Unusual kinetic role of a water-soluble iron(III) porphyrin catalyst in the oxidation of 2,4,6-trichlorophenol by hydrogen peroxide

Article abstract of DOI:10.1002/kin.20018

The oxidation of 2,4,6-trichlorophenol (TCP) to 2,6-dichloro-1,4-benzoquinone (DCQ) by hydrogen peroxide using iron(III) meso-tetra(4-sulfonatophenyl) porphine chloride, Fe(TPPS)Cl, as a catalyst was studied with stopped-flow UV-vis spectrophotometry and potentiometry using a chloride ion selective electrode. The observations are interpreted by a three-step kinetic model: the initial reaction of the catalyst with the oxidant (Fe(TPPS)⁺ + H₂O₂ → Cat′) produces an active intermediate, which oxidizes the substrate (Cat′ + TCP → Fe(TPPS)⁺ + DCQ + Cl^-) in the second step. The third step is the transformation of the catalyst into a much less active form (Cat′ → Cat″) and is responsible for the unusual kinetic phenomena observed in the system.

Full text of DOI:10.1002/kin.20018

Unusual Kinetic Role
of a Water-Soluble Iron(III)
Porphyrin Catalyst
in the Oxidation of
2
,4,6-Trichlorophenol
by Hydrogen Peroxide
1
2
´
GABOR LENTE, JAMES H. ESPENSON
1
Department of Inorganic and Analytical Chemistry, University of Debrecen, Debrecen, Hungary
Ames Laboratory and Department of Chemistry, Iowa State University, Ames, Iowa 50011
2
Received 26 June 2003; accepted 9 March 2004
DOI 10.1002/kin.20018
Published online in Wiley InterScience (www.interscience.wiley.com).
ABSTRACT: The oxidation of 2,4,6-trichlorophenol (TCP) to 2,6-dichloro-1,4-benzoquinone
(DCQ) by hydrogen peroxide using iron(III) meso-tetra(4-sulfonatophenyl) porphine chloride,
Fe(TPPS)Cl, as a catalyst was studied with stopped-flow UV–vis spectrophotometry and poten-
tiometry using a chloride ion selective electrode. The observations are interpreted by a three-
+
step kinetic model: the initial reaction of the catalyst with the oxidant (Fe(TPPS) + H O →
2
2
ꢀ
ꢀ
+
Cat ) produces an active intermediate, which oxidizes the substrate (Cat + TCP → Fe(TPPS)
−
+
DCQ + Cl ) in the second step. The third step is the transformation of the catalyst into a
ꢀ
ꢀꢀ
much less active form (Cat → Cat ) and is responsible for the unusual kinetic phenomena
observed in the system. ꢁC 2004 Wiley Periodicals, Inc. Int J Chem Kinet 36: 449–455, 2004
INTRODUCTION
toxic, persistent, and regarded as priority pollutants
for which efficient chemical treatment processes are
needed. Oxidative degradation is probably the most
advantageous reaction type for this purpose and sev-
eral different chemical methods have been reported.
[1–17]. These methods usually use H2O2 or KHSO5
as the stoichiometric oxidant and iron complexes as
catalysts. The most successful reported catalysts have
ligands with a set of four N-donor atoms in either a
planar or nonplanar geometry [1–6,9,12,13]. In addi-
tion to H2O2-dependent oxidation, TiO2-based systems
for photodegradation [8,10,11,14] and ozonization [7]
have also been thoroughly studied.
Chlorinated phenols are in widespread use as wood
preservatives, pesticides, fungicides, herbicides, insec-
ticides, or disinfectants, and they are also present in
the waste of paper mills. These compounds are highly
Correspondence to: James H. Espenson; e-mail: espenson@
iastate.edu.
Contract grant sponsor: Center for Catalysis, Institute for
Physical Research and Technology, Iowa State University.
Contract grant sponsor: Fulbright Program.
ꢁ
c 2004 Wiley Periodicals, Inc.

450
LENTE AND ESPENSON
2
,4,6-Trichlorophenol (TCP) is one of the most
Instruments HI 1131 combination pH electrode were
used connected to a Hanna Instruments pH302 pH-
meter. The electrodes were calibrated daily with stan-
dard NaCl solutions and standard buffers, respectively.
Kinetic experiments with the chloride ion selective
electrode were always carried out with a small amount
of NaCl added prior to the experiment (referred to
significant pollutants among chlorinated phenols
and is often used to test the efficiency of dif-
ferent oxidation methods [2,3,13]. In this paper,
we report our unusual kinetic observations using
iron(III) meso-tetra(4-sulfonatophenyl) porphine chlo-
ride, Fe(TPPS)Cl (structure shown in Scheme 1), as a
catalyst in the initial stage of the H2O2-dependent ox-
idation of TCP. This initial process is the transforma-
tion of TCP to 2,6-dichloro-1,4-benzoquinone (DCQ)
as shown in Eq. (1):
−
as [Cl ]0 in the figure captions) in order to avoid
badly defined voltage readings at the beginning of
the kinetic curves. Constant ionic strength was main-
tained with 0.1 M NaNO3; the same medium was
used in spectrophotometric studies. All experiments
were carried out in the dark because light (even flu-
orescent room light) accelerates further oxidation pro-
cesses significantly [17]. Our experience showed that
the monochromatic analyzing light beam of a scan-
ning spectrophotometer or stopped-flow instrument is
not intense enough to influence the reaction, but the
white light used in a diode-array spectrophotometer
corrupts the measurements. Nonlinear least squares fit-
ting was carried out by the software package Scientist
(1)
[18].
EXPERIMENTAL
Materials
RESULTS AND DISCUSSION
2
,4,6-Trichlorophenol purchased from Aldrich was pu-
rified by vacuum sublimation. Iron(III) meso-tetra(4-
sulfonatophenyl) porphine chloride was used as re-
ceived from Frontier Scientific (www.porphyrin.com).
Ion exchanged and ultrafiltered water from a Milli-
pore MILLI-Q purification system was used to prepare
the solutions. The concentrations of hydrogen peroxide
stock solutions were determined iodometrically.
Our observations on the stoichiometry of the reaction
(
1) in water were in agreement with an earlier re-
port which described the preparative use of reaction
1) in acetonitrile [1]. The kinetics of the process was
(
monitored by two independent methods: potentiome-
try using a chloride ion selective electrode and UV–vis
spectrophotometry. Typical kinetic curves are shown
in Figs. 1 and 2. The reactions were carried out in
air, a comparison with air-free experiments revealed
no significant differences. No reaction was observed in
a sample containing H2O2 and TCP in the absence of
the catalyst after a week.
Instrumentation and Computation
A Shimadzu UV-3101PC scanning spectrophotometer
and an Applied Photophysics SX-18MV stopped-flow
instrument were used in this study. In potentiomet-
ric experiments, a Weiss Research CL3005 combi-
nation chloride ion selective electrode and a Hanna
As shown in Eq. (1), the oxidation reaction produces
acid. This was confirmed by measurements with pH-
potentiometry. The initial pH of the unbuffered TCP
solution used in the experiments shown in Fig. 1 was
about 4.7 because of the acidity of TCP. The pKa of
◦
TCP was measured to be 6.15 ± 0.01 at 25.0 C by
combined pH-potentiometric and UV–vis spectropho-
tometric experiments. This value is an agreement with
the consideration that the electron-withdrawing effect
of the three chloro substituents makes TCP a much
stronger acid than unsubstituted phenol (pKa = 9.86
±
0.01 was measured under the same conditions). The
pH was shown to decrease during the oxidation reac-
tion (see Supplementary Material). In the experiment
shown as curve b in Fig. 1, the pH was 3.6 after 15 min.
A further, much slower decrease was detected and a
Scheme 1

ROLE OF CATALYST IN OXIDATION OF 2,4,6-TRICHLOROPHENOL BY HYDROGEN PEROXIDE
451
(up to about 10 min) are reported in this paper as they
revealed an unusual kinetic role of the catalyst. The
first phase was always reasonably separated from later,
slower processes.
When attempts were made to study reaction (1) at
pH ∼ 7 in phosphate buffer, no formation of chloride
ion was detected in an hour and no oxidation of TCP
wasconfirmedatall. Theprobablereasonforthisobser-
+
vation is that Fe(TPPS) quite rapidly forms the ꢀ-oxo
dimer (TPPS)Fe O Fe(TPPS) at pH 7, [19,20] which
seemstobecatalyticallyinactive. Thermodynamicdata
show that the dimer is not formed in significant con-
centrations below pH 6 [19,20].
The rise in the concentration of chloride ion during
the first 10 min was an exponential function of time
even in the absence of buffers, as shown in Fig. 1. This
strongly suggests that the pH change in the initial phase
(from about 5 to 3.5) does not have a significant effect.
Figure 1 Change of chloride ion concentration during the
catalytic oxidation of TCP. Line: best fits to exponential func-
A set of experiments was also carried out with the prior
addition of gradually increasing amounts of nitric acid.
Figure 3 shows that the pseudo first-order rate constant
didnot depend on the concentration of acidadded. Con-
sequently, the reaction rate is not influenced by the pH
in this range and conclusive experiments can be done
withoutusingbuffers. ThepH-independenceisnotvery
surprising as TCP, H2O2, and Fe(TPPS) do not have
known pH-dependent equilibria in this range.
It is notable that the pseudo first-order rate constant
kꢁ of the two traces shown in Fig. 1 is practically the
same for the two experiments having different catalyst
+
tions. [H2O2] = 47.3 mM; [TCP] = 1.61 mM; [Fe(TPPS) ]
−
=
1.4 ꢀM (a), 2.8 ꢀM (b); [Cl ]0 = 10 ꢀM; 0.1 M NaNO3;
◦
2
5.0 C.
final pH of about 2.8 was reached in 15 h. As seen
from the very end of curve b in Fig. 1, chloride ion
formation also continued on longer time scales, but its
rate was much slower than in the initial phase. These
observations are consistent with further oxidation on
extended time scales. Only results on the first phase
+
Figure 2 Stopped flow kinetic traces detected during the
Figure 3 Pseudo first-order rate constant as a function pH
during the catalytic oxidation of TCP. [H O ] = 48.3 mM;
catalytic oxidation of TCP. [H2O2] = 49.8 mM; [TCP] = 0
2
2
+
+
−
(
0
a), 0.055 mM (b), 0.110 mM (c); [Fe(TPPS) ] = 4.1 ꢀM;
[TCP] = 1.60 mM; [Fe(TPPS) ] = 3.1 ꢀM; [Cl ] = 10
0
◦
◦
.1 M NaNO3; 25.0 C; path length: 1 cm.
ꢀM; 0.1 M NaNO ; 25.0 C.
3

452
LENTE AND ESPENSON
concentrations, but the concentration of chloride ion
produced increases with increasing catalyst concentra-
tion. This unexpected finding is consistent with the cat-
alyst being the limiting reagent in some way. However,
it is also clear that the concentration of chloride ion
formed in the reaction, although only about 10–20%
of the initial TCP concentration, is more than a fac-
tor of 100 larger than the initial catalyst concentration,
which excludes the possibility that a stoichiometric re-
action of the catalyst is being monitored rather than a
catalytic process. Detailed experiments showed that kꢁ
ter using manual mixing. It is also notable that the range
of TCP concentrations used in the UV–vis experiments
was larger than in the potentiometric study because
useful experiments could be done at low TCP concen-
+
trations. It is known that Fe(TPPS) reacts with H2O2
directly resulting in multistep oxidative damage of the
porphyrin [21,22]. This is consistent with our find-
ings without TCP present. However, the absorbance
change is much slower (Fig. 2) because the presence of
oxidizable TCP inhibits the overall reaction between
+
Fe(TPPS) and H2O2, but not its first step.
+
is independent of the concentration of Fe(TPPS) , and
The spectrum of the parent complex features the
directly proportional to the concentration of H2O2 (see
Supplementary Material at http://www.interscience.
wiley.com/jpages/0538-8066/suppmat). The depen-
dence on the concentration of TCP is shown in Fig. 4.
Values of kꢁ could not be obtained at lower TCP con-
centrations because the detected curve was not expo-
nential under those conditions.
usual Soret band of iron porphyrins, ꢂmax 295 nm (ε
5
−1
−1
ꢀ
= 1.2 × 10 L mol cm ). The intermediates Cat
ꢀ
ꢀ
and Cat remain undetectable, consistent with the ap-
plication of the steady-state approximation for their
concentrations. The initial rate of absorbance change
was used to study the reaction quantitatively. The full
kinetic curves could not be fitted reasonably well to an
exponential function or other explicit integrated rate
equations used frequently in chemical kinetics. The
initial rate of absorbance change was first order with
UV–vis measurements with the stopped flow tech-
nique were also carried out to monitor the kinetics of
the reaction. The absorbance was followed at 395 nm,
which is the very intense Soret band of the porphyrin
+
respect to both H2O2 and Fe(TPPS) (see Supplemen-
5
−1
−1
catalyst (ε = 1.2 × 10 L mol cm ). TCP and DCQ
do not have significant absorption at this wavelength, so
thesemeasurementsgaveinformationonthestateofthe
catalyst. Figure 2 shows that a decrease in absorbance
was detected during the process even in the absence of
TCP. With increasing concentrations of TCP, the ab-
sorbance decrease became slower. In fact, when the
TCP concentration was larger (∼1.5 mM), the reaction
could also be followed by a scanning spectrophotome-
tary Material). The dependence of the initial rate on the
concentration of TCP is shown in Fig. 5. As noted ear-
lier, the presence of TCP slowed down the absorbance
change quite significantly.
We found that a simple three-step kinetic model
could interpret all the experimental observations. The
model is shown in Scheme 2.
Figure 5 Initial rate of absorbance change (395 nm) as a
function of TCP concentration during the catalytic oxida-
tion of TCP. Line: best fit to Eq. (7). [H2O2] = 23.8 mM;
Figure 4 Pseudo first-order rate constant as a function of
TCP concentration during the catalytic oxidation of TCP.
+
+
◦
Line: best fit to Eq. (5). [H2O2] = 48.5 mM; [Fe(TPPS) ] =
[Fe(TPPS) ] = 3.6 ꢀM; 0.1 M NaNO3; 25.0 C; path length:
−
◦
3
.1 ꢀM; [Cl ]0 = 10 ꢀM; 0.1 M NaNO3; 25.0 C.
1 cm.

ROLE OF CATALYST IN OXIDATION OF 2,4,6-TRICHLOROPHENOL BY HYDROGEN PEROXIDE
453
Scheme 2
In the first step, the reaction of the catalyst with
The formulas derived for kꢁ and the initial rate of
ꢀ
ꢀ
H2O2 produces intermediate Cat . Cat is the actual re-
active oxidizing species that reacts directly with the
substrate TCP in step 2, which also regenerates the
original form of the catalyst. Step 3 is a self-decay of
the concentration change of the catalyst derived on the
basis of Scheme 2 agree with the experimental obser-
vations. The best fit is indicated by solid lines in Figs. 4
and 5. The following parameters were determined with
ꢀ
−1 −1
and
the species Cat , which accounts for the deactivation of
least-squares fitting: k1 = 6.0 ± 0.7 L mol
s
5
−1
the catalyst. In this model, standard steady-state treat-
k2/k3 = (1.4 ± 0.2) × 10 L mol . The rate constants
k2 and k3 could not be resolved independently, as is typ-
ical for steady-state conditions [23]. It is reasonable to
ꢀ
ment [23] for the concentration of Cat yields
+
k1 · [H2O2] · [Fe(TPPS) ]
ꢀ
ask whether k might be evaluated directly from exper-
3
[
Cat ] =
(2)
k2[TCP] + k3
iments in the absence of TCP. Given that the parent cat-
alyst provides the monitored absorption that decreases
Then the steady-state differential equation for
ꢀ
ꢀꢀ
+
over time, and that intermediates Cat and Cat are un-
detectablylowconcentrations, suchexperimentswould
Fe(TPPS) is
+
d[Fe(TPPS) ]
provide k only.
−k3[Catꢀ]
1
=
=
The essence of the model given in Scheme 2 is
dt
+
that the catalyst is transformed to a less active form
k1 · [H2O2] · [Fe(TPPS) ]
k2[TCP] + k3
ꢀ
ꢀ
−k3
(3)
(Cat ) while the catalyzed oxidation reaction reaches
only about 10–20% conversion. As the reaction was
confirmed to proceed further on extended time scales,
Assuming that the concentration of TCP remains
constant throughout the process, which is not an unjus-
tifiable approximation based on the experimental data
this will be discussed later in more detail), the differ-
ential equation can be solved for the concentration of
the catalyst:
ꢀ
ꢀ
Cat isprobablynotcompletelyinactiveasanoxidation
catalyst, but its catalytic effect is certainly negligible
in the first phase of the reaction studied here. On the
basis of this model, an interesting prediction can be
made: adding a second portion of catalyst solution to a
reaction mixture in a potentiometric experiment after
10 min should give rise to a second round of chloride
ion formation that is similar to the initial one. Figure 6
shows that the experimental observations are in full
agreement with the predictions.
(
+
+
−kꢁ t
[
Fe(TPPS) ] = [Fe(TPPS) ]0 × e
(4)
where kꢁ is a pseudo first-order rate constant defined as
k1 · k3 · [H2O2]
kꢁ =
(5)
At this point it is necessary to explain why the
measured absorbance traces, unlike the rise in chlo-
ride ion concentration, were not exponential. The third
reaction in Scheme 2 is only the first step in a se-
quence of decomposition reactions [21,22]. Therefore
k2 · [TCP] + k3
In turn, the concentration of chloride ion as a function
of time can also be calculated:
k2
−
+
−kꢁ t
ꢀꢀ
[
Cl ] =
× [TCP]0 · [Fe(TPPS) ]0 × (1 − e
)
Cat is in fact an intermediate and its concentration
k3
does not change in the same exponential fashion the
(6)
+
concentration of Fe(TPPS) does. The non-zero final
ꢀ
ꢀ
absorbance in Fig. 2 clearly indicates that Cat and
further decomposition products also have some con-
tribution to the absorbance signal at this wavelength.
Consequently, the detected signal, which is the sum
of all contributions, does not change exponentially.
These complications are not present in the potentio-
metric experiments because chloride ion is monitored
selectively.
Finally, differentiation of Eq. (4) gives the initial
+
rate for the concentration of Fe(TPPS) :
ꢀ
ꢁ
+
d[Fe(TPPS) ]
+
=
−kꢁ · [FeTPPS ]0
dt
t=0
k1 · k3 · [H2O2]0
k2 · [TCP]0 + k3
+
=
−
[FeTPPS ]0
(7)

454
LENTE AND ESPENSON
ꢀ
suggest that Cat may be the end-on hydroperoxide
(
TPPS)Fe OOH. First, this is not an unlikely product
+
of the reaction between Fe(TPPS) and H2O2 with the
release of a single proton. Second, end-on hydroperox-
idesareveryoftenthekeycatalyticoxidantsindifferent
catalytic systems with iron complexes [24]. Finally, the
deactivation described in the third reaction of Scheme 2
does not involve any H2O2 or TCP. If H2O2 or TCP
were reactants in that step, the rate law would surely
have a different form. End-on hydroperoxides have a
well-known heterolytic bond cleavage giving iron(V)
oxo species and a hydroxide ion [24]. Thus the reaction
+
−
(
TPPS)Fe OOH → (TPPS)FeO + OH is not un-
reasonable in our system, and it does not involve H2O2
or TCP.
The second reaction of Scheme 2 almost certainly
does not happen as a single kinetic step. Based on
the known mechanisms of phenol oxidations [25–28],
it seems likely that the reaction produces a phenox-
ide radical via hydrogen atom abstraction from TCP
in the rate-determining step. The phenoxide radical is
then likely to react with hydrogen peroxide present in
large excess and release chloride ion forming the prod-
Figure 6 Sequential addition of catalyst during the catalytic
oxidation of TCP. [H2O2] = 94.8 mM; [TCP] = 1.42 mM;
+
[
(
[
Fe(TPPS) ] = 3.1 ꢀM (after starting the reaction), 6.1 ꢀM
total after the addition of the second portion of catalyst);
Cl ]0 = 25 ꢀM; 0.1 M NaNO3; 25.0 C.
−
◦
ꢀ
uct quinone in subsequent fast steps. If Cat is indeed
(
TPPS)Fe OOH, it may give water and the iron(IV)
It should also be noted that the pseudo first-order
oxo species (TPPS)FeO as the direct products of the
second reaction in Scheme 2. (TPPS)FeO may abstract
a hydrogen atom from another TCP giving (TPPS)Fe ,
thus completing the catalytic cycle in steps fast rel-
ative to the rate-determining step. However, because
kinetics does not yield information on reactions after
the rate-determining step, the previous considerations
serve only to show that it is possible to have a complete
catalytic cycle with chemically plausible steps assum-
derivation from Scheme 2 depends on the assumption
that the concentration of TCP does not change sig-
nificantly during the process. The maximum yield of
chloride ion was about 20% in the potentiometric ex-
periments used for the kinetic study, this implies that
the decrease in the TCP concentration during the pro-
cess is about 20% at most. One might argue that this is
too high to be considered as constant. However, there is
another reason why the approximation works well. The
actual rate of chloride ion formation can be calculated
from Scheme 2 as follows:
+
ꢀ
ing that Cat is the end-on hydroperoxide.
−
d[Cl ]
ꢀ
Some of the experiments were performed with facilities of the
Ames Laboratory of the U.S. Department of Energy, which
is operated by Iowa State University under contract W-7405-
Eng-82. We acknowledge helpful discussions with Prof.
W. S. Jenks and Mr. Youn-chul Oh.
=
=
k2 · [Cat ] · [TCP]
dt
+
k1 · k2 · [H2O2] · [FeTPPS ] · [TCP]
k2 · [TCP] + k3
(8)
This rate shows saturation with increasing concen-
trationofTCP. Thus, therateisinsensitivetothechange
in TCP concentration at high initial concentrations
and the pseudo first-order approximation works well
even with a substantial change in the TCP concentra-
tion. These observations also explain why the detected
curves were not exponential at low initial TCP concen-
trations.
BIBLIOGRAPHY
1
2
3
. Labat, G.; S e´ ris, J. L.; Meunier, B. Angew Chem, Int Ed
Engl 1990, 29, 1471.
. Sorokin, A.; Meunier, B. J Chem Soc, Chem Commun
1
994, 1799.
. Sorokin, A.; S e´ ris, J. L.; Meunier, B. Science 1995, 268,
163.
ꢀ
Cat is a reactive, steady-state intermediate present
1
at very low concentration levels and one can only
speculate about its chemical identity. Three hints
4. Sorokin, A.; De Suzzoni-Dezard, S.; Poullain, D.; No e¨ l,
J. P.; Meunier, B. J Am Chem Soc 1996, 118, 7410.

ROLE OF CATALYST IN OXIDATION OF 2,4,6-TRICHLOROPHENOL BY HYDROGEN PEROXIDE
455
5
6
7
8
. Shukla, R. S.; Robert, A.; Meunier, B. J Mol Catal, A
Chem 1996, 113, 45.
. Hadasch, A.; Sorokin, A.; Rabion, A.; Meunier, B. New
J Chem 1998, 45.
. Benitez, F. J.; Beltran-Hereida, J.; Acero, J. L.; Rubio,
F. J. Ind Eng Chem Res 1999, 38, 1341.
. Bauer, R.; Waldner, G.; Fallmann, H.; Hager, S.; Klare,
M.; Krutzler, T.; Malato, S.; Maletzky, P. Catal Today
H.; Morimoto, K.; Tatsumi, K. Environ Sci Technol
2003, 37, 1031.
17. Lente, G.; Espenson, J. H. Chem Commun 2003, 1962.
18. SCIENTIST, version 2.01; Micromath Software: Salt
Lake City, UT, 1995.
19. Fleischer, E. B.; Palmer, J. M.; Srivastava, T. S.;
Chatterjee, A. J Am Chem Soc 1971, 93, 3162.
20. El-Awady, A. A.; Wilkins, P. C.; Wilkins, R. G. Inorg
Chem 1985, 24, 2053.
1
999, 53, 131.
. Hemmert, C;Renz, M.;Meunier, B. JMolCatal, AChem
999, 137, 205.
9
21. Traylor, T. G.; Ciccone, J. P. J Am Chem Soc 1989, 111,
8413.
1
1
1
1
0. Li, X.; Cubbage, J. W.; Tetzlaff, T. A.; Jenks, W. S. J Org
Chem 1999, 64, 8509.
1. Li, X.; Cubbage, J. W.; Jenks, W. S. J Org Chem 1999,
22. Traylor, T. G.; Xu, F. J Am Chem Soc 1990, 112,
178.
23. Espenson, J. H. In Chemical Kinetics and Reaction
Mechanisms, 2nd ed.; McGraw-Hill: New York, 1995;
Ch. 4.
6
4, 8525.
2. Ichinobe, T.; Miyasaka, H.; Isoda, A.; Kimura, M.;
Hanabusa, K.; Shirai, H. React Funct Polym 2000, 43,
24. Kovacs, J. A. Science 2003, 299, 1024 and references
therein.
6
3.
1
3. Sen Gupta, S.; Stadler, M.; Noser, C. A.; Ghosh, A.;
Steinhoff, B.; Lenoir, D.; Horwitz, C. P.; Schramm, K.
W.; Collins, T. J. Science 2002, 296, 326.
25. Mihailovic, M. L; Cevkocic, Z. In The Chemistry of the
Hydroxyl Group; Patai, S. (Ed.); Wiley: London, UK,
1971; Ch. 10.
1
1
4. Mogyor o´ si, K.; Farkas, A.; D e´ k a´ ny, I.; Ilisz, I.; Dombi,
A. Environ Sci Technol 2002, 36, 3618.
5. Antonaraki, S.; Androulaki, E.; Dimotikali, D.; Hiskia,
A.; Papaconstantinou, E. J Photochem Photobiol, A
Chem 2002, 148, 191.
26. Colclough, N.; Smith, J. R. L. J Chem Soc, Perkin Trans
1994, 1139.
27. Jones, P.; Mantle, D; Wilson, I. J Inorg Biochem 1982,
17, 293.
28. Ferrari, R. P.; Laurenti, E.; Trotta, F. J Biol Inorg Chem
1999, 4, 232.
1
6. Fukushima, M.; Sawada, A.; Kawasaki, M.; Ichikawa,