Clean catalytic oxidation of 8-hydroxyquinoline to quinoline-5,8-dione with tBuOOH in the presence of covalently bound FePcS-SiO2 catalysts

Article abstract of DOI:10.1039/b923674k

The clean catalytic oxidation of 8-hydroxyquinoline (8-HQ) with tert-butyl hydroperoxide to quinoline-5,8-dione (QD), a molecular framework fragment of antitumor compounds, over silica-supported iron tetrasulfophthalocyanine catalysts (FePcS) is reported. The pronounced influence of the FePcS state (monomer vs. dimer) and the support (amorphous SiO₂vs. mesoporous MCM-41) on the catalytic activity and selectivity is revealed. Depending on the catalyst structure, turnover frequency values determined from the initial rates of 8-HQ consumption varied from 215 to 3570 h^-1.The effects of solvent, temperature, reagent concentrations and catalyst amounts on the substrate conversion and QD selectivity were studied to optimize the reaction conditions. With an optimal catalyst, the yield of the target product reached 66%. The truly heterogeneous nature of the catalysis was also demonstrated.

Full text of DOI:10.1039/b923674k

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/greenchem | Green Chemistry
Clean catalytic oxidation of 8-hydroxyquinoline to quinoline-5,8-dione with
tBuOOH in the presence of covalently bound FePcS–SiO2 catalysts
Olga V. Zalomaeva,a Alexander B. Sorokin*b and Oxana A. Kholdeevaa
Received 13th November 2009, Accepted 15th March 2010
First published as an Advance Article on the web 22nd April 2010
DOI: 10.1039/b923674k
The clean catalytic oxidation of 8-hydroxyquinoline (8-HQ) with tert-butyl hydroperoxide to
quinoline-5,8-dione (QD), a molecular framework fragment of antitumor compounds, over
silica-supported iron tetrasulfophthalocyanine catalysts (FePcS) is reported. The pronounced
influence of the FePcS state (monomer vs. dimer) and the support (amorphous SiO2 vs.
mesoporous MCM-41) on the catalytic activity and selectivity is revealed. Depending on the
catalyst structure, turnover frequency values determined from the initial rates of 8-HQ
consumption varied from 215 to 3570 h-1.The effects of solvent, temperature, reagent
concentrations and catalyst amounts on the substrate conversion and QD selectivity were studied
to optimize the reaction conditions. With an optimal catalyst, the yield of the target product
reached 66%. The truly heterogeneous nature of the catalysis was also demonstrated.
of 5-hydroxyquinoline (5-HQ) and 5-/8-aminoquinoline by
Introduction
bis(trifluoroacetoxy)iodobenzene (2.2 equiv.), with a quinone
Quinoline-5,8-dione (QD, Scheme 1) is a structural fragment of
yield of up to 88%.22 However, the low active oxygen content in
compounds that possess a wide spectrum of biological activity,
this oxidant (3.8%) leads to significant amounts of waste (the
and can be used as antitumor, antimalarial, antibacterial and
typical E-factor for these oxidation reactions is in the range 15–
antifungal drugs.1–9 Aminoquinone antibiotics, such as strepton-
25). On the other hand, several methods for the preparation of
igrin (bruneomycin) and lavendamycin (Scheme 1), are known
QD based on the oxidation of 8-hydroxyquinoline (8-HQ) have
to be effective agents for the treatment of cancers.8,10–13 A large
been also reported.23,24 The stoichiometric oxidation of 8-HQ to
number of studies have shown the crucial role of the quinoline-
QD with NaNO2, Na2S2O4 or K2CrO4 is a multi-step process,
5,8-dione ring in the antitumor activity of these drugs.1,14–17
producing the target product with a very low yield (16%).23a
Furthermore, QD has been used as a starting material in the
Potassium nitrosodisulfonate can also be applied for oxidation
synthesis of a series of antitumor agents developed by Bolognese
of 8-HQ.23b,23c
and co-authors.18–21
Substituted quinoline-5,8-quinones have been obtained by
the photooxygenation of substituted 8-HQ in the presence of
tetraphenylporphyrin in 50–82% yields.23d Sensitized photooxi-
dation of 8-HQ or 5-HQ has afforded QD in 64–70% yields.23e
So far, the only one-pot catalytic method for QD synthesis via
8-HQ oxidation has been published by Chauhan et al.24 The
authors used 5,10,15,20-tetraarylporphyriniron(III) chlorides as
catalysts and hydrogen peroxide as the oxidant, the yield of QD
being low to moderate (30%).
It has been shown that iron tetrasulfophthalocyanine (FePcS)
covalently bound to silica supports provides efficient catalysts
for the selective oxidation of aromatic compounds, especially
for the oxidation of functionalized phenols and naphthols with
tBuOOH to their corresponding quinones.25–29 The efficiency
of the iron porphyrins as catalysts in the oxidation of 8-HQ
demonstrated by Chauhan et al.24 has prompted us to suggest
that iron phthalocyanines, having a similar structure, could also
catalyze the oxidation of 8-HQ.
Scheme 1
Several approaches to the preparation of QD have been
reported in the literature, emphasizing the synthetic interest in
this transformation. The most effective procedure, suggested
by Barret and Daudon, involves the stoichiometric oxidation
In the present work, we report the oxidation of 8-HQ to QD in
the presence of FePcS catalysts that were supported on silica in
monomer or dimer forms (Fig. 1) using environmentally friendly
and economically attractive BuOOH and H2O2 oxidants. The
effects of the structure of the active site and the nature of
the support on the catalytic activity and selectivity have been
aBoreskov Institute of Catalysis, Pr. Ac. Lavrentieva 5, Novosibirsk,
630090, Russia. E-mail: khold@catalysis.ru; Fax: +7 383 330 8056;
Tel: +7 3833269433
bInstitut de Recherches sur la Catalyse, CNRS, 2, avenue A. Einstein,
69626, Villeurbanne Cedex, France.
t
E-mail: alexander.sorokin@ircelyon.univ-lyon1.fr; Fax: +33 472445399;
Tel: +33 472445337
1076 | Green Chem., 2010, 12, 1076–1082
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Table 1 Physicochemical properties of the Fe-containing catalysts
studied
Sample
Fe (wt%) Sa/m2 g-1 VMeb/cm3 g-1 dMec/nm
d
d
d-FePcS–SiO2 (1)
d-FePcS–SiO2 (2)
m-FePcS–SiO2
d-FePcS–MCM-41 0.2
m-FePcS–MCM-41 0.2
0.3
0.2
0.2
185
185
185
860
712
951
—
—
—
—
—
—
1.3
0.98
0.49
3.9
3.6
3.0
Fe-MMM-2
1.8
a BET surface area. b Specific mesopore volume. c Average mesopore
diameter. d Non-porous material.
Fig. 1 A schematic representation of FePcS supported on silica and
MCM-41 in monomer and dimer forms.
studied, and the reaction conditions optimized. The nature of
the catalysis in this system has also been studied.
Results and discussion
Catalyst preparation and characterization
It is known that iron phthalocyanine complexes can exist in
monomeric or m-oxo-dimeric forms, depending on the condi-
tions of their synthesis (solvent, concentration, etc.).30
Fig. 2 (A) DR UV-vis spectra of (a) d-FePcS–SiO2 (1), (b) d-FePcS–
SiO2 (2), (c) d-FePcS–MCM-41, (d) m-FePcS–MCM-41 and (e) m-
FePcS–SiO2. (B) The UV-vis spectrum of FePcS(Bu4N)4 (10-5 M) in
acetone.
Catalysts supported on amino-modified SiO2 containing
FePcS preferably exist in the m-oxo-dimeric form (d-FePcS–
SiO2, samples 1 and 2) and were prepared by two different
methods. Catalyst 1 was synthesized by the reaction of a
sulfonyl chloride derivative of iron tetrasulfophthalocyanine,
The broadness of the Q band in the UV-vis spectrum
characterizes the homogeneity of the FePcS distribution on
the surface.29 The full width at half maximum of the 640 nm
band is 160, 130 and 160 nm for 1, 2 and d-FePcS–MCM-
41, respectively. This result suggests a more homogeneous
distribution of FePcS on the surface of the SiO2 for 2 compared
to 1 and d-FePcS–MCM-41. On the other hand, the DR UV-vis
spectra of m-FePcS–SiO2 and m-FePcS–MCM-41 exhibit a Q
band at 670–680 nm, indicating a monomeric FePcS complex as
the major supported species (Fig. 2A). Note that FePcS(Bu4N)4
in acetone also exists mainly in the monomeric form, as can be
judged from the position of the Q band at 690 nm in the UV-vis
spectrum (Fig. 2B).
◦
FePc(SO2Cl)4, with amino-modified silica in pyridine at 80 C
by following a protocol described elsewhere.25 In the second
approach, FePcS anchoring was carried out under milder
conditions. Sulfonate groups of a tetrabutylammonium salt
of iron tetrasulfophthalocyanine, FePcS(Bu4N)4, activated with
triphenylphosphine ditriflate, rapidly reacted with the amino
groups of amino-modified SiO2 in CH2Cl2 at 25 ◦C to form
covalent sulfonamide bonds.29 By following this procedure, we
prepared 2. The immobilization of FePcS on the mesostructured
silicate MCM-41 was performed in the same way as for 1.
In order to prepare m-FePcS–SiO2 and m-FePcS–MCM-41,
catalysts containing monomeric forms of FePcS, a solution of
FePc(SO2Cl)4 in pyridine was stirred for 18 h to transform the m-
oxo-dimeric species to a dipyridine mononuclear complex, which
was then added to a suspension of amino-modified silica or
MCM-41.25 A sample of the mesoporous mesophase iron silicate
Fe-MMM-2 was prepared for comparison.31 The physicochem-
ical characteristics of the Fe-containing catalysts prepared and
used in this work are given in Table 1. The iron content in the
supported FePcS catalysts was typically 0.2–0.3 wt%.
Catalytic oxidation of 8-HQ
We examined the catalytic properties of the supported dimeric
and monomeric iron phthalocyanines in the selective oxidation
t
of 8-HQ with BuOOH (Scheme 2) and compared them with
those of homogeneous FePcS(Bu4N)4.
Diffuse reflectance UV-vis spectra of the supported FePcS
catalysts are presented in Fig. 2A. The spectra of d-FePcS–SiO2
(1 and 2) and d-FePcS–MCM-41 show an intensive Q band with
a maximum at 630–640 nm, indicating a dimeric FePcS complex
as the major species.25 The bands below 500 nm correspond to
metal to Pc ligand charge transfer. Upon dimerization, a more
sensitive p–p* transition of the phthalocyanine ligand (Q band,
between 600 and 700 nm) is blue-shifted with respect to that
of the monomeric form. A shoulder at 690 nm, indicative of
mononuclear species, suggests that a small amount of FePcS is
present in the monomer form (Fig. 2A).25
Scheme 2
Fig. 3 shows 8-HQ consumption and QD selectivity vs. time.
No induction period was revealed for both substrate conversion
and target product accumulation. One can see from Fig. 3A
that m-FePcS–SiO2, d-FePcS–MCM-41 and homogeneous
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1076–1082 | 1077
©

View Article Online
atmosphere, reagent concentrations, additives, etc.) on the
t
oxidation of 8-HQ by BuOOH over FePcS–SiO2 (1 and 2).
Firstly, we found that the reaction atmosphere (air or argon) does
not affect the conversion of 8-HQ and the yield of the quinone.
The addition of a typical chain radical inhibitor (ionol) to the
reaction mixture showed no retarding effect on the reaction
rate, indicating that the 8-HQ oxidation is not a chain radical
process. This is consistent with the fact that the reaction rate
is not influenced by the presence of O2. Increasing the reaction
temperature from 30 to 50 ◦C had no impact on the QD yield.
We found that the QD yield could be improved by decreasing
the 8-HQ concentration (Fig. 4). The addition of the substrate
in portions to the reaction mixture led to some decrease in the
QD yield. The substrate conversion and quinone yield increased
with increasing catalyst amount until a certain Fe/8-HQ ratio
was reached; then the QD yield had a tendency to decrease.
For 1 and [8-HQ] = 0.02 M, the optimal catalyst amount was
3.0 mol% Fe (Fig. 5A). Using this amount, 99% conversion of
8-HQ and 50% yield of QD were attained. Significantly, the
optimal amount of 2 was lower (0.8 mol% Fe), and even at a
higher 8-HQ initial concentration (0.1 M), a yield of QD as
high as 56% was obtained with complete substrate conversion
(Fig. 5B). Thus, comparing the QD yields at a similar substrate
conversion (close to 100%), we may conclude that, irrespective
of the reaction conditions employed, 2 was not only more active
Fig. 3 Kinetic profiles of 8-HQ oxidation in the presence of d-FePcS–
SiO2 (1) (ꢀ), d-FePcS–SiO2 (2) (᭺), d-FePcS–MCM-41 (᭹), m-FePcS–
SiO2 (♦) and m-FePcS(Bu4N)4 (ꢀ). Reaction conditions: [8-HQ] =
0.1 M, catalyst 0.2 mol% Fe, [tBuOOH] = 0.7 M, 30 ◦C, acetone.
m-FePcS(Bu4N)4 were significantly more active than both d-
FePcS–SiO2 samples. At the same time, 2 showed a higher
activity than 1.
The turnover frequency (TOF) values estimated from the
initial rates of 8-HQ consumption were 215 and 470 h-1 for 1 and
2, respectively. In turn, TOFs as high as 1400 h-1 were achieved
with d-FePcS–MCM-41 and FePcS(Bu4N)4. The highest activity
(TOF = 3570 h-1) was found for m-FePcS–SiO2. Thus, we may
conclude that both the state of the FePcS complex and the
porous structure of the support affect the catalytic activity. The
higher catalytic activity of 2 in comparison to 1 results, most
likely, from the more homogeneous FePcS distribution on the
support (vide supra), which makes the majorityofthe ironcentres
accessible by reactants.
Selectivities of QD formation for the different catalysts were
compared when their values achieved a constant level, after
1 or 2 hours depending on the catalyst, and increased in
the order d-FePcS–SiO2 (1) (30%) < FePcS(Bu4N)4 (37%) <
d-FePcS–MCM-41 (47%) < d-FePcS–SiO2 (2) (55%) < m-
FePcS–SiO2 (59%) (Fig. 3B). Hence, no general correlation
between the catalytic activity and QD selectivity was revealed.
Importantly, grafting FePcS onto silica allowed the attainment
of a superior selectivity towards the target product compared
to the homogeneous FePcS(Bu4N)4. In spite of the considerably
different reaction rates, m-FePcS–SiO2 and d-FePcS–SiO2 (2)
demonstrated a similar final selectivity to QD. However, the
evolution of selectivity for d-FePcS–SiO2 (2) was quite different
to those of the other catalysts (Fig. 3B). The reason for this
difference is not yet clear. The results obtained allow us to
suggest that the structure of the FePcS complex strongly affects
the activity of the catalyst, but not the selectivity. It is worth
noting that d-FePcS grafted onto MCM-41 was more active but
less selective than d-FePcS grafted on SiO2 (2). Therefore, the
use of a mesoporous support favours increasing the reaction rate
but leads to a decreasing QD selectivity.
Fig. 4 The effect of 8-HQ concentration on the oxidation of 8-HQ over
FePcS–SiO2 (1). Reaction conditions: catalyst 3 mol% Fe, [tBuOOH]/[8-
HQ] = 7, 30 ◦C, acetone, 6 h.
Importantly, the QD product was found to be stable under the
reaction conditions used. After the addition of a 7-fold excess of
tBuOOH to a mixture containing 0.05 mmol QD and 0.2 mol%
of 2 in 1 mL of acetone, no conversion of QD was observed over
3 h. Hence, we may conclude that parallel pathways lead to the
formation of the target and by-products.
Fig. 5 The effect of catalyst amount on the oxidation of 8-HQ
over FePcS–SiO2. Reaction conditions: [tBuOOH]/[8-HQ] = 7, 30 ◦C,
acetone, 6 h, (A) catalyst 1, [8-HQ] = 0.02 M and (B) catalyst 2, [8-HQ] =
0.1 M.
To optimize the reaction conditions for QD production, we
studied the effects of the reaction conditions (temperature,
1078 | Green Chem., 2010, 12, 1076–1082
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
but also more selective than 1. This is in line with the results
in H2O concentration until it reached a value of 0.3 M and then
tended to decrease (1, Fig. 8A). When a more concentrated 8-
HQ solution was used in the presence of 2, the selectivity to QD
also increased with the addition of supplementary water up to
[H2O] = 3 M (Fig. 8B). Meanwhile, as can be seen from Fig. 7B,
an additional increase in the oxidant concentration also allowed
an increase in QD yield.
shown in Fig. 3.
The stoichiometry of 8-HQ oxidation with BuOOH to pro-
t
duce QD is 1 : 2. However, the selectivity to QD increased with
increasing [tBuOOH]/[8-HQ] ratio until it reached 7 (Fig. 6).
The addition of tBuOOH in portions to the reaction mixture led
to some decrease of QD yield (Fig. 6A). A further increase in
the oxidant concentration resulted in a decay of the QD yield,
which could be explained by the enhancing of over-oxidation
reactions. Importantly, when using tBuOOH reactant in decane
t
instead of BuOOH in H2O, we observed a decrease of both
substrate conversion and product yield (Fig. 6B), which allows
the suggestion that water can participate in the pathway leading
to the formation of quinone according to mechanism proposed
in our recent paper.29b
Fig. 8 The effect of H2O concentration on the oxidation of 8-HQ
over FePcS–SiO2. Reaction conditions: [tBuOOH]/[8-HQ] = 5, 30 ◦C,
acetone, 6 h, (A) [8-HQ] = 0.02 M, catalyst 1 3 mol% Fe and (B) [8-HQ] =
0.1 M, catalyst 2 0.2 mol% Fe.
To summarize, the optimal reaction conditions for the produc-
tion of QD from 8-HQ were established as follows: [8-HQ] =
0.02 M, [tBuOOH] = 0.14 M, [H2O] = 0.3 M, catalyst 0.8 mol%
◦
Fe, acetone, 30 C. Table 2 presents the results of the catalytic
Fig. 6 The effect of [tBuOOH]/[8-HQ] molar ratio on the oxidation
of 8-HQ over FePcS–SiO2. Reaction conditions: 30 ◦C, acetone, 6 h,
(A) [8-HQ] = 0.02 M, catalyst 1 3 mol% Fe and (B) [8-HQ] = 0.1 M,
catalyst 2 0.2 mol% Fe. * tBuOOH was added in portions. ** A tBuOOH
solution in decane was used.
tests performed using different grafted FePcS catalysts under
these optimal conditions. For comparison, the data acquired in
a blank experiment (without any catalyst) with homogeneous
FePcS(Bu4N)4 and mesostructured iron silicate Fe-MMM-2 are
also given.
The reaction barely occurs without catalyst; only 15% of 8-HQ
was converted after 6 h, while no formation of QD was observed
(Table 2, entry 7). One can see that the highest yield (66%) of QD
was reached when using d-FePcS–SiO2 (2) (Table 2, entry 2). The
optimization of the reaction conditions allowed an increasing
A study of 8-HQ oxidation in different solvents showed
that the highest yields of QD were achieved in acetone, 1,2-
dichloroethane and toluene (Fig. 7). Acetonitrile and ethyl
acetate could also be considered suitable solvents. Importantly,
some water should be present to reach higher QD yields. As
was mentioned earlier, BuOOH in water was a better choice
than tBuOOH in decane. Furthermore, we found that at a fixed
concentration of tBuOOH, the QD yield increased with increases
t
Table 2 8-HQ oxidation in the presence of Fe-containing catalystsa
8-HQ
Entry Catalyst
Time/h conversion (%) QD yield (%)b
1
d-FePcS–SiO2 (1)
6
24
6
4
2
2
6
6
4
59
73
94
100
100
100
96
15
25
3
31 (52)c
51 (70)
66 (70)d
62 (62)
40 (40)
55 (55)
37 (39)
— (—)
— (—)
— (—)
2
3
4
5
6
7
8
9
d-FePcS–SiO2 (2)
m-FePcS–SiO2
d-FePcS–MCM-41
m-FePcS–MCM-41
e
FePcS(Bu4N)4
No catalyst
d-FePcS–SiO2 (2)f
Fe-MMM-2
6
a Reaction conditions [8-HQ] = 0.02 M, [tBuOOH] = 0.14 M, [H2O] =
0.3 M, catalyst amount 0.8 mol% Fe, acetone, 30 ◦C. b GC yield
based on the initial substrate; GC yield based on the substrate
consumed (selectivity) is given in parentheses. c The yield of the main
identified by-product, 6,7-epoxy-6,7-dihydroxyquinoline-5,8-dione, was
9%. d The yield of 6,7-epoxy-6,7-dihydroxyquinoline-5,8-dione was 16%.
e Homogeneous catalyst. f H2O2 was used instead of tBuOOH.
Fig. 7 The effect of solvent on the oxidation of 8-HQ over FePcS–
SiO2 (1). Reaction conditions: [8-HQ] = 0.02 M, catalyst 3 mol% Fe,
[tBuOOH] = 0.14 M, 6 h, 30 ◦C.
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1076–1082 | 1079
©

View Article Online
selectivity towards QD of up to 70% in the oxidation of 8-
HQ over the dimeric d-FePcS–SiO2 catalysts. The activity of 1
remained significantly lower, and 24 h were needed to obtain QD
in a 51% yield (Table 2, compare entries 1 and 2). On the other
hand, the alteration of the reaction conditions produced very
little effect on the results acquired using monomeric m-FePcS–
SiO2, for which the QD yield reached 62% (Table 2, entry 3).
The FePcS complexes supported on MCM-41 provided a
higher activity with respect to the same complexes anchored
onto amorphous silica (Table 2, entries 4 and 5 vs. entries 2
and 3). In the former case, a complete substrate conversion was
reached in 2 h, while 4 and 6 h were needed to obtain 100
and 94% conversion with m-FePcS–SiO2 and d-FePcS–SiO2 (2),
respectively. However, both MCM-41 supported catalysts were
less selective than the SiO2 supported examples. Importantly, the
majority of the supported FePcS catalysts demonstrated a supe-
rior catalytic performance in comparison to the homogeneous
catalyst FePcS(Bu4N)4 (Table 2, entry 6).
t
Fig. 9 The oxidation of 8-HQ with BuOOH over FePcS–SiO2 (2).
Reaction conditions: [8-HQ] = 0.1 M, catalyst 0.2 mol% Fe, [tBuOOH] =
0.7 M, 30 ◦C, acetone. (᭹) 8-HQ conversion after hot catalyst filtration.
The 1H NMR study showed the presence of a by-product in
the reaction mixture, which is tentatively identified as 6,7-epoxy-
6,7-dihydroquinoline-5,8-dione on the basis of GC-MS data
and its fragmentation pattern (see the Experimental section). In
principle, quinoline-5,8-dione-N-oxide, which could be expected
as the oxidation by-product, would have the same molecular ion
(m/z = 175). However, the presence of the peak at m/z = 106 due
to the loss of the OCC(O)CH fragment is only consistent with the
6,7-epoxy-6,7-dihydroquinoline-5,8-dione structure. The yield
of the by-product, calculated by integration of the corresponding
NMR signals, amounted to 5–16%, but we could not calculate
the total balance of products because of the formation of tars.
We have found that iron silicate Fe-MMM-2, containing
mostly site-isolated Fe ions,31 showed no catalytic activity in 8-
HQ oxidation with tBuOOH; only 3% substrate conversion was
observed after 6 h (Table 2, entry 9). This indicates the crucial
role of the FePcS complex in this specific oxidation reaction.
and selectivity of the oxidation. Noteworthy, m-FePcS–SiO2 was
more amenable to recycling. In the second cycle, the conversion
of 8-HQ was 66% and the selectivity of QD formation was 41%.
Increasing the reaction time to 24 h allowed 90% conversion
and 45% selectivity to QD to be achieved. However, good
recyclability of the supported FePcS–SiO2 catalysts remains a
challenge.
Experimental
General
Catalysts and materials. Iron tetrasulfophthalocyanine,
FePcS, was prepared according to a published procedure.34
Covalently bound FePcS–SiO2 and FePcS–MCM-41 mate-
rials were obtained by anchoring FePcS onto the surface
of amino-modified non-porous SiO2 (Aerosil, Degussa) and
mesoporous MCM-41 (synthesized according to a published
procedure35). Monomeric m-FePcS–SiO2 and m-FePcS–MCM-
41, and dimeric d-FePcS–MCM-41 and d-FePcS–SiO2 (1) were
prepared by following literature protocols,25 dimeric d-FePcS–
SiO2 (2) was prepared as described previously.29 Mesostructured
iron silicate Fe-MMM-2 was prepared by a sol-mesophase
route.31 FePcS(Bu4N)4 was obtained by replacement of the
sodium cation with tetrabutylammonium using tetrabutylam-
monium hydroxide.29 8-Hydroxyquinoline and tert-butyl hy-
droperoxide (70 wt% aqueous solution) were purchased from
Aldrich. The other reactants were obtained commercially and
used as received. The concentration of hydrogen peroxide (28–33
wt% aqueous solution) was determined iodometrically prior to
use.
t
When H2O2 was used as the oxidant instead of BuOOH, the
conversion of 8-HQ after 1 h was about 25%, but no QD was
found in the reaction mixture (Table 2, entry 8). Furthermore,
the phthalocyanine chromophore undergoes bleaching in the
presence of H2O2.32 Consequently, the combination of the FePcS
catalyst and the alkyl hydroperoxide oxidant is essential for the
selective oxidation of 8-HQ to QD.
In order to check for the possible contribution of a homo-
geneous catalytic reaction due to leached iron species into the
overall oxidation process, we performed hot catalyst filtration
experiments using 2 by following the methodology suggested by
Sheldon and co-workers.33 No further 8-HQ conversion and QD
formation in the filtrate solution was found after separation of
the solid catalyst (Fig. 9). This result clearly demonstrates that
the catalysis is of a heterogeneous nature and is not due to iron
species leached into solution from the catalyst surface.
Catalytic experiments and product analysis
We also studied the reusability of 2 under the optimal reaction
conditions. Unfortunately, both the activity and selectivity
decreased drastically in the second operation cycle. After 5 h, the
conversion of 8-HQ was only 25% and the formation of QD was
not observed. Although the phthalocyanine complex was much
more stable in the presence of tBuOOH than in the presence of
H2O2, a progressive oxidative degradation of the phthalocyanine
ligand cannot be excluded, leading to a decrease of the activity
Catalytic oxidations were performed in thermostated glass
vessels at 30–50 ◦C under vigorous stirring (500 rpm). Typically,
the reactions were initiated by adding 0.06–0.90 mmol of
tBuOOH or 0.7 mmol of H2O2 to a mixture containing 0.02–
0.10 mmol of substrate and 6–20 mg (0.04–1.00 mmol Fe) of
supported FePcS catalyst or 100 mL of a 0.0016 M solution of
FePcS(Bu4N)4 in acetone (0.16 mmol Fe) or 4.4 mg (1.6 mmol
1080 | Green Chem., 2010, 12, 1076–1082
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Fe) of Fe-MMM-2 in 1 mL of acetone or another solvent. For
recycling experiments, before their second use, the catalyst was
filtered off, washed with acetone and methanol, and dried at
100 ◦C. The oxidation products were identified by GC-MS and
1H NMR spectroscopy. The substrate conversions and product
yields were quantified by GC using biphenyl as the internal
standard, as well as by 1H NMR spectroscopy. The experiments
were performed at least in triplicate and were reproducible within
5%. The mass balances were in the range 80–85%. The remaining
15–20% was accounted for by tars.
achieved demonstrate a promising clean way for QD production
by a heterogeneous oxidation process.
Acknowledgements
We thank Mr V. Utkin for the product identification by
GC-MS, Mr A. V. Golovin for 1H NMR measurements,
Dr I. D. Ivanchikova for help with by-product identification
and M. E. Malyshev for the preparation of Fe-MMM-2. Dr
E. V. Kudrik is greatly acknowledged for the preparation of m-
FePcS–SiO2 and fruitful discussions. This research was partially
supported by the Russian Foundation for Basic Research
(grants 05-03-34760 and 09-03-93109), by Agence National de
Recherche (France, grant ANR-08-BLANC-0183-01) and by
the CNRS in the framework of IRCELYON - BIC Associated
European Laboratory.
Quinoline-5,8-dione. GC-MS (EI) m/z (relative int.): 159
(100, [M]+), 131 (51, [M - CO]+), 103 (72, [M - CO]+), 77 (27,
[M - C4H2O]+), 76 (48, [M - C4H3O]+), 50 (29).
1H NMR dH (250 MHz; CD3COCD3): 7.15 (1H, d, J =
10.4 Hz, H6(7)), 7.18 (1H, d, J = 10.4 Hz, H7(6)), 7.86 (1H, dd,
J1 = 8.0 Hz, J2 = 4.7 Hz, H3), 8.41 (1H, dd, J1 = 8.0 Hz, J2 =
1.5 Hz, H4), 9.02 (1H, dd, J1 = 4.7 Hz, J2 = 1.5 Hz, H2).
References
6,7-Epoxy-6,7-dihydroquinoline-5,8-dione. GC-MS
(EI)
1 (a) H. S. Song, Y. S. Song and K.-J. Lee, J. Heterocycl. Chem., 2006,
43, 1533; (b) R. H. J. Hargreaves, C. L. David, L. J. Whitesell, D. V.
LaBarbera, A. Jamil, J. C. Chapuis and E. B. Skibo, J. Med. Chem.,
2008, 51, 2492; (c) C. Morin, T. Besset, J.-C. Moutet, M. Fayolle,
M. Bru¨ckner, D. Limosin, K. Becker and E. Davioud-Charvet, Org.
Biomol. Chem., 2008, 6, 2731.
m/z (relative int.): 175 (100, [M]+), 159 (9, [M - O]+), 147 (12,
[M - CO]+), 131 (6, [M - O-CO]+), 119 (52, [M - 2CO]+), 106
(40, [M - C3HO2]+), 103 (16, [M - O-2CO]+).
2 (a) J. W. Lown and S.-K. Sim, Can. J. Chem., 1976, 54, 2563; (b) A. J.
Herlt, R. W. Rickards and J.-P. Wu, J. Antibiot., 1985, 38, 516.
3 J. W. Lown, Mol. Cell. Biochem., 1983, 55, 17.
4 Y. Take, K. Oogose, T. Kubo, Y. Inouye and S. Nakamura,
J. Antibiot., 1987, 40, 679.
5 I. A. Shaikh, F. Johnson and A. P. Grollman, J. Med. Chem., 1986,
29, 1329.
6 Y. Inouye, H. Matsumoto, R. Morishige, Y. Kitahara, A. Kubo and
S. Nakamura, Chem. Pharm. Bull., 1991, 39, 994.
7 M. Behforouz, J. Haddad, W. Cai and Z. Gu, J. Org. Chem., 1998,
63, 343.
8 T. Fryatt, D. T. Goroski, Z. D. Nilson, C. J. Moody and H. D. Beall,
Bioorg. Med. Chem. Lett., 1999, 9, 2195.
Instrumentation
GC analyses were performed using a Tsvet-500 gas chromato-
graph equipped with a flame ionization detector and DB-
5MS capillary column (30 m ¥ 0.25 mm). GC-MS analyses
of organic products were carried out using a Saturn 2000 gas
chromatograph equipped with a CP-3800 mass spectrometer
(30 m ¥ 0.32 mm WCOT fused silica CP-SIL 8 CB capillary
column) and Hewlett Packard 5973/6890 mass spectrometer
(30 m ¥ 0.25 mm cross-linked 5% PHME siloxane (0.25 mm
1
coating) capillary column HP-5MS). H NMR spectra of the
9 K. H. Yoo, E. Y. Yoon, Y. Y. Park, S. W. Park, Ch.-O. Lee, W. K. Lee,
D. Y. Chi and D. J. Kim, Bull. Korean Chem. Soc., 2001, 22, 1067.
10 A. D. Bolza´n and M. S. Bianchi, Mutat. Res., 2001, 488, 25.
11 I. S. Severina, N. V. Pyatakova, A. B. Postnikov, M. N. Preobrazhen-
skaya and Y. V. Khropov, Eur. J. Pharmacol., 2004, 483, 127.
12 M. G. Brazhnikova, V. I. Ponomarenko, I. N. Kovsharova, E. B.
Kruglyak and V. V. Proshlyakova, Antibiotiki, 1968, 13, 99.
13 D. L. Boger, M. Yasuda, L. A. Mitscher, S. D. Drake, P. A. Kitos and
S. C. Thompson, J. Med. Chem., 1987, 30, 1918.
reaction products were run on an AM 250 Bruker instrument.
DRS UV-vis and solution UV-vis spectra were run on a
Shimadzu UV-vis 2501PC spectrophotometer and a Specord
M40 spectrophotometer, respectively.
Conclusions
14 S. J. Gould and S. M. Weinreb, Fortschr. Chem. Org. Natur., 1982, 41,
77.
A clean and efficient procedure for the oxidation of 8-
hydroxyquinoline to quinoline-5,8-dione with BuOOH in the
15 J. Hadju, Met. Ions Biol. Syst., 1985, 19, 53.
t
16 M. M. Harding and G. V. Long, Curr. Med. Chem., 1997, 4, 405.
17 H. Soedjak, J. Hajdu, J. D. Raffetto, R. Cano, B. L. Bales, L. S.
Prasad and L. D. Kispert, Biochim. Biophys. Acta, Gen. Subj., 1997,
1335, 73.
presence of iron phthalocyanine supported on silica has been
presented. The catalysts with a dimeric active site, d-FePcS–
SiO2, demonstrated a higher selectivity than those with a
monomeric active site, m-FePcS–SiO2. Both mononuclear and
dinuclear FePcS complexes supported on mesoporous MCM-
41 showed a higher activity but a lower selectivity than the
corresponding complexes supported on non-porous SiO2. The
crucial factors affecting QD yield were the procedure for grafting
the FePcS complex onto the SiO2, the nature of the solvent, the
8-HQ concentration, the tBuOOH/8-HQ ratio and the catalyst
amount. Over the best catalyst (d-FePcS–SiO2 prepared under
mild conditions) and under the optimal reaction conditions,
the selectivity to quinone attained 70% at a 94% substrate
conversion. Although further work is still required to improve
the product yield and catalyst recyclability, the results already
18 A. Bolognese, G. Correale, M. Manfra, A. Lavecchia, O. Mazzoni,
E. Novellino, V. Barone, A. Pani, E. Tramontano, P. L. Colla, C.
Murgioni, I. Serra, G. Setzu and R. Loddo, J. Med. Chem., 2002, 45,
5205.
19 A. Bolognese, G. Correale, M. Manfra, A. Lavecchia, O. Mazzoni,
E. Novellino, P. L. Colla, G. Sanna and R. Loddo, J. Med. Chem.,
2004, 47, 849.
20 A. Bolognese, G. Correale, M. Manfra, A. Lavecchia, E. Novellino
and V. Barone, Tetrahedron, 2004, 60, 8189.
21 S. Castellano, A. Bertamino, I. Gomez-Monterrey, M. Santoriello,
P. Grieco, P. Campiglia, G. Sbardella and E. Novellino, Tetrahedron
Lett., 2008, 49, 583.
22 R. Barret and M. Daudon, Tetrahedron Lett., 1990, 31, 4871.
23 (a) Y. T. Pratt and N. L. Drake, J. Am. Chem. Soc., 1960, 82, 1155;
(b) D. W. Cameron, K. R. Deutscher and G. I. Feutrill, Aust. J. Chem.,
1982, 35, 1439; (c) H. J. Teuber and N. Go¨tz, Chem. Ber., 1954, 87,
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1076–1082 | 1081
©

View Article Online
1236; (d) J. Cossy and D. Belotti, Tetrahedron Lett., 2001, 42, 4329;
(e) A. S. Amarasekara, Synth. Commun., 1999, 29, 3063.
24 S. M. S. Chauhan, B. Karla and P. P. Mohaparta, J. Mol. Catal. A:
Chem., 1999, 137, 85.
25 (a) A. B. Sorokin and A. Tuel, New J. Chem., 1999, 23, 473; (b) A. B.
Sorokin and A. Tuel, Catal. Today, 2000, 57, 45.
B. S. Ramaswany, Inorg. Chim. Acta, 1982, 64, L87; (d) C. Ercolani,
M. Gardini, F. Monacelli, G. Pennesi and G. Rossi, Inorg. Chem.,
1983, 22, 2584; (e) R. Dieing, G. Schmid, E. Witke, C. Feucht, M.
Dreßen, J. Pohmer and M. Hanack, Chem. Ber., 1995, 128, 589; (f) C.
Ercolani and F. Monacelli, J. Porphyrins Phthalocyanines, 2001, 5,
668.
26 (a) C. Pe´rollier, C Pergrale-Mejean and A. B. Sorokin, New J. Chem.,
2005, 29, 1400; (b) C. Pergrale and A. B. Sorokin, C. R. Chim., 2000,
3, 803.
27 A. B. Sorokin, S. Mangematin and C. Pergale, J. Mol. Catal. A:
Chem., 2002, 182–183, 267.
31 M. N. Timofeeva, M. S. Mel’gunov, O. A. Kholdeeva, M. E.
Malyshev, A. N. Shmakov and V. B. Fenelonov, Appl. Catal., B,
2007, 75, 290.
32 N. d’Alessandro, L. Tonucci, M. Bressan, L. K. Dragani and A.
Morvillo, Eur. J. Inorg. Chem., 2003, 1807.
28 O. V. Zalomaeva, O. A. Kholdeeva and A. B. Sorokin, C. R. Chim.,
2007, 10, 598.
29 (a) O. V. Zalomaeva and A. B. Sorokin, New J. Chem., 2006, 30, 1768;
(b) O. V. Zalomaeva, I. D. Ivanchikova, O. A. Kholdeeva and A. B.
Sorokin, New J. Chem., 2009, 33, 1031.
30 (a) A. Hadasch, A. B. Sorokin, A. Rabion and B. Meunier,
New J. Chem., 1998, 22, 45; (b) S. Gupta, H. Huang and E. Yeager,
Electrochim. Acta, 1991, 36, 2165; (c) A. B. P. Lever, S. Licoccia and
33 R. A. Sheldon, M. Wallau, I. W. C. E. Arends and U. Schuchardt,
Acc. Chem. Res., 1998, 31, 485.
34 J. H. Weber and D. H. Busch, Inorg. Chem., 1965, 4,
469.
35 (a) A. Tuel and S. Gontier, Chem. Mater., 1996, 8, 114; (b) A. Corma,
M. T. Navarro and J. Perez-Pariente, J. Chem. Soc., Chem. Commun.,
1994, 147; (c) R. J. Davis, Z. Liu, J. E. Tabora and W. S. Wieland,
Catal. Lett., 1995, 34, 101.
1082 | Green Chem., 2010, 12, 1076–1082
This journal is
The Royal Society of Chemistry 2010
©