Heterogenized Fe(III) phthalocyanine: Synthesis, characterization and application in liquid-phase oxidation of phenol

Article abstract of DOI:10.1016/j.molcata.2009.11.017

Tetra-tert-butyl substituted FePc(acac) complexes were supported upon mesoporous molecular sieves MCM-41 and SBA-15 via adsorption and coordination bondings as well as by covalent grafting. Prepared materials were characterized by UV-vis, MALDI, N₂-BET, SEM, TEM, FTIR, ¹³C, ²⁹Si NMR and DSC-TG techniques. Supported FePcs exhibited high activity in liquid-phase oxidation of phenol with H₂O₂. The most resistant towards oxidative degradation were shown to be the covalently grafted FePc species.

Full text of DOI:10.1016/j.molcata.2009.11.017

Journal of Molecular Catalysis A: Chemical 319 (2010) 39–45
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Heterogenized Fe(III) phthalocyanine: Synthesis, characterization and
application in liquid-phase oxidation of phenol
a
a,b
a
a
Sergey V. Sirotin , Alexander Yu. Tolbin , Irina F. Moskovskaya , Sergey S. Abramchuk ,
a,b
a,∗
Larisa G. Tomilova , Boris V. Romanovsky
M.V. Lomonosov Moscow State University, Moscow 119991, Russia
a
b
Institute of Physiologically Active Compounds, Russian Academy of Sciences, Chernogolovka, Moscow Region 142432, Russia
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 24 September 2009
Received in revised form
Tetra-tert-butyl substituted FePc(acac) complexes were supported upon mesoporous molecular sieves
MCM-41 and SBA-15 via adsorption and coordination bondings as well as by covalent grafting. Prepared
13
29
materials were characterized by UV–vis, MALDI, N2-BET, SEM, TEM, FTIR, C, Si NMR and DSC–TG
techniques. Supported FePcs exhibited high activity in liquid-phase oxidation of phenol with H2O2. The
most resistant towards oxidative degradation were shown to be the covalently grafted FePc species.
2
4 November 2009
Accepted 26 November 2009
Available online 1 December 2009
©
2009 Elsevier B.V. All rights reserved.
Keywords:
MCM-41
SBA-15 molecular sieves
Iron(III) phthalocyanines
Hydroxylation of phenol
1
. Introduction
lently grafted by forming the siloxane bonds between the reactive
Pc molecules bearing (alkoxy) Si- containing functional substitutes
3
There has been growing interest in the use of iron phthalo-
and support surface OH groups. Earlier, FePcs symmetrically substi-
cyanines (Pcs) as liquid-phase oxidation catalysts [1,2]. They are
water-insoluble compounds and what is more important is that
they provide fairly reversible redox cycles of iron atoms. Immobi-
lization of FePcs on porous solids provides an easy way to separate
a catalytic material from reaction products and thereby to reuse
it. Application of high surface area matrices ensures the spatial
separation of catalytic species that behave very similarly to the
active centers in true molecular catalysis. In this respect, MCM-
tuted with eight (EtO) Si- functional groups have been successfully
3
grafted onto amorphous silica [7]. At the same time, the controlled
immobilization of Pc through single siloxane bond could be also
performed using monosubstitued derivatives.
The grafting of Pc via single bond provides some advantages
due to the uniformity of active centers. Indeed, this way excludes
different orientations of Pc molecules which may occur on using
polysubstituted derivatives.
4
1 and SBA-15 mesoporous molecular sieves seem to present the
In this paper we report the synthesis and characterization of
t
most promising materials for using them as matrices because of
their high surface, uniform pore size distribution and high thermal
stability.
There are several ways to immobilize Pcs: simple physical
adsorption [3], coordinative bonding to anchoring surface groups
tert-butylphthalocyanine ( PcFe) containing catalysts through the
t
immobilization of PcFe upon differently functionalized MCM-41
and SBA-15 sieves as catalysts for liquid-phase hydroxylation of
phenol by hydrogen peroxide.
[
[
4,5] and covalent bonding with surface silanol groups of a support
6]. The first two routes do not result in the chemical bonding and
2. Experimental
cannot prevent the Pc species from leaching whereas the third way
provides rather strong Pc immobilization. Complexes can be cova-
2.1. Syntheses
2.1.1. MCM-41 support
MCM-41 mesoporous molecular sieve was prepared using
the following protocol [8]. The reaction mixture of molar
composition 1 TEOS: 0.3 CTMABr:11 NH :60 EtOH:144 H O
∗ Corresponding author at: Chemistry Department, Lomonosov Moscow State
University, 1 Lenin Hills, Bld. 3, Moscow 119991, Russia. Tel.: +7 495 939 20 54;
fax: +7 495 939 35 70.
3
2
(TEOS = tetraorthoethoxysilane, CTMABr = cetyltrimethylammo-
E-mail address: bvromanovsky@mail.ru (B.V. Romanovsky).
nium bromide) was stirred at RT for 2 h, then the amorphous
1
381-1169/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2009.11.017

4
0
S.V. Sirotin et al. / Journal of Molecular Catalysis A: Chemical 319 (2010) 39–45
Scheme 1. Synthesis of covalently bonded Fe(III) phthalocyanine 3. (i) NaH/DMF, (CH3O)3SiCH2CH2CH2Cl; (ii) DBU/o-DCB, Fe(acac)3; (iii) SBA-15/toluene.
precipitate was separated, washed with water, dried and
air-calcined at 550 C.
(5 ml), NaH was added followed by stirring for 1 h. Then
3-(chloropropyl)-trimethoxysilan (0.24 ml, 1.220 mmol) was
added dropwise and the mixture was kept for 8 h (TLC control).
After completion of the reaction, phthalocyanine compounds
were precipitated by adding water followed by chromatographic
isolation of target ligand 1b to give 95 mg (79%).
◦
2.1.2. SBA-15 support
SBA-15 mesoporous molecular sieve was obtained by one-step
synthetic procedure [9]. 4 g of Pluronic P123 (EO20PO70EO20) was
added in 110 ml H O and stirred at RT until the dissolution was
ii) Synthesis of 2-(trimethoxysilylpropoxymethylbenzyloxy)-
9(10),16(17),23(24)-tri-tert-butylphthalocyanine Fe(III)(acac)
(2). To a solution of 1b (60 mg, 0.061 mmol) in o-DCB (10 ml),
DBU (0.2 ml) and Fe(acac)3 (32 mg, 0.091 mmol) were added
2
complete. Then 12.4 ml of 0.1 M HCl and 0.01 g of NH F were added,
4
◦
mixture was heated to 40 C. 9.4 ml (8.8 g) of TEOS was added at
vigorous stirring, then mixture was kept stirring for 72 h at 40 C.
◦
◦
Solid precipitate was washed by centrifugation until pH of 4–6.
followed by heating at 150 C for 1.5 h (UV–vis and TLC control).
After completion of the reaction, target complex was precip-
◦
Then it was dried at 90 C, calcined in the mixture of N and air at
2
◦
5
50 C for 24 h
itated by adding CH3OH to give 63 mg (91%). MS (m/e): 1188
+
[
M−acac+DHB] , 906 [M−acac–C14H23O Si+DHB].
4
2
.1.3. NH –MCM-41 support
2
The grafting of 3-APTES (3-aminopropyltriethoxysilane) onto
MCM-41 was performed according to the method described in [10].
g MCM-41 was dispersed in 50 ml of toluene followed by addition
t
2
.1.8. Fe Pc(acac)–Si–SBA-15 catalyst
3
00 mg of SBA-15 was dispersed in 10 ml of toluene fol-
lowed by addition of 59 mg of Fe Pc(acac)–Si(OMe) . The
mixture was refluxed for 24 h, then the powder was sepa-
rated, washed with toluene and ethanol to remove the ungrafted
Fe Pc(acac)–Si(OMe) , then dried at 80 C for 24 h. A pale green
1
t
3
of 1 ml of 3-APTES. The mixture was refluxed for 6 h and every 1.5 h
the ethanol formed was boiled away. The powder was washed with
◦
toluene and dried at 130 C.
t
◦
3
residue was obtained.
t
2
.1.4. Fe Pc(acac) sample
2
,9(10),16(17),23(24)-tetra-tert-butylphthalocyanine
iron(III)(acac) was prepared using the following procedures.
2.2. Characterization
To a solution of 2,9,16,23-tetra-tert-butylphthalocyanine ligand
[
1
(
1
11] (254 mg, 0.344 mmol) in o-dichlorobenzene (o-DCB) (10 ml),
Elemental analysis was performed using a Flash EA 1112 ana-
lyzer. Sample of catalyst was placed in a tin crucible and burnt in
pure oxygen. Gas products were analyzed by GLC. Iron content was
measured using AA analysis. A sample was treated by the mixture of
concentrated HF, HNO3 and HCl acids. Then the mixture was evap-
orated, and the residue was dissolved in HCl. This procedure was
done twice followed by adding to analyzer.
The BET surface area, the pore volume, and the average
pore diameter were calculated from the nitrogen adsorption
isotherms at 77 K recorded on an automated porosimeter ASAP
,8-diazabicyclo[5.4.0]undec-7-ene (DBU) (0.4 ml) and Fe(acac)3
◦
180 mg, 0.514 mmol) were added followed by heating at 150 C for
.5 h (UV–vis and TLC control). After the reaction was completed,
target complex was precipitated by adding CH OH to give 300 mg
(
3
98%) of target complex.
t
2
.1.5. Fe Pc(acac)/MCM-41 sample
t
The solution of Fe Pc(acac) (120 mg) in methylene chloride was
added to a thin uniform layer of MCM-41 (0.72 g) so as to cover all
particles. Then the slurry was dried at 80 C for 24 h which results
in a dark green residue.
◦
2000N Micromeritics. Before determination, all samples except
t
for NH –MCM-41 and Fe Pc(acac)–Si–SBA-15 were evacuated at
3
2
◦
−3
50 C and 10 mm Hg for 2 h.
t
2
.1.6. Fe Pc(acac)/NH –MCM-41 sample
FTIR spectra of the samples (KBr pellets) were recorded in the
2
t
−1
The procedure was the same as for Fe Pc(acac)/MCM-41 except
range of 400–4000 cm on a Nicolet Protege 460 E.S.P. Fourier-
transform spectrometer. The recorded IR spectra were processed
with the use of the OMNIC E.S.P. software package (Nicolet). The
bands characteristic of the surface OH groups of the supports and
grafted samples were recorded after evaluation.
that NH –MCM-41 was used as the support. A dark blue residue
was obtained.
2
t
2
.1.7. Fe Pc(acac)–Si(OMe)3 sample
Synthesis of 2-(trimethoxysilylpropoxymethylbenzyloxy)-
DSC–TG curves were registered in air in the regime of a lin-
◦
◦
9
(10),16(17),23(24)-tri-tert-butylphthalocyanine
Fe(III)(acac)
ear 10 C/min temperature increase in the range of 20–1000 C.
The recording and processing of the data were performed using
Universal Analysis software package.
included two stages (Scheme 1)
i) Synthesis of 2-(trimethoxysilylpropoxymethylbenzyloxy)-
(10),16(17),23(24)-tri-tert-butylphthalocyanine ligand (1b).
To the solution of 1a [12] (100 mg, 0.122 mmol) in DMF
The morphology of the particles was determined on a JEM-
2000FXII electron microscope in the SEM mode. TEM images were
collected using LEO912 AB OMEGA electron microscope.
9

S.V. Sirotin et al. / Journal of Molecular Catalysis A: Chemical 319 (2010) 39–45
41
Fig. 1. SEM microphotographies for SBA-15 material.
Fig. 2. TEM microphotographies for SBA-15 material.
Table 1
t
Characteristics of free supports and immobilized Fe Pc(acac).
2
3
Sample
Fe, wt.%
AAA
SBET, m /g
Pore volume, cm /g
Av. pore diameter, Å
CHNS
MCM-41
–
–
–
0.80
0.84
0.57
–
–
–
0.93
0.95
0.55
1010
760
820
n/a
n/a
500
0.68
0.28
0.84
n/a
n/a
0.62
27
18
65
n/a
n/a
65
NH2–MCM-41
SBA-15
t
Fe Pc(acac)/MCM-41
t
Fe Pc(acac)/NH2–MCM-41
t
Fe Pc(acac)–Si–SBA-15
Mass spectra of phthalocyanines (MALDI-TOF) were registered
on Autoflex II using 2,5-dihydroxybenzoic acid as a matrix.
NMR spectra of solid phase samples were collected on Bruker
Avance-II 400 with 9.4T magnetic field corresponding with the
added with vigorous stirring. Samples of 1 ml were periodically
taken for analyses from the reaction mixture. Reaction products
were analyzed using GC (30 m × 0.22 mm capillary SE-30 column).
The reaction catalysts were separated, washed with hot water
2
9
◦
working frequency Si = 79.49 MHz. The spectra were processed
(∼80 C), dried in air and then tested in the second cycle.
with TopSpin 2.1 software (Bruker). 29Si NMR spectra were col-
◦
lected for air dry samples using magical angle (54.7 C) rotation,
3. Results and discussion
5
kHz frequency and a 7 mm ZrO2 rotor. As an external reference
Si(CH3)4 was used, the collection of spectra was performed using
3
.1. MCM-41, NH –MCM-41 and SBA-15 supports
2
◦
29
one impulse consecution (90 C-impulse) Si and an impulse con-
secution with polarization transition to silica (cross-polarization
SEM image of as-synthesized MCM-41 material showed
1
29
H– Si) and a suppression of a spin–spin interaction. In the
this support to consist of 0.25–1.0 ␮m uniform microspheres.
For SBA-15 sieve, SEM reveals the amorphous needle-shaped
microstructures (Fig. 1). As seen, these needles are uniform and iso-
lated from each other that could facilitate easy access for entering
reagent molecules.
first case there were 128–512 scans and the time between
impulses was 60 s. In the second case there were 2048 scans and
the time was 5 s while the time of polarization transition was
2
ms.
The uniform pore sizes of SBA-15 can be directly evidenced by
TEM images (Fig. 2). On the other hand, the use of P123 as a template
leads to a worm-like channeled phase with 2D hexagonally ordered
structure.
2
.3. Catalytic tests
The hydroxylation of phenol was performed in a static system
at atmospheric pressure. Experiments were performed as follows.
Textural characteristics for all supports are summarized in
Water (10 ml), phenol (1 g) and catalyst (10 mg) were placed into a
Table 1. Surface area of MCM-41 as determined by N -BET method
was of 1010 m /g, and average pore diameter was estimated as
2
◦
2
3
0-ml flask and heated up to at 60 C. Then 2 ml of 25% H O was
2
2

4
2
S.V. Sirotin et al. / Journal of Molecular Catalysis A: Chemical 319 (2010) 39–45
t
Fig. 3. N2 adsorption isotherms (a) and pore size distribution (b) for MCM-41 (1), NH2–MCM-41 (2), SBA-15 (3) and Fe Pc(acac)–Si–SBA-15 (4).
its OH groups. As to the grafted 3-APTES, 3373 and 3296 cm 1
−
bands can be attributed to the asymmetrical and symmetrical NH
−
1
stretching vibrations in NH2 groups, and 1596 cm bands can be
ascribed to its deforming oscillations [13]. Bands in the range of
−
1
2
3
800–3000 cm are related to both propyl and ethoxy groups of
−
1
−1
(CH3 sym. stretch.), 2929 cm (CH2 asym.
-APTES: 2882 cm
stretch.), 2976 cm (CH3 asym. stretch.) [14,15].
−
1
Nitrogen wt.% was found to be of 3.1% that gives 2.2 mmol/g of
grafted 3-APTES. Total concentration of OH groups was calculated
from TG data providing the dehydroxylation of surface accounts for
◦
the weight loss in the range of 200–1200 C. For MCM-41 this value
was 3.5 mmol/g or 2.1 OH groups/nm2 which is close to the sur-
face concentration of 3-APTES. Generally, for MCM-41 material OH
concentration was shown to be in the range of 1.5–2.5 groups/nm2
Fig. 4. FTIR spectra for samples (1) MCM-41; (2) NH2–MCM-41; (3) 3-APTES. 29Si
[
16]. Given the measurement errors as well as the assumption men-
13
(
A) and C (B) NMR spectra of grafted 3-APTES.
tioned above, this can be considered only as a rough estimate.
The direct information on the interplay between grafting agents
and support surface can be gathered from 29Si NMR spectra
(Fig. 5A).
c.a. 3 nm. After grafting 3-APTES modifier, both specific surface and
pore volume of MCM-41 significantly decreased. This can be con-
sidered as the evidence (though indirect) for the occurrence of the
grafting, since the peaks at pore size distribution curves keep their
shapes and change only in absolute values (Fig. 3).
The chemical shift at −109 ppm is of highest intensity and is
characteristic for framework silica. The shoulder at −102 ppm is
related to silica atoms with terminal silanol groups. The peak at
−66 ppm can be attributed to the silica atoms of 3-APTES linked by
three bonds with the support. This value is −65 ppm in [17] and
−69 ppm in [18]. Peak at −60 ppm refers to the silica linked by two
bonds with the surface (−61 ppm in [18]). The propyl fragment of
3-APTES gives three peaks in 13C NMR spectra (Fig. 5B) attributed to
the first (9 ppm), second (26 ppm) and third (44 ppm) carbon atom
counting from silica [17].
FTIR spectra presented in Fig. 4 provide strong evidence of 3-
APTES grafting onto the matrix surface. After reaction of 3-APTES
−
1
modifier with the surface OH groups, their band at 3740 cm
disappear while the bands characteristic for the modifier appear.
−1
Complete disappearance of 3740 cm band implies that molecules
of modifiers cover all support surface and fully reacted with
Fig. 5. 29Si (A) and 13C (B) NMR spectra of grafted 3-APTES.

S.V. Sirotin et al. / Journal of Molecular Catalysis A: Chemical 319 (2010) 39–45
43
t
t
Fig. 6. MALDI-TOF mass spectra for Fe Pc(acac) (A) and H2 Pc–Si(OMe)3 (B).
+
t
Composition and textural characteristics of both sieve supports
and Fe Pc(acac) containing samples are given in Table 1. Substantial
acac ligand [M−acac] . For Fe Pc(acac)–Si(OMe) complex the spec-
3
t
trum of metal-free precursor (Scheme 1b) is shown as an example
(Fig. 6B). 985 m/z peak is attributed to [MH]+ molecular peak and
all of the characteristic fragment ion peaks are also detected: 862
[M−C H O Si] , 818 [M−C H O Si] , 802 [M−C H O Si], 699
increase in the average pore size of molecular sieve matrix on going
from MCM-41 to SBA-15 is due to the fact that the use of polymer
template P123 instead of CTMABr leads to a structure with quite
different morphologies and textures [19,20].
+
+
3
9
3
6
15
3
6
15
4
+
[M−C
H
24
O4Si] .
14
t
t
Taking into account the molecular size of Fe Pc(acac) with
UV–vis spectra of Fe Pc(acac)–Si(OMe)3 showed the band at
672 nm which can be attributed to monomeric form of phthalo-
cyanine with vibrational satellite at 616 nm.
large derivative, large pore SBA-15 matrix seems to be more
preferable for covalent bonding rather than MCM-41 material.
Noteworthy, uniform pore size distribution does not change
t
For Fe Pc(acac)–Si–SBA-15 sample, it was shown by means of
t
t
after Fe Pc(acac)–Si(OMe)3 grafting but there is a decrease of
FTIR (Fig. 7) that Fe Pc(acac) molecules are chemically bonded
the total pore volume. Since diameter of pores remains con-
stant it could be explained by partial blocking of pores by bulk
to SBA-15 surface. On the spectrum for the grafted sample the
−
1
band for the terminal silanol group (3740 cm ) has significantly
decreased. Though, in comparison with grafted 3-APTES sam-
ple it was not completely disappeared because the amount of
phthalocyanine molecules. On the other hand, the size of a single
t
Fe Pc(acac)–Si(OMe) complex is much smaller than the pore diam-
3
t
eter of the support. So, the blocking of pores most likely may occur
Fe Pc(acac)–Si(OMe)3 present in reaction mixture was not enough
t
by Fe Pc(acac) aggregations, which is well known even in diluted
to react with all surface OH groups of SBA-15. The bands in the
2
−1
solutions. Significant decrease of BET surface from 820 to 500 m /g
range of 2800–3000 cm can be attributed to the oscillations of
may also be connected with the pore blocking. At the same time,
most of Fe Pc(acac) molecules are very probable to reside within
CH2 and CH3 groups.
t
t
The amount of grafted Fe Pc(acac)–Si(OMe)3 was evaluated by
the support mesopores. In fact, the total amount of grafted com-
means of CHNS elemental analysis. 1.1 wt.% of N corresponds to
0.145 mmol/g of iron phthalocyanine grafted or 68 mol.% of ini-
tially added complex. Nitrogen wt.% was also used for calculation
of iron content (Table 1). The values estimated by such a way are
just slightly different from those determined by AAA.
2
plexes could not be grafted onto the outer surface (∼10 m /g), so
t
that it makes possible to consider Fe Pc(acac) species as located
mainly within the matrix pores.
t
t
For all Fe Pc(acac) containing samples, UV–vis reflectance spec-
3
.2. Fe Pc(acac)-containing catalysts
tra exhibit three peaks characteristic of phthalocyanine complexes
(Fig. 8). Bands in the range of 680–710 nm can be attributed
to monomeric form of Pcs, while absorbance in the range of
t
Structure of prepared Fe Pc(acac) complexes was evidenced by
t
means of MALDI-TOF technique (Fig. 6). Fe Pc(acac) complex gives
5
50–660 nm indicates the presence of their aggregates. Soret bands
7
92 m/z peak which is characteristic for molecular ion with cleaved
at 340–350 nm are well typical for both porphyrin and analo-
gous structures. Assumption on the aggregation is also supported
by the red shift of bands in comparison with that of free com-
t
plexes [21]. It is to note that freshly prepared Fe Pc(acac) and
t
Fe Pc(acac)–Si(OMe)3 are monomeric which was shown by means
of MALDI-TOF. However, their concentration on the support surface
t
could lead to the aggregation. In solid state individual Fe Pc(acac)
is already aggregated as the intensity of a band at 643 nm is much
higher than that of monomeric form. Chemical immobilization of
t
Fe Pc(acac) makes them mainly monomerized but nevertheless
some portion of a complex remains aggregated. This aggregation
effect may be resulted from the formation of ␮-oxo complexes
which can be revealed using the FTIR techniques as the absorbance
−
1
at 833 cm is commonly attributed to the Fe–O–Fe deformation
mode [22]. In our case the identification of this band is rather com-
−
1
plicated by very strong absorbance at 800 cm characteristic of
symmetric Si–O–Si oscillations [23]. However, the formation of ␮-
oxo complexes is hardly possible because a catalyst preparation
was performed in anhydrous solvents.
t
Fig. 7. FTIR spectra for SBA-15 (1) and Fe Pc(acac)–Si–SBA-15 (2).

4
4
S.V. Sirotin et al. / Journal of Molecular Catalysis A: Chemical 319 (2010) 39–45
As it has been shown in [24], the Pc molecules adsorbed upon
a matrix surface has on-edge orientation even at low coverages. If
this is the case, the 6th axial coordination position is well accessi-
ble for entering substrate molecules. In contrast, on immobilizing
Fe(III)Pc by coordination such an extra ligation is excluded since
this position is occupied by nitrogen atom of NH -anchor group.
2
Again, the 6th axial position remains unoccupied when the
t
covalent bonding of Fe Pc(acac) via siloxane groups occur (see
Fig. 10), and this results in increase of activity. However,
t
Fe Pc(acac)–Si–SBA-15 sample exhibits noticeably lower activity
t
than the catalyst with on-edge oriented Fe Pc(acac) complexes
though both samples display free axial coordination positions.
Probably, this effect is due to rather long and flexible spacers
t
through which Fe Pc(acac) species are anchored onto the surface.
On the one hand, these spacers are strongly believed to minimize
the influence of support walls on the FePc complexes and thus
to ensure their equal efficiency, regardless of whether they are
located on internal or outer surface. On the other hand, this makes
it possible that two neighboring Fe(III)Pcs form ␮-oxo complex-
like dimers. Such dimers expose no free extra coordination sites
and thereby are inactive. At the same time, these Fe(III)Pc dimers
have been shown to be more active than monomeric FePcs in oxi-
dation of 2,3,6-trimethylphenol oxidation [6] and allylic oxidation
of cyclohexene [25].
t
Fig. 8. UV–vis spectra for catalysts containing Fe Pc(acac) (1) MCM-41;
t
t
t
Data on activity of the catalysts tested in phenol oxidation and
their selectivity to dihydroxylation products are summarized in
Table 2.
(
2) Fe Pc(acac); (3) Fe Pc(acac)/MCM-41; (4) Fe Pc(acac)/NH2–MCM-41; (5)
t
Fe Pc(acac)–Si–SBA-15.
First, as it is seen from these data, the sum of catechol and
hydroquinone yields is substantially lower than the total phenol
conversion. It would be suggested that this difference may originate
from the very fast polymerization of benzoquinone as a primary
product of phenol oxidation to the black tar. However, the for-
mation of benzoquinone on phenol monohydroxylation has been
reported [5] to be suppressed in the presence of any N-containing
compounds, e.g., Pc complexes, and also in acidic media.
Further, there is no substantial difference in product distribu-
tion between the catalysts studied. At the beginning of reaction,
phenol seems to be oxidized in parallel to catechol and benzo-
quinone which was well documented for homogeneous systems
in [26].
Experimental kinetic curves of phenol conversion and the cal-
culated TOF values are presented in Fig. 9. It is to note, the blank
experiments showed the MCM-41 and SBA-15 supports were com-
pletely inactive in oxidation reaction.
t
Free Fe Pc(acac) complex is insoluble in aqueous media, so that
in this case the catalytic process could be unambiguously defined
as heterogeneous or, more exactly, as microheterogeneous one. In
fact, the phenol hydroxylation seems to occur on the outer surface
t
of finely dispersed particles of Fe Pc(acac) but the catalytic activity
as calculated per unit mass is rather small due to low total surface
of such grains.
t
The adsorption of Fe Pc(acac)on MCM-41 leads to dramatic
t
t
increase in activity. Even though the part of Fe Pc(acac) presents
Though the initial activities of all Fe Pc(acac) containing mate-
on the outer surface of support in less active aggregated form,
rials were high enough, they exhibit no activity on their reusing
after separation from reaction products obtained in the first oxi-
dation experiment. This loss in activity is commonly believed to
be resulted from the oxidative degradation of both free and het-
erogenized FePc complexes which occurs not only with H2O2 but
also with a mild oxidant as tert-butylhydroxyperoxide [27]. So that,
UV–vis spectra of the recovered Fe-containing catalysts showed no
adsorption bands characteristic of Pc core. Noteworthy, the Fe(III)
ions of these complexes do not seem to be the key factor of their
t
Fe Pc(acac)/MCM-41 sample is the most active.
t
Unexpectedly, Fe Pc(acac) complexes stabilized upon MCM-41
by coordination with surface amino groups exhibit very poor activ-
t
ity. On the contrary, the anchoring Fe Pc(acac) species via covalent
bonding results in appreciable increase of activity. These features
t
of immobilized Fe Pc(acac) are most probably due to specific envi-
ronment of central Fe atom in Pc molecules as it is illustrated in
Fig. 10.
t
t
t
Fig. 9. Oxidation of phenol over Fe Pc(acac) containing catalysts. (A) Curves of phenol conversion; (B) TOF values for samples (1) Fe Pc(acac); (2) Fe Pc(acac)/MCM-41; (3)
t
t
t
◦
Fe Pc(acac)/NH2–MCM-41; (4) Fe Pc(acac)–Si–SBA-15 (1 g phenol, 10 ml H2O, 2 ml 25% H2O2, 10 mg cat. or corresponding amount of Fe Pc(acac), 60 C).

S.V. Sirotin et al. / Journal of Molecular Catalysis A: Chemical 319 (2010) 39–45
45
t
t
t
Fig. 10. Location of FePcs on the support (1) Fe Pc(acac)/MCM-41; (2) Fe Pc(acac)/NH2–MCM-41; (3) Fe Pc(acac)–Si–SBA-15 (L = acac ligand, tert-butyl substituents of Pc are
omitted for clarity).
Table 2
t
Total phenol conversion and yields of dihydroxylation products on Fe Pc(acac) containing catalysts.
t
t
t
t
Catalyst and reaction time
Fe Pc(acac)
Fe Pc(acac)/MCM-41
Fe Pc(acac)/NH2-MCM-41
Fe Pc(acac)–Si–SBA-15
15 min
120 min
15 min
120 min
15 min
120 min
15 min
120 min
Phenol conversion, %
Catechol yield, %
Hydroquinone yield, %
9
0
0
26
8
1
30
13
3
55
23
9
2
1
0
19
5
0
7
0
0
33
12
0
autoxidation. In fact, we found Zn-containing polymerized Pc com-
plex to be stable toward H O . The synthesized Pc compounds that
were used for catalyst preparations bear electron-donor substitutes
and on being heterogenized show poor stability towards oxidation
References
2
2
[
[
[
1] N. Grootboom, T. Nyokong, J. Mol. Catal. A 179 (2002) 113.
2] O.V. Zalomaeva, O.A. Kholdeeva, A.B. Sorokin, C. R. Chimie 10 (2007) 598.
3] M. Alvaro, E. Carbonell, M. Esplá, H. Garcia, Appl. Catal. B 57 (2005) 37.
by H O .
[4] D.E. De Vos, P.A. Jacobs, Catal. Today 57 (2000) 105.
2
2
[
[
5] C.W. Lee, D.H. Ahn, B. Wang, J.S. Hwang, S.-E. Park, Micropor. Mesopor. Mater.
4–45 (2001) 587.
6] A.B. Sorokin, Alain Tuel, Catal. Today 57 (2000) 45.
One may expect that in contrast to these complexes with
electron-donor groups, FePcs with electron–acceptor substituents
would be more oxidation resistant [28] and thereby more stable
under the conditions of oxidation by hydrogen peroxide. However
this possibility requires further investigations.
4
[7] S. Mangematin, A.B. Sorokin, J. Porphyrins Phthalocyanines 5 (2001) 674.
[
8] M. Grün, K.K. Unger, A. Matsumoto, K. Tsutsumi, Micropor. Mesopor. Mater. 27
1999) 207.
9] J.M. Kim, Y.-J. Han, B.F. Chmelka, G.D. Stucky, Chem. Commun. (2000) 2437.
(
[
[
[
10] A. Cauvel, J. Renard, D. Brunel, J. Org. Chem. 62 (1997) 749.
11] L.G. Tomilova, E.V. Chernykh, V.I. Gavrilov, I.V. Shelepin, V.M. Derkacheva, E.A.
Luk’yanets, Zhurnal Obshcheiy Khimii (USSR) 52 (1982) 2304.
12] A.Yu. Tolbin, V.E. Pushkarev, G.F. Nikitin, L.G. Tomilova, Tetrahedron Lett. 50
(34) (2009) 4848.
4
. Conclusions
[
t
Fe(III) Pc(acac) adsorbed upon MCM-41 and covalently grafted
onto SBA-15 show high activity in phenol oxidation because of
easy accessibility of unoccupied axial position of Fe(III) atom in
[13] H. Juvaste, E.I. Iiskola, T.T. Pakkanen, J. Mol. Catal. A 150 (1999) 1.
[
[
14] R.S.S. Murthy, D.E. Leyden, Anal. Chem. 58 (1986) 1228.
15] K.C. Vrancken, P. Van Der Voort, I. Gillis-D’Hamers, E.F. Vansant, J. Chem. Soc.,
Faraday Trans. 88 (1992) 3197.
t
the phthalocyanine molecule. It is not the case when Fe Pc(acac)
forms the coordination bond with the grafted NH2 groups
which results in low activity. All Fe Pc(acac) containing catalysts
were shown to undergo fast oxidative degradation by hydrogen
peroxide.
[16] F. Fajula, D. Brunel, Micropor. Mesopor. Mater. 48 (2001) 119.
[17] W.A. Carvalho, M. Wallau, U. Schuchardt, J. Mol. Catal. A 144 (1999) 91.
t
[
[
18] T. Sasaki, C. Zhong, M. Tada, Y. Iwasawa, Chem. Commun. (2005) 2506.
19] D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka, G.D. Stucky,
Science 279 (1998) 548.
[
20] A. Galarneau, H. Cambon, F. Di Renzo, R. Ryoo, M. Choi, F. Fajula, New. J. Chem.
2
7 (2003) 73.
Acknowledgements
[
[
[
21] T. Buck, D. Wöhrle, G. Schulz-Ekloff, A. Andreev, J. Mol. Catal. A 70 (1991) 259.
22] A. Hadasch, A. Sorokin, A. Rabion, B. Meunier, New. J. Chem. 22 (1998) 45.
23] V.L. Zolobenko, S.M. Holmes, C.S. Cundy, J. Dwyer, Micropor. Mater. 11 (1997)
This work was supported by the Russian Foundation for Basic
Research, project Nos. 08-03-00544, 08-03-00753, the Federal
Special Program, grant 2008-10-1.3-07-47, and the Program of fun-
damental research of Russian Academy of Sciences “Development
of methods for synthesizing chemical compounds and creating new
materials”. S.V. Sirotin also acknowledges Haldor Topsoe A/S for
financial support. Authors thank Dr. Yu.G.Kolyagin for NMR mea-
surements and Dr. I.V. Kolesnik for assistance in SBA-15 synthesis.
83.
[
[
[
24] B.V. Romanovsky, O.M. Zakharova, V.Yu. Zakharov, R.E. Mardaleyshvili, Vestn.
Moscovsk. Un-ta. Ser. Khimiay 19 (1978) 524.
25] L.M. González, A.L. Villa de, P.C. Montes de, C.A. Sorokin, Tetrahedron Lett. 47
(
2006) 6465.
26] V.N. Sapunov, I.Yu. Litvintsev, A.I. Mikhailyuk, Kinetika i Kataliz 39 (1998) 365.
[27] S. Ernst, R. Gläser, M. Selle, Stud. Surf. Sci. Catal. 105 (1997) 1021.
28] I.V. Zhukov, V.E. Pushkarev, L.G. Tomilova, N.S. Zefirov, Russ. Chem. Bull., Int.
Ed. 54 (1) (2005) 189.
[