Inactivation path during the copper (II) catalyzed synthesis of Questiomycin A from oxidation of 2-aminophenol

Article abstract of DOI:10.1016/j.apcata.2012.09.009

The catalytic oxidation of 2-aminophenol (OAP) to 2-amino-3H-phenoxazin-3- one (APX, Questiomycin A) was the object of numerous studies partly due to antimicrobial properties of Questiomycin A and mostly because it can be used as a model for the synthesis of the naturally occurring antineoplastic agent Actinomycin D. Several copper complexes were used as dioxygen and/or substrates activators in order to mimic the activity of phenoxazinone synthase, but the reported assays failed to provide reasonable mechanistic features in media compatible with natural conditions. The main purposes of our work were to use simple copper salts to perform oxidation of OAP in oxygenated aqueous solutions and to develop a reaction scheme able to explain the low yields in APX along with the operational inactivation of the catalyst. A 11-step kinetic model able to describe the inactivation of copper(II) catalyst during oxidation of OAP to APX in oxygenated solutions was developed, and the rate constants for both catalytic and non-catalytic branch were estimated either experimentally or using a computing program for detailed kinetic simulation. It was demonstrated that the inactivation path can be assigned to formation of the stable bis(o-iminosemiquinonato)copper(II) complex, a compound reported as a moderate antimicrobial agent.

Full text of DOI:10.1016/j.apcata.2012.09.009

Applied Catalysis A: General 447–448 (2012) 74–80
Contents lists available at SciVerse ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Inactivation path during the copper (II) catalyzed synthesis of Questiomycin A
from oxidation of 2-aminophenol
Ca˘ta˘lina Olmazua, Mihaela Puiua,b,∗, Irina Babaligeaa, Adina Ra˘ducana, Dumitru Oanceaa
a Department of Physical Chemistry, University of Bucharest, 4-12 Elisabeta Boulevard, 030018, Bucharest, Romania
b LaborQ, University of Bucharest, 4-12 Elisabeta Boulevard, 030018 Bucharest, Romania
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 3 July 2012
Received in revised form 9 August 2012
Accepted 5 September 2012
Available online 14 September 2012
The catalytic oxidation of 2-aminophenol (OAP) to 2-amino-3H–phenoxazin-3-one (APX, Questiomycin
A) was the object of numerous studies partly due to antimicrobial properties of Questiomycin A and
mostly because it can be used as a model for the synthesis of the naturally occurring antineoplastic agent
Actinomycin D. Several copper complexes were used as dioxygen and/or substrates activators in order
to mimic the activity of phenoxazinone synthase, but the reported assays failed to provide reasonable
mechanistic features in media compatible with natural conditions. The main purposes of our work were
to use simple copper salts to perform oxidation of OAP in oxygenated aqueous solutions and to develop
a reaction scheme able to explain the low yields in APX along with the operational inactivation of the
catalyst. A 11-step kinetic model able to describe the inactivation of copper(II) catalyst during oxidation
of OAP to APX in oxygenated solutions was developed, and the rate constants for both catalytic and non-
catalytic branch were estimated either experimentally or using a computing program for detailed kinetic
simulation. It was demonstrated that the inactivation path can be assigned to formation of the stable
bis(o-iminosemiquinonato)copper(II) complex, a compound reported as a moderate antimicrobial agent.
Keywords:
2-Aminophenol
Oxidation
Questiomycin A
Kinetic modelling
Copper(II) complex
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
synthesis of the naturally occurring antineoplastic agent Actino-
mycin D [16].
Catalytic activation of dioxygen by copper ions in selective oxi-
dations of organic compounds was paid considerable attention over
at least four past decades mostly due to the abundance of cop-
per in biological systems [1–3]. Copper-containing enzymes are
usually acting as oxygen carriers in oxidation reactions of a large
variety of substrates (carbohydrates, amines, phenols etc.) [4,5].
The role of copper centers in these enzymes, along with their
detailed structure, is not entirely elucidated at present and may dif-
fer from enzyme to enzyme [6,7]. Mimicking systems using copper
complexes as dioxygen and/or substrates activators were widely
studied for understanding the complex pathways and mechanis-
tic features of biological oxidations [8–12]. In this context, the
oxidative coupling of 2-aminophenol (OAP) to the stable product,
2-amino-3H-phenoxazin-3-one (APX) through catalytic activation
of dioxygen/substrate was the subject of numerous studies [13–15]
partly due to antimicrobial properties of APX (also known as Ques-
tiomycin A) but mostly because it can be used as a model for the
Many works were dedicated in the past years to synthesis, char-
acterization of copper-, iron-manganese and cobalt (II) complexes
which were used to mimic the activity of phenoxazinone syn-
thase (PHS) [5,9,13], a two-copper centre oxidase isolated from
Streptomyces [7] which catalyzes the oxidation of substituted 2-
aminophenols to the phenoxazinone chromophore in the final step
of the Actinomycin D biosynthesis. It is now widely accepted that
the six-electron oxidative condensation of two molecules of OAP
to form APX occurs via 2-aminophenoxyl free radicals [15,17,18].
The generation of aminophenoxyl radical is supposed to occur cat-
alytically or in the absence of the catalyst (auto-oxidation), but the
following dismutation, addition, cyclization and oxidation steps do
not require catalyst assistance [19] (Scheme 1).
It is worth mentioning several works which proposed copper(I)
and (II) salts (CuCl, CuCl2, CuSO4, Cu(NO3)2, Cu(OCH3)2) and oxides
(CuO, Cu2O) [17,20] or Fe, Mn and Co(II) based complexes [5,13,21]
as mimic systems of PHS activity; most of them provided con-
versions of OAP to APX up to 90% [20,21]. However, the main
disadvantages of these assays are that the studies were achieved
in organic solvents (acetonitrile, dimethylformamide, methanol)
[13,15,20] at temperatures ranging from 40 to 60 ◦C [13,17,20],
which are not compatible to the natural conditions where PHS
displays its optimum activity [7]. Monocopper–dioxygen adducts
∗
Corresponding author at: Laboratory of Quality Control and Process Monitor-
ing, University of Bucharest, 4-12 Regina Elisabeta Boulevard, 030018 Bucharest,
Romania. Tel.: +40 21 4104888; fax: +40 21 4104888.
E-mail address: elenamihaela.puiu@g.unibuc.ro (M. Puiu).
0926-860X/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2012.09.009

C. Olmazu et al. / Applied Catalysis A: General 447–448 (2012) 74–80
75
NH
+
N
NH2
NH2
NH2
O
catalytic
O2
O2
3/2
2
+
.
HOO
+
+
3 H 2
O
OH
O
O
OH
OAP
APX
OAP
NH
+
NH
NH2
O
noncatalytic
Scheme 2. Overall equation of OAP oxidation to APX.
2
OH
O
KBr pellets on a BrukerTensor 37 spectrophotometer. High perfor-
mance liquid chromatography (HPLC) analysis of reaction mixtures
was performed on a HPLC-DAD system Agilent 1100 using a RP
chromatographic column Kinetex C18Phenomenex 100 mm length
and 2.1 mm inner diameter.
NH
NH
+
N
O
NH
O
noncatalytic
2
O
2
2
+
2
2
4
H2O
+
OH
O
APX
Scheme 1. Simplified model of kinetically significant oxidation steps of OAP.
2.3. Synthesysis of bis(o-iminosemiquinonato)copper(II) complex
The bis(o-iminosemiquinonato)copper(II) complex was
obtained as described in literature [4]: the solution of Cu(II)
salt was added in small portions to the ligand solution under
continuous stirring, so that the complexation always took
place with the excess OAP present. Each solution of 0.050 mM
CuCl2/CuSO4/Cu(CH3COO)2 in 10 ml of water was added drop-
wise to a colourless solution of 0.100 mM of OAP dissolved in
10 ml of ethanol (molar ratio Cu(II):OAP = 1:2). As OAP can be
rapidly oxidized by oxygen, in order to prevent the production of
o-iminosemiquinones, an argon stream was bubbled through the
solutions during the synthesis to ensure the absence of oxygen.
Coloured precipitates of Cu(II) complex formed instantaneously.
After 1.5 h stirring they were collected, washed with ethanol and
water and dried. The highest conversion of the Cu(II) salt to Cu(II)
complex was observed in the case of Cu(CH3COO)2 (60–70%) while
the lowest conversion was obtained for CuCl2 (10–15%).
have been postulated as intermediates in enzyme catalysis (notable
examples being dopamine ␤-monooxygenase and galactose oxi-
dase [22]), formulated as Cu(II) superoxo species [23]. Their
structures were demonstrated kinetically, spectroscopically and
by X-ray crystallography [8,22]. Many synthetic copper–oxygen
adducts were prepared following the activation of dioxygen when
binding to Cu(I) complexes with ␤-diketiminate ligands in order
to explain/predict the existence of several active intermediates
and the mechanistic features in copper catalyzed/mediated oxi-
dations of organic compounds [22,24]. In this work we proposed
and developed a reaction sequence for Cu(II) catalyzed oxidation of
OAP which includes as plausible active intermediates Cu(II) super-
oxo species and Cu–OAP complexes, whose occurrence is sustained
in literature data [4,23]. An inactivation step, assigned to the for-
mation of the stable bis(o-iminosemiquinonato)copper(II) complex
(also described in literature [4]), was added and experimentally
checked. The Cu(II)-catalyzed oxidation of OAP being followed in
air or dioxygen saturated solutions, the kinetic model took into
account the reactions involving dissolved dioxygen according to
the overall stoichiometry (Scheme 2):
2.4. Kinetic assays
The stability of OAP in the absence of oxygen was verified by
passing an argon stream through the reaction mixture, when no
significant transformation of OAP was observed. The initial concen-
tration of OAP in the reactor ranged within 1 × 10−4 and 3 × 10−3 M,
due to the low solubility in water of OAP and APX. Chemical
transformation of the substrate was monitored by product (APX)
analysis. For each experiment, 14–20 samples were removed from
reactor at different time intervals and the spectra of the reaction
mixture were recorded between 200 and 800 nm. HPLC analysis of
the reaction mixture indicates as the main reaction product APX,
other intermediates being accumulated in insignificant amounts
towards OAP and APX. The validity of Dalton’s law was assumed
when calculating the concentration of dioxygen in saturated aque-
ous solution [27]. The air or dioxygen stream was bubbled through
two thermostated saturation vessels containing water to avoid the
evaporation of the solvent from reactor. Under these conditions the
solution was saturated with dioxygen. All kinetic assays were run
at 30 ◦C. APX formation was monitored measuring the absorbance
at 434 nm (where OAP has no significant absorption) of the output
sample (ε = 23,200 cm−1 M−1 for APX) [14]. The pH of each reaction
mixture, established after copper salt addition, was also monitored
in the output stream, but no significant deviation from the initial
value (pH = 5.9–6.0) was noticed. Similar kinetic runs, carried out
in the absence of catalyst, indicated a significant transformation
with the formation of the same product. For the evaluation of the
kinetic parameters involved in the catalytic reaction mechanism,
the derivative d[APX]/dt assigned to the reaction rate of the cat-
alytic branch was obtained by subtracting the reaction rate of the
non-catalytic oxidation of OAP from the overall reaction rate at the
same conversion.
The proposed model takes into account the generation of reac-
tive oxygen species (ROS) such as HOO•, HO• and other highly
reactive radical species and starts with a set of rate constants
reported in literature for 2-aminophenol oxidation [18,19] and
generation of ROS species [25,26], others being obtained using an
optimization algorithm.
2. Experimental
2.1. Materials
OAP was obtained from Aldrich Chemical Co. APX (m.p. = 249 ◦C)
was synthesized upon the oxidation of 2-aminophenol with mer-
cury oxide followed by recrystallization from ethanol, as described
in several works [18,20]. Argon and gaseous dioxygen were pur-
chased from Linde (99.9% purity for argon, 99.6% for dioxygen).
Copper(II) salts: chloride, sulphate and acetate were of the highest
purity grade from Fluka. Other chemicals used were of analytical
reagent grade.
2.2. Apparatus
The ultraviolet–visible (Uv–Vis) measurements during the
copper-catalyzed oxidation of OAP were made on a Lambda 25
Perkin-Elmer spectrophotometer with a Peltier cell for tempera-
ture and stirring control. The infrared (IR) spectra of solid OAP, APX
and bis(o-iminosemiquinonato)copper(II) complex were recorded
in the wavelength range 4000–400 cm−1 at room temperature in

76
C. Olmazu et al. / Applied Catalysis A: General 447–448 (2012) 74–80
Fig. 1. Turnover frequency for three copper(II) salts: [CuCl2]0, [CuSO4]0,
[Cu(CH3COO)2]0 = 1 × 10−5 M, [OAP]0 = 1 × 10−3 M, T = 30 ◦C.
Fig.
3. Selwyn
proof
of
catalyst
inactivation:
[OAP]0 = 1 × 10−4 M,
[O2]dissolved = 1.17 × 10−3 M, T = 30 ◦C.
3. Results and discussion
The analysis of the extended progress curves (substrate or
product concentration vs. time) allows checking the catalyst inacti-
vation in excess of OAP, as in the case of the Selwyn test, originally
conceived for the enzyme inactivation [19,28]. Accordingly, in the
absence of inactivation, the graphs showing absorbance of APX (or
an amount directly proportional to concentration of APX) versus the
product [Cu(II)]0 × time, for the same OAP concentration at several
[Cu(II)]0 concentrations, should be superposable, which was not
observed in the studied system (Fig. 3).
The catalyst inactivation can be assigned to formation of bis(o-
iminosemiquinonato)copper(II) complex Cu(SH)2 (where SH is
aminophenoxyl radical), which is formed upon the interaction of
Cu(II) ions with OAP. The neutral square planar Cu(II) complex of
the stoichiometry Cu(SH)2 has been shown to be a diradical with a
singlet ground state [4,29]. The complex was characterized upon its
IR features against of those of OAP and APX, identifying the bands
previously reported in literature for Cu(SH)2 [4](Table 1).
It was observed that APX was obtained according to the mass
balance given by the stoichiometric equation (Scheme 2). In these
conditions OAP oxidation to APX catalyzed by copper salts can be
treated as a single reaction.
The reaction rates in initial conditions (2% conversion) were cal-
culated as the slopes of the linear parts of the extended kinetic
curves [APX] towards time and the turnover frequencies were
obtained by dividing the slopes values to salt concentrations (Fig. 1).
The catalytic efficiency introduced by different copper salts can
be assigned to anion effects resulting in incomplete dissociation
(copper acetate), ion pair formation (copper sulphate) and possibly
to some specific interactions.
Taking into account that the highest catalytic efficiency was
achieved with CuCl2, the following kinetic assays were focused on
oxidation of OAP catalyzed by this compound. It was observed that
the plots of initial rates against initial concentrations of OAP in
dioxygen-saturated solution exhibited a peak-shaped curve sim-
ilar to substrate inhibition pattern in enzymatic assays (Fig. 2).
A 90% catalytic conversion in dioxygen saturated solution was
achieved after 150 min. while the non-catalytic conversion was
around 30% in the same conditions. Compared to the high yields
obtained in time ranges up to 10 min. using several Cu(II) com-
plexes in non-aqueous media which were widely described in
literature [5,14,15,17], the low yields observed in aqueous media
provide another argument to sustain the hypothesis of a inactive
catalyst–substrate complex formation.
3.1. Kinetic modelling
According to recent reported data about oxidation of OAP [6,9]
and of several o-diphenols [2], an appropriate kinetic model can
be developed comprising the generation of ROS species during
the auto-oxidation of OAP in oxygenated solutions as well as for-
mation of free radical products via catalytic steps. It is generally
accepted that in solution Cu(II) is present as hexaaqua complex
(Cu(aq)6)2+ and during the reaction steps of the catalytic branch it
can be replaced by other ligands. In order to simplify the reaction
sequence, the water molecules are omitted from the formulas of
the involved copper(II) species.
k
1
SH2 + Cu2+−→ (CuSH)+ + H+
(1)
(2)
(3)
k
2
(CuSH)+ + O2−→ (CuO2)+ + SH•
k
•
3
+
(CuO2) −→ Cu2+ + O2
−
k
O•2 + H+−→ HOO•
(4)
(5)
4
−
k
5
SH2 + O2−→ SH• + HOO•
k
6
2SH•−→ P
(6)
(7)
k
7
2HOO•−→ H2O2 + O2
Fig. 2. Substrate inhibition of the initial reaction rate vs. initial concentration of
substrate [O2]dissolved = 1.17 × 10−3 M, T = 30 ◦C.
k
8
SH2 + H2O2−→ SH• + HO• + H2O
(8)

C. Olmazu et al. / Applied Catalysis A: General 447–448 (2012) 74–80
77
Table 1
IR spectral assignments of OAP, APX and Cu(II) complex (cm−1).
Compound
OAP
ꢀ˜ (C O)/cm−1
1226
ꢀ˜ (C N)/cm−1
1283
ꢀ˜ (C C)/cm−1
1605
ꢀ˜ (N H)/cm−1
ꢀ˜ (Cu N)/cm−1
ꢀ˜ (Cu O)/cm−1
3304
APX
1210
1273
1590
3413
3306
Cu(SH)2
1160
1296
1603
3286
620
422
599
526
k
[5,14,20], we conclude that the rate determining step is the for-
mation of aminophenoxyl radical.
9
H2O2 + HO•−→ HOO• + H2O
(9)
(10)
(11)
k
- Estimation of k8: as in the case of OAP auto-oxidation, the oxi-
dation of OAP in the presence of H2O2 displays a second-order
kinetics. Since the HPLC analysis of the reaction mixtures for both
auto-oxidation OAP and oxidation with H2O2 proved that APX is
the main reaction product and other intermediates are accumu-
lated in insignificant amounts, steps (5) and (8) can be treated as
single reactions, each being described by a single rate equation.
10
HOO• + H2O2−→ HO• + H2O + O2
k
11
(CuSH)+ + SH2−→ Cu(SH)2 + H+
where SH2 is OAP. Occurrence of the active intermediate (CuSH)+
is sustained by the formation (in absence of oxidizing agents)
of the inactive complex (Cu(SH)2). The formation of the highly
reactive species (CuO2)+ is supported by recent reported data
regarding the structure and reactivity of Cu(II)-superoxo adducts
[22–24]. The assumption that ROS such as HOO• and H2O2 are
released during the auto-oxidation is also sustained by litera-
ture reported data [25,26] and from observed effects of copper(II)
catalyst on OAP oxidation. In CuCl2 aqueous solution copper is
present as Cu2+, Cu(OH)+, CuCl+, CuCl2. The concentrations of these
species were calculated solving the system of equations resulted
from the equilibrium constants of formation (log KCu(OH)+ =
The step 6 includes several fast elementary steps (dismutations,
additions and oxidations occurring without catalyst assistance
[15]); therefore, it can be treated as a single step in the specified
working conditions (oxygen saturated solution). The constant k6
was fixed at a plausible value of 1 × 108 M−1 s−1 because varying
this parameter within 1 × 106–1 × 1010
M
−1 s−1(specific range of
rate constant for fast reactions in solution) does not change the
values of the optimised parameters.
−7.497,
log KCu(Cl)+ = 0.2,
log KCuCl = −0.26) [30] and
2
- Estimation of k11 the inactivation step of the copper(II) catalyst
assigned to formation of the inactive bis(o-iminosemiquinonato)
copper complex can be pursued using an isoconversional method
as described elsewhere for a first-order catalyst inactivation [28].
As long as the concentration of Cu(II) salt is much lower than
the substrates concentrations OAP and O2, a quasi-steady state is
attained and the time evolution of the system can be described by
a single overall reaction rate. The corresponding kinetic equation,
based on the quasi-steady state or quasi-equilibrium assump-
tions, coupled with the catalyst molar balance, takes the general
form:
the material balance equations, for [CuCl2]0 = 1 × 10−5 M and
pH = 6.0. It was found that (Cu(aq)6)2+ is the most stable species
(97%).
The simulation providing the kinetic behaviour of the entire sys-
tem has to start from a particular set of rate constants and initial
concentrations of the chemical species involved in the model. The
rate constants found in literature are summarized in Table 2.
In addition, two rate constants (k5 and k8) were estimated
experimentally following the experimental design and theoretical
reasoning depicted in several previous works [18,19,31]. Briefly, an
apparent first-order rate constant kꢀ5 was estimated by non-linear
regression analysis after fitting an apparent first-order growth
function on the progress kinetic curve [APX] vs. time obtained at
the oxidation of OAP with dioxygen in dioxygen-saturated solu-
tion. The second-order rate constant k5 was obtained as the ratio
of kꢀ5 and [O2]: k5 = (1.191 0.002) × 10−2 M−1 s−1with a signifi-
cance level ˛ = 0.05. In a similar manner was obtained k8, using
the progress kinetic curve [APX] vs. time obtained during oxida-
tion assay of OAP with hydrogen peroxide in a100-fold excess of
H2O2: k8 = (2.011 0.006) × 10−3 M−1 s−1with a significance level
˛ = 0.05.
d[APX]
dt
= [Cu(II)]0 × f [APX]
(12)
(
)
where f([APX]) is dependent only on the OAP conversion, if all other
operational variables are held constant.
Within the isoconversional method, several kinetic runs are
performed maintaining the initial concentrations of OAP and O2
constant and varying Cu(II)0 concentration. A family of extended
progress curves [APX] as a function of time, for different [Cu(II)]0
concentrations, is obtained. For a certain product concentration,
resulted at the same OAP conversion, designated as [APX]isoconv
,
different reaction times t1, t2, t3,. . . can be measured on each curve,
corresponding to [Cu(II)]0,1, [Cu(II)]0,2, [Cu(II)]0,3. . . concentrations.
Arbitrary values of [APX]isoconv can be selected as long as they
are present on all progress curves. At each point [Cu(II)]0,i–ti, the
reaction rate (d[APX]/dt)i can be calculated numerically by fitting
a convenient function (e.g. a polynomial) to each experimental
progress curve and taking its first derivative. Eq. (12) takes the form:
- Estimation of k5: it was previously shown that the oxidation
of OAP with dioxygen, in dioxygen saturated solution follows
apparent first-order kinetics [18] with respect to substrate con-
centration. The partial reaction order towards dioxygen also being
found equal to one [18,31] and taking into account Scheme 1
Table 2
ꢀ
ꢁ
d[APX]
dt
Rate constants in aqueous medium used in the optimisation procedure.
= f ([APX]isoconv) × [Cu(II)]∗
(13)
0,i
Rate constant
Value/M−1 s−1
Reference
i
where [Cu(II)]∗0,i is the total copper(II) concentration for run i, active
at time ti, considering that the remaining part of the catalyst is
converted into an inactive form. The time evolution of the active
k7
k9
k10
1.6 × 105
2.7 × 107
0.50
[21,22]
[14]
[21]

78
C. Olmazu et al. / Applied Catalysis A: General 447–448 (2012) 74–80
catalyst can be approximated in this case from the catalyst molar
balance:
[Cu(II)]∗0 + [Cu(SH)2] = [Cu(II)]0
(14)
where
[Cu(II)]∗ + [(CuSH)+] + [(CuO2)+] = [Cu(II)]∗
(15)
0
The terms [(CuSH)+] and [(CuO2)+] can be calculated from
the approximate steady-state conditions d[(CuSH)+]/dt ≈ 0 and
d[(CuO2)+]/dt ≈ 0, or:
k1•[Cu(II)]∗•[SH2] = k2•[(CuSH)+][O2] + k11[(CuSH)+][SH2]
k2[O2][(CuSH)+] = k3[(CuO2)+]
(16)
(17)
The equation system (15)–(17) allows calculating the concentra-
tions of actives intermediates (CuSH)+, and (CuO2)+ as well as the
concentration of active unbound catalyst Cu(II)∗ as function of the
total concentration of active catalyst Cu(II)∗0. The formation rate of
the inactive complex Cu(SH)2 can be written as:
Fig. 4. Estimation of k in1 in dioxygen/air saturated solutions according to the first
limiting case – Eqs. (25) and (26). [OAP]0 = 1 × 10−4 M, T = 30 ◦C, conversion ≈ 20%,
[CuCl2]0 ranging from 5 × 10−7 M to 1 × 10−6 M.
d[Cu(SH)2]
= k11[(CuSH)+][SH2]
(18)
dt
The concentration of the inactive complex can be calculated also
from Eq. (14)
On
the
other
hand,
assuming
that
k2k3[O2] ꢂ (k1k3 + k1k2[O2])[SH2], Eq. (22) becomes:
[Cu(SH)2] = [Cu(II)]0 − [Cu(II)]∗
(19)
0
d[Cu(II)]∗0
dt
k1k11[SH2]2[Cu(II)]∗
d[Cu(II)]∗0
−
dt
0
−
=
=
or
Deriving Eq. (19) with respect to time and equalling to Eq. (18) one
obtains:
k2[O2]
k1k11[SH2]2
[Cu(II)]∗0
k2[O2]
d[Cu(II)]∗0
= k11[(CuSH)+][SH2]
(20)
(27)
−
dt
where
After integration one obtains:
k1k3[Cu(II)]∗ [SH2]
k2k3[O2] + [SH2](k3k11 + k1k3 + k1k2[O2])
[(CuSH)+] =
(21)
0
∗
I
[Cu(II)]0,i = [Cu(II)]0,i × e−k
(28)
(29)
2
in2
with
Since the inactivation step (11) is much slower than the steps
(1) and (2) it can be neglected in the steady-state condition (16)
and consequently Eq. (20) becomes:
ꢂ
t
isoconv
k1k11
kin2
=
and I2
=
[SH2]2dt
k2[O2]
0
d[Cu(II)]∗0
dt
k1k3k11[SH2]2[Cu(II)]∗
0
The integrals I1 and I2 can be calculated numerically as the area
under the curves [OAP] or [OAP]2 as functions of time, after fitting
smooth functions (polynomials) on the experimental set of data of
the form [OAP] = [OAP]0 − 2[APX]. Replacing the concentration of
active catalyst from Eq. (25) in Eq. (13), one obtains:
−
=
(22)
k2k3[O2] + (k1k3 + k1k2[O2]) [SH2]
No simple procedure exists to integrate this equation in order
to obtain [Cu(II)]∗0 as a function of time when [SH2] changes in
time. However, two simpler limiting cases are obtained when
one of the two terms in the denominator can be neglected.When
k2k3[O2] ꢁ (k1k3 + k1k2[O2])[SH2], Eq. (22) takes the following
form:
(d[APX]/dt)i
soconv
I
1
· e−k
(30)
in1
= f [APX]
(
)
isoconv
[Cu(II)]0,i
d[Cu(II)]∗0
k1k3k11[SH2]
d[Cu(II)]∗0
−
dt
The inactivation constant kin1 can be estimated using a nonlinear
regression analysis of Eq. (30) with (d[APX]/dt)isoconv/[Cu(II)]0,i as
−
=
=
[Cu(II)]∗0 or
dt
k k + k k [O ]
(
)
1
3
1
2
2
dependent variable, I1 as independent variable and f [APX]
(
)
isoconv
k11[SH2]
and kin1 as estimated parameters, or using a linear regression anal-
ysis of the logarithmic form of the same equation. The evaluation
of the inactivation kinetic constant kin1 is illustrated in Fig. 4:
Alternatively, replacing the concentration of the active catalyst
from Eq. (28) in Eq. (13), one obtains:
∗
ꢀ
ꢀ
ꢁ
ꢁ
[Cu(II)]0
(23)
k
k
2
3
1 +
[O2]
Integration of Eq. (23) after separating the variables, and taking
into account that the concentration of the dissolved dioxygen was
maintained constant during the oxidation assays, gives:
∗
(d[APX]/dt)
[Cu(II)]0,i
I
2
in2
= f [APX]
· e−k
(31)
(
)
isoconv
ꢂ
ꢂ
Cu(II)
t
d[Cu(II)]∗0
[Cu(II)]∗0
k11
0
ꢃ
ꢄ
−
=
[SH2]dt
(24)
Following the same procedure, the linear regression analysis of
the logarithmic form of Eq. (31) is illustrated in Fig. 5. Both the
derivatives (d[APX]/dt)i and integrals I1 and I2 have been calculated
numerically.
1 + (k2/k3)[O2]
[Cu(II)]
0
0
For i set, from t = 0 to isoconversional time, tisoconv one obtains:
∗
[Cu(II)]0,i = [Cu(II)]0,i × e−k
(25)
I
1
in1
The results for kin1 (0.361 0.030 s−1 for O2 and
0.366 0.016 s−1 for air) and for kin2 (4165 169 s−1 for O2
and 4023 342 s−1 for air) indicate an apparent independence of
the inhibition of the dissolved oxygen concentration. The numeri-
cal integration and optimization of the rate constants involved in
where the inactivation constant kin1 and integral I1 are:
ꢂ
t
isoconv
k11
kin1
=
and I1
=
[SH2]dt
(26)
1 + (k2/k3)[O2]
0

C. Olmazu et al. / Applied Catalysis A: General 447–448 (2012) 74–80
79
Fig. 6. Simulated and experimental extended kinetic curves of the reac-
tion intermediate and products. [OAP]0 = 1 × 10−4 M, [O2]dissolved = 1.17 × 10−3 M,
[CuCl2]0 = 1 × 10−5 M, T = 30 ◦C .
Fig. 5. Estimation of k in2 in dioxygen/air saturated solutions according to the second
limiting case [Eqs. (28) and 29)]. [OAP]0 = 1 × 10−4 M, T = 30 ◦C, conversion ≈ 20%,
[CuCl2]0 ranging from 5 × 10−7 M to 1 × 10−6 M.
This explains the apparent independence of the inhibition con-
stant on the dioxygen concentration. For dioxygen saturated
solution the approximation is even rougher but still in accord
with the experimental behaviour. It can be concluded that only
the first limiting approximation k2k3[O2] ꢁ (k1k3 + k1k2[O2])[SH2],
although rough, can explain the results, while the second one,
predicting an inverse proportionality of the inhibition constant
with the oxygen concentration is in disagreement with the results.
The above approximations can partly explain the difference
between the measured inhibition constant kin1 = 0.366 M−1 s−1
the system of differential Eqs. (1)–(11) can help to discriminate
between these limiting cases and to establish the significance of
the inhibition constant.
3.2. Simulation assays
The time evolution of the reacting system was described by a
set of differential equations obtained from the reaction steps 1–11,
corresponding to the transformation rates for all species involved
in the proposed model (except dioxygen whose concentration is
kept constant in the working conditions). The optimisation process
was accomplished with Kintecus [32] on 20 different experimental
data sets.
and calculated value k11 = 0.048 M−1 s−1
.
4. Conclusions
The oxidation of 2-aminophenol with dissolved dioxygen in
aqueous solution to give 2-amino-3H-phenoxazin-3-one is cat-
alyzed by simple copper salts. The kinetics of the process indicates
the existence of an inhibition by the substrate, confirmed by the
characteristic variation of the initial reaction rate with substrate
concentration, which presents a maximum, and also by the Selwyn
test. A kinetic model implying 11 steps (catalytic and non-catalytic)
was proposed, where the inhibition step represents the forma-
tion of bis(o-iminosemiquinonato)copper(II) complex, isolated and
characterized. Several rate constants of the component steps
were taken from literature and the others were evaluated either
experimentally or numerically using an optimization procedure
according to Kintecus software. The inhibition constant was also
evaluated using an isoconversional procedure described earlier.
The agreement between the measured and simulated inhibition
constant proves that this method can rapidly provide mechanis-
tic insights of the studied system and accurately describe the time
evolution of active intermediates or side reaction products. The
relative simple and costly-efficient manner to detect and evaluate
the degree of operational inactivation of a catalyst in homogenous
media is suitable for routine analysis in batch/flow reactors where
the output signal of a stable reaction product can be measured, even
when dealing with a single reaction or with a reaction network.
3.3. Sensitivity analysis
The sensitivity analysis, accomplished with Atropos [33] at three
different time intervals in order to avoid over-parameterization
shows that all rate constants do influence the shape of the
simulated progress curves and consequently all proposed steps
can be considered kinetically significant. The optimized val-
ues are: k1 = (58.50 0.04) M−1 s−1
,
k2 = (2.18 0.04) M−1 s−1
,
k3 = (2.80 0.08) × 102 s−1
,
k4 = (5.100 0.024) × 106 M−1 s−1and
k11 = (4.80 0.05) × 10−2 M−1 s−1 (with ˛ = 0.05). The simulated
progress curves for the proposed active and inactive copper species
(CuSH)+, Cu(SH)2 together with both simulated and experimental
kinetic curves for APX are displayed in Fig. 6, for substrate con-
version around 90%. The agreement between experimental and
simulated kinetic curves of APX was noticed likewise on the other
experimental data sets.
The optimized parameters k1, k2, k3, k11 and concentrations [O2]
and [SH2] can be used to decide which terms in the denominator of
Eq. (21) and those subsequently derived from it can be neglected.
With present data Eq. (22) becomes simpler but still difficult to
integrate:
d[Cu(II)]∗0
dt
k1k11[SH2]2[Cu(II)]∗
k2[O2] + k1[SH2]
0
−
=
(32)
Acknowledgements
For air saturated solutions k2[O2] ≈ 4.9 × 10−4 s−1 and
k1[SH2] ≈ 5.9 × 10−3 s−1. If the first term in the denominator
is neglected as a rough approximation, one obtains kin1 ≈ k11, a
result obtained also from Eq. (26) since k2[O2]/k3 = 1.7 × 10−6 ꢁ 1.
This work was supported by the strategic grant POS-
DRU/89/1.5/S/58852 project “Postdoctoral programme for training
young scientific researchers”, co-financed by the European Social

80
C. Olmazu et al. / Applied Catalysis A: General 447–448 (2012) 74–80
Fund within the Sectorial Operational Programme for Human
Resources Development 2007–2013.
[16] S. Shimizu, M. Suzuki, A. Tomoda, S. Arai, H. Taguchi, T. Hanawa, S. Kamiya,
Tohoku J. Exp. Med. 203 (2004) 47–52.
[17] L. Prati, M. Rossi, N. Ravasio, J. Mol. Catal. 75 (1992) 347–355.
[18] M. Puiu, A. Ra˘ducan, I. Babaligea, D. Oancea, Bioproc. Biosyst. Eng. 31 (2008)
579–586.
[19] M. Puiu, M. Constantinovici, I. Babaligea, A. Raducan, C. Olmazu, D. Oancea,
Chem. Eng. Technol. 33 (2010) 414–420.
[20] T. Horváth, J. Kaizer, G. Speier, J. Mol. Catal. A: Chem. 215 (2004) 9–15.
[21] C. Mukherjee, T. Weyhermüller, E. Bothe, P. Chaudhuri, Compt. Rend. Chem. 10
(2007) 313–325.
References
[1] M. Ahmed, Coord. Chem. Rev. 254 (2010) 1918–1947.
[2] G. Busca, S. Berardinelli, C. Resini, L. Arrighi, J. Hazard. Mater. 160 (2008)
265–288.
[3] J. Aran˜a, V.M. Rodríguez López, E. Pulido Melián, M.I. Suárez Reyes, J.M. Don˜a
Rodríguez, O. González Díaz, Catal. Today 129 (2007) 177–184.
[4] N.V. Loginova, T.y.V. Koval’chuk, R.A. Zheldakova, N.P. Osipovich, V.L. Sorokin,
G.I. Polozov, G.A. Ksendzova, G.K. Glushonok, A.A. Chernyavskaya, O.I. Shadyro,
Bioorg. Med. Chem. Lett. 16 (2006) 5403–5407.
[22] D. Maiti, H.C. Fry, J.S. Woertink, M.A. Vance, E.I. Solomon, K.D. Karlin, J. Am.
Chem. Soc. 129 (2006) 264–265.
[23] A.M. Reynolds, B.F. Gherman, C.J. Cramer, W.B. Tolman, Inorg. Chem. 44 (2005)
6989–6997.
[24] N.W. Aboelella, S.V. Kryatov, B.F. Gherman, W.W. Brennessel, V.G. Young,
R. Sarangi, E.V. Rybak-Akimova, K.O. Hodgson, B. Hedman, E.I. Solomon, C.J.
Cramer, W.B. Tolman, J. Am. Chem. Soc. 126 (2004) 16896–16911.
[25] D.E.C. Benon, H.J. Bielski, A.B. Ross, J. Phys. Chem. Ref. Data 14 (1985)
1041–1100.
[26] I. Boutelet, S. Alexandre, J.-C. Vincent, Eur. J. Biochem. 223 (1994)
489–496.
[27] IUPAC Solubility Data Series in: Oxygen and Ozone, Pergamon Press, New York,
1981.
[5] T.M. Simandi, L.I. Simandi, M. Gyor, A. Rockenbauer, A. Gomory, Dalton Trans.
(2004) 1056–1060.
[6] W.A. Alves, S.A.d. Almeida-Filho, M.V.d. Almeida, A. Paduan-Filho, C.C. Becerra,
A.M.D.C. Ferreira, J. Mol. Catal. A: Chem. 198 (2003) 63–75.
[7] M. Le Roes-Hill, C. Goodwin, S. Burton, Trends Biotechnol. 27 (2009) 248–258.
[8] G. Battaini, A. Granata, E. Monzani, M. Gullotti, L. Casella, Biomimetic oxidations
by dinuclear and trinuclear copper complexes, in: R.v. Eldik, J. Reedijk (Eds.),
Advances in Inorganic Chemistry, Academic Press, 2006, pp. 185–233.
[9] M.R. Maurya, S. Sikarwar, T. Joseph, S.B. Halligudi, J. Mol. Catal. A: Chem. 236
(2005) 132–138.
[10] M. Rolff, J. Schottenheim, G. Peters, F. Tuczek, Angew. Chem. Int. Ed. 49 (2010)
6438–6442.
[11] D. Li, Y. Tong, J. Huang, L. Ding, Y. Zhong, D. Zeng, P. Yan, J. Mol. Catal. A: Chem.
345 (2011) 108–116.
[12] M. Rolff, J. Schottenheim, H. Decker, F. Tuczek, Chem. Soc. Rev. 40 (2011)
4077–4098.
[13] M. Hassanein, M. Abdo, S. Gerges, S. El-Khalafy, J. Mol. Catal. A: Chem. 287
(2008) 53–56.
[14] J. Kaizer, R. Csonka, G. Speier, J. Mol. Catal. A: Chem. 180 (2002) 91–96.
[15] I.C. Szigyártó, T.M. Simándi, L.I. Simándi, L. Korecz, N. Nagy, J. Mol. Catal. A:
Chem. 251 (2006) 270–276.
[28] D. Oancea, A. Stuparu, M. Nita, M. Puiu, A. Raducan, Biophys. Chem. 138 (2008)
50–54.
[29] N.V. Loginova, T.V. Koval’chuk, N.P. Osipovich, G.I. Polozov, V.L. Sorokin, A.A.
Chernyavskaya, O.I. Shadyro, Polyhedron 27 (2008) 985–991.
[30] J.D. Allison, D.S. Brown, K.J. Novo-Gradac, A Geochemical Assessment Model for
Environmental Systems: Version 4.0 User’s Manual, in: MINTEQA2/PRODEFA2,
United States Environmental Protection Agency, Office of Research and Devel-
opment, Washington, DC, 1999.
[31] D. Oancea, M. Puiu, Cent. Eur. J. Chem. 1 (2003) 233–241.
[32] J.C. Ianni, Kintecus Windows Version 4.5. www.kintecus.com
[33] J.C. Ianni, ATROPOS, Windows Version 1.0, 2004. www.kintecus.com.