Use of iron and manganese porphyrins in solution and immobilized on silica obtained by the sol-gel process as catalyst in the oxidation of organic substrates

Article abstract of DOI:10.1016/j.apcata.2011.07.019

The preparation and characterization of the metalloporphyrins [Fe(TAPP)]Cl, 5,10,15,20-tetrakis(acetalphenyl porphyrin) iron(III) chloride, and [Mn(TAPP)]Ac, 5,10,15,20-tetrakis (acetalphenyl porphyrin) manganese(III) acetate and of the solids obtained by

Full text of DOI:10.1016/j.apcata.2011.07.019

Applied Catalysis A: General 404 (2011) 120–128
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Use of iron and manganese porphyrins in solution and immobilized on silica
obtained by the sol–gel process as catalyst in the oxidation of organic substrates
Geani Maria Ucoskia, Kelly Aparecida Dias de Freitas Castroa, Kátia Jorge Ciuffib,
Gustavo Pimenta Riccib, Jacqueline Aparecida Marquesc, Fábio Souza Nunesa,
Shirley Nakagakia,∗
a Universidade Federal do Paraná, Departamento de Química- Laboratório de Bioinorgânica e Catálise CP 19081, CEP 81531-990, Curitiba, PR, Brazil
b Universidade de Franca, Departamento de Química, CP 82, CEP 14404-600, Franca, SP, Brazil
c Universidade Estadual de Ponta Grossa, Departamento de Química, CEP 84030-900, 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:
Received 4 April 2011
Received in revised form 12 July 2011
Accepted 15 July 2011
Available online 23 July 2011
The preparation and characterization of the metalloporphyrins [Fe(TAPP)]Cl, 5,10,15,20-
tetrakis(acetalphenyl porphyrin) iron(III) chloride, and [Mn(TAPP)]Ac, 5,10,15,20-tetrakis (acetalphenyl
porphyrin) manganese(III) acetate and of the solids obtained by their heterogenization in silica obtained
by the sol–gel process in acidic and basic media are reported. The compounds were characterized by
ultraviolet–visible absorption spectroscopy (UV–vis), X-ray diffraction (XRD), infrared spectroscopy
(FTIR), electron paramagnetic resonance (EPR), and thermogravimetric analysis (TGA). The catalytic
activity of the metalloporphyrins in solution and as heterogeneous catalysts concerning the oxidation
of organic substrates like cyclooctene and cyclohexane was investigated. The catalysts employed here,
which consist of reasonably low-cost metalloporphyrins with intermediate structure between first- and
second-generation porphyrins, furnished promising results. These novel solid materials offered benefits
such as higher product yields, compared to the first-generation catalysts, and inexpensive preparation
routes, compared to the cost involved in the production of second-generation metalloporphyrins.
© 2011 Elsevier B.V. All rights reserved.
Keywords:
Porphyrin
Silica
Sol–gel
Catalysis
Oxidation
Alkane activation
1. Introduction
dimerization, may also occur. These problems often lead to low cat-
alytic yields and make these homogeneous catalysts economically
The attainment of new efficient and selective catalysts for the
oxidation of organic substrates has aroused great scientific and
industrial interest. This has motivated the search for catalysts that
can be re-used, thereby making catalytic processes economically
viable and also minimizing possible environmental impacts caused
by the generation of residues.
Synthetic metalloporphyrins (MPs), especially the iron(III)
and manganese(III) complexes, are able to mimic the action of
cytochrome P-450 and have proven to be efficient and selective
catalysts for the oxidation of organic substrates [1–3]. Although
homogeneous systems based on MPs are effective for hydrocar-
bon hydroxylation and epoxidation, depending on their structure
synthetic MPs can undergo fast oxidative degradation under the
oxidation reaction conditions [1–8]. Moreover, the use of these
catalysts in homogeneous medium is limited because it is impos-
sible to recover and reutilize the MP at the end of the first reaction,
not to mention that a side reaction involving the catalyst, such as
unviable [9,10]. To minimize such difficulties, much effort has been
devoted to the immobilization of these complexes on rigid, inert,
and stable inorganic supports, so that more stable and selective
catalysts amenable to recovery and reuse can be achieved through
interaction between the MP and the support [6].
Among the supports that can be investigated for MP immobi-
lization, silica is often exploited because of its low cost, inertia, and
high functionalization capacity. Silica is an inorganic polymer con-
sisting of siloxane groups (Si–O–Si) in its interior and silanol groups
(Si–OH) on its surface [10–12]. The use of the sol–gel process for
silica preparation results in a highly pure solid under mild pressure
and temperature conditions [13]. For this reason, this methodol-
ogy has been frequently applied for synthesis of this support for
catalyst immobilization. The silica obtained by the sol–gel route is
a rigid support that allows for stable catalyst immobilization on
its surface, which inhibits catalyst dimerization as well as other
undesirable interactions [10,14,15].
In this work, we report the immobilization of the MPs
[Fe(TAPP)]Cl; 5,10,15,20-tetrakis(acetalphenyl porphyrin) iron(III)
chloride, and [Mn(TAPP)]Ac; 5,10,15,20-tetrakis (acetalphenyl por-
phyrin) manganese(III) acetate (Fig. 1), on silica obtained by the
∗
Corresponding author. Fax: +55 41 33613186.
E-mail address: shirleyn@ufpr.br (S. Nakagaki).
0926-860X/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.apcata.2011.07.019

G.M. Ucoski et al. / Applied Catalysis A: General 404 (2011) 120–128
121
troscopies. Because the FeP underwent dimerization during the
metallation process, it was treated with HCl [21]. To this end, HCl
(0.1 mol L−1) was slowly added to a MeOH solution containing a
known concentration of the FeP, until the electronic spectrum dis-
playing the characteristic bands of the FeP monomer was achieved.
The monomerized solid was exhaustively washed with water until
neutral pH, dried in a rotary evaporator, and characterized by
UV–vis, FTIR, and EPR spectroscopies. [Mn(TAPP)]Ac: UV–vis in
DCM, ꢀmax, nm (log ε): 474 (4.25), 578 (3.25), 618 (3.28) and
UV–vis in CH3OH 470 (4.23) 570 (3.30) 604 (3.23) (supplementary
information S2, Fig. S3); [Fe(TAPP)]Cl: UV–vis in DCM, ꢀmax, nm
(log ε): 422 (5.04) 508 (4.00) 690 (3.30) and UV–vis in CH3OH 420
(4.98) 532 (3.95) 682 (3.18) (supplementary information S2, Fig.
S4).
Fig. 1. Schematic representation of the metalloporphyrins used in this study:
[Mn(TAPP)]Ac = 5,10,15,20-tetrakis (acetalphenyl porphyrin) manganese(III)
acetate; [Fe(TAPP)]Cl = 5,10, 15, 20-tetrakis (acetalphenyl porphyrin) iron(III)
chloride.
hydrolytic sol–gel process conducted in either acidic or basic media
[16]. The catalytic activity of [Fe(TAPP)]Cl and [Mn(TAPP)]Ac in
cyclooctene and cyclohexane oxidation reactions using iodosyl-
benzene as oxygen donor was investigated in both homogeneous
solutions and heterogeneous medium. Although the free base por-
phyrin and its respective iron(III) and manganese(III) MPs had
already been previously synthesized and characterized by Nian
et al. [17], there are no literature reports on their immobilization or
on their catalytic activity in both homogeneous and heterogeneous
media.
2.4. Heterogenization of the MnP and the FeP on silica using the
sol–gel process
2.4.1. Basic catalysis – (SGB)
The silica solids containing the MP were prepared through
tetraethyl orthosilicate (TEOS) hydrolysis in isopropanol, employ-
ing NH3 as catalyst, in the presence of 1.10 × 10−5 mol MnP or FeP,
using H2O/Si (TEOS) = 50, NH3/ethanol = 10, and H2O/NH3 = 200
molar ratios [16]. The reaction mixture was kept at 40 ◦C, and the
solution was stirred for 30 min and kept at about 50 ◦C for com-
plete drying. After aging for about 30 days, the resulting solid was
extensively washed with dichloromethane and methanol in a Soxh-
let apparatus and dried at 70 ◦C. The MP loading in each solid was
quantified by measuring the amount of unloaded MP in the com-
bined reaction solvent and washings by UV–vis spectroscopy. The
pure silica matrix was also prepared in the absence of the MP, fur-
nishing a transparent colorless solid that was used in the catalytic
control reactions. The resulting solid products are herein referred
as SGBMn (green solid) and SGBFe (green solid) for the MnP and
FeP immobilized on silica, respectively. The solids were charac-
terized by UV–vis, FTIR, and EPR spectroscopies; X-ray diffraction;
thermogravimetric analysis; and physisorption analysis.
2. Experimental
2.1. Materials
All the compounds employed in this study were commercially
available with grade purity from Aldrich, Merck, or Fluka. Pyrrole
was distilled before use. Iodosylbenzene (PhIO) was prepared by
hydrolysis of iodosylbenzene diacetate following the method of
Sharefkin and Saltzman [18] and the purity was shown to be 98%
by iodometric titration.
2.2. Synthesis of the free base porphyrin [H2(TAPP)]
The free base porphyrin [H2(TAPP)] was synthesized and charac-
terizedas previouslydescribedin theliterature[17,19]. Theproduct
was characterized by UV–vis spectroscopy (UV–vis), infrared spec-
troscopy (FTIR), and 1H and 13C NMR (proton and carbon nuclear
magnetic resonance, respectively). UV–vis in DMF, ꢀmax, nm
(log ε): 422 (5.45), 516 (4.11), 554 (3.89), 592 (3.63), 650 (3.66)
(supplementary information, Fig. S1). FTIR data cm−1 (attribution).
3312 (N–H axial deformation) 2886 (C–H pyrrole axial deforma-
tion) 1602 (N–H pyrrole symmetric angular deformation) 1560,
1503, 1485 and 1470 (C C axial deformation) 1437 (C–H defor-
mation) 1353 (C–N amine axial deformation) 1237 (asymmetric
axial deformation due to the C–O–C acetal group at the porphyrin
ring meso position) 1102 (C–H deformation in ring plane) 1038
(symmetric axial deformation due to the C–O–C acetal group at
the porphyrin ring meso position) 936 (C–H pyrrole deformation)
797 and 729 (C–H out of plane angular deformation) and 668
(C–H angular deformation in ring plane). The 1H and 13C NMR
(proton and 13carbon nuclear magnetic resonance is presented in
supplementary information S1, Fig. S2).
2.4.2. Acid catalysis – SGA
The catalysts were prepared by the sol–gel process through
TEOS hydrolysis in tetrahydrofuran (THF), employing HCl as cat-
alyst, 1.10 × 10−5 mol MnP or FeP, and the following molar ratios:
H2O/Si (TEOS) = 4, THF/HCl = 308, and H2O/HCl = 92 [22]. The reac-
tion mixture was stirred for 1 h, at room temperature. After that,
the solids named SGAMnS (green solid) and SGAFeS (green solid),
resulting from MnP and FeP immobilization on silica (10 mg),
respectively, were obtained after complete dryness at about 50 ◦C
for 15 days. Two more solids were also prepared but dried in an
oven at about 50 ◦C for 1 day, namely SGAMnF (green solid) and
SGAFeF (reddish brown solid). All the four solids were extensively
washed with DCM and MeOH by Soxhlet procedure, followed by
drying at 70 ◦C. The MP loading was obtained by using UV–vis spec-
troscopy to measure the amount of MP in the combined washings.
Again, the pure silica matrix was also prepared in conditions simi-
lar to those described above, but in the absence of MP, furnishing a
transparent colorless solid that was also used in the catalytic con-
trol reactions. The obtained solids were characterized by the same
techniques employed for characterization of the catalysts prepared
under basic conditions.
2.3. Synthesis of the iron(III) and manganese(III) porphyrin
complexes
2.5. Homogeneous cyclooctene and cyclohexane oxidation by
PhIO catalyzed by the MnP and the FeP
The MPs were synthesized following the methodology by
described Adler and Longo [17,20], by metal(III) insertion into the
free-base porphyrin [H2(TAPP)] using ferrous chloride tetrahydrate
or manganese acetate tetrahydrate in dimethylformamide (DMF).
The products were characterized by UV–vis, FTIR, and EPR spec-
Cyclooctene and cyclohexane oxidation by PhIO catalyzed
by the prepared MPs were carried out in a 2 mL vial under
magnetic stirring, inside a dark chamber. The homogeneous

122
G.M. Ucoski et al. / Applied Catalysis A: General 404 (2011) 120–128
catalyst ([Mn(TAPP)]Ac or [Fe(TAPP)]Cl), PhIO, and the sub-
strate (cyclooctene or cyclohexane) (at 1:10:1000, 1:50:1000,
1:100:1000, 1:10:5000, or 1:50:5000 MP/PhIO/substrate ratio)
were mixed in the solvent mixture (dichloromethane/acetonitrile
(DCM/ACN) 1:1, v/v) under argon atmosphere, at room temper-
ature. The reaction mixture was stirred at room temperature for
1 h, in the absence of light. A sodium sulfite saturated solution
was added to the reaction, in order to eliminate excess PhIO and
to quench the reactions after the desired experiment time was
reached. The products were extracted from the reaction solu-
tion, transferred to a 2 mL volumetric flask, and analyzed by gas
chromatography using n-octanol (9.6 × 10−3 mol L−1) as internal
standard. The yields are based on PhIO (mol of product/mol of
oxidant, %). An investigation of the influence of reaction time on
product yield was also carried out using cyclohexane as substrate
in a 1:10:5000 MP/PhIO/substrate molar ratio, in order to deter-
mine the optimal conditions. Reactions times of 15, 30, 45, 60, 120,
360, 720, 1440, and 2880 min were investigated. Control reactions
(PhIO and substrate with no added MP) were performed using the
same procedure.
tion medium at 1:x:1000 and 1:x:5000 catalyst/oxidant/substrate
molar ratios, where x means a variable amount of oxidant and x = 1,
5, 10, 15, 20, 25, 30, 35, 40, 45, 50, or 100.
2.8. Characterization and apparatus
Qualitative and quantitative electronic spectra of solution sam-
ples were recorded on a Hewlett Packard-8452A-Diode Array
spectrophotometer, in the 200–900 nm range, using a 1 cm quartz
cuvette. Qualitative spectra of the solid samples were obtained by
diffuse reflectance on a VARIAN Cary 100 spectrophotometer.
FTIR spectra were registered on a Biorad 3500 GX spectropho-
tometer in the 400–4000 cm−1 range, using KBr pellets. KBr was
crushed with a small amount of the solid sample to be analyzed,
and the spectra were collected with a resolution of 4 cm−1 and
accumulation of 32 scans.
X-ray diffraction measurements of self-oriented film samples
were made placing the samples on neutral glass sample holders.
The measurements were performed in the reflection mode using a
Shimadzu XRD-6000 diffractometer operating at 40 k◦V and 40 mA
˚
(Cu-K␣ radiation ꢀ = 1.5418 A), with a dwell time of 1 /min.
Electron paramagnetic resonance (EPR) measurements of the
powder samples were carried out on an EPR BRUKER ESP 300E spec-
trometer (standard concavity: 4102-SP, frequency X band 9.5 GHz),
at 293 or 77 K, using liquid N2.
Thermogravimetric (TGA) measurements were performed on a
Mettler Toledo TGA/SDTA851, in nitrogen, at a 150 ␮L/min flow,
with a heating rate of 10 ◦C/min, from 30 to 1000 ◦C.
Products from the catalytic oxidation reactions were identified
using a Shimadzu CG-14B gas chromatograph (flame ionization
detector) equipped with a DB-WAX capillary column (J&W Scien-
tific).
Specific surface areas were determined by applying the BET
method to the corresponding nitrogen adsorption isotherms,
obtained by using a physical adsorption analyzer Micrometrics
ASAP 2020. The samples had been previously degassed for 1 h at
room temperature at a pressure lower than 50 mmHg.
2.6. Heterogeneous cyclooctene and cyclohexane oxidation by
PhIO catalyzed by the immobilized MnP and FeP
The solids prepared by MP immobilization on silica using the
sol–gel process in basic (SGB) or acidic media (SGA), namely
SGBMn, SGBFe, SGAMnS, SGAMnF, SGAFeS, and SGAFeF, were
investigated as heterogeneous catalysts in cyclooctene and cyclo-
hexane oxidation reactions by PhIO, employing a procedure similar
to that used in the case of homogeneous catalysis, using a 1:10:5000
catalyst/PhIO/substrate molar ratio. At the end of the reaction,
the products were extracted from the reaction suspension and
transferred to a volumetric flask. The catalyst solid was washed sev-
eral times with small amounts of the solvent mixture (ACN/DCM
1:1, v/v), followed by centrifugation. The extracted solvent was
transferred to the same volumetric flask. The products in the
extracted solution were analyzed by gas chromatography using
n-octanol (9.63 × 10−3 mol L−1) as internal standard. Control reac-
tions were carried out using this same procedure, and were
performed as follows: (a) substrate only, (b) substrate + PhIO, and
(c) substrate + PhIO + silicas (supports without MP).
3. Results and discussion
3.1. Synthesis and characterization of free base porphyrin
All heterogeneous catalysts were exhaustively washed and
dried for reuse in further reactions using the same procedure
described above.
The solvent recovered from the reaction media and the solvent
from the washing procedure were monitored by UV–vis, in order
to verify whether any MP was leached form the support during the
catalytic reaction and/or washing procedure.
[H2(TAPP)] and the complexes ([Mn(TAPP)] and [Fe(TAPP)])
UV–vis and FTIR techniques besides of the 1H NMR and EPR were
used to confirm the synthesis of the free base porphyrin [H2(TAPP)]
and metalloporphyrins (supplementary information S1 and S2).
3.2. Characterization of the silica-MP solids
2.7. UV–vis determination of the ideal quantity of oxidant to be
employed in the catalytic reactions
Silica matrix formation by the sol–gel process occurs through
alkoxysilane polymerization by hydrolysis, followed by TEOS con-
densation in alcohol and water. This synthesis can be catalyzed by
ammonium hydroxide solution (basic route) or hydrochloric acid
(acid route). MP addition to the reaction mixture during the process
generates solids that contain the target immobilized complex.
For the acid route, slow and fast drying of the silica obtained
by the sol–gel process was carried out, in order to verify whether
this variable influenced the MP immobilization rate on the silica
support. As for the basic route, only slow drying was accomplished,
since it is well known that silica synthesis occurs through formation
of spheres in basic medium, and that accelerated drying hinders
their formation [13,16,23–25].
A spectrophotometric study was conducted with the FeP and
MnP solutions, in order to determine the ideal quantity of PhIO to be
used in the homogeneous catalysis reactions. In a typical procedure,
an MP solution (5.5 × 10−5 mol L−1) in MeOH was prepared and
analyzed by UV–vis. PhIO was added in small increments, in order
to keep a controlled MP/PhIO molar ratio (1:1, 1:5, 1:10, or 1:100).
After each PhIO addition, the reaction solution was magnetically
stirred for 10 min and centrifuged for 2 min, to remove any traces
of insoluble particles, and a new UV–vis spectrum was recorded.
Reactions were carried out with both the MnP and the FeP, in a sim-
ilar way. In order to study the consumption of the catalytic species
generated by interaction between the MP and PhIO and to check
whether the porphyrin ring becomes more resistant to oxidation in
the presence of the substrate, cyclohexane was added to the reac-
Table 1 summarizes the MP loading on the various silicas pre-
pared in this work. The MP loading in the materials achieved
via MnP and FeP immobilization on silica in acidic medium fol-
lowed by the fast drying process, namely SGAFeF and SGAMnF, was

G.M. Ucoski et al. / Applied Catalysis A: General 404 (2011) 120–128
123
Table 1
MP immobilization on silica obtained by the sol–gel process and decomposed
organic matter content in TGA.
Solid
Immobilization
percentage (%)a
Decomposed
organic matter
Content (%) in
TGA
Loading (MP
concentration in
the silica) (mol g−1
% m/m
)
SGBMn
SGBFe
SGAMnS
SGAMnF
SGAFeS
SGAFeF
44
40
48
92
42
89
1.71
1.12
1.92
4.44
2.01
3.22
9.24 × 10−6
8.62 × 10−6
5.66 × 10−6
1.13 × 10−5
5.10 × 10−6
1.16 × 10−5
0.83
0.76
0.51
1.02
0.45
1.02
a
MP immobilized in the silica considering the employed initial mass.
higher compared with the MP loading in the materials obtained via
MP incorporation in acidic medium and submitted to slow dry-
ing (SGAFeS and SGAMnS) and with MP loadings detected in the
materials synthesized in basic medium (SGBMn and SGBFe). Con-
cerning the support SGA, the drying process probably influenced
the immobilization rate. This is because solvent molecules possibly
remained entrapped inside the cavity of the SGA solids that were
dried slowly (SGAFeS and SGAMnS), which may have led to compe-
tition between solvent and MP for the support. With respect to the
support preparation medium, the silica network formed via acid
catalysis must have facilitated MP incorporation since it enables
MP adhesion to the SGA surface as well as its entrapment inside
the silica network. In contrast, materials obtained by basic catalysis
consist of interconnected silica spherical particles, or monodisperse
spheres, which hinder MP entry into the SGB structure, thereby cul-
minating in lower MP loading [26]. So an MP can adhere to or entrap
into an SGB sphere only if it is hollow.
A parallel experiment which involved interaction between pure
silica gel obtained in the absence of MP using the sol–gel pro-
cess in acidic medium (SGA) with the MnP was conducted under
the same conditions adopted for preparation of the solid SGAMnS.
An MP loading of 1.26 × 10−6 mol g−1 and 22% immobilization was
achieved for this solid, designated SGA–MnP, which suggests that
MP addition after silica network formation resulted in MP surface
adsorption only. This was confirmed by the fact that SGA–MnP
washing removed much of the MP complex. Taken together, these
results demonstrate that MP addition during silica formation leads
to stronger MP immobilization on the surface and also inside the
silica network compared to MP addition after the silica network is
formed.
Fig. 2. EPR spectra of solid samples, obtained at 77 K (a) MnP, (b) SGAMnS, (c)
SGAMnF, (d) SGBMn, (e) FeP, (f) SGAFeF, (g) SGAFeS, and (h) SGBFe.
trum of the free FeP EPR (Fig. 2e), indicates that some FeP molecules
could undergo rhombic distortion after the immobilization process,
because of their interaction with the support surface. However, the
axial symmetry signal at g = 6.0 is still highly intense, showing that
this distortion is slight. Together, all these observations suggest that
FeP immobilization probably occurred on the silica support surface.
The FTIR spectra of the solids SGBMn, SGBFe, SGAMnS, SGAMnF,
SGAFeS, and SGAFeF and of the control silicas SGB and SGA dis-
played profile typical of silica matrix spectra. An intense and wide
band in the region of 3500 cm−1 was detected, which is due to axial
deformation of SiO-H groups present on the support surface. Other
important bands characteristic of silica were 1100 cm−1 ꢁSi–O–Si,
950 cm−1 ꢁSi–OH, and 790 cm−1 ıSi–O. Bands relative to the MP
were not seen because of the low MP concentration in the silica
[29].
The solids SGBMn, SGBFe, SGAMnS, SGAMnF, SGAFeS, and
SGAFeF were also analyzed by qualitative UV–vis spectroscopy,
and the presence of MP on the silica matrix was confirmed (Fig. 3).
This technique also showed that no demetallation took place dur-
ing the immobilization process, since only the bands of the MnP
and FeP monomers were observed. In the case of SGBMn, SGAMnS,
and SGAMnF, the MnP Soret bands were detected around 475 nm
(Fig. 3c, g, and h) and were redshifted compared to the free MnP
band (460 nm). This behavior can be attributed to the interaction
between the MnP and the support surface [30], which could occur in
four different ways, as reported in the literature: (a) MnP covalent
The X-ray powder difraction analysis of the solids SGBMn,
SGBFe, SGAMnS, SGAMnF, SGAFeS, and SGAFeF as well as of the
control silicas SGB and SGA exhibited the same profile, which was
typical of the XRD pattern of an amorphous solid with a charac-
teristic halo in the 20–30◦ (2Â) region (figure not shown). Since no
changes in the diffraction patterns were observed upon MP immo-
bilization onto SGB and SGA, it can be assumed that the MP lies on
the support surface [22,27].
The presence of immobilized MP in the solids SGBMn, SGBFe,
SGAMnS, SGAMnF, SGAFeS, and SGAFeF was also confirmed by
EPR analysis (Fig. 2). The solids SGBMn, SGAMnS, and SGAMnF
(Fig. 2b–d) were EPR-silent, in agreement with the presence of
manganese(III). These data support our previous conclusion that
after the metallation reaction the resulting MnP contained a
small amount of manganese(II) (Fig. 2a), which was efficiently
removed after the washings accomplished during SGBMn, SGAMnS,
and SGAMnF preparation. The solids SGBFe, SGAFeS, and SGAFeF
(Fig. 2f–h) presented EPR signals at g = 6.0 and g = 2.0, typical of
high-spin iron(III) in axial symmetry, as well as a signal at g = 4.3,
characteristic of high-spin iron(III) in rhombic symmetry [14,28].
The presence of the latter signal, which is absent from the spec-
Fig. 3. Qualitative UV–vis spectra of solid samples (a) SGA, (b) SGB, (c) SGBMn, (d)
SGAFeS, (e) SGAFeF, (f) SGBFe, (g) SGAMnS, and (h) SGAMnF.

124
G.M. Ucoski et al. / Applied Catalysis A: General 404 (2011) 120–128
Table 2
Table 3
Physicochemical properties of the materials obtained by MP immobilization on silica
obtained by the sol–gel process.
Cyclooctene oxidation by PhIO catalyzed by MP in homogeneous and heterogeneous
media.
˚
Sample
SBET (m2 g−1
)
Vp (cm3 g−1
)
Dp (A)
Catalyst
Run
Catalyst/PhIO/
cyclooctene
molar ratiob
Epoxide
yielda (%)
SGB
349.9511
219.9068
–
138.7637
209.8229
–
0.590353
1.091876
–
0.074887
0.116371
–
67.4784
198.6071
–
21.5869
22.1846
–
SGBMn
SGBFe
SGA
SGAMnS
SGAFeS
SGAMnF
SGAFeF
MnP
1
2
3
4
5
6
7
8
9
1:10:1000
1:10:5000
1:50:5000
1:10:5000
70
77
73
12
68
57
52
70
65
47
53
53
44
40
56
50
34
33
SGAMnS
First reuse
SGAMnF
First reuse
SGBMn
–
–
–
209.8311
0.113598
21.6551
BET surface area (SBET), average pore volume (Vp), and average pore diameter (Dp)
of the samples.
First reuse
FeP
10
1:10:1000
1:10:5000
1:50:5000
1:10:5000
11
12
13
14
15
16
17
18
binding to the silica network, generating Si–O–M–O–Si species; (b)
MnP encapsulation inside the silica polymeric matrix; (c) ␲ inter-
action between the MnP electron cloud and the support surface;
and (d) electrostatic interaction [31,32].
SGAFeS
First reuse
SGAFeF
First reuse
SGBFe
First reuse
For the solids SGAFeF and SGBFe (Fig. 3e and f), bands at 427 and
424 nm resembling the FeP Soret band were detected, respectively
(figure not shown). These observations suggest that FeP immo-
bilization probably took place on the silica, so no important ring
constraints were present after the sol–gel process [10,33,34]. On the
other hand, the Soret band was redshifted from 424 nm to 444 nm
in the electronic spectrum of the solid SGAFeS (Fig. 3d). This behav-
ior indicates that some interaction of the FeP or of the porphyrin
ring with the silica surface may have caused macrocycle ring dis-
tortions, in agreement with the EPR analyses. The SGBFe Q-band
(Fig. 3f) is evidence of dimerization, since this solid was obtained in
the presence of water traces. However, bearing in mind that ␮-oxo
dimeric species are EPR-silent due to antiferromagnetic coupling,
dimerization must have occurred to a small extent, because the
SGBFe EPR spectrum presented signals at g = 6.0, g = 2.0, and g = 4.3,
confirming the presence of monomeric FeP species in SGBFe, too
[17,21,26,35].
Thermogravimetric and differential thermal analyses (TGA and
DTA) (figure not shown) were performed for the free FeP and MnP
and for the solids SGBMn, SGBFe, SGAMnS, SGAFeS, SGAMnF, and
SGAFeF. There was mass loss between 30 to 135 ◦C, due to loss of
water and ethanol molecules weakly bound to the materials after
the condensation and polycondensation reactions [10,36]. Events
between 130 to 220 ◦C are related to the release of water and alco-
hol molecules entrapped in the support structure. This process
appeared in the DTA curves as an endothermic effect around 80 ◦C
for all the solids. The greatest water loss was observed for SGA and
indicates its more hydrophilic character. High-temperature pro-
cesses, between 200 to 1000 ◦C, were detected for SGBMn, SGBFe,
SGAMnS, SGAFeS, SGAMnF, and SGAFeF. These were attributed to
MP decomposition followed by silica structure condensation and
collapse, which resulted in SiO2 [10,26].
a
The epoxide yield was calculated on the basis of the PhIO amount used in the
reaction. Control experiment yielded 14% cyclooctenoxide in the case of homo-
geneous catalysis while 5% cyclooctenoxide was achieved in the case of control
heterogeneous catalysis.
b
Conditions – reactions were carried out in a DCM/ACN 1:1 mixture (v/v) at room
temperature, under argon.
low water content produce denser materials with small average
pore size, whist preparations carried out at high pH and high water
contents generated more porous materials [13,23].
The surface properties and pore sizes of the solids SBG, SBGMn,
and SBGFe were determined by the BET method, using nitro-
gen adsorption. The BET results (Table 2) demonstrated that the
sol–gel process performed in basic conditions furnished solids with
reduced surface area and increased pore diameter in the case of
SBGMn. So the manganese(III) ions do not contribute to any silica
surface area increment. It can be argued that the MnP is mostly
adsorbed onto the silica surface, thus covering a fraction of the
total support surface and leaving smaller area available for nitrogen
adsorption. This is also related with increased pore diameter [37].
An opposite effect was noticed for the solids prepared by acid
catalysis: the surface area was larger, while the pore size was little
affected. This can be an indication that the MPs were mostly incor-
porated into the silica matrix, and that their presence contributed
to significant silica matrix surface area enhancement. Therefore,
by controlling the reaction conditions, it is possible to alter the
material’s surface area and hence its application. Nevertheless, it
is important to highlight that this technique does not allow predic-
tion of the exact MP location on the support; i.e., on the surface or
as silica matrix component.
Organic matter (OM) decomposition in the solids SGBMn, SGBFe,
SGAMnS, SGAFeS, SGAMnF, and SGAFeF occurred to a small extent.
This fact can be rationalized when one takes into account the low
concentration of immobilized MP. In the case of the free MPs,
OM contents around 83 and 78% were detected for MnP and FeP,
respectively, while for the solids SGBMn, SGBFe, SGAMnS, SGAMnF,
SGAFeS, and SGAFeF the contents of decomposed OM were 1.71,
1.12, 1.92, 4.44, 2.01, and 3.22%, respectively. These results are sim-
ilar to the MP loading data calculated for the solids SGBMn, SGBFe,
SGAMnS, SGAMnF, SGAFeS, and SGAFeF (Table 1) [15].
The materials surface area and porosity, determined from nitro-
gen adsorption experiments, are listed in Table 2.
Comparison of the materials obtained by acid or basic cataly-
sis revealed that the SGB materials exhibited the largest surface
areas [26]. In general, preparations accomplished at low pH and
3.3. Catalytic oxidation reactions
Initially, the catalytic activities of the MnP and FeP (homoge-
neous catalysis) as well as of the solids SGBMn, SGBFe, SGAMnS,
SGAFeS, SGAMnF, and SGAFeF were investigated in the case of
cyclooctene oxidation (Table 3) by PhIO (oxidant). Cyclooctene is
frequently employed as diagnostic substrate for catalytic activity
evaluation because it is easily oxidized and cyclooctene oxide is
usually the sole product [28,38]. Next, the catalytic activities of
MnP, FeP, SGBMn, SGBFe, SGAMnS, SGAFeS, SGAMnF, and SGAFeF
were studied in the oxidation of a more inert alkane, namely cyclo-
hexane (Table 4), in order to gain information not only about
catalyst efficiency, but also about its selectivity, since cyclohex-
ane oxidation can yield the corresponding alcohol and/or ketone

G.M. Ucoski et al. / Applied Catalysis A: General 404 (2011) 120–128
125
Table 4
could favor substrate and oxidant approach to the metal center,
contributing to larger product formation. Moreover, the fact that
the result achieved for SGAMnS reuse (run 5) was very close to that
obtained for SGAMnF (run 6) indicates that SGAMnS washing and
drying for further reuse removed solvent molecules entrapped in
the solid silica obtained by the sol–gel process, thereby making the
active site more available for the access of PhIO and the substrate.
This result corroborates the thermogravimetric analysis, where the
solid SGAMnS, obtained by the sol–gel method under slow drying
and acid conditions, underwent the greatest water molecules and
hydroxyl groups loss.
The solids SGAFeS and SGAFeF (runs 13 and 15), produced by
the sol–gel process under acidic conditions, performed better com-
pared to the catalyst SGBFe (run 17), prepared by sol–gel in basic
medium. This result corroborates the SGBFe electronic spectrum
data (Fig. 3f), which had evidenced the presence of dimeric (␮-oxo)
species in this solid [26]. The existence of the latter species in the
catalyst should culminate in a less effective MP site for the catalytic
oxidation reaction.
Table 3 shows that all the solids containing immobilized MP dis-
played catalytic activity for cyclooctene oxidation, thus confirming
that the immobilization process and the experimental conditions
did not destroy or inactivate the MP. However, the yields from
heterogeneous reactions were generally lower than or similar to
those obtained in the case of homogeneous catalysis. Hence, the
MPs activity after immobilization was typically reduced. This fact
can be a consequence of the difficult access of the oxidant and
the substrate to the catalytic complex caused by support struc-
ture. Moreover, product adsorption onto the solid catalyst may
have occurred, diminishing its diffusion to the reaction solution and
leading to lower cyclooctene epoxide detection compared to the
amount that was actually produced. Notwithstanding these draw-
backs, the heterogeneous catalyst is advantageous due to its easy
recovery from the reaction solution and possibility of reuse, not to
mention that the MP active site on the support retained its activity
even after the recovery and washing procedures [5].
Cyclohexane oxidation by PhIO catalyzed by MP in homogeneous and heterogeneous
media.
Catalyst
Run
Catalyst/PhIO/
cyclohexane
molar ratiob
Cyclohexanol
yield (%)
Cyclohexanone
yielda (%)
MnP
1
2
3
4
5
6
7
8
1:10:1000
1:50:1000
1:100:1000
1:10:5000
27
10
6
35
24
20
5
7
10
8
4
2
1
8
3
3
5
4
8
2
1
3
1
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
SGBMn
First reuse
SGAMnS
First reuse
SGAMnF
First reuse
FeP
9
10
11
12
13
14
15
16
17
18
19
20
1:10:1000
1:50:1000
1:100:1000
1:10:5000
SGBFe
First reuse
SGAFeS
First reuse
SGAFeF
First reuse
4
2
a
The reactions yields were calculated on the basis of the PhIO amount used in
the reaction. Control experiments did not furnish any oxidation products for both
homogeneous and heterogeneous catalysis.
b
Conditions – reactions were carried out in a DCM/ACN 1:1 mixture (v/v) at room
temperature, under argon.
[4]. Experimental evidence reported in the literature indicates that
the active species in oxidation reactions carried out in homoge-
neous and heterogeneous media using iron(III) and manganese(III)
porphyrins are the high-valent intermediates FeIV(O)(P)•+ and
MnIV(O)(P)•+, respectively [4,14,21,39–52].
3.3.1. Cyclooctene epoxidation
The results from homogenous and heterogeneous cyclooctene
epoxidation by PhIO catalyzed by the MPs investigated herein are
summarized in Table 3.
Following cyclooctene oxidation, the supernatant of all the reac-
tions carried out in heterogeneous medium was analyzed by UV–vis
spectroscopy. No bands relative to MP were detected, indicating
that there was no catalyst leaching from the support during the oxi-
dation reactions. However, after the solid catalysts were washed
with MeOH and the solvent mixture ACN/DCM 1:1 prior to their
reuse, the electronic spectra of the combined washings exhibited
the Soret band. On the basis of quantitative UV–vis analyses, the
percentage of MP leaching from the silica, which was probably a
consequence of the use of MeOH during catalyst washing, was cal-
culated for all the solids. The leached MP values lay below 5%, except
for SGAMnS, for which an MP leaching of 30% was determined. This
value indicates that in the case of SGAMnS the MnP is more weakly
linked to the silica compared to the other solids prepared in this
work.
Except for SGAMnS, in the first catalyst reuse yields similar
to those achieved with the freshly prepared solid catalysts were
achieved, thereby confirming the stability of these solids in the
epoxidation reactions. This is a very favorable outcome of heteroge-
neous catalysis, since catalyst reuse is not yet possible in the case
of the homogeneous route due to difficult MP recovery from the
reaction medium [6].
In homogeneous medium, both MnP and FeP were active cata-
lysts for cyclooctene epoxidation (runs 1–3 and 10–12). The total
cyclooctene oxide yield achieved with the MnP (run 1–3) was
higher than that obtained with the FeP (run 10–12), irrespective
of the catalyst/oxidant/substrate molar ratio. The same trend was
observed in heterogeneous medium (runs 4–9 for MnP and 13–18
for FeP). The fact that the MnP furnished the best results in both
homogeneous and heterogeneous catalysis suggests that the active
species may be more stable compared to the active species gener-
ated in the case of the FeP in the conditions investigated here [7].
This is because the overlap of the filled iron orbitals with the empty
porphyrin orbitals is less effective compared to the overlap of the
manganese orbitals with the porphyrin orbitals [53].
Increasing PhIO molar ratio in relation to the MP and the sub-
strate (from 1:10:5000 (runs 2 and 11) to 1:50:5000 (runs 3 and
12)) did not alter product yield, suggesting that excess substrate
in the presence of excess oxidant prevents competitive reactions
between the MP active catalytic species and PhIO or the MP itself
[10].
The solids SGBMn (run 8) and SGAMnF (run 6) afforded yields
similar to that achieved in homogeneous medium (c.a. 70%),
whereas SGAMnS (run 4) furnished a lower yield (12%). However,
after SGAMnS was washed, dried, and recovered, 30% MnP leach-
ing was observed, and the recovered solid gave rise to 68% epoxide
yield (run 5). This result suggests that MnP rearrangement on the
silica surface or inside the silica pores might have occurred, making
the active species more available for catalysis. This rearrangement
3.3.2. Cyclohexane oxidation
Results from cyclohexane oxidation by PhIO catalyzed by the
supported MPs (heterogeneous catalysis) and by the MPs in solu-
tion (homogeneous catalysis) are given in Table 4. For this substrate,
it is generally accepted that the active species FeIV(O)(P)•+ or
MnIV(O)(P)•+ generated by reaction between the oxygen donor
PhIO and the MP complex abstracts a hydrogen atom from the

126
G.M. Ucoski et al. / Applied Catalysis A: General 404 (2011) 120–128
alkane and quickly collapses, regenerating the catalyst and forming
the corresponding alcohol [4,10,14,21,39–52].
Homogeneous catalysis was initially investigated by varying
the catalyst/oxidant/substrate molar ratios (1:10:1000, 1:50:1000,
and 1:100:1000), in order to assess the MP ring resistance
and the dependence of the reaction yields on PhIO concentra-
tion. A 1:10:5000 catalyst/oxidant/substrate molar ratio was also
employed, to check the influence of substrate excess on product
formation.
Large PhIO ratios (1:50:1000 and 1:100:1000) resulted in lower
alcohol and ketone yields (runs 2–3 and runs 12–13). Moreover,
under the oxidation reaction conditions utilized herein, the MnP
solution changed from the typical intense green color to brown,
suggesting that some catalyst decomposition took place. UV–vis
analysis of the reaction solution revealed that upon increasing PhIO
concentration a new band at 430 and 330 nm appeared in the spec-
trum of the MnP and of the FeP, respectively, with concomitant
reduction in the characteristic Soret band of the MP complex (figure
not shown). This is indication that new species are formed or that
the MPs undergo oxidative destruction during the reaction. The best
catalytic yields were achieved at low PhIO amounts (MP/PhIO molar
ratio = 1:10) (runs 1 and 4, runs 11 and 14). Besides, the reaction
solution color remained constant and the spectral characteristics
were maintained at 1:10 MP/PhIO molar ratio, evidencing the MPs
resistance to oxidative destruction in this condition. The low prod-
uct yields and the electronic spectrum changes suggestive of MP
oxidative destruction detected in the presence of larger PhIO quan-
tities (MP/PhIO molar ratios = 1:50 and 1:100) [21] were expected
since, unlike second-generation MPs [9], the MnP and the FeP pre-
pared herein do not bear substituents that protect the porphyrin
ring against oxidative degradation.
Fig. 4. Dependence of cyclohexane oxidation by PhIO catalyzed by the MnP on
reaction time.
With respect to the MnP in homogeneous medium, changing the
catalyst/oxidant/substrate molar ratio from 1:10:1000 (run 1) to
1:10:5000 (run 4) only resulted in slightly higher alcohol + ketone
yield (from 35% to 38%), but the reaction selectivity for the alco-
hol was considerably increased (alcohol/ketone ratio from 27/8
to 35/3). The alcohol produced at low catalyst/oxidant/substrate
molar ratio (1:10:1000) possibly acted as a competitive substrate,
with its consequent oxidation to ketone (run 1). Alcohol over oxi-
dation was minimized when the substrate concentration in the
reaction medium was raised (run 4). The same effect was not so
evident in the case of the FeP, a less efficient catalyst (runs 11 and
14).
For both MPs, the yields obtained during heterogeneous catal-
ysis (runs 5–10 and 15–20) were somewhat lower than those
observed for homogeneous catalysis, as previously observed in
the case of epoxidation. This validates the proposal that the sil-
ica matrix makes the access of the substrate to the catalytic site
difficult [5]. Although cyclohexane oxidation was lower in the
case of heterogeneous catalysis, the cyclohexane oxidation prod-
uct yields attest that even a less reactive carbon–hydrogen bond
can be hydroxylated using these heterogeneous catalytic systems
under mild conditions [21].
In the first reuse reaction, all the heterogeneous catalysts gave
rise to similar efficiency and selectivity toward cyclohexanol com-
pared to the freshly prepared catalysts, once again confirming the
reuse capacity of these solids.
The MnP was used in an investigation about the influence of
reaction time on product yield. For this purpose, the conditions in
which the best product yields were achieved during cyclohexane
oxidation by PhIO (Table 4, run 4) were utilized. Fig. 4 reveals that in
the optimal reaction conditions, the best yield was obtained within
the first hour of reaction, with selectivity for the alcohol. Within
the first 15 min, 33% alcohol had already been produced, and this
value remained virtually unchanged in the following 45 min (yield
at 1 h = 35%). Therefore, under ideal experimental conditions, the
Fig. 5. UV–vis spectrophotometric study of the MnP solution with successive PhIO
additions: (a) no substrate (b) with substrate (excess of 1000) and (c) with substrate
(excess of 5000).

G.M. Ucoski et al. / Applied Catalysis A: General 404 (2011) 120–128
127
oxidative degradation. A MnP/PhIO 1:25 molar ratio elicited 80%
decrease in Soret band intensity, corroborating the assumption that
the MnP is destroyed in the presence of larger PhIO quantities. It is
noteworthy that up to a 1:15 MnP/PhIO molar ratio the MnP degra-
dation was minimal, so this was an adequate condition to prevent
catalyst destruction.
In the presence of excess substrate (cyclohexane), namely at a
1:x:1000 catalyst/oxidant/substrate molar ratio (Fig. 5b), there was
slow diminishing of the Soret band intensity upon increasing PhIO
amounts. At a 1:25 MnP/PhIO molar ratio, the Soret band inten-
sity decreased about 55%, showing that the active catalytic species
was probably used to oxidize the substrate and, to a minor extent,
the porphyrin ring. The use of an even larger substrate amount
(1:x:5000) (Fig. 5c), which is the condition that furnished the high-
est yield in the case of cyclohexane oxidation, promoted Soret band
intensity decrease of only about 35% at a 1:25 MnP/PhIO molar
ratio. Taken together, these data demonstrate that catalyst lifetime
is improved in the presence of excess cyclohexane, corroborating
the previously discussed results (supplementary information S3).
3.5. Iron(III) porphyrin-PhIO solution characterization
A UV–vis study of solutions containing various FeP/PhIO molar
ratios was also accomplished, and results similar to those achieved
in the case of the MnP were obtained in both the absence and pres-
ence of excess cyclohexane (Fig. 6). In the absence of the substrate
there was gradual decrease in the intensity of the typical FeP Soret
band (Fig. 6a). In the presence of the substrate (Fig. 6b and c), there
was 40% reduction in the Soret band at a 1:25 FeP/PhIO molar ratio.
Therefore, the FeP was 16% more resistant than the MnP. At a 1:50
MP/PhIO molar ratio the MnP was completely destroyed, while 30%
FeP remained in the solution, thus confirming that the latter MP is
more resistant to the reaction medium, leading to larger product
formation. As previously observed for the MnP, the FeP resistance
to oxidation was enhanced in the presence of high substrate con-
centrations.
4. Conclusion
The free base porphyrin [H2(TAPP)] was synthesized and
complexed with manganese(III) and iron(III) ions, in order to
obtain new catalysts for oxidation reactions. The resulting MPs,
namely [Mn(TAPP)]Ac and [Fe(TAPP)]Cl, were immobilized on silica
obtained through the sol–gel process in acidic and basic conditions.
The immobilization efficiency depended on the drying process the
solids were submitted to.
Fig. 6. UV–vis spectrophotometric study of the FeP solution [Fe(TAPP)]Ac with suc-
cessive PhIO additions: (a) no substrate (b) with substrate (excess of 1000) and (c)
with substrate (excess of 5000).
prepared MP yields good catalytic results even for an unreactive
substrate like cyclohexane.
The MPs in solution and immobilized on silica were investigated
as catalysts for oxidation reactions of two hydrocarbons. In homo-
geneous medium, the MPs furnished good cyclooctene epoxidation
and cyclohexane hydroxylation yields, which demonstrated signif-
icant stability of these catalysts under strong oxidizing conditions.
Yields slightly lower or similar to those achieved in homogeneous
medium were obtained for the MPs immobilized on silica. These
results are very promising since heterogeneous catalysts, which
can be easily recovered from the reaction medium and reused,
are advantageous over homogeneous catalysts. In fact, recyclabil-
ity capacity and retained activity upon reuse were demonstrated
for all the solids prepared herein. Except for SGAMnS, which
underwent 30% leaching, all the solids presented no significant
percentage of MP leaching form the silica during the washings.
[Mn(TAPP)]Ac displayed better catalytic oxidation activity com-
pared with [Fe(TAPP)]Cl, suggesting that the catalytic species
involving the manganese complex might be more stable in the
reaction conditions investigated in this work. These results are
expected for porphyrin structures lying between the first and sec-
Finally, the fact that control cyclooctene and cyclohexane oxi-
dation reactions carried out with the SGB and SGA solids containing
no MP did not furnish any significant product yield led to the con-
clusion that the catalytic activity presented by all the investigated
solids is really due to the MP immobilized on the silica. In other
words, the MPs play an essential role in catalysis, promoting oxygen
transfer from the oxidant to the substrate [27].
3.4. Manganese(III) porphyrin-PhIO solution characterization
In order to determine the ideal MnP/PhIO molar ratio to be used
for the oxidation reactions carried out in homogeneous medium
and to monitor the MnP resistance to different oxidant quanti-
ties, solutions containing different MnP/PhIO molar ratios were
analyzed by UV–vis, at room temperature. Fig. 5a illustrates the
study conducted in the absence of substrate. Larger PhIO amounts
prompted reduction in the MnP Soret band, which characterizes

128
G.M. Ucoski et al. / Applied Catalysis A: General 404 (2011) 120–128
ond generation according to Dolphin’s classification [9]. It can
be concluded that the porphyrin structure obtained here is not
stable enough to ensure the high yields achieved with typical
second-generation MPs, such as [Fe(TPFPP)] (5,10,15,20–meso-
tetrakis(pentafluorophenyl)porphyrin) [54]. This is because in the
present case, the MP suffered some oxidative degradation, although
under specific controlled experimental conditions it can exhibit
good catalytic performance.
[17] J.L. Nian, L. Min, H.A. Kong, Inorg. Chim. Acta 178 (1990) 59–65.
[18] J.G. Sharefkin, H. Saltzman, Org. Synth. 5 (1963) 658.
[19] A.D. Adler, F.R. Longo, J.D. Finarelli, J. Goldmacher, J. Assour, L. Korsakoff, J. Org.
Chem. 32 (1967) 476.
[20] A.D. Adler, F.R. Longo, F. Kampas, F.J. Kim, J. Inorg. Nucl. Chem. 32 (1970)
2443–2445.
[21] J.T. Groves, T.E. Nemo, J. Am. Chem. Soc. 105 (1983) 6243–6248.
[22] H. Tanaka, T. Yamada, S. Sugiyama, H. Shiradoti, R. Hino, J. Colloid Interface Sci.
286 (2005) 812–815.
[23] L.L. Hench, J.K. West, Chem. Rev. 90 (1990) 33–72.
[24] R.J.P. Corriu, D. Leclereq, Angew. Chem. Int. Ed. 35 (1996) 1420–1436.
[25] T. Coradin, P.J. Lopez, ChemBioChem 3 (2003) 1–9.
[26] K.A.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.
Acknowledgments
[27] G.S. Machado, K.A.F. Castro, O.J. Lima, E.J. Nassar, K.J. Ciuffi, S. Nakagaki, Colloids
Surf. A 349 (2009) 162–169.
[28] M. Halma, A. Bail, F. Wypych, S. Nakagaki, J. Mol. Catal. A: Chem. 243 (2006)
44–51.
[29] S.T. Castaman, S. Nakagaki, R.R. Ribeiro, K.J. Ciuffi, S.M. Drechsel, J. Mol. Catal.
A: Chem. 300 (2009) 89–97.
[30] L. Barloy, J.P. Lallier, P. Battioni, D. Mansuy, New J. Chem. 16 (1992) 71–80.
[31] V.P. Barros, A.L. Faria, T.C.O. MacLeod, L.A.B. Moraes, M.D. Assis, Int. Biodeterior.
Biodegrad. 61 (2008) 337–344.
The authors acknowledge the financial support obtained from
CAPES (Coordenac¸ ão de Aperfeic¸ oamento de Pessoal de Nível Supe-
rior), CNPq (Conselho Nacional de Desenvolvimento Cientifico e
Tecnológico), FINEP (Financiadora de Estudos e Projetos), and
Araucária Foundation. We are also thankful for the technical sup-
port of Laboratório de Análises de Minerais e Rochas (LAMIR),
FUNPAR (Fundac¸ ão da Universidade Federal do Paraná) and UFPR
(Universidade Federal do Paraná).
[32] J.C. Biazzotto, H.C. Sacco, K.J. Ciuffi, C.R. Neri, A.G. Ferreira, Y. Iamamoto, O.A.
Serra, J. Non-Cryst. Solids 247 (1999) 134–140.
[33] F.S. Vinhado, C.M.C. Prado-Manso, H.C. Sacco, Y. Iamamoto, J. Mol. Catal. A:
Chem. 174 (2001) 279–288.
[34] S. Nakagaki, F.L. Benedito, F. Wypych, J. Mol. Catal. A: Chem. 217 (2004)
121–131.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.apcata.2011.07.019.
[35] C.M.C. Prado-Manso, E.A. Vidoto, F.S. Vinhado, H.C. Sacco, K.J. Ciuffi, P.R. Martins,
A.G. Ferreira, J.R. Lindsay-Smith, O.R. Nascimento, Y. Iamamoto, J. Mol. Catal. A:
Chem. 150 (1999) 251–266.
[36] H. Vrubel, K.J. Ciuffi, G.P. Ricci, F.S. Nunes, S. Nakagaki, Appl. Catal. A 368 (2009)
139–145.
[37] A.M. Youssef, A.A. Attia, Th. El-Nabarawy, Colloids Surf. A 125 (1997) 163–170.
[38] A.J. Appleton, S. Evans, J.R. Lindsay-Smith, J. Chem. Soc. Perkin Trans. 2 (1996)
281–285.
[39] J.T. Groves, J. Inorg. Biochem. 100 (2006) 434–447.
[40] W.J. Song, M.S. Seo, S.D. George, T. Ohta, R. Song, M.J. Kang, T. Tosha, T. Kitagawa,
E.I. Solomon, W. Nam, J. Am. Chem. Soc. 129 (2007) 1268–1277.
[41] F. Montanari, Pure Appl. Chem. 66 (1994) 1519–1526.
[42] D.G. Kellner, S.C. Hung, K.E. Weiss, S.G. Sligar, J. Biol. Chem. 277 (2002)
9641–9644.
[43] J.T. Groves, J. Lee, S.S. Marla, J. Am. Chem. Soc. 119 (1997) 6269–6273.
[44] R. Zhang, M. Newcomb, J. Am. Chem. Soc. 125 (2003) 12418–12419.
[45] Y. Iamamoto, M.D. Assis, O. Baffa, S. Nakagaki, O.R. Nascimento, J. Inorg.
Biochem. 52 (1993) 191–200.
[46] S. Nakagaki, Y. Iamamoto, O. Baffa, O.R. Nascimento, Inorg. Chim. Acta 186
(1991) 39–43.
[47] J.T. Groves, R.C. Haushalter, M. Nakamura, T.E. Nemo, B.J. Evans, J. Am. Chem.
Soc. 103 (1981) 2884–2886.
[48] W. Nam, Acc. Chem. Res. 40 (2007) 522–531.
[49] D. Mansuy, C. R. Chim. 10 (2007) 392–413.
[50] D. Mansuy, Pure Appl. Chem. 62 (1990) 741–746.
[51] N. Jin, J.L. Bourassa, S.C. Tizio, J.T. Groves, Angew. Chem. Int. Ed. 39 (2000)
3849–3851.
[52] W. Nam, I. Kim, M.H. Lim, H.J. Choi, J.S. Lee, H.G. Jang, Chem. Eur. J. 8 (2002)
2067–2071.
References
[1] G.R. Friedermann, M. Halma, K.A.F. Castro, F.L. Benedito, F.G. Doro, S.M. Drech-
sel, A.S. Mangrich, M.D. Assis, S. Nakagaki, Appl. Catal. A 308 (2006) 172–181.
[2] B. Meunier, Chem. Rev. 92 (1992) 1411–1456.
[3] J.R. Lindsay Smith, Y. Iamamoto, F.S. Vinhado, J. Mol. Catal. A: Chem. 252 (2006)
23–30.
[4] 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.
[5] S. Nakagaki, G.S. Machado, M. Halma, A.A.S. Marangon, K.A.F. Castro, N. Mattoso,
F. Wypych, J. Catal. 242 (2006) 110–117.
[6] G.S. Machado, K.A.F. Castro, F. Wypych, S. Nakagaki, J. Mol. Catal. A: Chem. 283
(2008) 99–107.
[7] B.S. Lane, K. Burgess, Chem. Rev. 103 (2003) 2457–2474.
[8] M.J. Nappa, C.A. Tolman, Inorg. Chem. 24 (1985) 4711–4719.
[9] D. Dolphin, T.G. Traylor, L.Y. Xie, Acc. Chem. Res. 30 (1997) 251–259.
[10] A.T. Papacídero, 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.
[11] W. Kaim, B. Schwederski, Bioinorganic Chemistry: Inorganic Elements in the
Chemistry of Life, Wiley, New York, 1994.
[12] E.J. Nassar, Y. Messaddeq, S.J.L. Ribeiro, Quim. Nova 25 (2002) 27–31.
[13] A.M. Buckley, M. Greenblatt, J. Chem. Educ. 71 (1994) 599–602.
[14] 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.
[15] 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.
[16] W. Stöber, A. Fink, E. Bohn, J. Colloid Interface Sci. 26 (1968) 62–69.
[53] L.J. Boucher, J. Am. Chem. Soc. 92 (1970) 2725–2730.
[54] C.K. Chang, F. Ebina, J. Chem. Soc. Chem. Commun. (1981) 778–779.