Synthesis and application of FeIII, NiII and Mn II complexes anchored to HMS as efficient catalysts for cycloalkane oxyfunctionalization

Article abstract of DOI:10.1016/j.molcata.2013.12.004

Methoxysilane Schiff-base pentacoordinate metal complexes, i.e. Fe[Sal(PMeO-Si)DPTA], Ni[Sal(PMeO-Si)DPTA] and Mn[Sal(PMeO-Si)DPTA], were synthesized and single site covalently anchored into the uniform porous texture of HMS (2-10 nm size) via condensation process. The correspondent supported catalysts (4.a, 4.b and 4.c, respectively) were characterized by FT-IR, SEM/EDS, XRD, TG, EPR and AAS analysis. In the catalytic tests, they showed high efficiency in the selective oxidation of cyclohexane using molecular O ₂ (overall conversion 40.7% with Cat. 4.a, 29.5% with Cat. 4.b and 26.2% with Cat. 4.c) under relatively mild condition in batch reactor. The Cat. 4.a system (HMS/3.a) exhibits highest TONs ca. 4.2 × 10³ with good selectivity ca. 70% (48% ketone selectivity). The reaction mechanism involves free radicals, as it was proved by the addition of PPh₃. Finally, these supported catalysts could be reuse up to several catalytic cycles.

Full text of DOI:10.1016/j.molcata.2013.12.004

Journal of Molecular Catalysis A: Chemical 383–384 (2014) 159–166
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
III
II
II
Synthesis and application of Fe , Ni and Mn complexes anchored to
HMS as efficient catalysts for cycloalkane oxyfunctionalization
∗
Kelly Machado, Pedro B. Tavares, Gopal S. Mishra
Laboratory of Materials, Centro de Química-Vila Real (CQ-VR), Department of Chemistry, University of Trás-os-Montes and Alto Douro (UTAD), Vila Real
5001-801, Portugal
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 8 April 2013
Received in revised form 31 October 2013
Accepted 1 December 2013
Available online 15 December 2013
Methoxysilane Schiff-base pentacoordinate metal complexes, i.e. Fe[Sal(PMeO-Si)DPTA], Ni[Sal(PMeO-
Si)DPTA] and Mn[Sal(PMeO-Si)DPTA], were synthesized and single site covalently anchored into the
uniform porous texture of HMS (2–10 nm size) via condensation process. The correspondent supported
catalysts (4.a, 4.b and 4.c, respectively) were characterized by FT-IR, SEM/EDS, XRD, TG, EPR and AAS
analysis. In the catalytic tests, they showed high efficiency in the selective oxidation of cyclohexane
using molecular O2 (overall conversion 40.7% with Cat. 4.a, 29.5% with Cat. 4.b and 26.2% with Cat. 4.c)
Keywords:
Supported catalyst
Cyclohexane
O2 oxidation
Batch reactor
Radical mechanism
under relatively mild condition in batch reactor. The Cat. 4.a system (HMS/3.a) exhibits highest TONs ca.
3
4
.2 × 10 with good selectivity ca. 70% (48% ketone selectivity). The reaction mechanism involves free
radicals, as it was proved by the addition of PPh3. Finally, these supported catalysts could be reuse up to
several catalytic cycles.
© 2013 Elsevier B.V. All rights reserved.
1
. Introduction
The activation C H bond of hydrocarbons with molecular oxy-
catalysts might be considered, namely due to their easy separation
from the product (by simple filtration), the possibility of recycling
them and the facility to control and tune the catalyst composition.
For the current process, we have chosen cyclohexane because it
oxidizes to alcohol and ketone products, also known as KA oil. These
product are further convert to adipic acid and caprolactam, which
are the key components for the production of urethane foams,
polyamide-6, lubricating additives and nylon-6 [18]. Typically, Co
slats have been used in the industrial cyclohexane oxidation either
gen is an important reaction since several industrial processes are
based on this transformation [1–4]. The use of molecular O2 in the
catalysis of alkanes oxidation is much attractive due to its low price,
readily availability in atmospheric air as “greener oxidant” [5] and
less toxic compared to other oxygen donors: H O , t-BuOOH, etc.
2
2
[
6,7]. In particular, Schiff bases ligands are able to stabilize many
different metals in various oxidation states [8–11], controlling the
performance of metals in a large range of valuable catalytic trans-
formations [12,13]. Nevertheless, homogeneous catalysts must be
separated later from the reaction mixture, which is a cost of the
process and difficult to recycle. To improve the method, one must
immobilize the metal complexes catalysts on solid supports, pre-
venting the requirement for laborious and inefficient extraction
processes. The activity of this type of catalyst is mainly dependent
upon the structure of complex, surface area of support and sub-
strate polarity which are the requirements for the reactants to be
in contact on the active catalyst centre [14]. In the case of using
high surface area mesoporous support, the inorganic complexes
have been anchored through the interaction between the silanol of
the supporting matrix and terminal reactive groups of the complex,
yielding supported hybrid catalysts [15–17]. Some such supported
by air O2 or pure O . At temperatures above 423 K low conver-
2
sions (ca. 4%) but 85% selectivity (KA oil) was achieved [19]. This
way, industrial cyclohexane oxidation uses inherently inefficient
methodology that necessitates repeated recycling of feedstock. The
alternative use of peroxides is expensive and is also accompanied
by the formation of toxic by-products [20–22]. The establishment
of a greener and more effective cyclohexane oxidation process by
the using of atmospheric oxygen is therefore a present need.
In past few years, metal cations or often oxides have been
incorporated on inorganic materials (e.g. zeolites, silica, alumina
or aluminophosphates), while promoters have been used to reduce
the induction period and to increase the selectivity for the targeted
product of cyclohexane [23–25]. Metal-substituted heteropoly
compounds (heteropolyoxometalates) catalysts demonstrated
good activity towards alkanes oxidation [26]. Re complexes over
SiO2 and Pd–Re over Al O3 have also demonstrated considerably
2
good catalytic activities for the oxidation of cycloalkane [27,28]. The
SiO2 supported di(ethylthio)alkane Pd complexes give 16.2% con-
version and 98% selectivity [29]. The Y-type zeolite anchored V and
∗ Corresponding author. Fax: +351 259350286.
E-mail addresses: mishrags@utad.pt, mishrags@gmail.com (G.S. Mishra).
1
381-1169/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcata.2013.12.004

1
60
K. Machado et al. / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 159–166
Scheme 1. Synthesis procedure of Schiff-base trimethoxysilane pentacoordinate metal complexes and single site anchoring into the uniform porous texture of HMS.
Cu salen complex catalysts were applied in the cyclohexane/H O2
2. Experimental
2
process and achieved 86.6% and 18.1% conversions with V and Cu
catalysts, respectively [30]. The Fe and Mn complexes were also
tested in the cyclohexane oxidation study and obtained 10.2% con-
version with Fe complex and t-BuOOH [31]. Previously, we had
also prepared same type of pentacoordinate Schiff base ligands
of V-complexes over SiO2 and tested in cyclohexane O2 oxida-
tion and obtained 36% conversion with 98% product selectivity
2.1. General materials and methods
A standard Schlenk technique was used for the synthe-
sis of Fe, Ni and Mn complexes under normal atmospheric
conditions. All the chemicals and solvents i.e. ethanol (EtOH),
n-pentane, toluene, diethyl ether (Et O), acetonitrile (MeCN),
2
[
32]. Some other maltolato type of V-complexes supported SiO2
tetrahydrofuran (THF) and cyclohexane (Cy-hx) were from Janssen
Chemicals. 3-Idopropyl trimethoxysilane (PMeO-Si), salicylalde-
hyde (Sal), bis(2,4-pentanedionato)cobalt, bis(aminopropyl)amine
(DPTA), potassium carbonate (K CO ), iron chloride anhydrous
were also applied and obtained relatively low conversion 15% [33].
Alumina supported hetero binuclear macrocyclic Co–V complex
and SiO2 anchored Schiff base Co complexes catalysts have been
investigated obtaining 20% conversion in both the cases [34,35].
These results show that the use of non precious metals in cat-
alysts is favourable due to low price, resistance to temperature
and light. Its strongly motivate the investigation of low cost cat-
alysts.
2
3
(FeCl ), nickel acetylacetonate (C10H14NiO ), manganese acetate
3
4
[Mn(CH COO) ], n-hydroxyphthalimide (n-NIPH) and tetraethy-
3
2
lorthosilicate (TOS) were from Sigma–Aldrich and they were used
as received.
Hence, the main target of this study is to synthesize and to apply
III
II
II
novel methoxysilane Schiff-base Fe , Ni and Mn complexes for
the easy anchoring (without any linker) over hexagonal meso-
porous silica (HMS) as heterogeneous catalyst. These catalysts are
tested for the gas–liquid oxidation of cyclohexane with molecular
O2 under relatively mild condition to obtain high conversion with
product selectivity.
2.2. Catalyst preparation
Ligand
2 (Scheme 1), Sal(PMeO-Si)DPTA, was prepared
as described earlier [32,36], starting from equimolar quan-
tities of bis(salicylidenimino-3-propyl)amine and 3-idopropyl
trimethoxysilane, in THF reflux, in the presence of K2CO3.

K. Machado et al. / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 159–166
161
2
.2.1. Synthesis of Fe[Sal(PMeO-Si)DPTA], 3.a
under reflux for 12 h in air. The obtained solids were washed in
Soxhlet with toluene to remove any non-anchored complexes. The
complex 3.a, 45.6 mg, 3.b, 46.0 mg and 3.c, 44.8 mg were anchored
into HMS. The final catalysts were recorded as Cat. 4.a (HMS/3.a),
Cat. 4.b (HMS/3.b) and Cat. 4.c (HMS/3.c). All the reaction manip-
ulations were carried out under N2 atmosphere due to sensitive
alkyloxy group of the metal complexes.
The complex 3.a was synthesized by adding FeCl .6H O
3
2
−
3
(
(
4.38 × 10 mmol) (in THF 30 mL solution) to a THF solution
−3
20 mL) of ligand 2 (4.99 × 10 mmol), with constant stirring. The
red mixture was heated and refluxed for 2 h. The final solution was
concentrated and upon addition of n-pentane a red solid precip-
itated. The solid was filtered, washed with n-pentane and dried
under vacuum to obtain 3.a complex (88% yield, Scheme 1). It is
soluble in polar organic solvents. Anal. calcd. for FeC26H37N O Si:
2.2.6. Solid catalyst characterization
3
5
−
1
C, 56.24; H, 6.66; N, 7.56. Found: C, 54.77; H, 6.12; N, 6.21. FT-IR
FT-IR (KBr pellet, cm ): broad shoulders were observed at
−1
−1
−1
and 1330 to 930 cm in all the supported
(
KBr pellet, cm ): ꢀ = 3010 [s, ꢀ (C H)]; 29,505 [s, ꢀ (C H)]; 1620
ca. 3715–3345 cm
and 1563 [s, (ꢀ C N) and s, (ꢀ C C)]; 1176 [s, (ꢀ Si OCH ); 1064
catalysts (4.a–4.c) due to HMS effect. Some additional peaks of
complexes on the support matrix were confirmed [s, (ꢀ C O at
3
◦
[
A
s, (ꢀ C N)] and 465 [ꢀ (FeO)]. EPR (CH Cl , at 25 C): g = 1.9982,
and A (in ×10 cm ) = 1.62 and 59.8. FAB -MS: m/z = 555.19
2
2
|
|
⊥
−4
−1
+
−1
−1
2260–2200 cm , NH bands at 1644–1540 cm and Si O bands
+
+
−1
[
(M+H) , 100], 410 [(M−(CH Si(OCH ) ) , 14], 1 [(Si(OCH ) ), 99].
980–955 cm . TGA: the supported Cat. 4.a to 4.c show slow weight
2
3
3
3
3
losses above ca. 380 to 435 K, most likely due to the release of mois-
ture. The second major weight loss of all the catalysts was observed
above ca. 500 K due to partial decomposition of organic groups
of metal complexes and complete weight loss of supported com-
plex on HMS was obtained at about 670 K. AAS analysis: the metal
content in the supported catalysts, Cat. 4.a, 4.b and 4.c, were mea-
sured in the range of 0.20–0.35 weight %. BET: surface area result
2
.2.2. Synthesis of Ni[Sal(PMeO-Si)DPTA], 3.b
The complex 3.b was synthesized by mixing a solution of
−3
C10H14NiO , m. wt. 256.91 (4.38 × 10 mmol) in THF (30 mL), to an
4
−
3
equimolar amount of ligand 2 (4.99 × 10 mmol) in THF (20 mL).
The green mixture was heated and refluxed for 2 h. The final solu-
tion was concentrated and upon addition of n-pentane a light green
solid precipitated. The solid was collected by filtration, washed with
n-pentane and dried under vacuo to obtain 3.b complex (85% yield,
Scheme 1). It is soluble in most of polar solvents but insoluble in
2
3
showsthat HMS has 695 m /g surfacearea (porevolume0.67 m /g).
2
After the anchoring, it was reduced i.e. for Cat. 3.a (474 m /g
3
2
and pore volume 0.53 m /g), for Cat. 3.b (756 m /g, pore volume
3
2
3
Et O. Anal. calcd. for NiC26H37N O Si: C, 56.04; H, 6.64; N, 7.54.
0.51 m /g) and for Cat. 3.c (789 m /g, pore volume 0.57 m /g). EPR:
for the catalysts samples 4.a–4.c (at room temperature) EPR spec-
2
3
5
−
1
Found: C, 55.19; H, 6.32; N, 7.26. FT-IR (KBr pellet, cm ): ꢀ = 3090
s, ꢀ (C H)]; 2832 [s, ꢀ (C H)]; 1644 and 1564 [s, (ꢀ C N) and s, (ꢀ
|| ⊥
trum with range of g = 1.9998–1.9825 and range of A and A (in
[
C
2
−
4
−1
C)]; 1136 [s, (ꢀ Si OCH ) and 1066 [s, (ꢀ C N)]. EPR (CH Cl , at
×10 cm ) = 153.2–164.7 and 59.6–61.8, were obtained, respec-
tively. SEM/EDS analysis: morphological and chemical analysis of
the Cat. 4.a–4.c showed bright spotted metal rich areas.
3
2
2
◦
||
⊥
−4
−1
+
5 C): g = 1.9982, A and A (in ×10 cm ) = 1.62 and 59.8. FAB -
+
+
MS: m/z = 557.19 [(M+H) , 100], 415 [(M−(CH Si(OCH ) ) , 14], 1.3
2
3
3
[
(Si(OCH ) ), 99].
3
3
2.2.7. Instrumentation
2.2.3. Synthesis of Mn[Sal(PMeO-Si)DPTA], 3.c
Infrared spectra (FT-IR, 4000–400 cm 1) were recorded on
−
The complex 3.c was synthesized by mixing a solution of
a Unicam Research Series, spectrophotometer in KBr pellets).
Elemental analyses (EA) were carried out on a Carlo Erba 1108 Ana-
lyzer by chromatographic combustion method. FAB mass spectra
(FB-MS) were obtained on a Bruker Daltonics APEX-QE spectropho-
tometer in electron spray ESI mode (7 T superconducting magnet).
The morphologies of the catalysts were analyzed by scanning
electron microscopy (SEM) on FEI Quanta 400, equipped with an
EDS detector (EDAX). X-ray diffraction patterns were collected in
Bragg–Bentano and in low angle geometries. The measurements
were performed on a Philips X’Pert MPD equipped with the ultra-
fast X’Celerator detector and secondary monochromator. A low
−
3
Mn(OAc) , m. wt. 173.01 (4.99 × 10 mmol) in THF (30 mL) to an
2
−
3
equimolar amount of ligand 2 (4.99 × 10 mmol) in THF (20 mL).
This reaction mixture was heated and refluxed for 2 h. The final
solution was concentrated and upon addition of n-pentane a
yellowish brown solid precipitated. The solid was collected by fil-
tration, washed with n-pentane and dried under vacuo to obtain
3
.c complex (86% yield, Scheme 1). It is soluble in most of polar
solvents but insoluble in Et O. Anal. calcd. for MnC26H37N O Si:
2
3
5
C, 56.35; H, 6.68; N, 7.58. Found: C, 56.16; H, 6.31; N, 6.60. FT-IR
−1
(
KBr pellet, cm ): ꢀ = 3066 [s, ꢀ (C H)]; 2935 [s, ꢀ (C H)]; 1653
◦
and 1574 [s, (ꢀ C N) and s, (ꢀ C C)]; 1142 [s, (ꢀ Si OCH ) and
angle accessory was used for 2Â measurements between 0.5 and 5 .
3
◦
||
⊥
1
072 [s, (ꢀ C N)]. EPR (CH Cl , at 25 C): g = 1.9982, A and A (in
A Micromeritices Gemini 2390 instrument was used for the deter-
mination of BET surface area and porosity of catalysts with N2 gas
physisorption to produce surface area and porosity results. A ther-
mogravimetric analysis was performed in a TA Instruments Q50
thermogravimetric analyzer. EPR spectra were recorded in the X
band Bruker ESP 300E spectrophotometer at room temperature and
calibrated with diphenylpicrylhydrazyl (dpph). Atomic Absorption
Spectroscopy (AAS) was used to determine the metal concentration
on a Perkin-Elmer 41000ZL spectrophotometer (5 mg of sample
2
2
−
4
−1
+
+
×
10 cm ) = 1.62 and 59.8. FAB -MS: m/z = 554.19 [(M+H) , 100],
+
4
16 [(M−(CH Si(OCH ) ) , 14], 1.2 [(Si(OCH ) ), 99].
2
3
3
3
3
2
.2.4. Preparation of HMS support
The HMS was prepared according to earlier published method
3
[
37]. Tetraethoxysilane (TEOS, 37.0 cm , 0.166 mol) was added to
3
3
a mixed solution of ethanol (88.1 cm , 1.51 mol), water (88.5 cm ,
3
4
.91 mol) and dodecylamine (10.3 cm , 0.0448 mol). During 24 h
the mixture was stirred at room temperature and the white pre-
cipitate obtained was vacuum-filtered and washed with deionized
was digested in HF + HNO ). The reaction products were analyzed
3
in gas chromatography (GC), FISONS GC-8000 series gas chro-
matograph equipped with a FID detector and a DB-WAX capillary
column (length: 30 m; internal diameter: 0.32 mm). The reaction
products were further identified by GC–MS using a Carlo-Erba
Auto/HRGC/MS instrument.
3
3
water (100 cm ) and ethanol (100 cm ). The precipitate was dried
and calcinated at 873 K for 24 h to remove the template and
obtained uniform porous texture of HMS (ca. 2–10 nm size). The
uniform porous texture structure was characterized by the XRD,
SEM and with SEM–EDS.
2.3. Heterogeneous catalysis method and product analysis
2
.2.5. Preparation of final catalyst
Each of the complexes 3.a, 3.b and 3.c (50 mg) was dissolved in
A stainless steel rocking type batch reactor (38 cm3 capacity),
dry toluene (50 mL). Then HMS (1000 mg) was added and heated
equipped with a gas inlet and a pressure gauge was used for the

1
62
K. Machado et al. / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 159–166
oxidation reactions. The inner temperature of the reactor was mon-
itored by a thermocouple. For each experiment, the reactor was
charged with the substrate and catalyst, closed, the air was removed
and O2 introduced. After the reaction, the catalyst was separated
from the liquid mixture by filtration, washed several times with
Fe complex
Ni complex
Mn complex
◦
acetonitrile and reactivated by air drying at 60 C for 8 h, for fur-
ther studies. The reaction products were quantitatively analyzed
by GC (30 ␮L of pentanone added as internal standard to 1.0 mL of
the filtered final reaction solution) and further identified by GC–MS.
Turn over numbers (TONs) was calculated as mole of product
per mole of supported metal complex. The overall conversion was
calculated as mol of product per mole of substrate and the selec-
tivity calculated as mol of particular product per mole of the total
products. Some additional experiments have been carried out for
the detection of alkylperoxide radicals (ROOH) [32]. At the end of
HMS/Fe catalyst
HMS/Ni catalyst
HMS/Mn catalyst
reaction, the final solution was treated with an excess of PPh prior
3
to GC analysis. The hydroperoxide still present is deoxygenated by
4
000
3500
3000
2500
2000
1500
1000
PPh3 to the alcohol ROH (with Ph PO formation), thus suppressing
3
-1
Wavenumber (cm )
the ROOH decomposition to both Cy-ol and Cy-one in GC. More-
over, some oxidation reaction of cyclohexane has been treated with
Fig. 1. FT-IR of metal complexes (3.a, 3.b and 3.c) and after anchoring of HMS as
supported catalysts (Cat. 4.a, 4.b and 4.c).
4
.0 mg of carbon (CBrCl ) or oxygen- (Ph NH) radical traps to con-
3 2
firm the presence of alkyl or hydroperoxide radicals.
0
,0
Fe/HMS
Ni/HMS
Mn/HMS
3
. Results and discussion
-0,2
-0,4
3.1. Supported catalysts: synthesis and characterization
-
-
0,6
0,8
Our first approach was the synthesis of Schiff base trimethoxysi-
III
II
II
lane pentacoordinate Fe
, Ni and Mn complexes for the
anchoring on mesoporous HMS support. These supported cat-
alysts were applied for the cyclohexane oxidation process as
recyclable catalysts. Hence, methoxysilane Schiff base ligand 2
was synthesized as according to F. Carré procedure [36], due to
terminal alkyloxy reactive groups for direct anchoring into sup-
port. The Schiff base ligand have two donor sets of phenolate
oxygen, two donor nitrogens and another tertiary amine nitrogen
atom that have been reacted with the terminal iodine of the 3-
iodopropyl-trimethoxysilane (obtained 87%, yield). The metal salts
were separately introduced at the next step of synthesis process
with ligand 2, for the preparation of three types of complexes,
i.e. Fe[Sal(PMeO-Si)DPTA], (yield 88%, 3.a), Ni[Sal(PMeO-Si)DPTA]
-1,0
-
-
-
1,2
1,4
1,6
350
400
450
500
550
600
650
700
Temperature (K)
Fig. 2. TGA analysis results of single site anchored metal complex into HMS as
supported catalysts (Cat. 4.a, 4.b and 4.c).
(
yield 85%, 3.b) and Mn[Sal(PMeO-Si)DPTA] (yield 86%, 3.c). The
proposed formulation for 3.a–3.c complexes has been supported
by elemental analysis, FT-IR, NMR, EPR and FAB-mass spectral
data. The lone pair of electron was confirmed in the EPR spectra
which shows the typical “g” value that was obtained in the range
All the anchored 3.a, 3.b and 3.c complexes show a broad
shoulder in FT-IR spectrum (Fig. 1), between; ꢀ = 3715–3345 and
ꢀ = 1330–930 cm , due to HMS effect. The presence of metal com-
plexes over HMS matrix was confirmed by the additional bands i.e.
−
1
|
|
⊥
−4
−1
of 1.9998–1.9825 and A and A (in ×10 cm ) in the range of
1
53.2–164.7 and 59.6–61.8. The Feb-Mass spectrum of each com-
C O at 2260–2200 and C N bands at 1644–1540 and Si O bands at
−
1
plex was characterized by an intense molecular ion, which is the
most abundant ion in the spectrum, as well as relatively abundant
doubly charged parent molecular ions. No polymeric species were
detected in any spectrum which suggests that either the complexes
are monomeric or any polymers present were destroyed upon sub-
limation or upon electron impact.
For the preparation of supported catalysts, a grafted reaction of
the OH of uniform porous texture HMS with the direct anchor-
ing of OCH3 of complexes (3.a, 3.b and 3.c) in toluene medium
by a condensation reaction was employed. Final supported hybrid
catalysts were obtained as Cat. 4.a, Cat. 4.b and Cat. 4.c. Since
methoxysilane is moisture sensitive, all the manipulations have
980–955 cm . The thermal stability of all the solid catalysts sam-
ples was analyzed by TGA (Fig. 2) showing 0.6% weight loss below
500 K due to moisture release, below the maximum reaction tem-
perature (i.e. 473 K, Fig. 6). Additional weight loss (0.9%) from 500 K
up to 700 K is due to complex decomposition. The BET surface result
2
shows that initially HMS has 695 m /g surface area with the pore
3
volume 0.67 m /g. After the anchoring of Cat. 3.a, it was reduced
2
3
474 m /g and pore volume 0.53 m /g. The surface areas for other
2
3
catalysts were obtained as 756 m /g, pore volume 0.51 m /g for
2
3
Cat. 3.b and 789 m /g, pore volume 0.57 m /g for Cat. 3.c. Thus sug-
gesting that the supported catalysts are stable in the range of the
reaction temperatures studied. The metal content presence of Fe,
Ni and Mn on the HMS were further confirmed in the AAS analysis
and obtained 0.20–0.35 wt.% of fresh catalysts. After the catalytic
reaction, the most active Cat. 4.a shows a small amount 1.32 wt.%
of metal leaching.
been carried out under N . Any residual and excess of complexes in
2
toluene were subsequently removed in the Soxhlet. To the best of
our knowledge, these supported catalysts are being described for
the first time (Scheme 1).

K. Machado et al. / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 159–166
163
Fig. 3. SEM image of HMS Fig. A of Cat. 4.a, Fig. B of Cat. 4.b, Fig. C of Cat. 4.c and Fig. D of SEM–EDS analysis of Cat. 4.a.
The morphology of all the supported catalysts has been stud-
ied in SEM instrument equipped with EDS detector. It has been
observed in the SEM images of Fe catalyst (Fig. 3A), Ni catalyst
1200
HMS
Fe@HMS
Mn@HMS
Ni@HMS
1
000
(
Fig. 3B) and Mn catalyst (Fig. 3C) over HMS, that the particles
either have round or hexagonal shape (mean particle size ca.
8
00
1
.20 ± 0.14 ␮m). No substantial difference was observed in the SEM
images, even though they are physically different in colour. There-
fore, the presence of Fe content was confirmed in the EDS analysis
600
(Fig. 3D). The main peaks are detected as Si at EK␣ = 1.72 keV, Fe at
EK␣ = 6.43 and 7.12 keV and Au (sample coating) at EK␣ = 2.20 keV.
No crystalline phases of these catalysts are observed in the XRD
spectra with Bragg–Bentano geometry (not shown). The low angle
XRD power patterns measured for HMS and supported catalysts
4
00
200
0
4
.a, 4.b and 4.c are shown in Fig. 4. All the catalyst and HMS sam-
ples exhibited a single low-angle (1 0 0) peak due to the long range
order, characteristic of the mesoporous structures. Additionally, the
intensity of d1 0 0 reflection of 4.a, 4.b and 4.c decreased and d1 0 0
spacing shifted to a higher value, compared with mesoporous HMS.
This effect is mainly due to the anchored metal complexes species
that occupy the internal pores of HMS by the Fe, Ni and Mn com-
plexes. Therefore, the pore volume of all the supported catalysts
should decrease.
1,5
2,0
2,5
2θ, degree
3,0
3,5
4,0
Fig. 4. X-ray diffraction spectra (XRD) of metal complexes into HMS as catalysts
Cat. 4.a, 4.b and 4.c).
(
cyclohexane by molecular O2 and obtained ketone in higher
amount and alcohol in lower amount. It was also found that
they serve as effective heterogeneous catalysts (typical high TONs
4.2 × 10 with 4.a, 3.4 × 10 with 4.b and 2.9 × 10 with 4.c) for
such a reaction without the need of using any additive. All the
reactions were performed under relatively mild and solvent free
conditions in a SS rocking type batch reactor. The Cat.4.a provides
the best conversion result of the cycloalkane 40.7% with overall
KA oil selectivity of 70% in comparison with the other catalysts
3
3
3
3.2. Catalytic oxidation study of cyclohexane
Three types of heterogeneous uniform porous texture HMS
supported complex catalysts HMS/Fe[Sal(PMeO-Si)DPTA], 4.a,
HMS/Ni[Sal(PMeO-Si)DPTA], 4.b and HMS/Mn[Sal(PMeO-Si)DPTA],
4
.c have been applied the in the oxidation reaction of neat

1
64
K. Machado et al. / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 159–166
Fig. 5. Effect of metal complexes on percentage conversion and selectivity in
the course of cyclohexane oxidation with O2 (reaction condition = cyclohexane,
Fig. 7. Pressure effect on the percentage conversion (a) and selectivity (b), of the
cyclohexane oxidation products in the presence of the Cat. 4.a (reaction condition:
2
7.82 m mol; temp. 423 K; time 4 h, p(O2) 10 bar; catalyst, 25 mg).
423 K temp., 25 mg catalyst and 27.8 mmol of substrate for 4 h).
overall
cy-one
overall
40
30
20
10
cy-ol
cy-one
5
0
0
by-products
cy-ol
by-products
4
30
2
0
0
1
50
40
30
20
0
0
6
5
4
3
2
1
0
0
0
0
0
400
425
450
Temperature (K)
475
2
4
6
8
10
12
14
Fig. 6. Temperature effect on the percentage conversion (a) and selectivity (b), of the
cyclohexane oxidation products in the presence of the Cat. 4.a (reaction condition:
Time (h)
1
0 bar p(O)2, 25 mg catalyst and 27.8 mmol of substrate for 4 h).
Fig. 8. Time effect on the percentage conversion (a) and selectivity (b), of the cyclo-
hexane oxidation products in the presence of the Cat. 4.a (reaction condition: 423
temp., 10 bar p(O)2, 25 mg catalyst and 27.8 mmol of substrate).
(
4.b and 4.c compare in Fig. 5, that lead to 29.5 and 26.2% conver-
sion s, respectively). Therefore, we focused our detailed catalytic
studies on the Fe based Cat.4.a. At the end of oxidation reaction, the
colour of Cat.4.a was turned from reddish to dark brown, substrate
colour from colourless to light yellow and a strong smell of product
was also observed. The final reaction mixture was confirmed first
in GC and then in GC–MS for identification of products.
overall
cy-one
4
0
0
cy-ol
by-products
3
20
Some blank experiments also demonstrate that these catalysts
are necessary for the cyclohexane oxidation to process. A low con-
version ca. 4% was observed when the reaction was carried out
without catalyst (analogues condition as Fig. 5). Another experi-
ment with HMS/ligand 2 (without metal) was also performed and
we obtained little improved conversion 7.6%. These results suggest
that not only metal is responsible for improving the catalytic activ-
ity but some other features also play a role in the catalytic oxidation
e.g. HMS and the structure of complex. The transition state of metal
also plays a key role to improve the catalytic efficiency.
1
0
0
0
6
50
40
30
20
On the basis of Fig. 5 results, the detailed catalytic study of
cycloalkane process has been carried out with best oxidation Cat.
10
20
30
40
50
60
Catalyst amount (mg)
4
.a. The best reaction condition has been optimized by various tem-
peratures (388–473 K), reaction O2 pressures (5–20 bar), reaction
times (2–14 h) and catalyst amounts (10–55 mg). All the catalytic
activity and selectivity results are summarized in Figs. 5–9.
Fig. 9. Amount of catalyst (Cat. 4.a) effect on the percentage conversion (a) and
selectivity (b), of the cyclohexane oxidation products in the presence of the (reaction
condition: 423 temp., 10 bar p(O)2 and 27.8 mmol of substrate for 4 h).

K. Machado et al. / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 159–166
165
Temperature effect on the oxidation of cyclohexane was exam-
ined by varying from 398 to 473 K, under constant condition:
7.8 mmol substrate, 4 h time, 10 bar p(O)2 and 25 mg catalyst
Fig. 6). The conversion of product increases slowly at 398 K (over-
all conversion 27.7%) beyond which it increases sharply until 423 K
overall conversion 40.7%). At higher temperature the overall con-
O
Cy-H
OH
2
(
(
step-2)
1/2 O
2
(
version drops due to the formation of unidentified black material
and more by-products (Fig. 6a). Thus in further studies the tem-
perature was not allowed to go above 423 K. At low temperature,
.
Cy-OO
Cy-OOH
3
98 K, the cyclo-ol selectivity was higher but after it drops with
the increasing of temperature (423–473 K). It is suggesting that
the Cy-OOH radicals (intermediate) were initially formed but they
were decomposed with the increasing of temperature. At the result,
high selectivity of cyclo-one was obtained at 423 K in compare to
cyclo-ol. These intermediates (Cy-OOH) were completely decom-
posed under deep oxidation at 473 K. This was confirmed when the
reaction mixture was treated with the excess of PPh3 before GC
analysis [38,39]. The selectivity dependence on the temperature is
Cat. FeIII
(
step-1)
(
step-3)
O
2
Cat. FeII
Cy-O.
II
Cat.Fe -OH
(step-4)
(
step-5)
2
-H O
OH
-R.
also shown in pressure effect, p(O ) on chemical reaction was also
2
investigated in the range from 5 up to 20 bar under constant reac-
tion conditions: 27.8 mmol substrate, 4 h time, 423 K temp., and
13.
. .
= Cy = cyclohexyl radical, Cy-OO = cyclohexyl peroxy radical
Where, C
6
H
2
5 mg catalyst (Fig. 7). The increase of pressure p(O ) from 5 up
2
Scheme 2. Proposed reaction mechanism for cyclo-alkane oxidation.
to 10 bar results in the enhancement of overall conversion from
.2 to 40.7%. Above 10 bar, it was slowly dropping overall con-
5
version 31.5% until at 20 bar for current catalytic system (Fig. 7a).
The increasing of main products (cyclo-one and cyclo-ol) selectivi-
ties were achieved with the pressure increase of molecular oxygen
has considerable catalytic activity after the consumption of O2. It
can be reused for further oxidation batches by admission of more
O2, even without isolating and reactivating the catalyst. However,
after being used, the catalyst could be reactivated by heating and
recycled, still displaying good catalytic activity. Thus, following
cyclohexane oxidation (analogous condition as Fig. 6) for 4 h, the
catalyst was separated, reactivated and recycled leading, under the
same reaction condition, to an overall 11% conversion reduced. The
preservation of some catalytic activity is consistent with the AAS
analysis of the supported catalyst that shows only a partial metal
leaching upon use. In fact, the freshly supported catalyst 4.a has
0.16 wt% Fe (see Section 3.1), whereas, after 4 h reaction (analogous
condition as Fig. 6), the amount of Fe has reduced to 0.14 wt%.
(
1
above 10 bar). The by-product selectivity was unchanged until
5 bar, then dropped. Therefore, the use of more O2 pressure is
helpful in terms of main products selectivity (Fig. 7b).
Reaction time effect, on the cyclohexane reaction was systemat-
ically investigated from 2 up to 14 h with most active Cat. 4.a under
constant reaction conditions: 27.8 mmol substrate, 10 bar p(O)2,
4
23 K, and 25 mg catalyst (Fig. 8). Initially 32.3% conversion was
obtained after 2 h time period, afterwards it increases rapidly and
conversion reaches ca. 40.7% at 4 h reaction time. After long reaction
time (14 h), lower conversion 38.5% (cyclo-one 9.0% plus cyclo-ol
7
.1% Fig. 8b) was observed. In the case of selectivity (Fig. 8b), the
best result (69.3%) was achieved after 4 h reaction period i.e. ketone,
8.2% and alcohol, 21.1%. The extension up to 14 h shows a negative
3.4. Proposed mechanism
4
effect on particularly main products selectivity, with concomitantly
increment of the by-products.
Catalyst 4.a amount effect on oxidation reaction was also inves-
tigated between 10 up to 55 mg to achieve reaction ideal conditions
The catalytic reaction mechanism is principally depending upon
the reaction condition i.e. temperature, pressure, liquid or gas
phase, the nature of the metal state and the oxidizing agent used
[38,40,41]. Therefore, the oxidation mechanism can be divided into
three main types: oxidation through coordination of the substrate,
catalytic oxygen transfer and oxidation via a free radical chain pro-
cess [40,41]. Although the detailed mechanisms of the reactions
are not clear, we believe the current catalytic proceed through
free radicals, when the final reaction solution treated with excess
PPh3 in the GC analysis [39]. The Cy-OOH intermediates (in the
reaction solution) were deoxygenated by PPh3 to Cy-OH (formed
(
constant conditions: 27.8 mmol substrate, 10 bar p(O) , 423 K and
2
4
h reaction time, Fig. 9). It was observed that increasing Cat. 4.a
amount from 10 up to 25 mg, the overall conversion increases from
7.7 up to 40.7%. Beyond this amount (25 mg) the overall conver-
3
sion decreases to 24.8% with 55 mg of catalyst (Fig. 6a). However,
the selectivity of main product, i.e. cyclo-one, remains always high
in compare to other products (Fig. 6a). The selectivity of other
desired product (cyclo-ol) and by-products has little change with
the increasing amount of catalyst. Therefore, the ratio between cat-
alyst and substrate is an also important factor to maintain a high
conversion of desire product.
Ph PO), thus suppressing Cy-OOH by the decomposition to both
3
ketone and alcohol in the chromatograph (see Section 2.3) [3,41]. It
was further confirm by some additional experiments with radical
traps i.e. liquid carbon-radical trap (CBrCl ) and oxygen-radical trap
3
(
Ph NH), analogues condition as Fig. 6. The main product conver-
2
3.3. Catalyst recycles study
sion ca. 11.3% was dropped, confirms the presence of alkylperoxide
intermediates.
•
In our catalytic systems, oxygen is a limiting reagent and the
Hence the proposed Scheme 2, the cyclohexyl radicals (Cy )
and cyclohexylperoxy radicals (Cy-OO ) can be primarily cre-
•
catalyst is still active at the end of the reaction and could be recy-
cled or reused even without reactivation. In fact, after performing
the oxyfunctization of cyclohexane under the analogous condition
as Fig. 6, the reactor was degassed, charged with more O2 until
ated by the reaction of molecular O2 (slowly induction period).
Homolytic cleavage of C H bond of cyclohexane (CyH) can occur
•
•
by H-abstraction by Cy-OO to generate a further Cy and the
hydroperoxide Cy-OOH (Step 1), whose homolytic decomposition
1
0 atm pressure. The reaction was allowed to proceed for a further
•
•
4
h period. The overall conversion of alkane then increased ca. 12%,
to Cy-O and further Cy-OO (Steps 2 and 3) can be catalyzed by
the Fe catalyst. Cy-ol can then be formed by H-atom abstraction
relatively to the first runs (40.7%), showing that the system still

1
66
K. Machado et al. / Journal of Molecular Catalysis A: Chemical 383–384 (2014) 159–166
•
from CyH by the Cy-OO (Step 4). The Cy-ol can be also formed
[13] A.R. Silva, M. Martins, M.M.A. Freitas, J.L. Figueiredo, C. Freire, B. de Castro, Eur.
J. Inorg. Chem. 10 (2004) 2027–2035.
•
by the decomposition of Cy-OO radicals (Step 5). Similar type of
mechanism is known to occur for metal catalysts with two oxida-
tion states of comparable stability such as Mn
[
[
14] S.S. Lin, H.S. Weng, Appl. Catal. A: Gen. 118 (1994) 21.
15] (a) P. Sutra, D. Brunel, Chem. Commun. 21 (1996) 2485–2486;
(b) S. Jana, B. Dutta, R. Bera, S. Koner, Langmuir 23 (2007) 2492–2496;
(c) C.J. Liu, S.G. Li, W.Q. Pang, C.M. Che, Chem. Commun. 1 (1997) 65–66.
16] C. Gilmartin, J.R.L. Smith, J. Chem. Soc., Perkin Trans. 2 (1995) 243–251.
17] (a) J. Sandee, L.A. van der Veen, J.N.H. Reek, P.C.J. Kamer, M. Lutz, A.L. Spek,
P.W.N.M. van Leeuwen, Angew. Chem. 38 (1999) 3231–3235;
II/III
IV
V
II/III
, V / , Co
and
I/II
Cu [35,40a,b,41]. It is also observe for current catalytic system
[
[
II/III
(
catalyst 4.a) having Fe
oxidation states.
(
b) S.R. Hall, C.E. Fowler, B. Lebeau, S. Mann, Chem. Commun. (1999) 201–202.
4
. Conclusions
[
18] J.R. Sheehan, Ullmann Encycropedia Industrial Organic Chemicals, vol. 8, Wiley-
VCH, Weinhelm, 1999, pp. 4575–4591.
Three potentially active trimethoxysilane pentacoordinate
complexes Fe[Sal(PMeO-Si)DPTA] [3.a], Ni[Sal(PMeO-Si)DPTA]
3.b] and Mn[Sal(PMeO-Si)DPTA] [3.c] were synthesized and char-
[19] (a) D.E. Mears, A.D. Eastman, in: A. Seidel (Ed.), Kirk-Othmer Encyclopedia of
Chemical Technology, vol. 13, 5th ed., John Wiley and Sons, New York, 2004,
pp. 706–710;
[
(
b) A.K. Suresh, M.M. Sharma, T. Sridhar, Ind. Eng. Chem. Res. 39 (2000)
acterized. These metal complexes were anchored into HMS without
any liker agent. All the supported catalysts 4.a, 4.b and 4.c were sys-
tematically tested in the cyclohexane O2 oxidation process under
relatively mild condition in a batch process. Typical high TONs
3958–3997;
(c) G. Lü, R. Zhao, G. Qian, Y. Qi, X. Wang, J. Suo, Catal. Lett. 97 (2004) 115–118.
20] R. Whyman, Applied Organometallic Chemistry and Catalysis, Oxford Univer-
sity Press, Oxford, 2001.
21] A.M. Kirillov, M.N. Kopylovich, M.V. Kirillova, E.Yu. Karabach, M. Haukka,
M.F.C.G. da Silva, A.J.L. Pombeiro, Adv. Synth. Catal. 348 (2006) 159–174.
[
[
3
3
4
2
.2 × 10 with Fe catalyst 4.a, 3.4 × 10 with Ni catalyst 4.b and
3
[22] J.M. Bregeault, Dalton Trans. 17 (2003) 3289–3302.
23] T.M. Salama, I.O. Ali, H.A. Gumaa, Microporous Mesoporous Mater. 113 (2008)
0–98.
[24] S.S. Lin, H.S. Weng, Appl. Catal. A: Gen. 118 (1994) 21–31.
.9 × 10 with Mn catalyst 4.c were obtained by these supported
[
catalysts. The supported Cat. 4.a showed highest overall conversion
9
4
0.7% with good selective of cyclo-one (48%). These results were
[
25] E.L. Pires, J.C. Magalhaes, U. Schuchardt, Appl. Catal. A: Gen. 203 (2000)
31–237.
26] H. Einaga, S. Futamura, Appl. Catal. B: Environ. 60 (2005) 49.
obtained under optimize condition (423 K, 10 bar p(O) , 25 mg cat-
2
2
alyst, 27.8 mmol of substrate) in a batch reactor. The TGA showed
that these catalysts were stable under the applied temperature
[
[27] G.S. Mishra, E.C.B. Alegria, L.R. Martins, A.J.L. Pombeiro, Appl. Catal. A: Gen. 285
2008) 92–100.
28] A.A. Susu, F.E. Enoh, A.D.F. Ogunye, J. Chem. Technol. Biotechnol. 30 (1980)
35–747.
[29] G.S. Mishra, S. Sinha, Catal. Lett. 125 (2008) 139–144.
(
(
473 K) and they can be recycled further more reactions. Experi-
[
ments with PPh and initiators provide some supporting testimony
3
7
in favour of a free-radical mechanism.
Acknowledgements
[
30] M.R. Maurya, A.K. Chandrakar, S. Chand, J. Mol. Catal. A: Chem. 270 (2007)
25–235.
2
[
31] G.C. Salomaõ, M.H.N. Olsen, V. Drago, C. Fernandes, L.C. Filho, O.A.C. Antunes,
Catal. Commun. 8 (2007) 69–72.
This work is financially supported by the Funda c¸ ão para
[32] G.S. Mishra, A. Kumar, S. Mukhopadhyay, P.B. Tavares, Appl. Catal. A: Gen. 384
2010) 136–146.
33] G.S. Mishra, A.J.L. Pombeiro, J. Mol. Catal. A: Chem. 239 (2005) 96–102.
[34] G.S. Mishra, M.J. Lakshmi, A. Kumar, J. Mol. Catal. A: Chem. 230 (1–2) (2005)
35–42.
(
a
Ciência
PTDC/EQU-ERQ/110825/2009, CQVR/PEst-C/QUI/UI0616/2011 and
Indian-Portuguese cooperation program-2013”. The authors express
e Tecnologia (FCT) of Portugal through projects:
[
“
[
[
35] A. KumarJr, G.S. Mishra, A. Kumar, Trans. Met. Chem. 28 (8) (2003) 913–917.
36] F. Carré, R.J.P. Corriu, E.L. Beltran, A. Mehdi, C. Reyé, R. Guilard, J. S y´ kora, A. van
der Lee, Dalton Trans. 16 (2003) 3211–3215.
gratitude to CACTI Centre Vigo, Spain, and UME/UTAD for catalysts
characterizations.
[
37] (a) P.T. Tanev, T.J. Pinnavaia, Science 267 (1995) 865;
(
2
b) B. Jarrais, C. Pereira, A.R. Silva, A.P. Carvalho, J. Pires, C. Freire, Polyhedron
8 (2009) 994–1000.
38] M. Nowothny, L.N. Pedersen, U. Hanefeld, T. Maschmeyer, Chem. Eur. J. 8 (2002)
724–3730.
39] M. Vennat, P. Herson, J.M. Bregeault, G.B. Shul’pin, Eur. J. Inorg. Chem. 5 (2003)
08–917.
References
[
[
[
[
[
1] Y. Ishii, S. Sakaguchi, T. Iwahama, Adv. Synth. Catal. 343 (2001) 393–427.
2] (a) G.A. Olah, A. Molnár, Hydrocarbon Chemistry, 3rd ed., Wiley, New York,
3
2
(
003, pp. 10;
b) C. Gabriele, C. Fabrizio, T. Ferrucio, Selective Oxidation by Heterogeneous
9
40] (a) C.A. Tolman, J.D. Druliner, M.J. Nappa, N. Herron, in: C.L. Hill (Ed.), Activation
Functionalization of Alkanes, Wiley, New York, 1989, pp. 303–360;
Catalysis, Springer Science + Bussniss Media, New York, 2001.
3] M.H. Valkenberg, W.F. Hölderich, Catal. Rev. 44 (2002) 321–374.
4] (a) R. Burch, D.J. Crittle, M.J. Hayes, Catal. Today 47 (1999) 229–234;
[
[
(
(
(
b) M. Hartman, S. Ernst (Eds.), Angew. Chem. Int. 39 (2000) 888–890;
c) M. Spielman, AIChE J. 10 (1964) 496–501;
d) C.A. Tolman, J.D. Druliner, P.J. Krisic, M.J. Nappa, W.C. Seidel, I.D. Williams,
(
b) R. Burch, M.J. Hayes, J. Mol. Catal. A: Chem. 100 (1995) 13–33.
5] K.T. Queeney, D.A. Chen, C.M. Friend, Am. Chem. Soc. J. 119 (1997)
945–6946.
[
[
S.D. Ittel, J. Mol. Catal. 48 (1988) 129–148;
6
(
(
e) A. Kumar, G.S. Mishra, A. Kumar, Trans. Met. Chem. 28 (2003) 913–917;
f) R. Pohorecki, J. Ba, W. Dyga, W. Moniuk, A. Podgorska, P. Zdrojkowski, Wierz-
6] T.F.S. Silva, G.S. Mishra, M.F.G. da Silva, R. Wanke, L.M.D.R.S. Martins, A.J.L.
Pombeiro, Dalton Trans. 42 (2009) 9207–9215.
7] K.M. Parida, M. Sahoo, S. Singha, J. Mol. Catal. A: Chem. 329 (2010) 7–12.
8] R. Kumar, S. Sithambaram, S.L. Suib, J. Catal. 262 (2009) 304–313.
9] D. Yang, M. Liu, W. Zhao, L. Gao, Catal. Commun. 9 (2008) 2407–2410.
chowski, Chem. Eng. Sci. 56 (2001) 1285–1391.
[
[
[
[
41] (a) U. Schuchardt, D. Cardoso, R. Sercheli, R. Pereira, R.S. Da Cruz, M.C. Mario,
D. Mandelli, E.V. Spinace, E.L. Pires, Appl. Catal. A: Gen. 211 (2001) 1–17;
(
b) K. Machado, A. Kumar, G.S. Mishra, J. Ind. Eng. Chem. (2013),
[
10] (a) P.G. Cozzzi, Chem. Soc. Rev. 33 (2004) 410–421;
http://dx.doi.org/10.1016/j.jiec.2013.09.055 (in press);
(
b) G.C. Salomão, M.H.N. Olsen, V. Drago, C. Fernandes, L.C. Filho, O.A.C. Antunes,
(c) G.S. Nunes, I. Mayer, H.E. Toma, K. Araki, J. Catal. 236 (2005) 55–61;
(d) E.I. Karasevish, Yu.K. Karasevish, Kinet. Catal. 41 (2000) 535–542;
(e) A. Maldotti, C. Bartocci, G. Varani, A. Molinari, P. Battioni, D. Mansuy, Inorg.
Catal. Commun. 8 (2007) 69–72.
[
[
11] T.F.S. Silva, M.F.G. da Silva, G.S. Mishra, L.M.D.R.S. Martins, A.J.L. Pombeiro, J.
Organomet. Chem. 696 (2011) 1310–1318.
12] L. Canali, D.C. Sherrington, Chem. Soc. Rev. 28 (1999) 85–93.
Chem. 35 (1996) 1126–1131.