Kinetics and mechanism of the ferroxime(II)-catalysed biomimetic oxidation of 2-aminophenol by dioxygen. A functional phenoxazinone synthase model

Article abstract of DOI:10.1039/b316543d

[Fe(Hdmg)₂(MeIm)₂] (1), referred to as ferroxime(II), is the precursor of a selective catalyst for the oxidative dehydrogenation of 2-aminophenol (Hap) to 2-amino-3H-phenoxazine-3-one (apx) by dioxygen under ambient conditions. The superoxoferroxime(III) species has been detected by ES-MS, and a 4-substituted 2-aminophenoxyl free radical by the ESR technique. The kinetics of the reaction was followed spectrophotometrically and by monitoring dioxygen uptake at constant pressure. According to the proposed mechanism, solvolysis of 1 is followed by O₂ binding to afford a superoxoferroxime, which abstracts an H-atom from Hap in the rate-determining step via an H-bonded intermediate, generating the free radical. This is supported by the observed primary deuterium kinetic isotope effect of 2.63. The system studied is a functional phenoxazinone synthase model.

Full text of DOI:10.1039/b316543d

View Article Online / Journal Homepage / Table of Contents for this issue
Kinetics and mechanism of the ferroxime(II)-catalysed biomimetic
oxidation of 2-aminophenol by dioxygen. A functional
phenoxazinone synthase model†
Tatiana M. Simándi, László I. Simándi,* Miklós Gyo˝r, Antal Rockenbauer and
Ágnes Gömöry
Chemical Research Center, Institute of Chemistry, Hungarian Academy of Sciences,
P.O. Box 17, H-1525 Budapest, Hungary. E-mail: simandi@chemres.hu
Received 19th December 2003, Accepted 16th February 2004
First published as an Advance Article on the web 27th February 2004
[Fe(Hdmg)2(MeIm)2] (1), referred to as ferroxime(), is the precursor of a selective catalyst for the oxidative
dehydrogenation of 2-aminophenol (Hap) to 2-amino-3H-phenoxazine-3-one (apx) by dioxygen under ambient
conditions. The superoxoferroxime() species has been detected by ES-MS, and a 4-substituted 2-aminophenoxyl
free radical by the ESR technique. The kinetics of the reaction was followed spectrophotometrically and by
monitoring dioxygen uptake at constant pressure. According to the proposed mechanism, solvolysis of 1 is followed
by O2 binding to afford a superoxoferroxime, which abstracts an H-atom from Hap in the rate-determining step via
an H-bonded intermediate, generating the free radical. This is supported by the observed primary deuterium kinetic
isotope effect of 2.63. The system studied is a functional phenoxazinone synthase model.
3H-phenoxazine-3-one (apx).4f The electropolymerisation of
Introduction
Hap yields insoluble films and some apx.4g
The catalytic activation of dioxygen by transition metal
We have previously investigated the catecholase5 and
complexes continues to attract interest1 with special emphasis
on elucidation of the reaction mechanisms via the modelling
of oxidases, oxygenases, dehydrogenases and related metallo-
enzymes. Important groups of substrates used in these studies
have been phenol and catechol derivatives, which provide
valuable mechanistic information and are sometimes also of
phenoxazinone synthase6 activity of cobaloxime(), and the
catecholase activity of [bis(N-methylimidazole)bis(dimethyl-
glyoximato)iron()], for which we proposed the name
ferroxime(). We have reported detailed mechanistic schemes
based on kinetic and structural results.3e,7
We have now found that ferroxime() is a functional model
synthetic utility. Iron-containing catechol dioxygenases are
for phenoxazinone synthase, converting 2-aminophenol (Hap)
involved in the metabolism of aromatic compounds and
to 2-amino-3H-phenoxazine-3-one (apx). In the present paper
low-molecular iron complexes are intensively studied as models
we report kinetic studies on this functional model system.
of the enzyme active centre.2,3 In addition to dioxygenase
activity, enzymes and models also exhibit catecholase activity,
a feature of biological importance.
Experimental
Related to catecholase activity is the ability to promote
[Fe(Hdmg)2(MeIm)2], referred to as ferroxime(),7a was
the oxidation of 2-aminophenol derivatives to 2-amino-3H-
synthesised by the procedure reported in the literature.8
phenoxazine-3-ones. The enzyme phenoxazinone synthase is
2-Aminophenol (Reanal, Hungary) was recrystallised from
PriOH (mp 177 ЊC). 2-Amino-3H-phenoxazin-3-one (mp
254–256 ЊC; lit. 259 ЊC9) was prepared by the cobaloxime()
catalysed oxidation of 2-aminophenol with O2.6b
involved in the last stages of the biosynthesis of Actinomycin
D4a,b a naturally occurring antineoplastic agent, used clinically
for the treatment of certain types of cancer.4c According to
Scheme 1, the 2-aminophenol derivative A is catalytically
The progress of the reaction was monitored by HPLC
analyses, using a Waters 410 instrument with a 991 Diode Array
Spectrophotometric Detector (column: Nucleosil C-18; 10 µm,
250 × 4.6 mm; eluent methanol–1% acetic acid (77 : 23)).
Calibration was based on UV/VIS spectra recorded for
authentic Hap and apx on a Hewlett-Packard 8453 Diode
Array Spectrophotometer. The reaction was also followed by
spectrophotometry and by the volume of O2 absorbed.
Rates of dioxygen uptake were measured in a constant
pressure gas-volumetric apparatus described earlier.5b The
reaction was started by dropping the solid ferroxime() catalyst
precursor into a vigorously stirred, thermally equilibrated
methanol solution of Hap from a sample holder manipulated
from the outside. The rate of O2 absorption was independent of
the stirring rate, excluding eventual diffusion control effects.
The oxidation was also followed by recording the time
evolutions of UV/VIS spectra on a Hewlett-Packard 8453 diode
array spectrophotometer.
Scheme 1 Last stage of the biosynthesis of Actinomycin D.
converted to AD in the presence of dioxygen and phen-
oxazinone synthase, which is a type 2 copper-containing
oxidase isolated from Streptomyces.4d The mechanism of this
six-electron oxidative coupling reaction has been studied using
2-aminophenol (Hap) derivatives as substrates.4d Nishinaga and
co-workers.4e have studied the Co(salen) catalysed oxidation
of substituted Hap derivatives. Speier and co-workers have
reported the TEMPO-initiated oxidation of Hap to 2-amino-
ESR spectra of reacting solutions were recorded on a Bruker
ELEXYS E500 CW-EPR spectrometer, keeping the solution
saturated with O2 throughout the experiments. Spectra were
accumulated over the initial 30 min of the reaction.
† Electronic supplementary information (ESI) available: Table S1:
Initial rates of dioxygen uptake as a function of catalyst, substrate
and dioxygen concentration. See http://www.rsc.org/suppdata/dt/b3/
b316543d/
D a l t o n T r a n s . , 2 0 0 4 , 1 0 5 6 – 1 0 6 0
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 2 0 0 4
1056

View Article Online
Fig. 1 Time evolution of the UV/Vis spectra after the solvolysis phase: [Fex]o =0.595 mM; [Hap]o =3.12 mM; T =25 ЊC, solvent MeOH. The wide
band at 400–440 nm shows accumulation of apx. Spectra 1–6 (from bottom up) were taken at 60 s intervals, spectra 7–11 at 380, 470, 580, 720 and
880 s, respectively. The inset is an expanded view of the 515–560 nm section, where spectra 1 and 2 show a decrease in absorbance due to solvolysis
before the absorbance starts to increase owing to apx formation.
Electrospray ionisation mass spectrometry (ES-MS) was run
on a PE-Sciex API 2000 triple quadrupole instrument. Mass
spectra were recorded in the mass range of m/z 1000–10 using
the electrospray ionisation technique.
Throughout this paper, M = mol dmϪ3
.
Results and discussion
Product and stoichiometry of 2-aminophenol oxidation
The oxidation of Hap to apx by O2 catalysed by [Fe(Hdmg)2-
(MeIm)2] (eqn. (1)) takes place at room temperature and 1 bar
O2 pressure in MeOH as solvent.
(1)
Fig. 2 Time dependence of apx concentration (HPLC data) and O2
absorption by the catalytic solution (volumetric data): [Fex]o =0.50
mM; [Hap]o =16.01 mM; in MeOH, T =25 ЊC.
where (i) is ferroxime, MeOH, 25 ЊC.
We followed reaction (1) by three techniques, viz. (i) spectro-
photometrically, (ii) by measuring the volume of dioxygen
absorbed by the solution, (iii) by monitoring the amount of
Hap and apx by HPLC. In addition, electrospray ionisation
mass spectrometry (ES-MS) was used to identify the species
present in solution during the reaction.
The time evolution of the UV-Vis spectra is shown in Fig. 1.
The series of spectra reveal a biphasic reaction. In the first
phase, up to about 1.5 min, [Fe(Hdmg)2(MeIm)2] undergoes
solvolysis7a (eqn. (2)), apparent from the decay of the 530 nm
band (see inset). Both complexes 1 and 2 as well as free MeIm
have been detected by ES-MS during the reaction.
ESR spectroscopy
The ESR signal of the O2-saturated, reacting solution, gener-
ating the apx product was monitored over the initial 30 min
period of the experiment. Noise reduction by the accumulation
of 200 successive spectra was necessary to obtain a good
quality spectrum, which is shown in Fig. 3. This corresponds to
ؒ
a single free-radical intermediate, Hapx , the spectrum of which
has been successfully simulated in terms of two N-atoms (1 in
ortho and 1 in para position) and three H-atoms (2 in the NH2
group and 1 in ortho position), using the software described in
ref. 10 (for the formula of Hapx , see the reaction mechanism)
The ESR spectrum of Hapx showed up immediately after the
ؒ
ؒ
(2)
This is followed by the oxidation of Hap to 2-amino-
phenoxazine-3-one (apx) in the second phase, as is shown by
the increase of the 410–440 nm band characteristic of apx.
Isosbestic points are observed at 280 and 298 nm during the
oxidation stage, indicating that a single reaction is taking place.
The amount of O2 absorbed by the catalytic solution and
the concentrations of Hap and apx have been simultaneously
monitored as a function of time. The results are shown in
Fig. 2.
These experiments directly demonstrate the validity of
stoichiometric eqn. (1): the molar amount of O2 absorbed for
1 mol of apx formed is 1.50 0.02 at any measured point. An
important feature is that O2 is reduced to water (rather than
H2O2) in the catalytic process.
Fig. 3 ESR spectrum of a solution in which the catalytic reaction (1) is
taking place (a) and the simulated spectrum (b) using the parameters
g=2.006, 1×aN =3.1 G, 1×aN =3.3 G, 3×aH =3.5 G ([Fex]o =1.0×10Ϫ2
M; [Hap]o =0.1 M; solvent methanol, T =25 ЊC, O2 atmosphere).
D a l t o n T r a n s . , 2 0 0 4 , 1 0 5 6 – 1 0 6 0
1057

View Article Online
Table 1 ES-MS peaks from the reacting mixture
Species
Molecular mass
Peak (M ϩ Hϩ) or Mϩ
Intensity
MeIm
82.11
109.13
212.21
368.08
400.08
450.14
83
110
213
368
400
450
Strong
Strong
Strong
Medium
Weak
2-Aminophenol (Hap)
2-Amino-3H-phenoxazine-3-one (apx)
[FeIII(Hdmg)2(MeIm)]ϩ
[FeIII(Hdmg)2(MeIm)(O2)]ϩ
[FeIII(Hdmg)2(MeIm)2]ϩ
Strong
reactants were mixed and O2 was admitted (about 1 min). The
initial poor resolution gradually improved as the number of
accumulated spectra increased. The signal persisted for the
period of observation at a steady state level, then a slow
decrease of the signal intensity was observed.
Electrospray ionisation mass spectrometry
ES mass spectra of the reacting solution were run at intervals.
The major peaks identified are listed in Table 1. Iron complexes
show up as iron() species. Both complexes 1 and 2 involved in
solvolysis equilibrium (2) can be observed. A weak signal due to
superoxoiron() has been detected, but it was not present in
solutions prepared under N2.
Kinetic measurements
Dioxygen absorption curves were recorded as a function of
time in methanol as solvent. From these curves the initial rates
(Vin) of dioxygen uptake were determined as a function of cata-
lyst and substrate concentration as well as dioxygen partial
pressure. The primary results are shown in the ESI† (Table S1;
each value shown is the average of three individual runs repro-
ducible to within 4%). At constant Hap concentration and O2
pressure the initial rate was found to be proportional to the
catalyst concentration (Fig. 4). When varying the concentration
of dioxygen, a saturation type curve was observed (Fig. 5). This
yielded a straight line upon plotting 1/Vin against 1/[O2]o. A
similar behaviour was found when varying the Hap concen-
tration at constant catalyst and dioxygen concentration (Fig. 6).
Fig. 5 Dependence of the initial rate of dioxygen absorption on the
dioxygen concentration. Conditions: [Fex]o =6.85×10Ϫ4 M, [Hap]o =
1.01×10Ϫ2 M, T =25 ЊC.
Fig. 6 Dependence of the initial rate of dioxygen absorption on the
substrate concentration. Conditions: [Fex]o =4.79×10Ϫ4 M, [O2]=
1.1×10Ϫ3 M, T =25 ЊC.
(4)
(5)
Fig. 4 Dependence of the initial rate of dioxygen absorption on the
ferroxime() concentration. Conditions: [Hap]o =1.0×10Ϫ2 M, [O2]=
1.1×10Ϫ3 M, T =25 ЊC.
The kinetic pattern emerging from Figs. 4–6 can be described
by the empirical rate law in eqn. (3), where A, B and C are
constants.
(6)
(7)
(8)
(3)
The observed kinetic behaviour is consistent with the
reaction mechanism in eqns. (4)–(12), where for brevity FeII,
FeIII and FeIV denote the square-planar ferroxime moieties,
[FeII(Hdmg)2], [FeIII(Hdmg)2]ϩ and [FeIV(Hdmg)2]2ϩ, respect-
ively, and N = MeIm. Coordinated MeOH is omitted except
in step (4).
[NFeIVO] ϩ Hap
[NFeII] ϩ bqmi ϩ H2O (fast) (9)
D a l t o n T r a n s . , 2 0 0 4 , 1 0 5 6 – 1 0 6 0
1058

View Article Online
Table 2 Rate and equilibrium constants determined from the fitting of eqn. (15) to the rate dataa
Source
10Ϫ3K1/MϪ1
K2/MϪ1
10Ϫ5K1K2/MϪ2
103k3/sϪ1
103slope/sϪ1
Fig. 5 (Vin vs. [O2])
1.70 0.09
1.62 0.08
n.a.
77.3 4.5
81.4 3.9
n.a.
1.31
1.32
n.a.
8.67 0.31
8.31 0.31
n.a.
n.a.
n.a.
3.05
2.84
n.a.
Fig. 6 (Vin vs. [Hap]o)
Fig. 4 (Vin vs. [Fex]o)
Calcd from eqn. (15)
Cobaloxime/Hap/O2 systemb
n.a.
n.a.
0.72
15.6 0.07
a T = 25 ЊC, solvent MeOH. b Ref. 6b.
In the steady state, the solvolysis of [FeIIN2] is complete,
therefore, the first term in eqn. (13) can be neglected. The
concentrations of species [NFeIII(O2H)] and [NFeIVO] are
assumed to be negligible owing to their high reactivity.
The simplified mass balance will thus have the form of
eqn. (15).
[Fe]o = [NFeII] ϩ [NFeIII(O2Ϫ)] ϩ [X]
(15)
Combination of the mass balance for the ferroxime species
with the expressions for equilibria (5) and (6) leads to kinetic
equation (16) for the mechanism (4)–(12), which is in agreement
with the empirical rate law of eqn. (3):
According to mechanism (4)–(12), one of the MeIm ligands
of catalyst precursor 1 is lost via solvolysis, producing catalyst
2, which then reversibly binds a dioxygen molecule, affording
a superoxo–iron() complex. Hydrogen-bonding of Hap to
the external superoxo oxygen leads to intermediate X, which
undergoes intramolecular H-atom transfer in the rate-
(16)
ؒ
determining step, generating the 2-aminophenoxyl radical, ap ,
and an (hydrogen peroxo)iron() species. The order of binding
of O2 and Hap has been deduced from the facts that (i)
ferroxime() is reactive toward O2 and undergoes oxidation to a
µ-peroxodiferroxime() complex, but (ii) Hap does not bind to
ferroxime() under N2 as shown by the lack of spectral changes.
Thus O2 binds first and Hap can only bind to coordinated
where [Fe]o and [Hap]oare the initial concentrations of added
ferroxime() and 2-aminophenol, respectively. It is assumed
that the decrease of [Hap]o is negligible while the initial rate of
O2 uptake is being measured.
The rate constant k3 and equilibrium constants K1and K2
have been determined by nonlinear least squares fitting of eqn.
(16) to the curves in Figs. 5 and 6. The good agreement between
the measured and calculated slopes of the straight line in Fig. 4
provides additional support for kinetic equation (16) and the
proposed reaction mechanism (4)–(12). The results are col-
lected in Table 2.
ؒ
O2 (superoxo ligand). The ap formed in step (7) undergoes
rapid oxidation in step (8) to the benzoquinone monoimine
(bqmi) intermediate, and is therefore not detectable by ESR
spectroscopy.
ؒ
In step (8) an electron from ap is utilised for the proton-
assisted reductive heterolytic splitting of the O–O bond of 5,
producing a reactive oxoiron() species (6). The latter rapidly
oxidises a second Hap molecule in step (9) as required by the
observed stoichiometry.
Using eqn. (16), the limiting initial rate reached at sufficiently
high Hap concentration can be expressed as in eqn. (17).
Vin,lim = k3[Fe]o
(17)
Steps (4)–(9) indicate the route for the formation of bqmi but
further steps are required to explain the formation of apx via
ؒ
the free-radical intermediate Hapx . A straightforward path is
Kinetic isotope effect
the rapid dimerisation of bqmi to H2apx, step (10), followed by
hydrogen atom abstraction step (11) and electron transfer step
(12). Apparently, the ratio of rate constants of these steps
ensures a detectable concentration of Hapx as shown by the
ESR experiments.
The oxidation of Hap to apx by ferricyanide, mercuric or
lead() oxide is known to involve bqmi as an intermediate,
which is the source of dimeric species.11
In the catalytic process steps (5) and (6) can be regarded as
pre-equilibria and (7) as rate-determining step. Reversibility is
often encountered in dioxygen binding to iron(), which is in
support of step (5). Hydrogen bonding to the superoxo ligand
is similar to that observed in the oxidation of 3,5-di-tert-
butylcatechol catalysed by the analogous cobaloxime()5 and
ferroxime().7a The ferryl species formed is reactive enough to
oxidise a second Hap molecule. To derive the kinetic equation
corresponding to the proposed mechanism, we need the mass
balance for the ferroxime species and for Hap, which can be
expressed from eqns. (13) and (14), respectively.
According the proposed mechanism, the rate-determining step
is intramolecular H-atom transfer within the hydrogen-bonded
intermediate X. To test this hypothesis, we have made an
attempt at detecting the kinetic isotope effect associated with
rupture of the O–H bond in Hap. For this objective, we
measured the initial rate spectrophotometrically under a given
set of conditions in CH3OD and in CH3OH as solvents. It can
be assumed that in CH3OD the phenolic OH of Hap undergoes
nearly complete exchange for OD. Consequently, if the rate-
determining step (7) consists in breaking of the –O–H (–O–D)
bond, a kinetic isotope effect (KIE) should be observed upon
going from CH3OH to CH3OD as solvent. According to stat-
istical mechanics, the isotope effects on equilibrium constants
K1 and K2 involved in kinetic equation (16) can be calculated
from the appropriate partition functions, which in turn can be
estimated from infrared frequencies of the bonds around the
reaction centre. Application of the standard procedure12a yields
negligible isotope effects for both K1 and K2. Thus, at identical
reactant concentrations in the two solvents, the ratio of the
initial rates VH/VD can be expressed from kinetic equation (16)
as eqn. (18):
ؒ
[Fe]o = [FeIIN2] ϩ [NFeII] ϩ [NFeIII(O2Ϫ)] ϩ
[X] ϩ [NFeIII(O2H)] ϩ [NFeIVO] (13)
[Hap]o = [Hap] ϩ X ϩ bqmi ϩ apx
(14)
VH/VD = k3H/k3D
(18)
D a l t o n T r a n s . , 2 0 0 4 , 1 0 5 6 – 1 0 6 0
1059

View Article Online
The KIE for the ferroxime()-catalysed oxidation of Hap is
k3H/k3D = 2.63 0.10. According to theoretical calculations,12b
for a nonlinear transition state of the type shown, the predicted
Acknowledgements
This work was supported by the Hungarian Science Fund
(OTKA Grants T 029036 and T 034282).
interval of KIE corresponds to k3H/k3D
≈ 1–3 for an
O ؒ ؒ ؒ H ؒ ؒ ؒ O bond angle of 120Њ. This result is in support of
rate-determining H/D-atom transfer (7) from the 2-amino-
phenol OH/OD group to the external O-atom of the superoxo-
iron() moiety, as depicted by Scheme 2.
References
1 Bioinorganic Catalysis, ed. J. Reedijk and E. Bouwman, Marcel
Dekker, New York, 2nd edn., 1999.
2 (a) L. I. Simándi, Catalytic Activation of Dioxygen by Metal
Complexes, Kluwer Academic Publishers, Dordrecht, Boston,
London, 1992, p. 400; (b) Advances in Catalytic Activation of
Dioxygen by Metal Complexes, Kluwer Academic Publishers,
Dordrecht, Boston, London, 2003, pp. 336.
3 (a) T. Funabiki, in Advances in Catalytic Activation of Dioxygen by
Metal Complexes, ed. L. I. Simándi, Kluwer Academic Publishers,
Dordrecht, Boston, London, 2003, ch. 4, pp. 158–226; (b)
O. Gentschev, N. Moller and B. Krebs, Inorg. Chim. Acta, 2000, 300,
442; (c) M. Merkel, F. K. Muller and B. Krebs, Inorg. Chim. Acta,
2002, 337, 308; (d ) L. Que, Jr., in Bioinorganic Catalysis, ed.
J. Reedijk and E. Bouwman, Marcel Dekker, New York, 2nd edn.,
1999, pp. 269–321; (e) L. I. Simándi, T. M. Simándi, Z. May and
G. Besenyei, Coord. Chem. Rev., 2003, 245, 87.
4 (a) U. Hollstein, Chem. Rev., 1974, 74, 625; (b) E. Katz and
H. Weissbach, J. Biol. Chem., 1962, 237, 882; (c) E. Frei, Cancer
Chemother. Rep., 1974, 58, 49; (d ) C. E. Barry III, P. G. Nayar and
T. B. Begley, Biochemistry, 1989, 28, 6323; (e) K. Maruyama,
T. Moriguchi, T. Mashino and A. Nishinaga, Chem. Lett., 1996, 819;
( f ) J. Kaizer, R. Csonka and G. Speier, J. Mol. Catal. : Chemical,
2002, 180, 91; (g) D. Goncalves, R. C. Faria, M. Yonashiro and
L. O. S. Bulhoes, J. Electroanal. Chem., 2000, 487, 90.
5 (a) L. I. Simándi, T. Barna, Gy. Argay and T. L. Simándi, Inorg.
Chem., 1995, 34, 6337; (b) L. I. Simándi and T. L. Simándi, J. Chem.
Soc., Dalton Trans., 1998, 3275.
6 (a) L. I. Simándi, T. Barna, L. Korecz and A. Rockenbauer,
Tetrahedron Lett., 1993, 34, 717; (b) L. I. Simándi, T. Barna and
S. Németh, J. Chem. Soc., Dalton Trans., 1996, 473.
7 (a) T. L. Simándi and L. I. Simándi, J.Chem. Soc., Dalton Trans.,
1999, 4529; (b) L. I. Simándi and T. L. Simándi, J. Inorg. Biochem.,
2001, 86, 432.
8 I. W. Pang and D. V. Stynes, Inorg. Chem., 1977, 16, 590.
9 J. F. Corbett, Spectrochim. Acta, 1965, 21, 1411.
Scheme 2 Transition state (X) of the rate-determining H-atom
transfer step (7).
Conclusion
The major mechanistic features of the ferroxime()-catalysed
oxidation of Hap to apx, derived from kinetic, ES-MS, ESR
and KIE data point to a free-radical process occurring via a
ternary catalyst–dioxygen–substrate complex X, in which the
phenolic OH group transfers an H-atom to a superoxoiron().
This feature is due to the rigid H-bonded square-planar
equatorial geometry of the catalyst complex. Only one of the
two axial MeIm ligands is lost in MeOH by solvolysis, vacating
an axial position for O2 to bind. No simultaneous coordination
of the substrate is possible in this situation, therefore, an even-
tual electron-transfer path is blocked. The favoured oxidation
route is thus H-atom transfer, consistent with the observed
kinetic isotope effect. The overall reaction resembles the
last stage of the biosynthesis of Actinomycin D, therefore,
ferroxime() can be regarded as a functional model of
phenoxazinone synthase. Ferroxime() exhibits catecholase
activity, too, via a similar mechanism and intermediates.
The H-atom transfer mechanism exemplified by the present
system has also been demonstrated for other catalytic reactions
of the catecholase type.5–7 It may be the preferred reaction path
in cases when complexes of square-planar macrocyclic ligands
operate in the presence of an axial ligand, which enhances the
O2-binding ability of the metal but prevents ternary complex
formation by blocking substrate binding in axial position.
Work is in progress to further assess the scope of this mech-
anistic pattern.
10 A. Rockenbauer and L. Korecz, Appl. Magn. Reson., 1996, 10, 29.
11 (a) Comprehensive Organic Chemistry, The Synthesis and Reactions
of Organic Compounds, ed. D. H. R. Barton and W. D. Ollis,
Pergamon Press, Oxford, New York, Toronto, Sydney, Paris,
Frankfurt, 1979, vol. 4, Heterocyclic Compounds, ed. P. G.
Sammes, pp. 1068–1078; (b) Y. Yano, M. Ikuta, Y. Amamiya and
T. Nabeshima, Chem. Lett., 1991, 461.
12 (a) F. Ruff and G. Csizmadia, Study of Organic Reaction
Mechanisms. Application of Kinetic Methods (in Hungarian),
Nemzeti Tankönyvkiadó: Budapest, 2000, ch. 8; (b) R. A. More
O’Ferrall, J. Chem. Soc. B, 1970, p. 785.
D a l t o n T r a n s . , 2 0 0 4 , 1 0 5 6 – 1 0 6 0
1060