Manganese porphyrin in solution and heterogenized in different materials mediates oxidation of hydrocarbons by iodosylbenzene

Article abstract of DOI:10.1016/j.molcata.2013.06.022

We prepared a free base porphyrin (HP) containing hydroxy and methoxy groups in the mesophenyl substituents of the porphyrin ring. We then inserted Mn(III) ion into the resulting HP, to obtain MnP. We used silica synthesized by the sol-gel process to immobilize the MnP in either acidic or basic medium, which afforded the solids MnPA and MnPB, respectively. We also conducted the solvothermal Mn(III) ion insertion into HP, which furnished the self-structured solid MnPS. Because the MnP was catalytically active in homogeneous medium, we investigated the activity of all the prepared solids as heterogeneous catalysts in the oxidation of various hydrocarbons. MnPA, MnPB, and MnPS presented similar or higher activity than the homogeneous MnP in the oxidation of cyclooctene, cyclohexene, cyclohexane and n-heptane. Concerning alkane oxidation, the heterogeneous catalysts were more selective for the alcohol than the MnP in homogeneous solution. Moreover, the solid catalysts favored the oxidation of linear alkane, with selectivity toward the primary alcohol. We were able to recover all the solids at the end of the reaction and reuse them.

Full text of DOI:10.1016/j.molcata.2013.06.022

Journal of Molecular Catalysis A: Chemical 378 (2013) 263–272
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Manganese porphyrin in solution and heterogenized in different
materials mediates oxidation of hydrocarbons by iodosylbenzene
Gabriel Kaetan Baio Ferreiraa, Kelly Aparecida Dias de Freitas Castroa,
Guilherme Sippel Machadoa, Ronny Rocha Ribeirob, Katia Jorge Ciuffic,
Gustavo Pimenta Riccic, Jacqueline Aparecida Marquesd, Shirley Nakagakia,∗
a Laboratório de Bioinorgânica e Catálise and Departamento de Química—Centro Politécnico, Universidade Federal do Paraná, Curitiba 81531990 (UFPR)
PR, Brazil
b Laboratório de EPR da Região Sul, Departamento de Química—Centro Politécnico, Universidade Federal do Paraná (UFPR), Curitiba, PR, Brazil
c Unifran–Universidade de Franca, Franca, SP, Brazil
d Departamento de Química, UEPG, Ponta Grossa, PR, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history:
We prepared a free base porphyrin (HP) containing hydroxy and methoxy groups in the mesophenyl
substituents of the porphyrin ring. We then inserted Mn(III) ion into the resulting HP, to obtain MnP. We
used silica synthesized by the sol–gel process to immobilize the MnP in either acidic or basic medium,
which afforded the solids MnPA and MnPB, respectively. We also conducted the solvothermal Mn(III)
ion insertion into HP, which furnished the self-structured solid MnPS. Because the MnP was catalytically
active in homogeneous medium, we investigated the activity of all the prepared solids as heterogeneous
catalysts in the oxidation of various hydrocarbons. MnPA, MnPB, and MnPS presented similar or higher
activity than the homogeneous MnP in the oxidation of cyclooctene, cyclohexene, cyclohexane and n-
heptane. Concerning alkane oxidation, the heterogeneous catalysts were more selective for the alcohol
than the MnP in homogeneous solution. Moreover, the solid catalysts favored the oxidation of linear
alkane, with selectivity toward the primary alcohol. We were able to recover all the solids at the end of
the reaction and reuse them.
Received 20 February 2013
Received in revised form 28 June 2013
Accepted 29 June 2013
Available online xxx
Keywords:
Porphyrin
Heterogeneous catalysis
Oxidation
Silica
Self-assembly
© 2013 Elsevier B.V. All rights reserved.
1. Introduction
cells and energy [11], and catalysts of various chemical reactions
including oxidation processes [12].
Macrocyclic ligands such as porphyrins, phthalocyanines, and
porphycenes can form complexes with a variety of transition metals
in different oxidation states [1,2]. These ligands are often used as
model compounds in studies involving biological systems, because
they can mimic the catalytic activity of cytochrome P-450, catalase,
and lignin peroxidase, among other enzymes [3]. The selectivity and
efficiency of biological systems has motivated studies that employ
model compounds—these models help investigate the mechanisms
of oxidation and reduction reactions, aiming to obtain efficient size-
and shape-selective catalytic centers [3–6].
From a structural viewpoint, synthetic metalloporphyrins (MPs)
resemble the active center of heme-containing biological systems.
Therefore, MPs are frequently investigated as model compounds
and are applied in such areas as photodynamic therapy [7,8], chem-
ical sensors [9], construction of supramolecular structures [10],
MPs are catalytically active in both homogeneous (when the
MP is soluble in the reaction medium) and heterogeneous systems
(when the MP complex is not in the same phase as the substrate and
the reaction solvent, due to its immobilization onto different solids
or because it is structured as an insoluble material) [5,13–29].
In homogeneous systems, MPs usually undergo oxidative
destruction of the porphyrin ring [2] which, together with other
factors [3–6], contributes to low product yields. In contrast, in
the biological system the protein matrix isolates the catalytic site,
inhibiting oxidative destruction of the macrocyclic ligand and giv-
ing high product yields; another enzyme is necessary to catabolize
the heme ligand [30].
In academic biomimetic studies, the oxidative destruction of MP
can be reversed through the rational synthesis of more robust por-
phyrin ligands. Alternatively, the catalyst can be transferred from
the reaction solution to a solid phase, making catalysis a heteroge-
neous process [1–6]. Catalyst heterogenization can be achieved by
immobilizing the MP onto a rigid, inert support such as porous glass
[31], cationic clays like montmorillonite [32], lamellar anionic
synthetic compounds including layered double hydroxides
∗
Corresponding author. Tel.: +55 41 33613180; fax: +55 41 3361 3186.
E-mail addresses: shirleyn@ufpr.br, shirleynb17@gmail.com (S. Nakagaki).
1381-1169/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcata.2013.06.022

264
G.K.B. Ferreira et al. / Journal of Molecular Catalysis A: Chemical 378 (2013) 263–272
[13,33,34], and hydroxy salts [35], among other matrices
[36–38].
hydroxylation using O2-ascorbate system and the metal ion in the
porphyrins significantly affected the catalytic efficiencies of these
hybrid materials [52].
More recently, a new strategy has been proposed for catalyst
heterogenization: the soluble catalytic species is converted into
an insoluble solid via a synthetic process that involves the for-
mation of a supramolecular network. This process generates a
microporous solid, the so-called metal-organic framework, or MOF
[14,15,29,39–41].
Both of the abovementioned strategies have provided effec-
tive heterogeneous catalysts. These strategies have overcome two
major drawbacks of homogeneous catalysis: (i) catalyst deactiva-
tion caused by oxidative destruction of the catalytic species and (ii)
formation of inactive species via side reactions [2–4,6].
A further advantage of heterogeneous catalysis is the possi-
bility to recover and reuse the catalyst—the support-macrocyclic
complex system is more stable than the macrocyclic complex in
solution, making the catalytic process more profitable and allowing
for future technological applications. Moreover, the solid catalyst
is usually more efficient than its homogeneous counterpart: the
catalyst-support interaction and the porous structure of the MOF
solid give rise to unusual selectivity [53].
This paper uses a manganese(III) porphyrin (MnP) obtained
from the ligand 5,10,15,20-tetrakis free base (4-hydroxy,3-
methoxyphenyl) porphyrin (HP) as homogeneous and heteroge-
neous catalyst in oxidation reactions. We compare the catalytic
activity of the target MnP in three different systems: in homoge-
neous medium, supported onto silica gel obtained by the sol–gel
process, and arranged in an insoluble structured solid (Fig. 1).
Porphyrins and other macrocyclic molecules are good build-
ing blocks for supramolecular structures: they have a rigid planar
core measuring 1 nm2. The peripheral positions of this core, espe-
cially the meso position of the macrocyclic ring, can easily bind to a
large variety of chemical functions. Additionally, these macrocycles
can coordinate metal ions of different oxidation states, providing
a large number of algorithms to construct distinct supramolec-
ular architectures, coordination polymers, and porous crystalline
structures [14]. MOFs are the best-known class of materials orig-
inating from covalent interactions between these building blocks
and metallic units (such as metal ions or clusters), culminating in
mono (1D), bi (2D), or three-dimensional (3D) porous crystalline
coordination polymers with high surface area and well-defined
pores and channels [16,42]. The combination of these character-
istics with the well-known features of MPs – inherent stability,
unique optical properties, and synthetic versatility – enables MOFs
to act as sensors, molecular separators, and catalysts of heteroge-
neous processes [15]. When these solids are used as heterogeneous
catalysts, their structure, which consists of channels and pores,
becomes important: the channels and pores confer the macrocycle
catalytic efficiency as well as size- and shape-selectivity toward
certain substrates [43–46].
Immobilization of catalytic species onto silica obtained by the
hydrolytic sol–gel route is also an efficient tool to obtain highly
pure, rigid, and inert catalytic solids The hydrolytic sol–gel pro-
cess described by Stöber [47], which employs basic catalysis, often
generates solid silica with spherical features [48], but this specific
morphology has not been reported to change the catalytic results
obtained with MPs immobilized onto silica [49,50]. Recently Ucoski
et al. immobilized two neutral metalloporphyrins [Mn(TAPP)]Ac
and [Fe(TAPP)]Cl (TAPP = 5,10,15,20-tetrakis(acetalphenyl por-
phyrin)) on silica obtained through the sol–gel process in acidic
and basic conditions. Yields slightly lower or similar to those
achieved in homogeneous medium were obtained for the MPs
immobilized on silica. The two solid were easily recovered from
the reaction medium and reused, and the recyclability capacity
and retained activity upon reuse were demonstrated [50]. Farhadi
et al. reported that Mn(III) (TDCPP)Cl, was covalently bound to a
silica prepared by the sol–gel process and used as photocatalyst in
a heterogeneous process for the selective and efficient oxidative
decarboxylation of ␣-arylacetic acids with H2O2 at room tem-
perature. The activity of the solid catalyst in the heterogeneous
photocatalytic system was higher than that of a corresponding
homogeneous one. The catalyst was reused several times without
loss of its activity and selectivity [51]. Cai et al. prepared three types
of hybrid materials based on silica-metalloporphyrins with Fe, Co
and Mn porphyrin [APTCPP] (5-(4-allyloxy)phenyl-10,15,20-tri(4-
chlorophenyl) porphyrin) also using the sol–gel process involving
a thiol-ene polymerization reaction of the free base porphyrin with
3-mercaptopropyltrimethoxysilane. They investigated the solids as
catalysts for the aerobic oxidation of cyclohexane. They found that
the prepared hybrid materials were more efficient catalyst than
the analogous non-supported metalloporphyrins for cyclohexane
2. Experimental
All the solvents and reagents were purchased from Aldrich,
Merck, or Fluka and were used without purification, unless oth-
erwise stated. N,N-Dimethylformamide (DMF) was distilled under
reduced pressure and stocked over 4-Å molecular sieves. Pyrrole
was distilled under reduced pressure prior to use. Alkenes (cis-
cyclooctene and cyclohexene) were purified on an alumina column
before use.
2.1. Synthesis
2.1.1. Free base porphyrin (HP), manganese porphyrin (MnP) and
the self-structured material (MnPS)
The free base porphyrin (HP, [H2(THMPP)]−[5,10,15,20-
tetrakis (4-hydroxy, 3-methoxyphenyl)porphyrin]) was syn-
thesized by the Adler–Longo’s methodology [54,55]. Man-
ganese(III) ion (manganese(II) acetate, metal ion/HP molar
ratio = 10) was inserted into the purified HP under acetic
acid reflux, for 6 h [56] giving the dark green complex
named MnP ([Mn(THMPP)]−[5,10,15,20-tetrakis (4-hydroxy,3-
methoxyphenyl)porphyrinato] manganese(III) acetate). HP and
MnP were characterized by UV–vis and FTIR (Figs. S1 and S2a,
supplementary material). Perpendicular microwave polarization
X-band electron paramagnetic resonance (EPR) recorded for MnP
showed no absorptions, as expected for the Mn(III) ion inserted into
the porphyrin complex.
Vanillin was employed to synthesize the tetra substituted por-
phyrin because it is a cheap aldehyde in comparison to others
containing hydroxyl groups aldehydes (almost four times cheaper
than 4-hydroxybenzaldehyde). Our choice in a hydroxyphenyl por-
phyrin is based on the easiness of deprotonation of phenol groups
both to form coordination polymers with metal centers and to cova-
lently bind to silica during sol–gel process.
In a solvothermal reaction, HP and solid manganese(II) acetate
(metal ion/HP molar ratio = 10) were reacted in DMF for 48 h, result-
ing in a black solid, MnPS. This material was characterized by
reflectance UV–vis spectroscopy (which displayed bands relative
to MnP), FTIR spectroscopy, EPR, powder X-ray diffraction (PXRD),
and textural analysis by the BET method.
2.1.2. MnP immobilization onto silica—materials MnPA and
MnPB
A MnP solution in methanol (22 mmol L−1), tetraethyl ortho-
silicate (TEOS) (13 mmol), and deionized water (10 mL) were
mixed in an Erlenmeyer flask, under magnetic stirring. Alco-
hol (ethanol for the acidic method or isopropyl alcohol for the

G.K.B. Ferreira et al. / Journal of Molecular Catalysis A: Chemical 378 (2013) 263–272
265
Fig. 1. Schematic representation of the manganese porphyrin (MnP, top, middle) and of the solids resulting from the heterogenization process: solid manganese porphyrin
in silica prepared by the sol–gel process (MnPA and MnPB, bottom left, silica network represented by the grey circle with some metalloporphyrins encapsulated into its pores
or adsorbed onto the surface); insoluble manganese porphyrin solid prepared by solvothermal process (MnPS, bottom right, tridimensional material and single sheet view).
basic method) (50 mmol) was then added to the flask, together
with the hydrolysis-condensation catalyst (hydrochloric acid for
the acidic or ammonium for the basic route) (2.5 mmol). The
TEOS/alcohol/water molar ratio was kept at 1:4:40. The system was
maintained under magnetic stirring for 30 min, at 60 ◦C, followed by
aging for 25 days, at room temperature. The obtained glass solids,
MnPA (obtained by acid catalysis) and MnPB (originated from basic
catalysis), were exhaustively washed with methanol in a Sohxlet
extractor for 24 h and dried under reduced pressure. The MnP load-
ing on the silica (mol of MnP per gram of silica) was indirectly
obtained: the amount of MnP removed from the silica during the
washing procedures was determined by quantitative UV–vis anal-
ysis of the washing solutions; this amount was then subtracted
from the initial quantity of MnP used in the immobilization pro-
cedure. MnPA and MnPB were characterized by UV–vis and FTIR
spectroscopies, EPR, and textural analysis.
equipment was operated at 40 kV and 40 mA, using Cu K˛ radiation
−1
˚
(ꢀ = 1.5418 A) and a dwell time of 1 min
.
Textural analyses were carried out on the basis of the cor-
responding nitrogen (Air–liquid, 99.999%) adsorption at −196 ◦C,
obtained from a static volumetric apparatus (Micromeritics ASAP
2020 adsorption analyzer). The samples (0.2 g) were degassed at
200 ◦C for 24 h (p < 0.133 Pa). The specific surface area was calcu-
lated by the BET method, and the total pore volume was estimated
from the amount of nitrogen adsorbed at a relative pressure of 0.95.
All the products from the catalytic oxidation reactions were
identified using an Agilent 6850 gas chromatograph (flame ion-
ization detector) equipped with a DB-WAX capillary column with
a length of 30 m and internal diameter of 0.25 mm (J&W Sci-
entific). The oven temperature program used to determine the
oxidation products from cis-cyclooctene, cyclohexene, and cyclo-
hexane started at 100 ◦C. The temperature was then increased
to 150 ◦C at 10 ◦C min−1, followed by further temperature rise to
200 ◦C at 50 ◦C min−1, which was maintained for 1 min. To deter-
mine the products originated from n-heptane, the temperature
program started at 70 ◦C, followed by a temperature elevation to
100 ◦C at 5 ◦C min−1, maintained for 1 min, and further temperature
increase to 200 ◦C at 20 ◦C min−1, kept for 1 min.
2.2. Catalytic activity tests
All the materials were tested as catalysts in the oxidation
of different hydrocarbon substrates (cis-cyclooctene, cyclohex-
ene, cyclohexane, and n-heptane), using iodosylbenzene (PhIO,
prepared according to Saltzman and Sharefkin and periodically
checked for purity by iodometric titration [57]) as oxidant and a
catalyst/oxidant/substrate molar ratio of 1:10:1000. The reactions
were carried out using dichloromethane/acetonitrile (DCM/ACN)
1:1 (v/v) as solvent mixture, for 1 h, at room temperature, under
argon atmosphere. Products were quantified by gas chromatogra-
phy using n-octanol as internal standard.
3. Results and discussion
3.1. HP, MnP, and MnPS
Condensation of pyrrole and vanillin via acid catalysis using
Adler–Longo’s methodology [54] furnished the free base porphyrin
HP as a purple solid; its UV–vis spectrum displayed the typical
Soret band at 426 nm (ε = 41,700 L mol−1 cm−1), as well as the four
Q-bands at 520, 558, 596, and 654 nm, characteristic of free base
porphyrins (Fig. S1a, supplementary material).
The UV–vis spectrum of the green solid MnP presented a Soret
band at 476 nm, due to the bathochromic shift undergone by man-
ganese(III) porphyrins upon Mn(III) ion insertion into the free base
porphyrin. The set of bands under 430 nm and the two Q-bands at
580 and 622 nm resulted from alteration in the microsymmetry of
the porphyrin macrocycle from D2h to D4h, a consequence of the
presence of the Mn(III) ion in the porphyrin core [55,58–61] (Fig.
S1b, supplementary material).
2.3. Characterization and apparatus
The UV–vis spectra were registered on a Varian Cary 100 Bio
Spectrophotometer in the 200–800 nm range. The liquid samples
were analyzed using a cell with 1-cm path length. A Teflon® sup-
port was employed for the solid samples.
The FTIR spectra were recorded on a Biorad 3500 GX spec-
trophotometer in the 400–4000 cm−1 range. The KBr pellets were
obtained by crushing the solids (1 mg) with spectroscopic grade KBr
(100 mg). All the spectra were collected with a resolution of 4 cm−1
and accumulation of 32 scans.
Room temperature and 77-K perpendicular microwave polar-
ization X-band EPR was measured on a Bruker EMX microX
spectrometer; 77-K parallel microwave polarization X-band EPR
experiments were conducted on a Bruker Elexsys E500 spectrom-
eter equipped with a Bruker ER4116DM dual mode resonator.
For the PXRD analysis, self-oriented films were placed on
neutral glass sample holders. The patterns were obtained on a
Shimadzu XRD-6000 diffractometer, in the reflection mode. The
The perpendicular microwave polarization X-band EPR of MnP
did not evidence any absorption, as expected for manganese(III)
porphyrins (data not shown) [62].
Mn(III) ion insertion into HP under solvothermal conditions
gave a black insoluble solid (MnPS) with UV–vis spectral profile
similar to the one achieved for the MnP obtained under conven-
tional metal insertion conditions (solvent reflux). The reflectance
UV–vis spectrum of MnPS revealed a Soret band at 426 nm,

266
G.K.B. Ferreira et al. / Journal of Molecular Catalysis A: Chemical 378 (2013) 263–272
Table 1
attesting to the presence of the MnP in MnPS (Fig. S2, supplemen-
tary material).
Textural properties of the silica-based solids.
The FTIR spectrum of MnPS (Fig. S3, supplementary material)
presented the same pattern as the FTIR spectrum of MnP, con-
firming the presence of the MnP in the structure of MnPS. Bands
relative to acetate as a counterion also appeared, as observed for
MnP (difference between asymmetric and symmetric stretches
equal to 172 cm−1 for MnP and 172 cm−1 for MnPS) [63,64] as
expected for a manganese(III) porphyrin. The band at 1653 cm−1
may refer to C=O stretching from DMF molecules present in MnPS
[65], adsorbed onto the walls of the material and/or coordinated to
central metal site of manganese porphyrins [66,67].
The perpendicular microwave polarization X-band EPR spectra
of MnPS at room temperature and 77 K were similar: there was
intense absorption around g = 2, characteristic of species contain-
ing Mn(II) centers (Fig. 2a). In contrast, the EPR spectra of MnP,
in the traditional EPR analysis (perpendicular polarization CW-
EPR) shows no signal, as expected, due to the presence of only
Mn(III) central species that are “silent” to this technique in the
perpendicular mode. When MnPS is analyzed by parallel EPR (elec-
tron paramagnetic resonance, in which an oscillating magnetic
field is applied parallel to the static magnetic field—parallel polar-
ization CW-EPR[68]), transitions both for Mn(II) and Mn(III) are
observed and these data suggested that manganese(III) species are
also present in MnS solid. The EPR technique of parallel polar-
ization is an useful tool for observing EPR spectra from integer
electron spin systems such as found in many transition metals
(Mn(III)). On the other hand conventional perpendicular mode
EPR allows observation of half integer spin systems (Mn(IV),
Mn(II), etc.) These signals suggest that MnPS, displaying Mn(III)
in the core of the macrocycle (Fig. 2b) (a situation that does
not generate any signal in perpendicular microwave polariza-
tion EPR), also contains Mn(II) species binding the macrocycles
together in a supramolecular arrangement that resembles MOFs
[41,62,68,69].
MnPA
MnPB
SiA
SiB
MnP Loading (mol g−1
% MnP Immobilization
)
5.11 × 10−6
47
383
0.29
30.14
3.05 × 10−5
98
317
0.84
105.9
SBET (m2 g−1
)
368
0.28
29.8
315
0.74
100.4
Pore volume (cm3 g−1
Pore size (Å)
)
to those depicted in Fig. 3d and e, except for the appearance of an
amorphous halo, attributed to the action of DMF on MnPS. There-
fore, the robust crystalline MnPS maintained its structure after
cycles of drying and resolvation with DMF [14].
3.2. Materials based on silica—MnPA and MnPB
We obtained the MnP loadings in the materials MnPA and MnPB
(mol of MnP per gram of silica) indirectly: first, we determined the
amount of MnP removed from the silica during the washings of
MnPA and MnPB by quantitative UV–vis analysis of the methanol
washing solutions; we then subtracted this amount from the initial
quantity of MnP used in the immobilization procedure, which fur-
nished the following loadings MnPA: 5.11 × 10−6 mol of MnP g−1
of silica and MnPB: 3.05 × 10−5 mol g−1 of silica. We ground MnPA
and MnPB and analyzed the powders by UV–vis diffuse reflectance
spectroscopy. MnPA and MnPB displayed the same spectral profile
as MnP, confirming the presence of the MnP in the silica network;
no MnP demetallation occurred during the immobilization pro-
cedure, irrespective of the medium (acidic or basic) (figure not
shown).
We obtained sixfold higher MnP loading in basic medium. The
use of acid or basic catalyst during the hydrolytic sol–gel route
accounted for the different MnP loadings in MnPA and MnPB. In
basic medium, the hydroxyl groups at the phenyl rings of the MnP
are probably deprotonated (vanillin residue has pKa = 7.78), facili-
tating participation of the MnP in the hydrolysis and condensation
reactions leading to the silica network.
We did not detect any signals around g = 2 in the perpendicular
microwave polarization X-band EPR analysis of MnPA and MnPB at
room temperature. Hence, MnPA and MnPB contain stable Mn(III)
species [62,74].
The FTIR spectra of MnPA and MnPB and of the control solids
SiA and SiB (silica generated via the same routes but in the absence
of the MnP) were very similar (not shown)—only bands relative
to silica deformation and stretching appeared; the MnP did not
contribute to the spectra. MnPA and MnPB contained low con-
centrations of the MnP (less than 0.4% in mass); therefore, we
were only able to visualize the intense Si–O bands. FTIR data:
cm−1 (attribution)—956 (Si–OH), 1080 (Si–O–Si), 3465 (Si–OH)
[63].
The PXRD analysis (Fig. 3) of the synthesized compounds
revealed distinct patterns. The diffractograms of the starting salt
Mn(OAc)2 and of HP (Fig. 3a and b, respectively) agreed with the
patterns reported in the literature [70]. MnP displayed a peak at
2Â = 8.58◦ (Fig. 3c), attributed to organization of the planar MnP
after its insertion into the support used for the PXRD analysis–these
planes can diffract the X-ray radiation, generating the diffraction
peak. However, this diffraction did not directly relate to a structure
with effective diffraction planes, as evidenced by the absence of
secondary diffraction peaks in the analysis.
The PXRD pattern of MnPS (Fig. 3d) was different: compared
with the diffractogram of MnP, the first peak shifted to 2Â = 8.76◦,
˚
corresponding to a distance of 10.08 A. We ascribed this signal as
the (0 0 1) peak [71]. We also detected the peaks (0 0 2), (0 0 3), and
(0 0 5) [70], which show that MnPS presents more diffraction planes
and is a more crystalline material than MnP. MnPS also displayed
short-distance peaks in the 2Â range from 32◦ to 39◦, relative to
Interestingly, incorporation of the MnP during formation of the
silica gel network did not change the textural properties of the sil-
ica (compare MnPA with SiA and MnPB with SiB in Table 1), which
contrasts with results of other works [50]. The surface area of MnPA
and MnPB was around 350 m2 g−1. We had expected a larger sur-
face area for the material obtained by basic catalysis, MnPB, rather
than for the material synthesized by acid catalysis, MnPA [49,50].
However, MnPA required shorter gelation time, which culminated
in a less ordered silica network. Compared with MnPB, the ran-
dom arrangement of the silicate units in MnPA resulted in a larger
surface area.
˚
distances around 2.50 A; these peaks probably refer to the distance
between two metal ions in the structure and suggest that the crys-
talline structure of MnPS is different from the starting structure of
MnP.
To evaluate the robustness of the synthesized MnPS crystalline
network, we exhaustively dried MnPS under vacuum at room tem-
perature [39], to eliminate traces of the solvent DMF, and analyzed
the resulting solid by PXRD (Fig. 3e). Compared with the diffrac-
togram in Fig. 3d, the diffractogram in Fig. 3e did not evidence
any significant modifications. Thus, the structure of MnPS was not
destroyed during the drying process, and its long-range order was
preserved [72,73]. Next, we suspended MnPS in DMF for 6 h, for
resolvation, and recorded a new PXRD diffractogram while MnPS
was still wet (Fig. 3f). Again, we obtained a PXRD pattern similar
According to Table 1, basic catalysis afforded more porous
materials with larger pore size. Basic catalysis generally furnishes
highly branched structures; i.e., spherical particles, and large inter-
cluster pores remain after solvent removal, due to steric effects.

G.K.B. Ferreira et al. / Journal of Molecular Catalysis A: Chemical 378 (2013) 263–272
267
Fig. 2. 77-K X-band EPR MnPS spectra. (a) Perpendicular polarization. The intense absorption around 3500 G, characteristic of Mn(II) species, is due to a ꢁmS
=
1 selection
rule transition. (b) Parallel polarization. The ꢁmS
= 4 transition, present as an absorption around 900 G, is due to Mn(III) species. The Mn(II) species also produce ꢁmS = 2
absorptions in the parallel polarization EPR experiment (B, dashed rectangle).
Fig. 3. Powder X-ray diffraction patterns of (a) Mn(OAc)2, (b) HP, (c) MnP, (d) MnPS after synthesis, (e) MnPS after de-solvation under vacuum, and (f) MnPS after re-solvation
with DMF.
Consequently, spheres with high degree of monodispersity are
expected in this case. On the other hand, acid catalysis yields
weakly branched systems and denser materials [75], but it does not
favor monodispersity; the material is irregular, which justifies the
larger surface area and the smaller pore size and volume of MnPA
[76–78].
According to the IUPAC classification, the nitrogen adsorp-
tion/desorption isotherms of MnPA and MnPB (Fig. S4 and S5,
supplementary material) are Type-IV [79]. Nevertheless, MnPA and
MnPB have hysteresis H2 and H1, respectively. H2 hysteresis is
characteristic of disordered/interconnected pores, but it can also
indicate ink-bottle pores; H1 hysteresis relates to ordered and
cylindrical pores open at both ends [79,80]. Again, the shorter gela-
tion time required by MnPA when compared with MnPB explains
these results.
3.3. Catalytic activity of the synthesized materials
We tested all the synthesized solids containing MnP in their
structures (MnPS, MnPA, and MnPB) as heterogeneous catalysts
in the oxidation of hydrocarbons; we used MnP in homogeneous
medium for comparison (it is noteworthy that MnP was not totally
soluble in the solvent mixture ACN:DCM). We also conducted con-
trol reactions without catalyst (homogeneous control reaction)
or using pure silica gel prepared by the sol–gel process without
immobilized MnP (heterogeneous control reaction). We investi-
gated processes involving single oxygen transfer only—we used
iodosylbenzene (PhIO, insoluble in the reaction medium) as the
oxidant, and the model compounds acted as monooxygenases. We
evaluated cyclic alkenes (cis-cyclooctene and cyclohexene) and
alkanes (cyclohexane and heptane) as substrates (Fig. 4).
Fig. 4. Schematic representation of the oxidation of hydrocarbons catalyzed by the MnP studied herein (clockwise): cis-cyclooctene, cyclohexene, cyclohexane, and n-heptane.

268
G.K.B. Ferreira et al. / Journal of Molecular Catalysis A: Chemical 378 (2013) 263–272
Table 2
MnPA and MnPB furnished lower product yield than homoge-
neous MnP. After immobilization, the catalytic species in MnPA and
MnPB probably became less accessible to the reactants as compared
with the MnP in solution [83].
Yields of cis-cyclooctene epoxide from cis-cyclooctene oxidation by PhIO catalyzed
by MnP systemsa.
Catalyst
Cyclooctene yield (%)b
MnPA, with six times lower MnP loading than MnPB, afforded
larger product yield (35% and 27%, respectively, see Table 2). The
textural properties of the silica-based MnPA and MnPB (Table 1)
justify these results: MnPA has higher surface area and smaller pore
size and volume. Hence, catalysis probably occurs on the surface of
the MnPA catalyst, and not within its pores and interstices. UV–vis
analysis of the supernatants of the reactions catalyzed by the silica-
based catalysts did not detect any MnP, so the MP did not leach from
either MnPA or MnPB.
MnP
MnPS
MnPS reuse
MnPA
MnPA reuse
MnPB
54
60
55
35
31
27
23
MnPB reuse
PhIOc
4.5
4.6
Silica + PhIOd
e
Mn(OAc)2
17
a
Reactions carried out in duplicate. Reaction conditions: argon atmosphere,
absence of light, magnetic stirring, room temperature, 60 min, DCM/ACN 1:1 (v/v)
solvent mixture.
The solid MnPS provided product yield higher than those
obtained with the silica-based catalysts MnPA and MnPB and sim-
ilar to that achieved with the MnP in homogeneous medium.
Insoluble MnPS and MnP should have similar composition: MnPS
should consist of MnP units structured in such a way that an insol-
uble solid exists in the reaction medium. The solid MnPS proved to
be advantageous over the silica-based MnPA and MnPB for the fol-
lowing reasons: (i) whereas the MnP is diluted in the silica, due to
the low catalyst loading in MnPA and MnPB, the surface of the solid
MnPS consists of MnP. In other words, in MnPS the MnP is totally
available to the approach of the substrate and the oxidant, so the
reactants do not need to permeate into the structure of the solid to
react with the active catalytic species. (ii) It is possible to recycle and
reuse MnPS in more than one reaction, because MnPS is robust. To
conduct the reuse reactions, we recovered MnPS from the reaction
medium after the first use; we then washed, activated, and reused
the recovered catalyst in a second reaction. Recycled MnPS gave
virtually the same product yield as the fresh MnPS and the MnP in
homogenous catalysis (Table 2). Indeed, FTIR analysis of MnPS after
the first and second reaction confirmed its robustness: the recycled
MnPS displayed the same spectral pattern as the freshly prepared
MnPS (Fig. S3b, supplementary material). A catalytic test with man-
ganese(II) acetate tetrahydrate – Mn(OAc)2·4H2O – was performed
to evaluate the possible activity of Mn(II) centers. Epoxide of 17%
yield was obtained (Table 2), showing that divalent Mn ion in solu-
tion can catalyze epoxidation and maybe, those present in MnPS
can also have catalytic activity, since they are not blocked by met-
alloporphyrin ligands with their coordination spheres saturated,
what does not allow reactants approximation.
b
Yields based on PhIO.
Homogeneous control reaction.
Heterogeneous control reaction.
Control reaction using manganese(II) acetate tetrahydrate as catalyst in the
c
d
e
same reaction conditions.
3.3.1. Oxidation of cis-cyclooctene
First, we used cis-cyclooctene as diagnostic substrate. Con-
trary to cyclohexene, which gives epoxide, alcohol, and/or ketone
as products, cis-cyclooctene oxide is the only product gener-
ated during metalloporphyrin-catalyzed cis-cyclooctene oxidation
[5,18]. This happens because proton abstraction from the allyl-
lic position of cis-cyclooctene generates a less stable radical than
proton abstraction from the vinyllic position of this same sub-
strate [18]. Indeed, the cyclooct-2-en-1-yl radical (allylic) can
assume an unfavorable configuration, because trans interactions
occur between the hydrogen atoms. To maintain the unsatura-
tion and the radical structure in the same plane, the chain of
the cyclooct-2-en-1-yl radical becomes distorted and, thus, less
stable.
At a catalyst/oxidant/substrate molar ratio of 1:10:1000, MnP
gave 54% cis-cyclooctene oxide yield compared with 4.5% achieved
in the control reaction conducted in the absence of MnP (Table 2).
Therefore, MnP really catalyzed cis-cyclooctene oxidation by PhIO.
In the case of the heterogeneous system MnPS, the activity of the
MnP was virtually maintained, but it decreased when MnPA and
MnPB were used as catalysts.
In homogeneous medium, MnP afforded cis-cyclooctene oxide
yield similar to those obtained with classic metalloporhyrins
(MPs) such as metal(III) tetraphenylporphyrins, [M(TPP)] [81],
the so-called first-generation porphyrins [3]. However, MnP
provided lower product yield than those achieved with second-
generation MPs (often between 80% and 100% epoxide yield).
Second-generation MPs contain bulky groups (such as phenyl,
alkoxy, and tert-butyl) or electron withdrawing substituents [3]
at the ortho position of the meso-phenyl substituents in the por-
phyrin macrocycle [12]. These substituents activate the catalytic
species, furnishing higher product yields than the first-generation
MPs. The substituents in second-generation MPs also protect the
catalyst against deactivation; more specifically, they avoid for-
mation of non-catalytic species like dimeric species as well as
catalyst self-destruction due to contact between the activated
and the non-activated MP species [2–6]. We can consider the
MnP studied in this paper as a first generation-like catalyst: its
macrocyclic structure does not contain substituents that confer
enough protection against deactivation effects, so the high-valence
manganyl porphyrin (Mn(V)(O)Por), assumed as the catalytically
active species [82] in this case, is less stable and gives prod-
uct yields comparable to those attained with first-generation
MPs [2–6].
MnPA and MnPB were also recycled for oxidation of cyclooctene
and the yields obtained for epoxide formation were 31% (MnPA)
and 23% (MnPB). These results show that manganese porphyrin
immobilized onto silica still remains its catalytic activity.
We performed UV–vis analysis of the supernatant of each reac-
tion that used the solid MnPS as catalyst. The reaction supernatant
was colorless, indicating that the MnP did not leach from MnPS
during catalysis. Therefore, this solid is robust and resistant to
the catalytic conditions. Moreover, the absence of MnP in the
supernatant of the reaction confirmed that the cis-cyclooctene
oxidation by PhIO catalyzed by MnPS was a truly heterogeneous
process. However, when we exposed MnPS to methanol, a small
amount of the MnP solubilized in this solvent, and the solu-
tion acquired a green color. We calculated the loss of MnP in
methanol and estimated it to be about 2%. Nevertheless, this
mass loss did not interfere in the catalytic result because we
conducted the cis-cyclooctene oxidation reaction in a different
solvent.
Hydrogen peroxide, a green oxidant, was not investigated
as oxidizing agent in this work because with manganese por-
phyrins it is necessary the use of
a co-catalyst [21,84,85].
Also, for heterogeneous reactions, solids MnPS, MnPA and
MnPB promote disproportionation of H2O2, not allowing its
use.

G.K.B. Ferreira et al. / Journal of Molecular Catalysis A: Chemical 378 (2013) 263–272
269
3.3.2. Oxidation of cyclohexene
The oxidation of cyclohexene can take place at the vinyllic posi-
tion, to give the epoxide, or at the allylic position, to produce the
relative alcohol or ketone products. Therefore, cyclohexene oxida-
tion poses a new challenge to researchers in the area of catalysis:
scientists have sought to enhance its kinetics and promote dif-
ferent selectivity, once there is competition between the allylic
and vinyllic routes [18–20,86,87]. In this paper, we investigated
three different conditions regarding the amount of oxidant and
substrate used during cyclohexene oxidation-conditions A, B, and
C (see Table 3). All the studied MnP materials exhibited catalytic
activity in the oxidation of cyclohexene by PhIO (Table 3).
In conditions A and B, MnP in homogeneous medium was selec-
tive toward the epoxide (Table 3, runs 1 and 2), as expected for
catalysis using MPs [81]. In the absence of a catalyst, though, allylic
oxidation is usually favored; indeed, the homogeneous reaction
control (without MnP) conducted under condition B gave an epox-
ide/allylic product ratio equal to 0.21 (Table 3, run 16), while MnP
afforded an epoxide/allylic product ratio equal to approximately
7.0, corroborating the selectivity of the latter catalyst toward the
epoxide. It is noteworthy that conversion values (the amount of
substrate converted in product) for catalyzed reactions in condition
B are higher than those obtained from control reactions, proving
that catalytic activity is due to metalloporphyrins molecules.
The oxidation of alkenes at the vinyllic position occurs via
heterolytic transfer of the oxygen atom from the oxidant to the sub-
strate. This route is favored in the case of PhIO, because this oxidant
prevents the formation of radical species [19,20,56,88–92].
Condition C (Table 3, run 3), which employed a higher amount
of substrate, significantly decreased the epoxide yield (44% to 31%
from B to C, runs 2 and 3, respectively) and increased the formation
of allylic products (from 6.6% to 44%, runs 2 and 3, respectively). This
change in products ratio can be considered as a drop in the catalytic
activity of MnP, because allylic products usually result from non-
catalytic processes. Here, addition of larger volumes of cyclohexene
makes the reaction solution more apolar, lowering the solubility of
the MnP in the reaction medium. The reaction probably follows a
non-catalytic pathway, which justifies the lower catalytic yields. In
fact, the yields of allylic products obtained in run 3 were similar to
those obtained in the control reaction (run 17).
Fig. 5. Dependence of the cyclohexanol yield on the composition of the solvent
mixture.
O2 adsorption and favoring allylic products. In spite of the larger
surface area of MnPA, O2 adsorption onto this catalyst is more dif-
ficult: its pores are irregular and interconnected, which indicates
the presence of ink-bottle pores, as discussed in the paragraph on
nitrogen adsorption/desorption isotherms.
In condition B, the results attained with MnPS and homoge-
neous MnP were similar (Table 2, runs 5 and 2, respectively). As
discussed for cis-cyclooctene oxidation, in MnPS the MnP is present
as an organized structure, which favors the catalytic activity of this
manganese(III) porphyrin.
We also investigated the main advantage of heterogeneous cata-
lysts over homogeneous ones–catalyst reuse–in condition B, using
MnPS (Table 3, runs 7 and 8). The recycled MnPS furnished total
product yields similar to those obtained with the fresh MnPS under
the same condition (Table 3, run 5), with high selectivity for the
epoxide. Hence, MnPS is a robust and reusable solid catalyst.
Condition B gave the highest epoxide/allyllic products ratio
for all the catalysts (homogeneous MnP and heterogeneous
MnPS, MnPA, and MnPB) and also the highest conversion values,
showing that this condition allowed for the best catalytic per-
formance among the tested conditions and favored the vinyllic
product.
In the case of heterogeneous catalysis (MnPS, MnPA, and MnPB),
conditions A and C (Table 3, runs 4, 6, 9, 11, 12, and 14) afforded
epoxide as well as a large amount of ketone. Despite exhaustive
deaeration of the reaction vessels with argon purge, dioxygen still
remained adsorbed onto the pores and surfaces of the catalysts.
The presence of O2 in the reaction medium favors formation of
radical species, which quickly react with the substrate and gen-
erate the allylic radical, leading to the ketone [93]. By doubling the
amount of PhIO (going from condition A to condition B), the epox-
ide yields increased for all the heterogeneous catalyst. Therefore, a
greater amount of oxidant is necessary for the reaction to take place
between the heterogenized MnP and the oxygen donor, leading to
effective formation of the active catalytic species.
3.3.3. Oxidation of cyclohexane
The inert nature of saturated C–H bonds and the different
chemo- and regioselectivities achieved during alkane oxidation
using different oxidants and catalysts makes alkane functional-
ization a challenging task [12,94]. In this study, we employed
cyclohexane, to probe the chemoselectivity (formation of cyclohex-
anol, c-ol, and/or cyclohexanone, c-one) and n-heptane, to monitor
both the chemo (production of heptanol or heptanone) and regio-
selectivity (oxidation at different positions of the carbon chain)
of the prepared catalysts. We did not investigate diols, aldehydes,
carboxylic acids, or other n-heptane oxidation products.
Concerning the oxidation of cyclohexane, all the studied cat-
alysts were selective for the alcohol (Table 4, runs 1 to 4), with
c-ol/c-one ratios ranging from 5 to 16. The control reactions did
not yield any product (Table 4, runs 5 and 6), showing that it is
difficult to oxidize cyclohexane in the absence of a catalyst.
As expected for metalloporphyrin-catalyzed cyclohexane oxida-
tion, the homogeneous catalyst MnP (run 1) was selective toward
the alcohol [93]. Indeed, this MnP provided greater yields than
first-generation MPs ([Fe(TPP), 8% yield for cyclohexanol, in DCM
[20,31,95,96]; [Mn(TAPP)], 27% yield cyclohexanol [3,50]).
In an attempt to achieve better performance in the catalytic
oxidation of cyclohexane in homogeneous medium, we studied
how the DCM/ACN volume ratio affected the process (Fig. 5). We
found that the pure solvents (ACN or DCM) led to the worst results.
As observed in the case of cis-cyclooctene oxidation, MnPA fur-
nished higher product yields than MnPB in all the conditions. Again,
the textural properties of MnPA and MnPB justify these results: in
MnPA, the MnP is located at the surface of the silica, facilitating
the approach of PhIO and the substrate to the catalytic site, cul-
minating in slightly better product yields. In the case of MnPB, its
pores are ordered/cylindrical and open at both ends, facilitating

270
G.K.B. Ferreira et al. / Journal of Molecular Catalysis A: Chemical 378 (2013) 263–272
Table 3
Yields of oxidation products from cyclohexene oxidation by PhIO catalyzed by MnP systemsa.
e
Catalyst
Run
Conditionb
Product yields (%)c
Oxide
Ratiod
Conversion (10−3
)
Alcohol
Ketone
MnPf
1
2
3
4
5
6
7
8
A
B
C
A
B
C
B
B
A
B
C
A
B
C
A
B
C
A
B
C
39
44
31
24
46
24
38
36
12
17
21
8.5
14
20
8.1
8.9
14
12
10
17
6.1
3.0
5.4
4.5
3.0
8.1
3.2
2.6
6.0
4.2
10
5.0
1.7
5.0
9.0
3.0
7.4
8.5
5.5
8.0
19
3.3
39
26
7.0
48
4.6
5.9
37
8.8
42
30
3.6
21
66
40
48
55
26
32
1.56
6.98
0.70
0.77
4.6
0.43
4.87
4.23
0.27
1.3
0.40
0.24
2.6
0.77
0.11
0.21
0.25
0.19
0.32
0.42
5.46
9.73
2.23
4.15
10.5
2.24
8.70
8.31
3.65
5.12
2.08
2.85
3.50
1.42
5.01
6.38
1.81
4.80
5.70
1.64
MnPS
1st reuse
2nd reuse
MnPA
9
10
11
12
13
14
15
16
17
18
19
20
MnPB
PhIOg
Silica + PhIOh
a
Reactions carried out in duplicate. Reaction conditions: argon atmosphere, absence of light, magnetic stirring, room temperature, 60 min, DCM/ACN 1:1 (v/v) solvent
mixture. bConditions relate to the different catalyst/PhIO/cyclohexene molar ratios used in the experiments: A—1:10:1000; B—1:20:1000; C—1:20:5000.
c
Yields based on PhIO-oxide = cyclohexenoxide, alcohol = cyclohexen-1-ol, ketone = cyclohexen-1-one.
Ratio between epoxide yield and the sum of alcohol and ketone yields.
Conversion is defined by the total amount of substrate that is converted in products and is dimensionless. For conversion values determination, it was considered that
d
e
2 mols of PhIO are necessary to form 1 mol of ketone.
f
Homogeneous catalysis.
Homogeneous control reaction.
Heterogeneous control reaction.
g
h
Table 4
Yields of oxidation products from alkane oxidation by PhIO catalyzed by MnP systemsa.
Catalyst
Run
Cyclohexane oxidation product yield (%)b
Run
n-Heptane oxidation product yield (%)c
c-ol
21
12
6.3
5.0
–
–
c-one
Ratio c-ol/c-one
16
9.2
5.2
7.1
–
c-ol (total)
13
3.4
2.6
2.1
–
c-one (total)
TSI index
53.5
59.4
53.8
52.3
–
MnPd
MnPS
MnPA
MnPB
PhIOe
Silicaf
1
2
3
4
5
6
1.3
1.3
1.2
0.7
–
7
8
9
10
11
12
1.1
1.1
1.2
1.0
–
–
–
–
–
–
a
Reactions carried out in duplicate. Reaction conditions: argon atmosphere, absence of light, magnetic stirring, room temperature, 60 min, DCM/ACN 1:1 (v/v) solvent
mixture.
b
Yields based on PhIO–c-ol = cyclohexanol and c-one = cyclohexanone.
Yields based on PhIO–c-ol (total) = total heptanol yields percentage resulted from the alcohol at carbon position 1, 2, 3, of the linear alkane and c-one (total) = total
c
heptanone yields percentage resulted from the alcohol at carbon position 2 and 3, of the linear alkane. TS index = terminal selectivity index.
d
Homogeneous catalysis.
Homogeneous control reaction.
Heterogeneous control reaction.
e
f
Because the MnP is more soluble in ACN, we had expected that
increased ACN content in the solvent mixture would favor the cat-
alytic activity. However, the substrate and PhIO are poorly soluble
in ACN, which accounts for the lower product yields. As for DCM, the
substrate cyclohexane is highly soluble in this solvent, whereas the
MnP and PhIO have low solubility in this same solvent. In the end,
DCM/ACN 4:6 v/v was the solvent mixture that afforded the best
cyclohexanol yields, 26%, which is five percentage points above the
standard conditions used in this study (Table 4, DCM/ACN 1:1 v/v,
run 1) [97]. We did not detect any significant formation of ketone
in most of the solvent proportions used here (values below 1.4%).
As expected, the heterogeneous catalysts MnPS, MnPA, and
MnPB led to lower product yields than homogeneous MnP. Two fac-
tors operate in the case of MnPA and MnPB: the difficult interaction
between the MnP diluted in the solid matrix (Table 4, runs 3 and 4)
and the inertia of cyclohexane when compared with unsaturated
substrates [2]. Again, MnPS (run 2) was the best heterogeneous
catalyst for substrate oxidation, for the same reasons already dis-
cussed in items 3.3.1 and 3.3.2—the high catalyst density per mass
of solid catalyst as compared with the silica-based solids MnPA and
MnPB.
3.3.4. Oxidation of n-heptane
The structure of n-heptane, the most inert substrate investigated
in this paper, contains five secondary carbons, which are more labile
to oxidation, and two primary carbons (carbons at position 1 and
7), which are more inert to the oxidation reaction due to thermo-
dynamic factors and the low stability of the radical formed at these
positions [94]. Both thermodynamics and statistics are expected
to influence the distribution of the products. Yields of 3- and 2-
heptanol should be similar; the yield of 4-heptanol should be half
the yield achieved for 3- and 2-heptanol. There should not be signif-
icant formation of 1-heptanol, because of the higher inertia of the
C–H bonds of the primary carbons at positions 1 and 7 of n-heptane
[53].

G.K.B. Ferreira et al. / Journal of Molecular Catalysis A: Chemical 378 (2013) 263–272
271
As in the case of cyclohexane, all the catalysts used here were
selective for alcohol (Table 4, runs 7 to 10), mainly MnP in homoge-
nous system (run 7). The catalytic yields obtained with n-heptane
were low, confirming that this linear alkane is inert to oxidation
[2]. The homogeneous and heterogeneous control reactions (runs
11 and 12, respectively) did not afford any products, as expected
for the oxidation of a highly inert substrate.
the plane of the porphyrin ring. In the case of the MnP, this config-
uration restricts the approach of the substrate to the metal center,
explaining the higher TSI values [99,100].
4. Conclusion
In an attempt to solve problems related to homogeneous
catalysis mediated by metalloporphyrins, several strategies were
developed by various groups in the last three decades, such as
addition of electron-withdrawing groups at the meso-phenyl sub-
stituints, and heterogenization onto different inert matrices or
by self-assembly methodologies. Diverting chemical functional-
ization options, that would modify the intrinsic characteristic of
the hydroxyl–methoxyphenyl substituted porphyrin applied in
this work, heterogenization techniques were developed with suc-
cess. Immobilization of the manganeseporphyrin onto silica by the
sol–gel process afforded materials with different compositional,
textural and morphological characteristics and reactivities towards
oxidation of different hydrocarbons.
Self-assembly of manganese(III) porphyrins and Mn(II) centers
resulted in an insoluble interesting solid, that showed catalytic
activities similar to those obtained in homogeneous systems and
high selectivity for epoxidation of alkenes and hydroxylation of
alkanes. This material has the advantage of being recyclable and
robust, what goes with characteristic of desired new materials, such
as coordination polymers and metal-organic frameworks.
Interestingly, all the catalysts based on the MnP used in this
work were more selective for 1- and 2-heptanol, and not for 2-
and 3-heptanol, as expected if one considers the statistical and
thermodynamic factors [94]. To better illustrate this selectivity,
we introduced an index that correlates the linear alkane oxidation
at carbons 1 and 2, the terminal selectivity index (TSI), defined as
the ratio between the sums of the yields of 1- and 2-heptanol and
the total alcohol yield obtained in the oxidation of n-heptane. This
index is similar to the primary selectivity index set by Cook et al.
[98]; the difference is that the latter authors only considered oxida-
tion at the primary carbon, whilst TSI takes into account both highly
hindered and less hindered positions (Table 4). Heterogeneous
catalytic systems promote the terminal oxidation of linear hydro-
carbons when (i) the structure of the catalyst is able to stabilize the
less thermodynamically favored intermediate and (ii) the geometry
of the solid catalyst only allows for the approach of the terminal car-
bon of the substrate to the active catalytic species. Thomas et al. [53]
described that aluminophosphate molecular sieves (AlPOs) doped
with transition metals and bearing pores and channels can catalyze
the oxidation of alkanes in air. The latter system was highly selec-
tive for the terminal carbon of alkanes, because the channels in the
solid allowed the substrate to arrive at the catalytic species in an
end-on approach.
Porphyrin systems in homogeneous medium can also promote
the oxidation of alkanes at terminal positions. Cook et al. [98]
reported that the manganese(III) porphyrin [Mn(TTPPP)] was selec-
tive toward the primary alcohol when it was used as catalyst for
hexane oxidation in a homogenous system. In this case, the authors
attributed the selectivity for the primary carbon to steric factors
within the porphyrin itself, and not to a cage formed by a solid
matrix. The 2,4,6-triphenyl phenyl substituent at the meso pos-
itions of [Mn(TTPPP)] hindered the approach of the substrate by
creating a structure similar to the cavities found in solid supports.
As a result, only the terminal carbon of the linear alkane was able
to approach the metal center, to give 1-hexanol.
Acknowledgments
The authors acknowledge the financial support obtained from
CAPES (Coordenac¸ ão de Aperfeic¸ oamento de Pessoal de Nivel Supe-
rior), CNPq (Conselho Nacional de Desenvolvimento Científico e
Tecnológico), FINEP (Financiadora de Estudos e Projetos), and
Araucária Foundation. We are also thankful for the technical sup-
port of FUNPAR (Fundac¸ ão da Universidade Federal do Paraná) and
UFPR (Universidade Federal do Paraná).
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/
j.molcata.2013.06.022.
Using [Mn(TTMPP)] as catalyst, a porphyrin containing three
methoxy groups instead of the three phenyl substituents in
[Mn(TTPPP)], Nappa and Tollman [99] showed catalytic selectivity
for the carbon at position 2 of hexane, because the substituents in
[Mn(TTMPP)] were not as bulky as the substituents in [Mn(TTPPP)].
Therefore, thermodynamic factors operated in this case: because
carbon 2 was accessible, it was oxidized rather than the primary
carbon. We applied TSI for [Mn(TTMPP)] and found a value of 52.0%
against the 53.5% reached for MnP. Hence, extremely bulky groups
like the ones in Mn(TTPPP) or moderately bulky substituents like
the ones in Mn(TTMPP) are not necessary to achieve oxidation at
positions 1 or 2.
A TSI around 40% was expected for a thermodinamically cor-
rected statistical distribution of the products (contribution of 40%
from C2 and none from C1). We obtained TSI values higher than
40% for all the investigated catalysts, a value that was also reported
for [Mn(TTMPP)], a recognized manganese(III) porphyrin for the
selective oxidation of heptane. MnP has methoxy groups only at
one meta position of its phenyl rings, thereby posing lower steric
hindrance to the approach of the substrate as compared with
[Mn(TTMPP)]. In the case of porphyrins that can form atropisomers,
the statistical distribution that leads to the most stabilized spatial
configuration is the one that presents three of the four phenyl rings
pointing upward and one phenyl ring facing downward relative to
References
[1] B. Meunier, Bull. Soc. Chim. Fr. 4 (1986) 578–594.
[2] D. Mansuy, Pure Appl. Chem. 62 (1990) 741–746.
[3] D. Dolphin, T.G. Traylor, L.Y. Xie, Acc. Chem. Res. 30 (1997) 251–259.
[4] F. Bedioui, Coord. Chem. Rev. 144 (1995) 39–68.
[5] D.R. Leanord, J.R. Lindsay-Smith, J. Chem. Soc., Perkin Trans. 2 1 (1991) 25–30.
[6] B. Meunier, Chem. Rev. 92 (1992) 1411–1456.
[7] E.D. Sternberg, D. Dolphin, Tetrahedron 54 (1998) 4151–4202.
[8] M. Ethirajan, Y. Chen, P. Joshi, R.K. Pandey, Chem. Soc. Rev. 40 (2011) 340–362.
[9] M. Biesaga, K. Pyrzyn’ska, M. Trojanowicz, Talanta 51 (2000) 209–224.
[10] I. Beletskaya, V.S. Tyurin, A.Y. Tsivadze, R. Guilard, C. Stern, Chem. Rev. 109
(2009) 1659–1713.
[11] M.G. Walter, A.B. Rudine, C.C. Wamser, J. Porphyrins Phthalocyanines 14
(2010) 759–792.
[12] M. Costas, Coord. Chem. Rev. 255 (2011) 2912–2932.
[13] F. Wypych, G.A. Bubniak, M. Halma, S. Nakagaki, J. Colloid Interface Sci. 264
(2003) 203–207.
[14] K.S. Suslick, P. Bhyrappa, J.H. Chou, M.E. Kosal, S. Nakagaki, D.W. Smithenry,
S.R. Wilson, Acc. Chem. Res. 38 (2005) 283–291.
[15] J.Y. Lee, O.K. Farha, J. Roberts, K.A. Scheidt, S.B.T. Nguyen, J.T. Hupp, Chem. Soc.
Rev. 38 (2009) 1450–1459.
[16] O.K. Farha, J.T. Hupp, Acc. Chem. Res. 43 (2010) 1166–1175.
[17] G.S. Machado, G.C.G. Arizaga, F. Wypych, S. Nakagaki, J. Catal. 274 (2010)
130–141.
[18] A.J. Appleton, S. Evans, J.R. Lindsay-Smith, J. Chem. Soc., Perkin Trans. 2 (1995)
281–285.
[19] J.R. Lindsay-Smith, P.R. Sleath, J. Chem. Soc., Perkin Trans. 2 (1982) 1009–1015.
[20] J.T. Groves, T.E. Nemo, J. Am. Chem. Soc. 105 (1983) 6243–6248.

272
G.K.B. Ferreira et al. / Journal of Molecular Catalysis A: Chemical 378 (2013) 263–272
[21] P.C. Cooke, J.R. Lindsay-Smith, J. Chem. Soc., Perkin Trans. 1 (1994) 1913–1923.
[22] F. Doro, J.R. Lindsay Smith, A.G. Ferreira, M.D. Assis, J. Mol. Catal. A: Chem. 164
[62] M.T. Caudle, C.K. Mobley, L.M. Bafaro, R. Lobrutto, G.T. Yee, T.L. Groy, Inorg.
Chem. 43 (2004) 506–514.
(2000) 97–108.
[63] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination Com-
pounds parts A and B, fifth ed., Wiley Interscience Publication, Hoboken, NJ,
2000.
[23] A.A. Guedes, J.R. Lindsay Smith, O.R. Nascimento, D.F.C. Guedes, M.D. Assis, J.
Braz. Chem. Soc. 16 (2005) 835–843.
[24] S. Evans, J.R. Lindsay Smith, J. Chem. Soc., Perkin Trans. 2 (2001) 174–180.
[25] O. Leal, D.L. Anderson, R.G. Bowman, F. Basolo, R.L. Burwell Jr., J. Am. Chem.
Soc. 97 (1975) 5125–5129.
[64] G.B. Deacon, R.J. Phillips, Coord. Chem. Rev. 33 (1980) 227–250.
[65] E-Y. Choi, P.M. Barron, R.W. Novotny, H-T. Son, C. Hu, W. Choe, Inorg. Chem.
48 (2009) 426–428.
[26] P. Battioni, J.-P. Lallier, L. Barloy, D. Mansuy, J. Chem. Soc., Chem. Commun.
(1989) 1149–1151.
[66] A.M. Shultz, O.K. Farha, J.T. Hupp, S.T. Nguyen, J. Am. Chem. Soc. 131 (2009)
4204–4205.
[27] P. Battioni, J.F. Bartoli, D. Mansuy, Y.S. Byun, T.G. Traylor, J. Chem. Soc., Chem.
Commun. (1992) 1052–1053.
[28] D. Mansuy, Coord. Chem. Rev. 125 (1993) 129–141.
[29] S. Nakagaki, G.K.B. Ferreira, G.M. Ucoski, K.A.D.F. Castro, Molecules 18 (2013)
7279–7308.
[30] T. Matsui, M. Iwasaki, R. Sugiyama, M. Unno, M. Ikeda-Saito, Inorg. Chem. 49
(2010) 3602–3609.
[31] S. Nakagaki, A.R. Ramos, F.L. Benedito, P.G. Peralta-Zamora, A.J.G. Zarbin, J.
Mol. Catal. A: Chem. 185 (2002) 203–210.
[32] A.M. Machado, F. Wypych, S.M. Drechsel, S. Nakagaki, J. Colloid Interface Sci.
254 (2002) 158–164.
[33] M. Halma, F. Wypych, S.M. Drechsel, S. Nakagaki, J. Porphyrins Phthalocya-
nines 6 (2002) 502–513.
[34] S. Nakagaki, M. Halma, A. Bail, G.G.C. Arízaga, F. Wypych, J. Colloid Interface
Sci. 281 (2004) 417–423.
[35] G.S. Machado, F. Wypych, S. Nakagaki, Appl. Catal., A 413–414 (2012) 94–102.
[36] S. Nakagaki, K.A.D.F. Castro, G.S. Machado, M. Halma, S.M. Drechsel, F.
Wypych, J. Braz. Chem. Soc. 17 (2006) 1672–1678.
[37] S. Nakagaki, A.S. Mangrich, F. Wypych, Inorg. Chim. Acta 254 (1997) 213–217.
[38] A.T. Papacıdero, L.A. Rocha, B.L. Caetano, E. Molina, H.C. Sacco, E.J. Nassar, Y.
Martinelli, C. Mello, S. Nakagaki, K.J. Ciuffi, Colloids Surf., A 275 (2006) 27–35.
[39] A.M. Shultz, O.K. Farha, J.T. Hupp, S.B.T. Nguyen, J. Am. Chem. Soc. 131 (2009)
4204–4205.
[40] A. Dhakshinamoorthy, M. Alvaro, Y.K. Hwang, Y.K. Seo, A. Corma, H. Garcia,
Dalton Trans. 40 (2011) 10719–10724.
[41] D.J. Tranchemontagne, J.L. Mendoza-Cortes, M. O’Keeffe, O.M. Yaghi, Chem.
Soc. Rev. 38 (2009) 1257–1283.
[42] S. Motoyama, R. Makiura, O. Sakata, H. Kitagawa, J. Am. Chem. Soc. 133 (2011)
5640–5643.
[43] P. Bhyrappa, K.S. Suslick, J. Am. Chem. Soc. 119 (1997) 8492–8502.
[44] M.E. Kosal, K.S. Suslick, J. Solid State Chem. 152 (2000) 87–98.
[45] J. Chou, M.E. Kosal, H.S. Nalwa, N.A. Rakow, K.S. Suslick, The Porphyrin Hand-
book, Academic Press, New York, 2000.
[67] S. Roy, C.B. George, M.A. Ratner, J. Chem. Phys. C 116 (2012) 23494–23502.
[68] K.A. Campbell, M.R. Lashley, J.K. Wyatt, M.H. Nantz, R.D. Britt, J. Am. Chem.
Soc. 123 (2001) 5710–5719.
[69] M.A. Vázquez-Fernández, M.R. Bermejo, M.I. Fernández-García, G. González-
Riopedre, M.J. Rodríguez-Doutón, M. Maneiro, J. Inorg. Biochem. 105 (2011)
1538–1547.
[70] M. Halma, K.A.D.F. Castro, V. Prévot, C. Forano, F. Wypych, S. Nakagaki, J. Mol.
Catal. A: Chem. 310 (2009) 42–50.
[71] R. Makiura, S. Motoyama, Y. Umemura, H. Yamanaka, O. Sakata, H. Kitagawa,
Nat. Mater. 9 (2010) 565–571.
[72] T. Ohmura, A. Usuki, K. Fukumori, T. Ohta, M. Ito, K. Tatsumi, Inorg. Chem. 45
(2006) 7988–7990.
[73] D.W. Smithenry, S.R. Wilson, K.S. Suslick, Inorg. Chem. 42 (2003) 7719–7721.
[74] G.R. Friedermann, A.S. Mangrich, M. Halma, K.A.D.F. Castro, F.L. Benedito,
F.G. Doro, S.M. Drechsel, M.D. Assis, S. Nakagaki, Appl. Catal., A 308 (2006)
172–181.
[75] C.J. Brinker, G.W. Scherer, Sol–Gel Science: The Physics and Chemistry of
Sol–Gel Processing, first ed., Academic Press, San Diego, 1990.
[76] A.M. Buckley, M. Greenblatt, J. Chem. Edu. 71 (1994) 599–602.
[77] L.L. Hench, J.K. West, Chem. Rev. 90 (1990) 33–72.
[78] G.S. Machado, K.A.D.F. Castro, O.J. Lima, E.J. Nassar, K.J. Ciuffi, S. Nakagaki,
Colloids Surf., A 349 (2009) 162–169.
[79] K.S.W. Sing, D.H. Everett, R.A.W. Haul, L. Moscou, R.A. Pierotti, J. Rouquerol, T.
Siemieniewska, Pure Appl. Chem. 57 (1985) 603–619.
[80] A. Grosman, C. Ortega, Langmuir 21 (2005) 10515–10521.
[81] J.T. Groves, T.E. Nemo, R.S. Myers, J. Am. Chem. Soc. 101 (1979) 1032–1033.
[82] J.T. Groves, W.J. Kruper, R.C. Haushalter, J. Am. Chem. Soc. 102 (1980)
6375–6377.
[83] S. Nakagaki, G.S. Machado, M. Halma, A.A.S. Marangon, K.A.D.F. Castro, N.
Mattoso, F. Wypych, J. Catal. 242 (2006) 110–117.
[84] T. Tatsumi, M. Nakamura, H. Tominaga, Catal. Today 6 (1989) 163–170.
[85] M. Moghadam, V. Mirkhani, S. Tangestaninejad, I. Mohammdpoor-Baltork, H.
Kargar, J. Mol. Catal. A: Chem. 288 (2008) 116–124.
[46] D.W. Smithenhy, S.R. Wilson, K.S. Suslick, Inorg. Chem. 42 (2003) 7719–7721.
[47] W. Stöber, A. Fink, E. Bohn, J. Colloid Interface Sci. 26 (1998) 62–69.
[48] X. Liang, J. Xiao, B. Chen, Y. Li, Inorg. Chem. 49 (2010) 8188–8190.
[49] K.A.D.F. Castro, M. Halma, G.S. Machado, G.P. Ricci, G.M. Ucoski, K.J. Ciuffi, S.
Nakagaki, J. Braz. Chem. Soc. 21 (2010) 1329–1340.
[86] J.T. Groves, D.V. Adhyam, J. Am. Chem. Soc. 106 (1984) 2177–2181.
[87] E.C. Zampronio, M.C.A.F. Gotardo, M.D. Assis, H.P. Oliveira, Catal. Lett. 104
(2005) 53–56.
[88] T. Oyama, Mechanisms in Homogeneous and Heterogeneous Epoxidation
Catalysis, first ed., Elsevier, Amsterdam, 2008.
[50] G.M. Ucoski, K.A.D.F. Castro, K.J. Ciuffi, G.P. Ricci, J.A. Marques, F.S. Nunes, Appl.
Catal., A 404 (2011) 120–128.
[51] S. Farhadi, Saeid, A. Zabardasti, M.H. Rahmati, J.Chem. Res. 3 (2011) 157–160.
[52] J.H. Cai, J.W. Huang, P. Zhao, Y.J. Ye, H.C. Yu, L.N. Ji, J. Sol–Gel Sci. Technol. 50
(2009) 430–436.
[53] M.J. Thomas, R. Raja, G. Sankar, R.G. Bell, Acc. Chem. Res. 34 (2001) 191–200.
[54] A.D. Adler, F.R. Longo, J.D. Finarelli, J. Goldmacher, J. Assour, L. Koakoff, J. Org.
Synth. 32 (1967) 476.
[89] J.T. Groves, D.V. Subramanian, J. Am. Chem. Soc. 106 (1984) 2177–2181.
[90] J.T. Groves, M.K. Stern, J. Am. Chem. Soc. 110 (1980) 8628–8633.
[91] T.G. Traylor, A.R. Miksztal, J. Am. Chem. Soc. 111 (1989) 7443–7448.
[92] M.J. Nappa, R.J. McKinney, Inorg. Chem. 27 (1988) 3740–3745.
[93] J.R. Lindsay-Smith, Y. Iamamoto, F.S. Vinhado, J. Mol. Catal. A: Chem. 252
(2006) 23–30.
[94] J.M. Thomas, J.C. Hernandez-Garrido, R. Raja, R.G. Bell, Phys. Chem. Chem.
Phys. 11 (2009) 2799–2825.
[55] J.L. Nian, L. Min, H.A. Kong, Inorg. Chim. Acta 178 (1990) 59–65.
[56] E. Baciocchi, F. d’Acunzo, C. Galli, M. Loele, J. Chem. Soc., Chem. Commun.
(1995) 429–430.
[57] H. Saltzman, J.G. Sharefkin, Org. Synth. 5 (1963) 60–61.
[58] H. Kobayashi, T. Higuchi, Y. Kaizu, H. Osada, M. Aoki, Bull. Chem. Soc. Jpn. 48
(1975) 3137–3141.
[59] W. Zheng, N. Shan, L. Yu, X. Wang, Dyes Pigm. 77 (2008) 153–157.
[60] J.B. Allison, R.S. Becker, J. Chem. Phys. 32 (1960) 1410–1417.
[61] L.J. Boucher, J. Am. Chem. Soc. 92 (1970) 2725–2730.
[95] S. Nakagaki, C.R. Xavier, A.J. Wosniak, A.S. Mangrich, F. Wypych, M.P. Cantão,
I. Denicoló, L.T. Kubota, Colloids Surf., A 168 (2000) 261–276.
[96] Y. Iamamoto, Y.M. Idemori, S. Nakagaki, J. Mol. Catal. A: Chem. 99 (1995)
187–193.
[97] Y. Iamamoto, M.D. Assis, K.J. Ciuffi, H.C. Sacco, L. Iwamoto, A.J.B. Melo, O.R.
Nascimento, C.M.C. Prado, J. Mol. Catal. A: Chem. 109 (1996) 189–200.
[98] B.R. Cook, T.J. Reinert, K.S. Suslick, J. Am. Chem. Soc. 108 (1986) 7281–7286.
[99] M.J. Nappa, C.A. Tolman, Inorg. Chem. 24 (1985) 4711–4719.
[100] L.K. Gottwald, E.F. Ullman, Tetrahedron Lett. 36 (1969) 3071–3074.