction
alysts for the peroxidative oxidation [6–8], or carboxylation [6]
of some alkanes.
In spite of the very rich coordination chemistry that has been
expanded for rhenium [1–4], the use of complexes with this
metal as catalysts is still an underdeveloped field of research.
However, the wide application of [(Me)ReO3] (MTO) in oxida-
tion catalysis [4–6] clearly demonstrates the ability of this metal
to form highly active catalysts for oxidation reactions of olefins
and other unsaturated substrates.
Recently we have found that some rhenium complexes with
O- or N-ligands can catalyze, in homogeneous systems, alkane
functionalization reactions, under mild or moderate conditions.
Hence, MTO and other Re oxides [6], benzoylhydrazido- and
benzoyldiazenido-Re complexes [7], and trispyrazolylmethane
and pyrazole complexes of Re [8], can act as homogeneous cat-
For these Re systems, dioxygen was not an adequate oxidant,
but we have observed that, when supported on suitably modified
silica gel, some bis(maltolato)oxovanadium complexes could
catalyze the partial oxidation, with O2, of cyclic [9–11] and
linear [12] alkanes to the corresponding alcohols and ketones.
Hence, the extension of this type of study to rhenium complexes,
by immobilizing them on a silica gel support, towards the devel-
opment of rhenium heterogenous catalytic systems (with the
normal advantages over the homogenous ones e.g. in terms of
easy catalyst separation and recovery) that could operate with
the ideal “green” oxidant, i.e. dioxygen, for oxidative function-
alization reactions of alkanes, constitute main aims of the current
study. A contribution towards widening the use of Re complexes
in catalysis is also an objective of this work.
We have selected cyclohexane as the substrate in view of the
relevance of its partially oxidized products (cyclohexanol and
cyclohexanone) for the production of adipic acid and caprolac-
∗
Corresponding author. Fax: +351 218464455.
E-mail address: pombeiro@ist.utl.pt (A.J.L. Pombeiro).
1381-1169/$ – see front matter © 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2008.01.022

G.S. Mishra et al. / Journal of Molecular Catalysis A: Chemical 285 (2008) 92–100
93
tam, which are further used in the manufacture of nylon-6,6 and
nylon-6, respectively [13–17]. Normally, Co compounds (cobalt
naphthenate or cobalt acetate) have been used for the industrial
oxidation reaction, with dioxygen as oxidant, at a temperature
above 150 ◦C, and only a low conversion (ca. 4%) is achieved to
obtain a high selectivity (ca. 85%) towards a mixture of cyclo-
hexanone and cyclohexanol [17]. Studies have been undertaken
to develop new selective catalytic systems with a higher activity
under mild conditions and using different oxidizing agents (e.g.
hydrogen peroxide, t-butyl hydroperoxide, molecular oxygen
and ozone). Heterogeneous catalysts are often oxides or metal
cations incorporated in inorganic supports (e.g. silica, alumina,
zirconia, active carbon, zeolites or aluminophosphates), while
promoters can be used to reduce the induction period and to
increase the selectivity for the target products [18–24]. Other
oxidation catalysts are transition metal-substituted heteropoly
compounds (heteropolyoxometalates), which can show activity
in the oxidation of alkanes [14,24–28].
In the current study, the new pyrazole rhenium com-
pound [ReClF{N2C(O)Ph}(Hpz)2(PPh3)] 4 (Hpz = pyrazole),
whose synthesis and characterization are reported herein
for the first time, and other rhenium complexes, i.e.
[ReOCl3(PPh3)2] 1, [ReCl2{N2C(O)Ph}(Hpz)(PPh3)2] 2 and
[ReCl2{N2C(O)Ph}(Hpz)2(PPh3)] 3, have been supported on
3-aminopropyl functionalized silica and used as heteroge-
neous catalysts in an efficient and selective oxidation of
cyclohexane (mainly to cyclohexanone), by dioxygen in the
absence of solvent and additives, under relatively mild reaction
conditions.
Analysis. Found: C, 45.4; H, 3.3; N, 9.8%.
ReC31H26N6POClF.1/2CH2Cl2 requires C, 45.3; H, 3.5;
N, 9.6.
FT-IR (KBr pellet): 3239 [s, ␯(NH)], 1588 [s, ␯(C O)], 1538
[vs, ␯(N N) N2COPh and/or ␯(N C) Hpz], 430 [w, Re–F].
[1H] NMR (CDCl3): δ 15.18 and 12.66 [s, 1H + 1H, H(1)
Hpz], 7.95 [dd, 2H, 3JHH = 7.2, 4JHH = 2.0, o-COPh], 7.63 [m,
1H, H(3) or H(5) Hpz], 7.62–7.55 [m, 6H (o- or m-PPh3)], 7.53
[m, 1H, p-COPh], 7.50 [m, 1H, H(3) or H(5) Hpz], 7.42 [m, 2H,
m-COPh], 7.38–7.32 [m, 9H, {(m- or o-) + p}-PPh3], 7.29 [s,
1H, H(3) or H(5), Hpz], 7.0 [m, 1H, H(3) or H(5), Hpz], 6.35 [q,
1H, JHH = 2.4, H(4), Hpz], 6.18 [q, 1H, JHH = 2.4, H(4) Hpz].
1
31P-{ H} NMR (CDCl3): δ 4.43 [d, 2JPF = 26.8].
1
13C-{ H} NMR (CDCl3): δ 171.12 [s, C O], 139.69 and
137.93 [s, C(3) Hpz], 134.0 [d, JCP = 9.8, Co or Cm (PPh3)],
133.9 [s, Cp (COPh)], 132.14 [s, C(5) Hpz], 132.0 [s, Co or Cm
(COPh)], 131.46 [s, C(5) Hpz], 130.7 [d, JCP = 2.5, Cp (PPh3)],
129.91 [d, JCP = 9.4, Ci (PPh3)], 129.76 [s, Ci (COPh)], 128.78
[s, Cm or Co (COPh)], 128.14 [d, JCP = 10.6, Cm or Co (PPh3)],
106.94 and 106.50 [s, C(4) Hpz].
[13C] NMR (CDCl3): 171.12, 139.69 and 137.93 [d,
JCH = 184.2 and 187.4 C(3) Hpz], 134.0 [dd, JCH = 158.9,
Co or Cm (PPh3)], 133.9 [d, JCH = 135.6, Cp (COPh)],
132.14 [s, JCH = 162.6 C(5) Hpz], 132.0 [d, JCH = 172.5, Co
or Cm (COPh)], 131.46 [d, JCH = 157.6 C(5) Hpz], 130.7
[dd, JCH = 159.1, Cp (PPh3)], 129.91, 129.76, 128.78 [dm,
JCH = 150.8, Cm or Co (COPh)], 128.14 [dd, JCH = 162.5, Cm
or Co (PPh3)], 106.94 and 106.50 [d, JCH = 183.5 and 186.2,
C(4) Hpz].
FAB+-MS: m/z 772 ([M]+), 639 ([M–N2COPh]+),
753 ([M–F]+), 737 ([M–F–Cl]+), 704 ([M–Hpz]+), 571
([M–N2COPh–Hpz]+), 517 ([M–N2COPh–Hpz–F]+).
2. Experimental
2.1. Synthesis of the rhenium complexes (1–3)
2.3. Immobilization of the Re complexes on chemically
modified silica
Complexes [ReCl2{␩2-N,O-N2C(O)Ph}(PPh3)2] [29],
[ReCl2{N2C(O)Ph}(Hpz)(PPh3)2] 2 (Hpz = pyrazole) [30],
[ReCl2{N2C(O)Ph}(Hpz)2(PPh3)] 3 [30], and the precur-
sor [ReOCl3(PPh3)2] 1 [31,32] were prepared according
to published procedures. Reaction of 1 with an excess of
PPh3 and PhCONHNH2 led to the formation of [ReCl2{␩2-
N,O-N2C(O)Ph}(PPh3)2] [29], whereas complexes 2 and 3
were synthesized by treatment of the latter with pyrazole or
tris(pyrazolyl)methane (HCpz3) [30] (reactions a–d, Scheme 1).
We have used the wet-impregnation method [33] for
immobilization of the rhenium complexes on the surface of 3-
aminopropyl modified silica gel (available from Sigma). Each of
the above rhenium complexes (25 mg) was separately dissolved
in methanol (10 mL), the solution added to the 3-aminopropyl
modified silica (1.0 g) and the mixture refluxed at 80 ◦C for 5 h.
The modified silica gel with the supported rhe-
nium complexes was filtered-off, washed several times
with methanol, dried at 40 ◦C for 6 h, under dini-
trogen, and then at 50 ◦C for 4 h, in an oven. After
drying we found that 19 mg of [ReOCl3(PPh3)2] 1,
21 mg of [ReCl2{N2C(O)Ph}(Hpz)(PPh3)2] 2, 20 mg
2.2. Synthesis of the pyrazole fluoro-rhenium complex 4
The newly synthesized pyrazole fluoro-rhenium complex
[ReClF{N2C(O)Ph}(Hpz)2(PPh3)] 4 was obtained (reaction e,
Scheme 1) upon a slow addition of a methanol (15 mL) solu-
tion of Tl[BF4] (24 mg, 0.084 mmol) to a dichloromethane
of [ReCl2{N2C(O)Ph}(Hpz)2(PPh3)]
[ReClF{N2C(O)C6H5}(Hpz)2(PPh3)]
3
or 23 mg of
were separately
4
(20 mL) solution of [ReCl2{N2C(O)Ph}(Hpz)2(PPh3)]
3
loaded per gram of 3-aminopropyl functionalized silica sup-
port. The colours of the 3-aminopropyl functionalized silica
supported catalysts are as follows: light green for 1 and yellow
for 2–4.
(57 mg, 0.073 mmol). The mixture was stirred under reflux for
ca. 48 h and filtered to remove TlCl. Concentration under vac-
uum of the dark green solution followed by slow addition of
Et2O resulted in the precipitation of complex 4 as a dark green
solid which was filtered-off, washed with Et2O and dried under
vacuum (0.040 g, 56% yield).
The immobilization of the metal compounds possibly occurs
via hydrogen-bonding between the NH2 (and/or OH) groups
of the silica support and ligands of the complexes (halides, ben-

94
G.S. Mishra et al. / Journal of Molecular Catalysis A: Chemical 285 (2008) 92–100
Scheme 1.
zoyldiazenide and/or pyrazole). Anchoring by covalent bond via
the NH2 moieties [34,35] is not believed to occur in our cases
since the Re complexes do not react with n-propylamine under
the immobilization reaction conditions.
Although in the IR spectrum (KBr pellet) of the final sup-
ported catalyst the bands of the complex are buried under
those of the support, one can observe a broad shoulder (at
ca. 3300–3150 cm−1) of the very strong ␯(N–H) band (maxi-
mum at ca. 3470 cm−1) of the amino-modified silica support
that possibly can be assigned to the H-bonded groups involved
in H-bonding between the complex and the 3-aminopropyl
silica.
NMR data, assignments and coupling constants common to
1
the 13C–{ H} NMR spectra are not repeated. Coupling con-
stants are in Hz; abbreviations: s = singlet, d = doublet, t = triplet,
q = quartet, m = complex multiplet; br = broad, dd = doublet
or doublets, dm = doublet of multiplets. Atom labelling for
the Hpz ligand: HN(1)N(2)C(3)HC(4)HC(5)H. Gas chro-
matography was carried out with a FISONS Instruments
GC 8000 series gas chromatograph equipped with a FID
detector and a DB-WAX capillary column (length: 30 m;
internal diameter: 0.32 mm). Positive-ion FAB mass spec-
tra were obtained on a Trio 2000 Fisons spectrometer by
bombarding 3-nitrobenzyl alcohol (NBA) matrices of the
samples with 8 keV (ca. 1.28 × 1015 J) Xe atoms. Mass cal-
ibration for data system acquisition was achieved using CsI,
GC–MS analyses were performed by using this spectrometer
with a coupled gas chromatograph Carlo Erba Instruments,
Auto/HRGC/MS. The C, H, N elemental analyses were car-
ried out by the Microanalytical Service of the Instituto
2.4. Instrumentation
Infrared spectra (4000–400 cm−1) were recorded on a Jasco
FTIR-430 spectrometer in transmission mode using KBr pel-
lets; wavenumbers are in cm−1; abbreviations: vs = very strong,
s = strong, m = medium, w = weak, br = broad. 1H, 31P and
13C NMR spectra were recorded on a Varian Unity 300
spectrometer at ambient temperature; δ values are in ppm
relative to SiMe4 (1H or 13C) or H3PO4 (31P). In the 13C
´
Superior Tecnico (IST). Thermogravimetric analysis (TGA)
of the catalyst was performed at a Setaram TG/DTA/DSC-
92 instrument at the Chemical Engineering Department,
IST.

G.S. Mishra et al. / Journal of Molecular Catalysis A: Chemical 285 (2008) 92–100
95
2.5. Typical oxidation reaction procedure
The oxidation reactions were preformed in a high pressure
stainless steel cylindrical rocking batch reactor (22 mL capac-
ity), provided with a gas delivery inlet and a pressure gauge.
The inside temperature was controlled by an on/off controller
with a suitable thermocouple for temperature sensing. In typical
conditions, 3.0 mL (27.8 mmol) of neat cyclohexane and 20 mg
of supported rhenium catalyst were used for 8 h. At the end of
the reactions, the supported catalysts, whose colours (see above)
had turned into light brown, were separated from the solutions
by filtration, using a filter paper, and washed three times with
acetone. The supported catalysts could be reactivated for further
use by heating in an air oven at 60 ◦C for 6 h. In the experiments
with unsupported Re catalysts, the reactions were carried out
in a similar way to that of the supported ones, but by using the
former (1.2 mg) instead of the latter (20 mg) catalysts.
The oxidation products were analyzed by gas chromatogra-
phy (GC) (30 ␮L of cyclopentanone added as internal standard
to 1.0 mL of the filtered final reaction solution). The temperature
of injection of the chromatograph (see above) was 240 ◦C. The
initial temperature was maintained at 100 ◦C for 1 min, then
raised 10 ◦C/min to 180 ◦C, and held at this temperature for
1 min. Helium was used as the carrier gas. They were further ana-
lyzed by GC–MS measurements. The turnover numbers (TONs)
(moles of product per mole of catalyst) and yields (moles of
product per mole of cyclohexane) were estimated. Blank exper-
iments with silica gel or the 3-aminopropyl modified silica gel,
but without the Re catalyst, were also performed, giving max-
imum conversions to cyclohexanol and cyclohexanone of 1.7
and 0.6%, respectively (under the same conditions of entry 3,
Table 1). This low, although detectable, activity can result, at
least in part, from the effect of the stainless steel reactor walls
[36], but this was not investigated further, in the current study,
in view of its low significance.
Fig. 1. Effect of temperature on the conversion of cyclohexane, upon oxidation
by molecular O2, in the presence of [ReClF{NNC(O)C6H5}(Hpz)2(PPh3)] 4
supported on 3-aminopropyl modified silica (p(O2) = 13.6 atm, time = 8 h). Point
numbering corresponds to runs of Table 1.
anol, the main products. As a result of the chemical reaction, the
colour of the supported catalysts changes to brown.
After being used, the supported catalysts can be reactivated
upon washing (three times) with acetone and drying in an air
oven at 60 ◦C for 6 h, showing usually a comparable activity to
that of the initial run.
The effects of various factors (temperature, oxygen pressure,
reaction time, amount of catalyst, presence of a co-catalyst)
on the catalytic activity and selectivity, as well as the activity
towards other substrates, were further studied (Tables 1–3 and
Figs. 1–7) for the case of the supported complex 4 which pro-
3. Results and discussion
The supported complexes (1–4) act as selective catalysts
(or catalyst precursors) for the oxidation of neat cyclohex-
ane, by molecular oxygen, to cyclohexanol and, mainly,
cyclohexanone, under relatively mild conditions, without any
solvent or additive (Table 1). The pyrazole fluoro-complex
[ReClF{N2C(O)Ph}(Hpz)2(PPh3)] 4 provides the most active
system, in a typical run (entry 3, Table 1) leading, after 8 h
reaction time, to yields (moles of product/mole of cyclohex-
ane) of cyclohexanone and cyclohexanol of 8.3 and 2.5% (10%
overall for these products, within a total 11.3% of cyclohexane
conversion, including the by-products), for p(O2) = 13.6 atm, at
150 ◦C. The activity corresponds to a TON (estimated simply as
the number of moles of cyclohexanone + cyclohexanol per mole
Re-complexloadedonthesilica)of5.0 × 103 andismuchhigher
than those of the other catalysts (see entries 18–20, Table 1).
The strong smell at the end of the reaction, when carried out
in the presence of the catalyst, indicated the products formation,
what was further confirmed by GC and GC–MS analyses that
allow to quantify the amounts of cyclohexanone and cyclohex-
Fig. 2. Effect of temperature on the selectivity of the oxidation of cyclohexane
by molecular O2, in the presence of [ReClF{NNC(O)C6H5}(Hpz)2(PPh3)] 4
supported on 3-aminopropyl modified silica (p(O2) = 13.6 atm, time = 8 h). Point
numbering corresponds to runs of Table 1.

96
G.S. Mishra et al. / Journal of Molecular Catalysis A: Chemical 285 (2008) 92–100
Table 1
Cyclohexane oxidation with dioxygen catalyzed by supported rhenium complexesa
Entry
Catalyst
p(O2)b (atm)
Time (h)
Catal (mg)
Temperature (◦C)
Conversion (%)c
Cyclohexanone
Cyclohexanol
Others
Overall
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
1
2
3
13.6
13.6
13.6
13.6
13.6
5.5
8
8
8
8
8
8
8
8
8
8
4
6
12
24
8
8
8
20
20
20
20
20
20
20
20
20
20
20
20
20
20
10
30
50
20
20
20
100
125
150
175
200
150
150
150
150
150
150
150
150
150
150
150
150
150
150
150
2.2
4.1
8.3
9.7
11.7
1.3
4.3
6.4
10.4
12.8
1.9
1.3
2.2
2.5
2.7
3.3
0.8
2.2
2.8
2.9
3.3
0.5
1.4
2.3
3.2
1.1
3.5
3.7
2.3
2.1
3.0
0.6
0.6
0.5
1.1
1.9
0.0
0.4
0.5
1.2
2.2
0.0
0.1
1.4
3.5
0.0
0.8
1.6
0.0
0.0
0.2
4.1
6.9
11.3
13.5
16.9
2.1
6.9
9.7
14.5
18.3
2.4
8.2
11.0
16.3
19.0
13.6
13.6
13.6
13.6
13.6
13.6
13.6
13.6
13.6
13.6
3.8
5.3
10.0
11.2
2.4
11.7
13.1
3.0
13.7
17.9
3.5
16.0
18.4
5.3
8
8
8
5.0
5.6
7.1
8.8
a
Supported Re catalyst = 20 mg, cyclohexane = 27.8 mmol, in an autoclave (22 mL capacity).
Measured at 25 ◦C (1 atm = 1.01 bar = 101 kPa).
Percentage molar yield (moles of product per mole of cyclohexane).
b
c
Table 2
Effect of various heteroaromatic carboxylic acids (as co-catalysts) on the cyclohexane oxidation with dioxygen catalyzed by [ReClF{NNC(O)C6H5}(Hpz)2(PPh3)]
4a
Co-catalyst
Entry
Conversion (%)b
Cyclohexanone
Cyclohexanol
Overall
3-Amino-2-pyrazinecarboxylic acid
2,6-Pyrazinedicarboxylic acid
2,3-Pyrazinedicarboxylic acid
Pyrazinecarboxylic acid
1
2
3
4
5
10.3
10.6
11.5
12.7
8.8
3.1
3.0
3.4
3.3
2.8
14.3
14.9
15.6
16.8
11.9
Picolinic acid
a
Same conditions as those indicated for the entry 3 in Table 1.
Percentage molar yield (moles of product per mole of cyclohexane).
b
vides the most active catalytic system. TGA analysis of this
freshly supported catalyst shows that it is stable up to ca. 200 ◦C
(Fig. 8).
The unsupported Re complexes do not lead to amounts of
products considerably above those obtained in blank assays,
what suggests the involvement of a heterogeneous catalytic
process, the Re complexes, under homogeneous reaction con-
ditions, conceivably undergoing decomposition into inactive
species.
3.1. Effect of temperature
Table 3
Oxidations of various cycloalkanes and of benzene with dioxygen catalyzed by
The temperature effect on the oxidation of cyclohexane was
examined in the range from 100 to 200 ◦C (Table 1, entries 1–5,
and Fig. 1, for p(O2) = 13.6 atm and 8 h reaction time). The reac-
tion temperature was not allowed to go beyond 200 ◦C due to the
formation of an unidentified black material. Although the con-
version of cyclohexane to cyclohexanol increases rather slowly,
a sharp increase occurs towards cyclohexanone (11.7% yield at
200 ◦C, with an overall conversion of 16.9%). In order to test if
this behaviour could reflect the conversion of cyclohexanol to
cyclohexanone, we have checked if this alcohol could be oxi-
supported [ReClF{NNC(O)C6H5}(Hpz)2(PPh3)] 4a
Entry
Substrate
Conversion (%)b
Cyclohexanone
Cyclohexanol
1
2
3
4
Cyclopentane
Cycloheptane
Cyclooctane
Benzene
4.7
7.5
8.8
0.0
2.8
3.8
3.3
0.5
a
Same conditions as those for the entry 3 in Table 1.
Percentage molar yield (moles of product per mole of cyclohexane).
b

G.S. Mishra et al. / Journal of Molecular Catalysis A: Chemical 285 (2008) 92–100
97
Fig. 3. Effect of O2 pressure on the conversion of cyclohexane, upon oxidation
by molecular O2, in the presence of [ReClF{NNC(O)C6H5}(Hpz)2(PPh3)] 4
supported on 3-aminopropyl modified silica (temperature = 150 ◦C, time = 8 h).
Point numbering corresponds to runs of Table 1.
Fig. 5. Effect of time on the selectivity of the oxidation of cyclohexane by molec-
ular O2, in the presence of [ReClF{NNC(O)C6H5}(Hpz)2(PPh3)] 4 supported
on 3-aminopropyl modified silica (temperature = 150 ◦C, p(O2) = 13.6 atm).
Point numbering corresponds to runs of Table 1.
dizedtotheketonebyperforminganexperimentat175 ◦C, under
identical conditions to those of run 4 (Table 1), but using cyclo-
hexanol as the substrate instead of cyclohexane. We have then
observed a conversion of the alcohol into the ketone of only 4.3%
(49% selectivity), what suggests that most of the ketone derived
from the oxidation of the alkane is not formed via oxidation of
free cyclohexanol.
Thedependenceoftheselectivityonthetemperatureisshown
in Fig. 2 which indicates that a maximum selectivity towards
cyclohexanone is achieved at ca. 150 ◦C.
3.2. Effect of dioxygen pressure
The effect of O2 pressure on cyclohexane oxidation was
studied in the range from 5.5 to 19 atm p(O2) at constant tem-
perature (150 ◦C) and time (8 h) (entries 3 and 6–10, Table 1).
The increase of this pressure results in an enhancement of the
conversions of cyclohexane (Fig. 3), e.g. the overall conversion
increases from 2.1 to 18.3% upon changing p(O2) from 5.5 to
19 atm, what is consistent with the promotion of the solubility
of this gas in the system. The selectivity towards cyclohexanone
Fig. 4. Effect of O2 pressure on the selectivity of the oxidation of cyclohexane
by molecular O2, in the presence of [ReClF{NNC(O)C6H5}(Hpz)2(PPh3)] 4
supported on 3-aminopropyl modified silica (temperature = 150 ◦C, time = 8 h).
Point numbering corresponds to runs of Table 1.
Fig. 6. Effect of the amount of 3-aminopropyl modified silica with supported
[ReClF{NNC(O)C6H5}(Hpz)2(PPh3)] 4 on the selectivity of the oxidation
of cyclohexane by molecular O2 (temperature = 150 ◦C, p(O2) = 13.6 atm,
time = 8 h). Point numbering corresponds to runs of Table 1.

98
G.S. Mishra et al. / Journal of Molecular Catalysis A: Chemical 285 (2008) 92–100
versions towards the desired cyclohexanone and cyclohexanol
products does not greatly increase after the first 8 h (e.g. the
overall conversion of cyclohexane into the ketone and the alco-
hol increases from 10.8 to 14.4%, for 8–24 h reaction time).
Moreover, extending the reaction time above 8 h results in a
decrease of the selectivity (Fig. 5) towards such products, con-
comitant with a more pronounced increase of the formation of
by-products.
3.4. Effect of the amount of supported catalyst
The increase of the amount of the supported
[ReClF{N2C(O)Ph}(Hpz)2(PPh3)] 4 catalyst leads to a
growth of the overall conversion (3.5–18.4% for 10–50 mg,
respectively, entries 3 and 15–17, Table 1). The selectivity
towards the main product, i.e. cyclohexanone, remains nearly
invariant, ca. 70%, but the relative amount of by-products
increases (Fig. 6).
3.5. Effect of co-catalysts
Some heteroaromatic acids were tested as possible
co-catalysts (Table 2). Pyrazinecarboxylic acid promotes con-
siderably the reaction, increasing the overall conversion (from
11.3 to 16.8%, entries 3, Table 1, and 4, Table 2). A promoting
effectofthesameacidwasalsoobservedbyus[11,12]forrelated
alkane oxidations catalyzed by supported V-complexes, and by
Fig. 7. Selectivities of the oxidations of various cycloalkanes and benzene
with molecular O2, in the presence of [ReClF{NNC(O)C6H5}(Hpz)2(PPh3)]
4
supported on 3-aminopropyl modified silica (temperature = 150 ◦C,
p(O2) = 13.6 atm, time = 8 h, catalyst = 20 mg).
also increases with p(O2), but only until 14 atm, while that for
cyclohexanol decreases with the pressure increase due to the
promotion of the relative amount of by-products. Hence, the use
of O2 pressures above ca. 14 atm is not advantageous in terms
of selectivity.
¨
Shul’pin and Suss-Fink [37,38] in the peroxidative oxidation
of alkanes, in various homogeneous systems with V-species.
2,3-Pyrazinedicarboxylic acid, 2,6-pyrazinedicarboxylic acid
and 3-amino-2-pyrazinecarboxylic acid have a less pronounced
effect, while picolinic acid is almost inactive (11.9% overall
conversion vs. 11.3% in the absence of this acid). Picolinic acid
is also less effective than pyrazinecarboxylic acid [11,37,38] in
the above V-catalyzed alkane oxidations.
3.3. Effect of the reaction time
The effect of the reaction time was studied (entries 3 and
11–14, Table 1) in the range 4–24 h (at constant temperature,
150 ◦C, and p(O2) = 13.6 atm) and it was observed that the con-
3.6. Oxidation of other cycloalkanes
The supported [ReClF{N2C(O)Ph}(Hpz)2(PPh3)] 4 com-
plex was also tested as a catalyst for the oxidation, with O2,
of some other cycloalkanes (cyclopentane, cycloheptane and
cyclooctane) and benzene (Table 3, Fig. 7). In the cases of these
cycloalkanes, the corresponding cycloalkanones and cycloalka-
nols are formed with conversions and selectivities comparable
to those for cyclohexane oxidation (Fig. 7). In contrast, the
oxidation of benzene only leads to phenol in a rather lower yield.
3.7. Mechanism
Alkane oxidation reactions can occur [39] via different types
of mechanisms. In the current study, the involvement of a free-
radical process is supported by the inhibition of the catalytic
activity when performing the reactions in the presence of bro-
motrichloromethane or diphenylamine, i.e. a carbon- and an
oxygen-radical trap, respectively.
Fig. 8. Thermogravimetric analysis of 3-aminopropyl modified silica with sup-
ported [ReClF{NNC(O)C6H5}(Hpz)2(PPh3)] 4 catalyst: (a) weight loss and (b)
heat flow.
Hence, the oxidation is proposed to proceed via the cyclo-
hexyl radical (Cy•) obtained by reaction of cyclohexane (CyH)

G.S. Mishra et al. / Journal of Molecular Catalysis A: Chemical 285 (2008) 92–100
99
with O2 upon homolytic C–H bond cleavage (slow induction
period). OxidationofCy• byO2 wouldformthealkylperoxyrad-
ical CyOO• which could react in two ways: (i) decomposition to
the final products, i.e. cyclohexanol (CyOH) and cyclohexanone
(Cy(–H) O), also with formation of O2; (ii) H-atom abstrac-
tion from the alkane to form the alkyl hydroperoxide CyOOH,
whose homolytic decomposition to alkyloxy (CyO•) (O–O bond
cleavage)andalkylperoxy(CyOO•)(O–Hrupture)couldbecat-
alyzed by a rhenium catalyst. The alcohol (CyOH) could then be
formed either by H-abstraction from the alkane (CyH) by CyO•
or (see (i) above) by decomposition of the CyOO• to yield also
cyclohexanone.
This type of mechanism has been proposed [10–12,39,40]
for catalysts in which the metal has two available oxidation
states of comparable stability, a property that is not unusual
in Re coordination chemistry [1]. The involvement of the
hydroperoxide CyOOH is substantiated by the promotion of the
detected amount of CyOH with a decrease of that of the ketone
Cy( H) O, when the final reaction solution is treated with an
excess of PPh3 prior to GC analysis, according to a method
reported by Shul’pin [41,42]. The CyOOH still present in the
final reaction solution (which did not undergo the above men-
tioned decomposition in (ii)) is then deoxygenated by PPh3 to
give CyOH (with formation of phosphine oxide), thus eliminat-
ing the CyOOH decomposition to both CyOH and Cy( H) O
in the gas chromatograph [41,42]. In the absence of PPh3, the
overall GC detected amounts of cyclohexanol and cyclohex-
anone comprise those derived from CyOOH during the reaction
(see (ii) above) and those formed upon decomposition, at the
GC operating conditions, of the remaining CyOOH in the final
reaction solution.
Another point that deserves a comment is that in the present
study cyclohexanone always predominates over cyclohexanol,
what shows that the ketone can also be formed by other reactions
apart from those mentioned above. They can include (i) metal-
assisted decomposition of CyOOH to Cy( H) O, or (ii) metal-
assistedoxidationofthealcoholCyOHtotheketoneCy( H) O.
In our case, pathway (ii) does not appear to be a relevant one, in
view of the observed limited conversion of the alcohol into the
ketone, under typical reaction conditions.
Reactions of those types have been proposed for cyclohexane
oxidation with O2 catalyzed by a photoexcited polyoxode-
catungstate immobilized on amorphous or MCM-41 silicas [43],
which also usually lead to a higher amount of the ketone in
comparison with the alcohol. A different CyOOH decomposi-
tion path, via adsorbed intermediates without formation of free
radicals, has been recently proposed [44], on the basis of kinetic
studies, for the CyH oxidation with O2 on MnAPO-5 catalysts
(APO-5 = aluminophosphate molecular sieves). However, it is
not expected to be followed in the current study since it would
account for a ketone/alcohol product ratio of only 0.5.
oxidant) to cyclohexanone (the major product) and cyclo-
hexanol, under moderate conditions. The fluoro-complex
[ReClF{N2C(O)Ph}(Hpz)2(PPh3)] 4 provides the system with
the best activity and selectivity towards those products, reaching,
under typical conditions (pO2 = 13.6 atm, at 150 ◦C, 8 h reaction
time), an overall conversion of 11% and an overall selectivity
of 95%. The conversion increases to ca. 16–17% upon increas-
ing the O2 pressure to 19 atm or duplicating the catalyst amount
or using pyrazinecarboxylic acid (an heteroaromatic acid) as a
promoter. The higher activity of the fluoro-complex 4, in com-
parison with the related chloro-complexes 1–3, is in accord
with the observed [8] behaviours for the oxidation of ethane
(to acetic acid or acetaldehyde) and of cyclohexane (to cyclo-
hexanone and cyclohexanol), and possibly concerns the overall
stronger electron-donor character [45] of the fluoride ligand,
relatively to chloride, thus promoting the oxidation (by O2) of
the complex to a higher metal oxidation state, more favourable
to the alkane reaction. A related behaviour was reported [46]
for cycloalkanes oxidation catalyzed by the metaloporphyrins
[Mn(X)(TTP)] [X = halide, OH−, CH3COO−; TTP = meso-
tetrakis(p-tolyl)porphyrinato dianion], the fluoro–Mn complex
acting as the best catalyst.
The reaction is proposed to proceed via a radical mechanism
involving both carbon- and oxygen-centred radicals, as substan-
tiated not only by assays with radical traps, but also by the
inactivity (or very low activity) of the complex 4 to catalyze the
oxidation of benzene. The oxidations with O2 of the related C5,
C7 and C8 cycloalkanes, as catalyzed by complex 4, lead sim-
ilarly to the corresponding cycloalkanones and cycloalkanols,
the former also as the main products, showing the generality of
the catalytic behaviour of that complex for the oxidation of such
a type of substrates (cycloalkanes). Further catalytic studies with
those cycloalkanes and also with linear alkanes are under way.
Acknowledgements
This research work has been partially supported by the
Foundation of Science and Technology (FCT), Portugal under
POCI 2010 program (FEDER funded). One of the authors
G.S.M. express gratitude to FCT for postdoctoral fellowship.
The authors thank Dr. C. Henriques for kindly having performed
the TGA analysis and Mr. I. Marques for running the FAB-MS
and the GC–MS analysis.
References
[1] A.J.L. Pombeiro, M.F.C.G. da Silva, R.H. Crabtree, in: R.B. King (Ed.),
Encyclopedia of Inorganic Chemistry, vol. IX, Wiley, 2005, pp. 5499–5515.
[2] C. Roma˜o, in: R.B. King (Ed.), Encyclopedia of Inorganic Chemistry, vol.
VII, Wiley, 2005, pp. 4647–4692.
[3] C. Pettinari, C. Santini, in: J.A. McCleverty, T.J. Meyer (Eds.), Compre-
hensive Coordination Chemistry II, vol. I, Elsevier, Oxford, 2004, pp.
159–210.
[4] C.C. Roma˜o, F.E. Ku¨hn, W.A. Herrmann, Chem. Rev. 97 (1997) 3197.
[5] W.A. Hermann, F.E. Ku¨hn, Acc. Chem. Res. 30 (1997) 169.
[6] M.V. Kirillova, A.M. Kirillov, P.M. Reis, J.A.L. Silva, J.J.R. Frau´sto da
Silva, A.J.L. Pombeiro, J. Catal. 248 (2007) 130.
[7] A.M. Kirillov, M. Haukka, M.F.C. Guedes da Silva, A.J.L. Pombeiro, Eur.
J. Inorg. Chem. 11 (2005) 2071.
4. Conclusions
The work shows that pyrazole rhenium complexes, when
supported on 3-aminopropyl functionalized silica gel, cat-
alyze the cyclohexane oxidation with O2 (the ideal “green”

100
G.S. Mishra et al. / Journal of Molecular Catalysis A: Chemical 285 (2008) 92–100
[8] E.C.B. Alegria, M.V. Kirillova, L.M.D.R.S. Martins, A.J.L. Pombeiro,
Appl. Catal. A: Gen. 317 (2007) 43.
[28] Y. Taniguchi, T. Hayashida, H. Shibasaki, D. Piao, T. Kitamura, T. Yamaji,
Y. Fujiwara, Org. Lett. 1 (1999) 557.
[9] G.S. Mishra, J.J.R. Frau´sto da Silva, A.J.L. Pombeiro, J. Mol. Catal. A:
Chem. 265 (2007) 59.
[10] G.B. Shul’pin, G.S. Mishra, L.S. Shul’pina, T.V. Strelkova, A.J.L.
Pombeiro, Catal. Commun. 8 (2007) 1516.
[11] G.S. Mishra, A.J.L. Pombeiro, J. Mol. Catal A: Chem. 239 (2005) 96.
[12] G.S. Mishra, A.J.L. Pombeiro, Appl. Catal. A: Gen. 304 (2006) 185.
[13] R. Mokaya, M. Poliakoff, Nature 437 (2005) 1243.
[14] U. Schuchardt, D. Cardoso, R. Sercheli, R. Pereira, R.S. da Cruz, M.C.
Guerreiro, D. Mandelli, E.V. Spinace, E.L. Pires, Appl. Cat. A: Gen. 211
(2001) 1.
[29] J. Chatt, J.R. Dilworth, G.J. Leigh, V.D. Gupta, J. Chem. Soc. (1971) 2631.
[30] E.C.B. Alegria, L.M.D.R.S. Martins, M.F.C.G. da Silva, A.J.L. Pombeiro,
J. Organomet. Chem. 690 (2005) 1947.
[31] G.W. Parshall, L.W. Shive, F.A. Cotton, Inorg. Synth. 17 (1977) 110.
[32] J. Chatt, A.G. Row, J. Chem. Soc. (1962) 4019.
[33] D. Kumar, C.C. Landry, Micropor. Mesopor. Mater. 98 (2007) 309.
[34] D.M. Haddleton, D.J. Duncalf, D. Kukulj, A.P. Radigue, Macromolecules
32 (1999) 4769.
[35] P. Battioni, J.F. Bartilli, Y.S. Byun, T.G. Traylar, D. Mansuy, Chem. Com-
mun. (1992) 1051.
[15] A.E. Shilov, G.B. Shul’pin, Chem. Rev. 97 (1997) 2879.
[16] R. Whyman, Applied Organometallic Chemistry and Catalysis, Oxford
University Press, 2001.
[17] D.E. Mears, A.D. Eastman, in: A. Seidel (Ed.), Kirk-Othmer Encyclopedia
of Chemical Technology, vol.13, 5th ed., Wiley, 2004, pp. 706–710.
[18] P. Du, J.A. Moulijn, G. Mul, J. Catal. 238 (2006) 342.
[19] P. Selvam, S.K. Mohapatra, J. Catal. 238 (2006) 88.
[20] R. Zhang, Z. Qin, M. Dong, G. Wang, J. Wang, Catal. Today 110 (2005)
351.
[21] G.S. Mishra, A. Kumar, J. Mol. Catal. A: Chem. 192 (2003) 275.
[22] C.B. Almquist, P. Biswas, Appl. Catal. A: Gen. 214 (2001) 259.
[23] D.L. Vanoppen, P.A. Jacobs, Catal. Today 49 (1999) 177.
[24] H. Einaga, S. Futamura, Appl. Catal. B: Environ. 60 (2005) 49.
[25] M. Misono, Chem. Commun. (2001) 1141.
[36] J. Hao, H. Cheng, H. Wang, S. Cai, F. Zhao, J. Mol. Catal. A: Chem. 271
(2007) 42.
[37] G.B. Shul’pin, Y.N. Kozlov, G.V. Nizova, G. Su¨ss-Fink, S. Stanislas, A.
Kitaygorodskiy, S. Kulikova, Perkin Trans. 2 (2001) 1351.
[38] G. Su¨ss-Fink, S. Stanislas, G.B. Shul’pin, G.V. Nizova, H. Stoekli-Evans,
A. Neels, C. Bobillier, S. Claude, Dalton Trans. (1999) 3169.
[39] M. Hartman, S. Ernst, Angew. Chem. Int. 39 (2000) 888.
[40] A. Kumar, G.S. Mishra, A. Kumar, Trans. Met. Chem. 28 (2003) 913.
[41] G.B. Shul’pin, Comptes Rendus Chimie 6 (2003) 163.
[42] G.B. Shul’pin, J. Mol. Catal. A: Chem. 189 (2002) 39.
[43] A. Maldotti, A. Molinari, G. Varani, M. Lenarda, L. Storaro, F.
Bigi, R. Maggi, A. Mazzacani, G. Sartori, J. Catal. 209 (2002)
210.
[44] B. Mode´n, Bi-Z. Zhan, J. Dakka, J.G. Santiesteban, E. Iglesia, J. Catal. 239
(2006) 390.
[45] A.J.L. Pombeiro, Eur. J. Inorg. Chem. (2007) 1473.
[46] J. Poltowicz, J. Haber, J. Mol. Catal. A: Chem. 220 (2004) 43.
[26] M.V. Kirillova, J.A.L. da Silva, J.J.R. Frau´sto da Silva, A.J.L. Pombeiro,
Appl. Catal. A: Gen. 332 (2007) 159.
[27] G. Su¨ss-Fink, L. Gonzalez, G.B. Shul’pin, Appl. Catal. A: Gen. 217 (2001)
111.