Combination of two catalytic sites in a novel nanocrystalline TiO 2-iron tetrasulfophthalocyanine material provides better catalytic properties

Article abstract of DOI:10.1039/b507211e

Mesoporous titania nanocrystals containing iron tetrasulfophthalocyanine (FePcS) have been synthesised by a one-pot hydrolytic process from a modified Ti alkoxide; the novel hybrid catalyst was efficient in heterogeneous oxidation of 2,3,6-trimethylphenol and β-isophorone suggesting a cooperative effect between TiO₂ and FePcS catalytic sites. The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2005.

Full text of DOI:10.1039/b507211e

View Article Online / Journal Homepage / Table of Contents for this issue
L E T T E R
Combination of two catalytic sites in a novel nanocrystalline
TiO –iron tetrasulfophthalocyanine material provides better
catalytic properties
2
´
Mirvat Beyrhouty, Alexander B. Sorokin,* Stephane Daniele* and
Liliane G. Hubert-Pfalzgraf
Institut de Recherches sur la Catalyse, UPR 5401, Universite´ Claude Bernard Lyon 1,
avenue A. Einstein, 69626 Villeurbanne Cedex, France. E-mail: daniele@catalyse.cnrs.fr
2
Received (in Montpellier, France) 23rd May 2005, Accepted 7th July 2005
First published as an Advance Article on the web 8th August 2005
Mesoporous titania nanocrystals containing iron tetrasul-
fophthalocyanine (FePcS) have been synthesised by a one-pot
hydrolytic process from a modified Ti alkoxide; the novel hybrid
catalyst was efficient in heterogeneous oxidation of 2,3,6-tri-
methylphenol and b-isophorone suggesting a cooperative effect
The FT-IR analyses of 2 showed the absorption peaks
related to nOH and new nSO stretching vibrations between
3
ꢀ
1
1260 and 1000 cm (in comparison to [FePc(SO
reveals that sulfonic acid groups are chemisorbed as sulfonates
onto the TiO nanoparticles. XPS spectra recorded from 0 to
1200 eV indicated that the hybrid material contained iron
3 4
H) ]Cl). This
2
2
between TiO and FePcS catalytic sites.
[
Fe2p 3/2 ¼ 709.1 eV], titanium [Ti2p 3/2 ¼ 458.8 eV], nitrogen
Nano-sized functional hybrid materials have become a
highly innovative research field. The ability to tailor the
organic part combined with the properties of the nano-sized
inorganic matrixes is of great interest for potential applications
in electronics, optics and catalysis. In particular, an association
of a metal complex with a catalytically active inorganic support
could provide a catalyst having improved catalytic properties
due to the combination of two different catalytic sites in one
material.
[N1s ¼ 400.3 (50%), 398.8 (50%) eV], sulfur [S2p ¼ 167,3 eV],
oxygen [O1s ¼ 532.1 (18%), 530.2 (82%) eV], carbon [C1s ¼
284.6 eV] and an atomic ratio Ti2p/Fe2p ¼ 23. The two O1s
2ꢀ
peaks at 532.1 and 530.2 eV are ascribed to the O contribu-
tions of sulfonate and oxide ligands, respectively. The binding
energy values of the O1s and S2p peaks of the sulfonate groups
are different in comparison to the free ligand [O1s ¼ 531.1 eV,
S2p ¼ 168,1 eV], suggesting SO
3 2
covalent bonding onto TiO .
7
According to the assignment made by Gosh et al., based on
all-electron ab initio HF calculations on the metal-free phtha-
locyanine, the N1s peak at 400.3 eV is assigned to the four
pyrrolic nitrogens (the inner ones bonded with two carbons
and the central Fe atom). The absence of a chlorine peak
indicates that under sol–gel conditions it was replaced by
hydroxo or oxo ligands. Diffuse reflectance UV–Vis spectro-
scopy of 2 confirmed the presence of intact tetrasulfophthalo-
cyanine. The intense Q-band at 641 nm suggests that
phthalocyanine exits predominantly in dimeric form while
the shoulder at 696 nm could be attributed to the monomer
form. Such dimerisation has already been observed for alumi-
Recently we reported the synthesis of nanocrystalline parti-
cles of titania at low temperature (100 1C) by a sol–gel process
using titanium alkoxide and bromide ammonium salts as
1
catalysts. This synthetic approach could allow to introduce
acid-containing compounds into TiO particles via the synth-
2
esis of hybrid nanoparticles from modified titanium alkoxides.
Taking into account the catalytic properties of FePcS in the
2
selective oxidation of alkynes and aromatic compounds as
3
4
well as in the oxidative degradation of organic pollutants this
complex containing four sulfonic groups seemed to be a
suitable candidate. Photocatalytic degradation studies of pol-
lutants using either Co(II)-tetrasulfophthalocyanine grafted on
8
nium tetrasulfophthalocyanine chloride in sol–gel processing.
ꢀ1
5
TiO2 via a silane reagent or polycrystalline TiO2 samples
6
impregnated with Cu(II)-phthalocyanine were reported in the
The loading of the hybrid material (214 mmol g or 19.2 wt%
of phthalocyanine), determined by ICP-MS on iron, is high
literature. However, to the best of our knowledge, none
stemmed from direct covalent bonding of functionalized
phthalocyanine on nanoparticulate TiO2 material. In this
work, we report a practical one-pot sol–gel preparation of
surface-modified nanocrystalline titania particles from hetero-
leptic alkoxide and their catalytic activities in the aerobic
oxidation of b-isophorone.
and could be easily modulated by adding various equivalents
during the hydrolysis. XRD patterns, obtained at
i
of Ti(OPr )
4
ꢀ
1
the scan rate of 0.021 2y s , show that 2 contains anatase
(around 85%) as the main crystalline phase along with broo-
kite. Average particle sizes, estimated from line-broadening
according to the Scherrer equation, gave 5.4 and 12 nm, for
anatase and brookite phases, respectively. The morphology of
the particles was examined by transmission electron micro-
scopy and revealed highly agglomerated nanocrystals of 6–8
nm in length (Fig. 2).
Iron(III) tetrasulfophthalocyanine chloride [FePc(SO
was used to substitute alkoxide ligands (Fig. 1). The reaction
3 4
H) ]Cl
i
between [FePc(SO
2
3
H)
-propanol for 12 h resulted in a blue-green crystalline material
H) ]Cl). Elemental analysis
4 4
]Cl and excess of Ti(OPr ) in refluxing
Fig. 3 shows the nitrogen adsorption–desorption isotherms
and the corresponding pore size distribution curve (inset) for
desorbed sample 2 (200 1C/6 h). The BET surface area was 250
1
(yield ¼ 60% based on [FePc(SO
3
4
gave a ratio Ti/Fe ¼ 40. UV–Vis and FT-IR spectra indicated
that the structure of phthalocyanine is retained upon modifica-
tion. 1 was hydrolysed in pure water in the presence of
tetrabutylammonium bromide salt to give material 2 which
was characterised by elemental analysis, FT-IR, XRD, TEM,
BET, UV–Visible and XPS techniques.
ꢀ
1
ꢀ1
2
m g (243 m g for TiO obtained by the same procedure)
3
1
and the total pore volume was 0.33 cm g (determined by BJH
method). The isotherm, a type IV with a hysteresis of type H2,
shows that 2 is a well-characterised ordered mesoporous
2
system with interconnected particles, different from TiO (less
T h i s j o u r n a l i s & T h e R o y a l S o c i e t y o f C h e m i s t r y a n d t h e
C e n t r e N a t i o n a l d e l a R e c h e r c h e S c i e n t i f i q u e 2 0 0 5
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 2 4 5 – 1 2 4 8
1245

View Article Online
2
Fig. 1 Preparation of FePcS–TiO (2).
rigid texture). This self-assembly process gave an average pore
diameter with a narrow distribution of about 5 nm, compar-
able with the size of the phthalocyanine ligand (2 ꢁ 2 nm).
The catalytic activity of 2 has been determined in the TBHP
oxidation of 2,3,6-trimethylphenol (TMP) into 2,3,6-trimethyl-
quinone (TMQ), a precursor of vitamin E industrially obtained
mol% of FePc 55% yield of KIP was obtained. Manganese
porphyrin complexes were the most efficient catalysts to pro-
vide 93% KIP yield. Recently, Murphy and Baiker reported an
efficient homogeneous oxidation of b-IP in the presence of
Cu(salen) (100% conversion, 67% yield of KIP). However,
only 52% conversion and less than 6% KIP yield was obtained
1
1
by p-sulfonation with H
tion with MnO . While only 20% conversion and 1% TMQ
2
SO
4
followed by stoichiometric oxida-
2
in the best case with the Cu(salen) catalyst supported on SiO .
2
This very low activity and poor selectivity in the heterogeneous
oxidation were explained by the different reaction pathways
yield were obtained with TiO2 after 2 h, 2 provided 96%
conversion and 57% yield of TMQ under identical reaction
1
2b
imposed by the isolated immobilised catalytic sites.
Addi-
9
conditions. Using H O , 2 was less active (44% conversion)
tionally, metal(salen) complexes were not stable against leach-
ing of the metal from the solid silica matrix.
We were delighted to observe that b-IP was successfully
2
2
but high selectivity (84%) was achieved. In the presence of
TiO , TMP conversion was only 15%. It should be noted that
conventional FePcS–SiO catalysts were not active in oxidation
2
2
oxidised by O
isomerisation product formed from the enolate intermediate.
In the presence of FePcS–SiO b-IP was oxidised by dioxygen
2
to give KIP, allylic alcohol 3 and a-IP, the
3
of TMP using H O as oxidant. These data show that
14
2
2
combination of two catalytically active FePcS and Ti-based
sites in one material could result in co-operative catalysis
providing an improved catalyst.
2
to afford 38% yield of KIP (Table 1). In the presence of TiO2
only 2% yield of KIP was obtained, a-IP being the main
product. FePcS supported on TiO2 performed much better
than FePcS supported on SiO
yield. It is notable that the combined yield of KIP and 3 was
8% for 2 (FePcS–TiO ) compared to 45% for FePcS–SiO
2 2
or TiO alone giving 57% KIP
7
2
2
suggesting cooperativity of the two catalytic sites in the oxi-
dation.
Taking into account the importance of recyclability of
heterogeneous catalysts we studied the recycling of 2 and the
possibility of metal leaching into solution at 1 mmol b-IP scale
and 10 mL reaction mixture. After completing the first run the
catalyst 2 was isolated by centrifugation, washed with MeCN
(2 ꢁ 25 ml) and dried. On reuse, we observed a gradual
decrease of catalytic activity, 77 and 64% conversion in runs
2 and 3, respectively (Table 1). The yield of KIP was also lower,
33% (run 2) and 27% (run 3). In order to determine possible
Fe leaching in solution we isolated the solution after run 1 by
filtration (0.45 mm filter). The residual Fe content in this
solution of 7 ppm indicated that 9.9% of fixed FePcS was
dissolved in the reaction mixture. However, taking into ac-
count the very small particle size of 2 we can not exclude the
In order to confirm the superior catalytic properties of 2 we
studied the aerobic oxidation of b-isophorone (b-IP), an im-
portant intermediate for the preparation of flavors and fra-
grances fine chemicals. The transformation of b-IP to epoxide
1
0
11
has been described. Several homogeneous and heteroge-
1
2
neous catalysts for the oxidation of b-IP to ketoisophorone
KIP) have been published. Homogeneous phthalocyanine and
porphyrin complexes of different transition metals were used
(
1
3
3
for aerobic b-IP oxidation in the presence of Et N. With 1.3
Fig. 2 TEM image of 2 obtained on a JEOL 2010 electron micro-
scope.
Fig. 3 Nitrogen adsorption–desorption isotherms of samples 2 (m)
and TiO (J) and pore size distribution vs. radius (inset).
2
1
246
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 2 4 5 – 1 2 4 8

View Article Online
a
Table 1 Heterogeneous oxidation of b-IP
molecular self-assembly hydrolytic process which results in a
new hybrid material with well-ordered mesoporous structure
and high surface area. This catalyst containing two different
catalytic sites exhibited promising results in heterogeneous
aerobic oxidation of cyclic olefins exemplified by b-isophorone.
b
Yield (% 3)
Catalyst
Conversion (%)
KIP
a-IP
2
2
2
2
2
(FePcS–TiO
(run 2)
2
)
99
77
64
68
77
97
92
57
33
27
29
40
2
21
9
10
16
20
15
17
46
22
(run 3)
c
6
7
d
Experimental
9
TiO
2
4
All manipulations were performed under dry argon using
was purchased from Al-
i
Schlenk tube techniques. Ti(OPr )
FePcS–SiO
2
38
7
4
a
Reaction conditions: substrate (0.1 mmol), catalyst (15 mg, 1 mol%
to substrate), triethylamine (0.05 mmol), dimethyl sulfoxide (1 mL),
drich. Solvents were purified by standard methods and stored
over molecular sieves. FT-IR spectra were recorded as Nujol
mulls on a Perkin-Elmer Paragon 500 FT-IR spectrometer.
UV–Visible spectra were carried out on an UNICAM UV2-
100 spectrometer. The DR UV–Vis spectra were recorded on a
Lambda 35 Perkin-Elmer spectrophotometer. BET measure-
ments were performed on a Micromeritics ASAP 2010. Powder
X-Ray diffraction data were obtained with a Siemens D 5000
diffractometer using the CuKa radiation. XPS experiments
were performed with an Escalab 200R (VG Scientific) spectro-
meter using the monochromated AlKa radiation as excitation
source. TEM images were collected on a JEOL 2010 micro-
scope. Analytical data were determined by an inductively
coupled plasma–mass spectrometry method.
b
c
6
2
0 1C, O atmosphere, 24 h. Determined by GC. After 5 h of
reaction, the solid catalyst was separated by filtration and the solution
d
was allowed to react for 19 h (total reaction time ¼ 24 h). Subsequent
oxidation run with the catalyst separated after 5 h of reaction.
presence of the smallest catalyst particles in the solution after
filtration. In a separate experiment the catalyst was filtered
after 5 h of reaction when the conversion was 48%, KIP yield
was 25% and a-IP yield was 4%. The solution was allowed to
react further. After 19 h (24 h total reaction time) the conver-
sion was 68% (þ20% as compared with 5 h reaction time). The
yield of KIP increased from 25 to only 29% (þ4%) while the
yield of a-IP increased from 4% to 15% (þ11%). Comparison
of these data with those obtained from standard oxidation
Materials synthesis
(
(
Table 1, run 1) shows significant decreases of catalytic activity
ꢀ31%) and KIP yield (ꢀ28%) after removal of the supported
Synthesis of anhydrous [FePc(SO
FePcS (600 mg in 50 mL of water), prepared according to
the modified Weber–Busch procedure, was converted to the
tetrabutylammonium salt by the treatment with 5 mL of a 40%
tetrabutylammonium hydroxide aqueous solution followed by
H) ]Cl. Sodium salt of
3 4
1
5
catalyst, and 5% increase in a-IP yield. The recycled catalyst
was used for the subsequent cycle. Similarly to run 2, in this
second oxidation the conversion was 77%, the yield of KIP
was 40% and the yield of a-IP was 17% thus demonstrating a
good reproducibility of recycling. These results suggest that the
oxidation of b-IP to KIP occurs mainly at the supported
catalyst and the reaction of isomerisation of b-IP to a-IP
occurs in solution containing triethylamine base. The FePcS–
the extraction with CH
2
Cl
2
(6 ꢁ 50 ml). The organic phase was
dried and the obtained material was dissolved in 200 mL of
acetonitrile. 37% HCl (2 mL, about 6-fold excess with respect
to the sulfonate group) was added dropwise under stirring. The
precipitate of [FePc(SO
and dried at 60 1C for 24 h in vacuo. [FePc(SO
further dried by the azeotropic distillation of ethanol. Finally,
3
H)
4
]Cl was isolated by centrifugation
H) ]Cl was
TiO catalyst recovered after three successive oxidations ex-
2
3
4
hibited practically the same diffuse reflectance (DR) UV–Vis
spectrum as that of the initial supported catalyst indicating no
significant degradation (Fig. 4).
Although further work is still required to understand the
mechanism of synergistic action, to optimise this novel oxida-
tion catalyst in terms of selectivity and scope and to improve its
recyclability, the KIP yield already obtained with 2 provides
the basis for a heterogeneous oxidation of such an important
substrate as b-isophorone.
the product was dried in vacuo at 60 1C for 24 h (435 mg, 91%
ꢀ
1
yield). FT-IR (Nujol, cm ): 3423br, 3173 [nO–H]; 1771w,
1715m, 1604m, 1504m [n(CQN, CQC)]; 1330w, 1306w;
1258w, 1227m, 1182s, 1169s, 1147s, 1107s, 1089s, 1052m,
3
1027s [n(SO )]; 963m, 929m, 844w, 762w, 748m, 700s, 677w,
650s, 630s, 598m, 591m, 566m, 543w.
Synthesis of 1. In a typical preparation, 0.74 g (2.6 mmol) of
and 0.194 g (0.21 mmol) of [FePc(SO
In summary, iron(III) phthalocyanine was successfully cova-
i
Ti(OPr )
4
3
H)
4
]Cl were
2
lently grafted onto crystalline TiO nanoparticles by a simple
refluxed in 2-propanol for 12 h. After filtration, the resultant
dark green solution was concentrated under vacuum. Cooling
it down to ꢀ4 1C resulted in a dark blue-green crystalline
material 1 (0.85 g, 60%/Fe). Anal. Found: Fe, 0.45; Ti,
one-pot sol–gel process. This method avoids preliminary mod-
ifications of either the support or the metal complex that are
usually necessary in conventional procedures. Tetrasul-
fophthalocyanine acts as a bridging ligand and controls the
ꢀ
1
1
1
1
6
5.15%. FT-IR (Nujol, cm ): 1775w, 1772m, 1685w, 1606m,
581w, 1512m, 1463w [n(CQN, CQC)]; 1336w, 1325m,
308w; 1263w, 1227w, 1186w, 1161w, 1146m [n(SO )]; 702w,
96w, 648m, 632m, 601w, 595w, 567w, 524w, 508w, 468w
3
[
n(M–O, M–N)].
Synthesis of 2. In a typical hydrolysis process, 1 g of 1 (83.1
mmol) in 5 mL of 2-propanol was added dropwise to 25 mL of a
n
4
N( Bu) Br (0.4 g, 1.24 mmol) aqueous solution under vigorous
stirring at boiling temperature. The mixture was heated under
reflux for 2 h and was centrifuged to give a dark blue-green
solid. The as-prepared precipitate (denoted 2) was washed
several times with deionised water, until the solution was
colorless, and ethanol (25 mL) and was dried at 70 1C over-
ꢀ
1
Fig. 4 DR UV–Vis spectra of initial FePcS–TiO
oxidations (light line).
2
(bold line) and after
night. Anal. Found: Fe, 1.20%. FT-IR (KBr, cm ): 3342br
[nO-H]; 1722m, 1617s, 1539m, 1481w, 1401m [n(CQN,
3
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 2 4 5 – 1 2 4 8
1247

View Article Online
CQC)]; 1254w, 1218w, 1187m, 1141w, 1101w, 1072w, 1032m
n(SO )]; 721s, 553vs, 478vs [n(M–O, M–N)].
References
[
3
1
G. Goutailler, C. Guillard, S. Daniele and L. G. Hubert-Pfalzgraf,
J. Mater. Chem., 2003, 13, 342, and references therein.
C. Perollier and A. B. Sorokin, Chem. Commun., 2002, 1548.
´
(a) A. B. Sorokin and A. Tuel, New J. Chem., 1999, 23, 473; (b) A.
B. Sorokin, S. Mangematin and C. Pergrale, J. Mol. Catal. A:
Chem., 2002, 182–183, 267; (c) S. V. Barkanova and O. L. Kaliya,
J. Porphyrins Phthalocyanines, 1999, 3, 180.
Catalytic tests
2
3
The reaction products were identified and quantified by GC-
MS (Hewlett-Packard 5973/6890 system; electron impact ioni-
zation at 70 eV, He carrier gas, 30 m ꢁ 0.25 mm cross-linked
4
(a) A. B. Sorokin, J.-L. Se
´
ris and B. Meunier, Science, 1995, 268,
163; (b) A. B. Sorokin and B. Meunier, Chem. Eur. J., 1996, 2,
5
5
0
% PH ME siloxane (0.25 mm coating) capillary column, HP-
1
1
4
MS) and GC (Agilent 4890D system, N carrier gas, 15 m ꢁ
2
308; (c) B. Meunier and A. B. Sorokin, Acc. Chem. Res., 1997, 30,
70; (d) X. Tao, W. Ma, J. Li, Y. Huang, J. Zhao and J. C. Yu,
.25 mm crosslinked 5% PHME siloxane (0.25 mm coating)
capillary column, HP-5MS) methods.
Chem. Commun., 2003, 80; (e) M. Bressan, N. d’Alessandro, L.
Liberatore and A. Morvillo, Coord. Chem. Rev., 1999, 185–186,
3
85.
Typical procedure for the oxidation of TMP. 2 (8 mg, 2
mol%) was added to a solution of TMP (80 mmol) in 1,2-
dichloroethane (DCE, 4 mL), then four 15.8 mL portions of a
5
6
7
A. P. Hong, D. W. Bahnemann and M. R. Hoffmann, J. Phys.
Chem., 1987, 91, 2109.
G. Mele, G. Ciccarella, G. Vasapollo, E. Garcia-Lopez, L.
Palmisano and M. Schiavello, Appl. Catal. B, 2002, 38, 309.
A. Gosh, J. Fitzgerald, P. G. Gassman and J. Almlof, Inorg.
¨
3
.8 M TBHP in DCE were added to the mixture at reaction
times of 0, 0.5, 1, 1.5 h (total oxidant amount: 240 mmol). The
reaction was carried out at 30 1C for 2 h. The products were
identified by GC-MS and quantified by GC using authentic
samples.
Chem., 1994, 33, 6057.
8
9
H. Zhan, W. Chen and M. Wang, Mater. Lett., 2003, 57, 1108.
For TMP oxidation on Ti-based catalysts, see: O. A. Kholdeeva,
N. N. Trukhan, M. P. Vanina, V. N. Romannikov, V. N. Parmon,
J. Mrowiec-Bialon and A. B. Jarzebski, Catal. Today, 2002, 75,
2
03, and references therein.
Typical procedure for the oxidation of b-IP. Catalyst (1
mol%) was added to a solution of b-IP (0.1 mmol) in dimethyl
sulfoxide (1 ml) containing 0.05 mmol of triethylamine. The
reaction was performed at 60 1C under oxygen atmosphere (2
bar) for 24 h. The products were identified by GC-MS and
quantified by GC using authentic samples. Recycling experi-
ments were carried out with 1 mmol b-IP keeping the same
reagent ratio.
1
0
(a) R. Hutter, T. Mallat and A. Baiker, J. Catal., 1995, 157, 665,
and references therein; (b) T. Honma, M. Nakajo, T. Mizugaki, K.
Ebitani and K. Kaneda, Tetrahedron Lett., 2002, 43, 6229.
11 E. F. Murphy, M. Schneider, T. Malat and A. Baiker, Synthesis,
2001, 547.
1
2
(a) E. F. Murphy, T. Malat and A. Baiker, Catal. Today, 2000, 57,
1
15; (b) E. F. Murphy and A. Baiker, J. Mol. Catal. A: Chem.,
002, 179, 233, and references therein.
2
1
1
3
4
N. Ito, T. Etoh, H. Hagiwara and M. Kato, Synthesis, 1997, 153.
M. Constantini, A. Dromard, M. Jouffret, B. Brossard and J.
Varagnat, J. Mol. Catal., 1980, 7, 89.
(a) J. H. Weber and D. H. Busch, Inorg. Chem., 1965, 4, 469; (b) A.
Hadasch, A. B. Sorokin, A. Rabion and B. Meunier, New J.
Chem., 1998, 22, 45.
Acknowledgements
1
5
The authors thanks Mr W. Desquesnes for his help in inter-
pretation of nitrogen adsorption–desorption isotherms.
1
248
N e w J . C h e m . , 2 0 0 5 , 2 9 , 1 2 4 5 – 1 2 4 8