An efficient copper(III) catalyst in the four electron reduction of molecular oxygen by l-ascorbic acid

Article abstract of DOI:10.1016/j.molcata.2010.10.026

The anionic [Cu^III(LH_-4)]^- complex (L = 8,17-dioxa-1,2,5,6,10,11,14,15-octaaza-tricyclo[13.3.1] eicosane-3,4,12,13- tetrone) is one of the rare copper(III) compounds, which exhibits high stability in basic aqueous solution. This copper(III) compound catalyses the reduction of molecular oxygen by l-ascorbic acid at pH 7.97, in Tris-HCl buffer solution at the ionic strength of 0.1 M (NaCl). Stoichiometry of the reaction (2:1.07 for l-ascorbic acid:dioxygen) indicates formation of water and dehydroascorbic acid as primary products. Based on detailed kinetic measurements, the rate equation-d[AscH^-]/dt = k_obs[AscH^-] ^0.5[Cu^III(LH_-4)][O₂]^0.5 was obtained. The proposed mechanism includes a fast redox pre-equilibrium between the copper(III) centre and its reduced, copper(II) form, induced by the presence of l-ascorbate. The equilibrium constant K₁′ at pH 8 (8.78 ± 3.11 × 10^-3) and k values for the forward and backward reactions (1.18 ± 0.68 × 10⁶ and 1.46 ± 0.82 × 10⁸ M^-2 s^-1, respectively) were determined by stopped-flow technique, following the decrease in absorbance of the copper(III) form at 550 nm. In the presence of molecular oxygen, re-oxidation of the copper(II) form of the catalyst takes place, based on cyclic voltammetry (CV) measurements. The decrease of the Cu(II) → Cu(III) oxidation and the subsequent increase of the Cu(III) → Cu(II) reduction current peaks in the CV spectrum, when argon is exchanged to dioxygen atmosphere, indicate a relatively fast oxidation rate for [Cu ^II(LH_-4)]^2-. The determined ΔS ^? (-41 ± 2 Jmol^-1 K^-1) for the catalytic reaction indicate an associative mechanism for the formation of the catalytically active copper-oxygen species that will react with the l-ascorbate to yield dehydroascorbate as product.

Full text of DOI:10.1016/j.molcata.2010.10.026

Journal of Molecular Catalysis A: Chemical 334 (2011) 77–82
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
An efficient copper(III) catalyst in the four electron reduction of molecular
oxygen by l-ascorbic acid
a,∗
b
b
c
c
˙
József S. Pap , Łukasz Szywriel , Magdalena Rowi n´ ska-Zyrek , Konstantin Nikitin , Igor O. Fritsky ,
b
Henryk Kozłowski
Department of Chemistry, University of Pannonia, Wartha V. u. 1., H-8201 Veszprém, Hungary
Faculty of Chemistry, University of Wrocław, 14, F. Joliot-Curie Str., 50383 Wrocław, Poland
Department of Chemistry, National Taras Shevchenko University, 01033 Kiev, Ukraine
a
b
c
a r t i c l e i n f o
a b s t r a c t
The anionic [CuIII(LH 4)] complex (L = 8,17-dioxa-1,2,5,6,10,11,14,15-octaaza-tricyclo[13.3.1] eicosane-
3
−
Article history:
Received 29 June 2010
Received in revised form 26 October 2010
Accepted 31 October 2010
Available online 10 November 2010
−
,4,12,13-tetrone) is one of the rare copper(III) compounds, which exhibits high stability in basic
aqueous solution. This copper(III) compound catalyses the reduction of molecular oxygen by l-
ascorbic acid at pH 7.97, in Tris–HCl buffer solution at the ionic strength of 0.1 M (NaCl).
Stoichiometry of the reaction (2:1.07 for l-ascorbic acid:dioxygen) indicates formation of water
and dehydroascorbic acid as primary products. Based on detailed kinetic measurements, the
Keywords:
Copper(III)
l-Ascorbic acid
Catalytic oxidation
Dioxygen reduction
rate equation–d[AscH ]/dt = kobs[AscH ] [CuIII(LH−4)][O2]
−
−
0.5
0.5
was obtained. The proposed mechanism
includes a fast redox pre-equilibrium between the copper(III) centre and its reduced, copper(II) form,
ꢀ
−3
induced by the presence of l-ascorbate. The equilibrium constant K1 at pH 8 (8.78 ± 3.11 × 10 ) and k val-
6
8
−2 −1
ues for the forward and backward reactions (1.18 ± 0.68 × 10 and 1.46 ± 0.82 × 10 M
s , respectively)
were determined by stopped-flow technique, following the decrease in absorbance of the copper(III) form
at 550 nm. In the presence of molecular oxygen, re-oxidation of the copper(II) form of the catalyst takes
place, based on cyclic voltammetry (CV) measurements. The decrease of the Cu(II) → Cu(III) oxidation
and the subsequent increase of the Cu(III) → Cu(II) reduction current peaks in the CV spectrum, when
II
2−
.
argon is exchanged to dioxygen atmosphere, indicate a relatively fast oxidation rate for [Cu (LH−4)]
‡
−1 −1
The determined ꢀS (−41 ± 2 Jmol
K ) for the catalytic reaction indicate an associative mechanism
for the formation of the catalytically active copper–oxygen species that will react with the l-ascorbate
to yield dehydroascorbate as product.
©
2010 Elsevier B.V. All rights reserved.
1
. Introduction
oxygen-activating iron enzymes, it is well established that the cat-
alytic cycle proceeds via iron(IV) transient species. Also, for the
copper-containing peptidylglycine ␣-amidating monooxygenase
(PHM) [4] and dopamine ␤-monooxygenase (D␤M) [5], interme-
Direct reaction between a closed electron shell bearing organic
substrate and the triplet ground state of dioxygen is normally
spin-forbidden. Therefore, electron flow from the substrate to the
dioxygen molecule is often assisted by the active site of metalloen-
zymes, which lower the activation energy need of the oxidation.
When dioxygen is used by the enzyme as the ultimate electron
acceptor molecule, reactive oxo-, peroxo-, or oxyl-intermediates
may form, which then will decompose to hydrogen peroxide, or
water as side products by gaining electrons from the substrate
molecule which itself will degrade to the oxidized product and
restore thereducedformof the enzymatic active site. Incase ofiron-
containing enzymes, like the diiron-containing soluble methane
monooxygenase (sMMO) [1] or the heme- [2], or nonheme [3]
•
−
diates formulated as Cu(II)–OOH and Cu(II)–O /Cu(III) O have
been proposed. Hamilton et al. suggested the possible role of cop-
per(III) in monocopper enzymes, such as galactose oxidase [6],
however, this oxidation state is generally considered to be inacces-
sible in biological systems. Trivalent complexes of copper, on the
other hand, with bis(biureth)- [7], bis(oxamide)- [7], peptide- [8],
dithiocarbamate- [9], are well characterized, and extensive stud-
ies on electron transfer reactions of copper(III) peptide complexes
[10] and copper(III) imine–oxime complexes [11] have been carried
out. High-valent copper complexes occur in biomimetic chemistry
[12], and oxidizing ability of copper(II)–OOH and dicopper(II)–OOH
[
13] species is widely studied. Other macrocyclic [14] and digly-
cylethylenediamine copper(III) complexes [15] have been reported
to oxidize ascorbic acid. No example can be found, where copper(III)
shows catalytic activity in oxidation reactions against a substrate,
∗ Corresponding author. Tel.: +36 88624720.
E-mail address: jpap@almos.vein.hu (J.S. Pap).
1
381-1169/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2010.10.026

7
8
J.S. Pap et al. / Journal of Molecular Catalysis A: Chemical 334 (2011) 77–82
where P is the barometric pressure, p is the water vapour pres-
◦
sure, t is the temperature in C and DO is the concentration of
−
1
dissolved dioxygen in mg L . Concentration of the molecular oxy-
gen in solution was changed by mixing different volumes of air-
and nitrogen-saturated buffer solutions in a sealed cuvette. Thus
only the initial rates (up to 10–15% conversion) could be used for
further calculations to determine the partial order for dioxygen.
3. Results and discussion
Scheme 1. Structure of the catalyst [CuIII(LH−4)]−
.
Reaction of l-ascorbic acid with dioxygen in the presence of
CuIII(LH )] . Catalytic reactions were carried out in buffered
−
[
−4
although the presence of copper(III) ion is assumed in some cases
16]. It can be presumed, that a stable copper(III) complex with the
appropriate ligand environment can possibly act as a catalyst that
cycles between the copper(III)/copper(II) states.
aqueous solutions at an optimum pH 8 for the following reasons. At
pH ≤ 7 the catalyst decomposes within a few cycles, whereas above
pH 8, the ascorbic acid (pKa values are 4.17 and 11.6) is present
[
−
both in the monoanionic (AscH ) and in some extent, the dian-
ionic (Asc2 ) form, that may alter the observed mechanism. At pH
8, on the other hand, these complications do not occur, and we can
assume a dominant reaction mechanism between the copper(III)
−
Template synthesis of remarkably stable square-planar cop-
per(III) complexes based on hydrazide and hydrazide/oxime
ligands have been reported [17]. Here we present one of these com-
III
−
−
plexes, [Cu (LH )] (L = 8,17-dioxa-1,2,5,6,10,11,14,15-octaaza-
catalyst and the AscH anion.
−4
tricyclo[13.3.1]eicosane-3,4,12,13-tetrone, Scheme 1) [17a] as a
catalyst in the oxidation reaction of l-ascorbate anion by molecular
oxygen.
In order to determine the stoichiometry of the catalytic reaction,
dioxygen uptake has been measured gas-volumetrically. During
the oxidation of l-ascorbic acid to dehydroascorbic acid (Asc) half
equivalent of dioxygen was consumed compared to the trans-
formed l-ascorbic acid (see Fig. S1). Based on this, the two-electron
oxidation of an ascorbate molecule results in the four-electron
reduction of a dioxygen molecule according to Eq. (3) similarly to
the native l-ascorbate oxidase, a blue multicopper oxidase enzyme
[19].
2
. Experimental
Materials. l-Ascorbic acid and the salts that were used for buffer
preparation and for cyclic voltammetry were commercial products
of reagent grade and were used without any further purification.
III
Double distilled water was used as solvent. The K[Cu (LH 4)]
−
−
[
CuIII(LH 4)]
−
−
2Asc + 2OH−
catalyst has been synthesized as it was reported earlier for
2
AscH + O2
−→
(3)
LiK[Cu(LH )] ·10H O [17a] except that instead of LiOH·H O and
−
4
2
2
2
To be noted, that the observed excess of O2 consumption (apart
KOH, only KOH was used during the synthesis.
from the ∼5% error of the measurement) can be ascribed to the
oxidation of ascorbate catalysed by some unidentified impurity in
the solutions (vide infra). Since the stoichiometry of the l-ascorbic
acid oxidation catalysed by various metal ions is 1:1 with respect to
l-ascorbic acid and dioxygen [20], such reactions will cause some
increase in the observed dioxygen consumption.
Instruments. Electronic absorption spectra were recorded on a
Beckman DU 650 spectrophotometer connected to a temperature
◦
controller (± 0.1 C accuracy). The pH of solutions was measured
with a digital pH meter (Schott). Cyclic voltammograms (CVs) of
the complex and l-ascorbate solutions were obtained in water con-
taining 0.1 M NaClO4 as supporting electrolyte at 298 K. Hanging
mercury drop electrode (HMDE) was used as a working electrode,
−
Kinetic measurements. The decrease in AscH concentration was
followed by UV–vis spectroscopy measuring the intensity of the
3
M calomel electrode (CE) and a platinum wire used as a reference
−
−1
−1
band ascribed to AscH at 265 nm (ε = 14,360 M cm ). Among
the applied conditions that are summarized in Table 1, the l-
and counter electrode, respectively. Scan rates ranging from 100
−1
to 800 mV s and a potential range from +0.200 to −1.250 V were
−
ascorbic acid is present predominantly as AscH [21]. During
applied.
calculations, only the monoanionic form has been considered to
interact with the catalyst. Fig. 1 illustrates the spectral changes
during a typical oxidation reaction.
Kinetic measurements. Reactions were performed in 1 cm path-
length quartz cuvettes, in 3 mL solutions. The partial orders for
the reactants and the catalyst were determined under pseudo-
first-order conditions, with systematically varying the initial
The initial catalyst concentration can be determined from its
−
1
−1
−
III
−
absorption band at 550 nm (ε = 7480 M cm ), before the addi-
concentrations of AscH , [Cu (LH )] , or dioxygen at 298 K,
−4
−
tion of AscH (Fig. 1, solid line). This absorption can be assigned
while the initial concentration of the other two reactants and the
pH were maintained constant. Stirring rate had no effect on the
3+
as an N → Cu charge-transfer band as it was reported earlier
[17a]. (Note, that in contrast with a typical kinetic experiment,
concentration of the catalyst was increased to illustrate the spec-
tral changes at 550 nm in Fig. 1. This way, the catalyst/dioxygen
ratio is higher, than in a normal kinetic assay, where the dioxygen
concentration is at least two orders of magnitude higher, than the
catalyst concentration that is enough to sustain the pseudo first-
order conditions.) Upon addition of the substrate (Fig. 1, dotted
line) the 550 nm band slightly decreases and a new band at 265 nm
III
−
reaction rates. The concentration of [Cu (LH )] was essentially
−4
unchanged before and after the reactions, based on its 550 nm
absorption band intensity. Air-saturated and buffered (50 mM
Tris–HCl) aqueous solutions were used, the pH was set to 7.97 and
the ionic strength was fixed at 100 mM with NaCl. The reaction rates
and kobs values were calculated by integration method (up to 75%
conv.). For the determination of the activation parameters, temper-
ature was varied between 288 and 308 K. Dioxygen concentration
was calculated based on empirical formula [18] using Eq. (1), for
−
−
from the AscH can be seen. After more than 90% of the AscH is
transformed, the original spectrum is almost completely restored
◦
◦
◦
t < 30 C and Eq. (2) for 30 C < t < 50 C, respectively.
DO = 0
(Fig. 1, dashed line). These spectral changes are attributed to the
.678(P − p)
(1)
(2)
partial reduction of copper(III) to copper(II) by the added ascor-
bate. We obtained more information about this first equilibrium
step using stopped-flow technique with the exclusion of dioxygen,
as it will be presented later. Electrochemical experiments (CV) also
3
5 + t
DO = 0
.827(P − p)
49 + t

J.S. Pap et al. / Journal of Molecular Catalysis A: Chemical 334 (2011) 77–82
79
Table 1
Kinetic data for the copper(III)-catalysed reaction of l-ascorbate and dioxygen at pH 8.
◦
−
−5
[CuIII(LH.4) ] (10 M)
−
−6
−4
−d[AscH−]/dt (10 M s
−9
−1
−9
−1
−1 −1
kobs (M )
Exp.
Temp. ( C)
[AscH ] (10 M)
[O2] (10 M)
)
V0 (10 M s
)
s
1
2
3
4
5
6
7
8
9
25
25
25
25
25
25
25
25
25
25
25
25
25
25
3.42
4.86
7.92
2.39
2.39
2.39
2.39
2.39
2.39
2.39
2.39
1.44
2.39
3.82
4.78
5.74
6.69
2.60
2.60
2.60
2.60
2.60
2.60
2.60
2.60
2.60
2.60
2.60
2.60
2.60
2.60
9.02 ± 0.10
10.93 ± 0.08
15.06 ± 0.17
15.53 ± 0.08
17.31 ± 0.15
18.78 ± 0.20
20.40 ± 0.22
22.99 ± 0.21
11.38 ± 0.14
17.31 ± 0.15
27.76 ± 0.49
33.24 ± 0.64
40.97 ± 0.69
45.47 ± 0.76
40.02 ± 0.44
40.68 ± 0.31
43.91 ± 0.50
41.79 ± 0.20
43.71 ± 0.39
42.45 ± 0.46
40.86 ± 0.44
40.90 ± 0.37
46.27 ± 0.57
43.28 ± 0.39
43.18 ± 0.76
41.21 ± 0.79
42.52 ± 0.71
40.86 ± 0.68
9.30
17.03 ± 0.47
10.56
13.18
16.78
21.27
11.22
10.77
10.89
10.95
10.84
10.65
10
11
12
13
14
a
4
2.26 ± 1.64
15
16
17
18
19
20
21
22
23
24
25
25
25
25
25
25
25
25
15
20
30
35
8.36
8.54
8.45
8.69
9.34
9.64
9.78
11.67
10.70
10.97
10.31
2.39
2.39
2.39
2.39
2.39
2.39
2.39
4.78
4.78
4.78
4.78
0.33
0.43
0.61
0.90
1.30
1.45
2.00
3.19
2.86
2.37
2.21
5.86 ± 0.12
7.92 ± 0.32
8.34 ± 0.21
11.90 ± 0.36
13.25 ± 0.53
13.99 ± 0.28
15.45 ± 0.32
18.54 ± 0.30
26.70 ± 0.41
50.49 ± 1.01
62.74 ± 1.13
20.10 ± 0.33
31.93 ± 0.50
65.51 ± 1.36
86.95 ± 1.57
a
2
1/2
2
Mean value of the kinetic constant kobs and its standard deviation ꢁ(kobs) were calculated as kobs = (˙iωki/˙iωi) and ꢁ(kobs) = ˙iωi(ki − kobs) /(n − 1)˙iωi) , where ωi = 1/ꢁ .
i
support a fast reduction of copper(III) to copper(II) by ascorbate
Further evidence on the half-order for the substrate is seen
in Fig. 3, where the reaction rates of the individual experiments
(Table 1, experiments 1–8) calculated according to Eq. (4) are
plotted against the square root of the initial concentrations of l-
ascorbate. The linear dependence clearly indicates a partial order
of one half in the substrate.
(
vide infra).
The time sequence of a typical catalytic reaction can be seen
−
0.5
in Fig. 2. Plots of 2[AscH ] vs. time can be fitted with a straight
line under pseudo first-order conditions (see Fig. 2 for example)
in agreement with a half-order dependence with respect to the
ꢀ
substrate. The pseudo first-order rate constant, k can be calculated
Runs were performed at various initial catalyst concentra-
tions (Table 1, experiments 9–14). Reaction rates show a linear
dependence on copper(III) concentration as can be seen in Fig. 4,
indicating first-order in the catalyst (m = 1). Linear fit to data points
crosses the ordinate axis above zero that can be interpreted as a
contribution from the autoxidation reaction of the l-ascorbate to
the overall reaction rate. Addition of EDTA efficiently inhibits the
from the slopes of these fitted lines for the individual experiments
performed at various initial substrate concentrations. The reaction
rate can be given according to Eq. (4) for each individual experiment
(
Table 1, experiments 1–8):
−
−
d[AscH ]
0.5
ꢀ
−
=
k [AscH ]
(4)
dt
−
III
−
oxidation of AscH in the absence of [Cu (LH )] , as it was found
−
4
where k = k [CuIII(LH ) ] [O ] .
ꢀ
−
m
n
earlier for similar systems [22]. In fact, this “background” reaction
obs
−4
2
Fig. 1. Changes in copper(III) concentration in the presence of l-
ascorbate and dioxygen in water (pH 7.97, ꢂ = 0.1 M with NaCl, 298 K,
Fig. 2. Plot of l-ascorbate concentration and its square root as a function of time
III
−
−6
−
−5
M
(pH 7.97, ꢂ = 0.1 M, 298 K, [Cu (LH−4) ] = 2.39 × 10 M, [AscH ]i = 7.92 × 10
III
−
−5
−
−5
−4
−4
[Cu (LH−4) ]i = 2.65 × 10 M, [AscH ]i = 8.92 × 10 M and [O2] = 2.6 × 10 M).
and [O2] = 2.6 × 10 M, see Table 1, exp. 3).

8
0
J.S. Pap et al. / Journal of Molecular Catalysis A: Chemical 334 (2011) 77–82
Rapid mixing of various amounts of ascorbate solutions to a
solution of the catalyst and concomitantly following the decrease
in the 550 nm absorption band allows us to determine the equilib-
ꢀ
rium constant. The K1 can be transformed to K1 according to Eq.
(7) among the applied reaction conditions.
K [H O]
ꢀ
1
2
−
K
=
(7)
1
[
OH ]
ꢀ
The rate constant for the forward reaction (k1 ) can be obtained
directly from the initial rate for copper(III) decay in the presence of
−
1
–50 equivalent of AscH . Typical SF curves are shown in Fig. S3.
The values for the apparent rate constants and equilibrium con-
stants are summarized in Table S1. The extent of the reduction of
copper(III) to copper(II) is small as it can be seen from the value of
−
3
the equilibrium constant at pH 8 (8.78 ± 3.11 × 10 ).
Cyclic voltammetry measurements. It was previously reported
[
17a] that the complex undergoes a reversible one-electron reduc-
tion according to Eq. (8) in aqueous solution.
Fig. 3. Reaction rate as a function of square root of the initial l-ascorbate concen-
−
−5
−
E
2−
tration (pH 7.97, ꢂ = 0.1 M, 298 K, [AscH ] is varied between 3.42–21.27 × 10 M,
CuIII(LH )] + e ꢀ [Cu (LH )]
1/2
II
[
(8)
III
−
−6
−4
−4
−4
[
Cu (LH−4) ] = 2.39 × 10 M and [O2] = 2.6 × 10 M, see Table 1, exp. 1–8).
In our system the E
was −0.09 V vs. a 3 M calomel electrode
1/2
−
1
is a catalytic oxidation caused by the metal ion content of the water
used.
at a 200 mV s
scan rate (Fig. 5, solid line spectrum). The peak
currents (|ia| = 0.855 ␮A and ic = 0.866 ␮A, respectively) show the
full reversibility of this redox couple among the experimental con-
ditions. When the solution is exposed to air, the picture changes
significantly (Fig. 5, dashed line spectrum).
Performing the catalytic reactions at different dioxygen pres-
sures (Table 1, experiments 15–21) the initial reaction rates show
linear dependence on the square root of the initial dioxygen con-
centrations (Fig. S2). This strongly indicates half-order dependence
with respect to dioxygen (n = 0.5). Based on our results, the over-
all rate law can be written at constant pH, temperature and ionic
strength according to Eq. (5):
The current peak for the reduction step increases, due to the
compensational enhancement of the anodic current resulting from
the fast re-oxidation of copper(II) by dioxygen. The oxidation peak
almost totally disappears due to the same effect, and another oxi-
dation peak appears at higher potentials (0.18 V, see Fig. 5, dashed
line), possibly originating from a copper(III)-dioxygen adduct, that
can be proposed as an intermediate in the catalytic process (vide
−
−
d[AscH ]
0.5
−
−
0.5
=
kobs[AscH ] [CuIII(LH ) ][O ]
(5)
−4
2
dt
Temperature dependence of the kinetic constant kobs was mea-
−
infra). Addition of AscH to the solution in a 1:10 ratio restores
‡
sured in order to determine the activation parameters. The ꢀH
the original peaks (Fig. 5, dotted line), whereas the peak at ∼0.18 V
−
1
‡
calculated from Eyring’s equation is 50.3 ± 2.2 kJ mol , and the ꢀS
is missing, indicating that the copper(III)-dioxygen adduct rapidly
−1
−1
is −41 ± 2 Jmol
K at 298 K. Summary of the rate constants and
−
reacts with the added AscH .
other parameters for the individual assays are given in Table 1.
Stopped-flow (SF) kinetics. In nitrogen-saturated, buffered solu-
tions the copper(III)/copper(II) equilibrium according to Eq. (6) can
Proposed mechanism. A reaction mechanism that fits the kinetic,
SF kinetic, UV–vis spectroscopic and CV data is shown in Scheme 2.
As it was shown in Fig. 1, the initial concentration of copper(III)
−
be measured in the presence of different concentrations of AscH .
ꢀ
is decreased by the added ascorbate. The K1 constant of the cop-
−
K1
2−
[CuIII(LH 4)] + AscH− + OH ꢀ2[Cu (LH )] + Asc + H O (6)
−
II
2
−
−4
2
Fig. 5. CV spectra of [CuIII(LH−4)] under argon (solid line), in the presence of
air (dashed line) and in the presence of 10 equivalents of l-ascorbate and air
(dotted line), in 0.1 M NaClO4 solution against 3 M KCl calomel electrode with
−
Fig. 4. Plot of the reaction rate as
a function of the initial catalyst con-
−
−5
centration (at pH 7.97, ꢂ = 0.1 M, 298 K, [AscH ] = 10.65–11.22 × 10 M,
III
−
−6
−4
−1
−
−3
III
−
−3
[
9
Cu (LH−4) ] = 1.44–6.69 × 10
M
and [O2] = 2.6 × 10 M, see Table 1, exp.
200 mV s scan rate, T = 298 K, [AscH ] = 10 × 10 M, [Cu (LH−4) ] = 10 M, and
−4
–14).
[O2] = 2.6 × 10 M).

J.S. Pap et al. / Journal of Molecular Catalysis A: Chemical 334 (2011) 77–82
81
sity and does not change significantly over the course of the
reaction. This is due to transformation of dehydroascorbic acid to
other products that are described in the literature [23]. The rel-
−
1
atively high activation enthalpy (50.3 ± 2.2 kJ mol at 298 K) can
originate from the repulsion between the negatively charged react-
ing intermediates.
4. Conclusions
−
−
Mechanism of the 2e oxidation of AscH to Asc and the
Scheme 2. Suggested mechanism for the copper(III)-catalysed reaction of l-
ascorbate and dioxygen at pH 8.
−
concomitant 4e reduction of dioxygen to water catalysed by
III
−
Cu (LH 4)] (Scheme 1) consists of three general stages: (1) acti-
−
[
vation (reduction) of the metal site by the substrate, (2) reduction
of dioxygen by the metal centre to give a reactive copper–oxygen
intermediate and (3) quenching the intermediate by the substrate.
Although the stoichiometry of the reaction is the same as it was
established for the native ascorbate oxidases, our catalytic system
cycles through a completely different mechanism compared to the
tricopper active site of the native enzymes. In the enzyme the cop-
per(II)/copper(I) redox couple is involved, whereas in our system
copper(III) and copper(II) species occur in the catalytic process, due
per(III)/copper(II) equilibrium in the presence of various amounts
of ascorbate was calculated from the SF results under nitrogen
atmosphere. When the solution is exposed to air, the absorption
band at 550 nm grows quickly, i.e. copper(II) is oxidized to cop-
per(III) and after all the ascorbate has reacted the absorbance
approaches to its original value. Re-oxidation of copper(II) to cop-
per(III) is supported by the results from the cyclic voltammetry
experiments, too.
Based on the observed rate law, among catalytic conditions,
one molecule of substrate and dioxygen contributes to the for-
mation of two catalytically active species, each of which contains
one copper. This species then can react with the substrate in a
fast step to produce Asc and thus closing the catalytic cycle. We
believe that the catalytic cycle starts with two pre-equilibria in
which formally a peroxodicopper(III) intermediate is formed which
then can dissociate to mononuclear copper(III)-oxyl radical anions
to the ability of the ligand to stabilise copper(III). The O is reduced
2
to water, similarly to several monooxygenase enzymes. A unique
III
−
feature of the [Cu (LH 4)] catalyst is that upon its reduction, the
−
square planar ligand environment is no longer favoured, thus bind-
ing an additional ligand at the apical position and the formation
of a square pyramidal coordination sphere is likely to take place.
This is represented in the suggested mechanism based on differ-
ent kinetic and spectroscopic methods. Difference in the preferred
coordination geometry of the reduced and oxidized metal ion may
be a useful activation/de-activation switch in catalytic reactions
with copper(III) complexes.
in the rate-limiting step (k ). These species are proposed on the
3
analogy of peroxodicopper(II) complexes that are well known in
the literature [12] and the negative activation entropy for the
−
1
−1
reaction (−41 ± 2 Jmol
K at 298 K) is in accordance with an
associative mechanism for the dioxygen activation. Since cop-
per(III) intermediates like the ones proposed in this mechanism
have never been experimentally described, their presence is hypo-
thetic. Applying steady-state treatment for this mechanism, where
d[(LH 4)Cu (O )Cu (LH
of product formation and the following rate equation (10) can be
Acknowledgements
Financial support of the Marie Curie Foundation, the Univer-
sity of Wroclaw and the Hungarian National Research Fund (OTKA
PD75360) is gratefully acknowledged.
III
III
4−
) ]/dt = 0, Eq. (9) will describe the rate
−
2
−4
obtained:
Appendix A. Supplementary data
−
−
d[AscH ]
0.5
=
{k [(LH )CuIII(O )CuIII(LH−4
)
4−]}
(9)
3
−4
2
dt
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.molcata.2010.10.026.
−
−
d[AscH ]
dt
References
ꢀ
ꢁ
ꢁ
ꢁ
ꢁ
ꢂ
−
2
−
−
k1[CuIII(LH−4) ] [AscH ][OH ]
[1] E.G. Kovaleva, M.B. Neibergall, S. Chakrabarty, J.D. Lipscomb, Acc. Chem. Res.
40 (2007) 475–483.
+ k−2[(LH−4)CuIII(O )CuIII(LH−4
4−
[2] (a) M. Sono, M.P. Roach, E.D. Coulter, J.H. Dawson, Chem. Rev. 96 (1996)
2841–2887;
k k
2
)
]
2
3
=
× [O ] ×
2
k
+ k3
k
−1
[Asc][H O] + k [O ]
−
2
2
2
2
(b) B. Mennier, S.P. Visser, S. Shaik, Chem. Rev. 104 (2004) 3947–3980;
(
c) T.M. Makris, S.G. Sliger, I. Schlichting, Chem. Rev. 105 (2005) 2253–2277.
3] (a) M.M. Abu-Omar, A. Loaiza, N. Hontzeas, Chem. Rev. 105 (2005) 2227–2252;
b) M. Costas, M.P. Mehn, M.P. Jensen, L. Que Jr., Chem. Rev. 104 (2004) 939–986.
(
10)
[
[
(
4] (a) J.P. Klinman, Chem. Rev. 96 (1996) 2541–2562;
k1[CuIII(LH−4) ] [AscH ][OH ] ꢁ k [(LH )CuIII(O2)
−
2
−
−
When
−2
−4
(b) P. Chen, J. Bell, B.A. Eipper, E.I. Solomon, Biochemistry 43 (2004) 5735–5747.
CuIII(LH 4) ] and k [Asc][H O] ꢁ k [O ] (this can be presumed
4−
[5] W.A. Francisco, N.J. Blackburn, J.P. Klinman, Biochemistry 42 (2003) 1813–
819.
−
−1
2
2
2
1
based on the SF kinetic results, as K is small, but k1 and k are
1
−1
[
[
6] G.R. Dyrkacz, R.D. Libby, G.A. Hamilton, J. Am. Chem. Soc. 98 (1976) 626–628.
7] J.J. Staggerda, J.J. Bour, P.J.M.W.L. Birker, Inorg. Chem. 10 (1971) 1202–1205.
large), at constant pH and [Asc], Eq. (10) simplifies to Eq. (11),
where C = [OH ] /{[Asc][H O]}
−
0.5
0.5
:
[8] (a) D.W. Margerum, K.L. Chellappa, F.P. Bossu, G.L. Burce, J. Am. Chem. Soc. 97
1975) 6894–6896;
2
(
ꢃ
−
(b) D.W. Margerum, in: T.E. King, H.S. Mason, M. Morrison (Eds.), Oxidases and
Related Redox Systems, Pergamon Press, Oxford, 1982, pp. 193–206;
(c) L.L. Diaddario, W.R. Robinson, D.W. Margerum, Inorg. Chem. 22 (1983)
1021–1025;
−
d[AscH ]
k k
C × [AscH−]0.5[CuIII(LH ) ][O ]
2
3
−
0.5
=
K1
−4
2
dt
k
+ k3
−2
(
11)
(
d) M.R. McDonald, F.C. Fredericks, D.W. Margerum, Inorg. Chem. 36 (1997)
Eq. (11) satisfies the kinetic data and the deduced kinetic
3119–3124;
0
.5
0.5 0.5
(e) S.K. Burke, Y. Xu, D.W. Margerum, Inorg. Chem. 42 (2003) 5807–5817.
9] (a) P.J. Beurskens, J.A. Cras, J.J. Steggerda, Inorg. Chem. 7 (1968) 810–813;
rate law (Eq. (5)), if kobs = (K k k ) /(k + k3)
C
. UV–vis spec-
1
2
3
−2
[
troscopic results indicates that the [Asc] is constant, since its
characteristic absorption band around 300 nm is very low in inten-
(
b) G. Hogarth, A. Pateman, S.P. Redmond, Inorg. Chim. Acta 306 (2000)
232–236.

8
2
J.S. Pap et al. / Journal of Molecular Catalysis A: Chemical 334 (2011) 77–82
[
10] (a) D.W. Margerum, G.D. Owens, Metal Ions in Biological Systems, Vol. 12, Mar-
cel Dekker, New York, 1981, p. 75;
(b) I.O. Fritsky, H. Kozłowski, O.M. Kanderal, M. Haukka, J. S´ wi a˛ tek-Kozłowska,
E. Gumienna-Kontecka, F. Meyer, Chem. Commun. (2006) 4125–4127.
[18] Standard Methods for the Examination of Water and Wastewater, 12th ed.,
American Public Health Association, New York, 1965, pp. 408–410.
[19] (a) A. Messerschmidt, A. Rossi, R. Ladenstein, R. Huber, M. Bolognesi, G. Gatti,
A. Marchesini, R. Petruzzelli, A. Finazzi-Agro, J. Mol. Biol. 206 (1989) 513–529;
(b) A. Messerschmidt, R. Ladenstein, R. Huber, M. Bolognesi, L. Avigliano, R.
Petruzzelli, A. Rossi, A. Finazzi-Agro, J. Mol. Biol. 224 (1992) 179–205.
[20] (a) P.A. Seib, B.M. Tolbert, Ascorbic Acid: Its Chemistry, Metabolism and Uses,
Adv. Chem. Ser. Am. Chem. Soc., Washington, DC, 1982;
(
(
(
b) J.M. Anast, D.W. Margerum, Inorg. Chem. 21 (1982) 3494–3501;
c) W. Levason, M.D. Spicer, Coord. Chem. Rev. 76 (1987) 45–120;
d) D.W. Margerum, Pure Appl. Chem. 55 (1983) 23–34.
[11] (a) F.M. Al-Sogair, Y. Sulfab, Trans. Met. Chem. 18 (1993) 19–22;
b) A. Hussein, Y. Sulfab, M. Nasreldin, Inorg. Chem. 28 (1989) 157–160.
(
[
[
12] E.A. Lewis, W.B. Tolman, Chem. Rev. 104 (2004) 1047–1076.
13] (a) D. Maiti, A.A. Nerducci Sarjeant, K.D. Karlin, Inorg. Chem. 47 (2008)
8
736–8747;
(
(
(
b) D. Maiti, R. Lucas, A.A. Narducci Sarjeant, K.D. Karlin, J. Am. Chem. Soc. 129
2007) 6899–6998;
c) K. Itoh, H. Hayashi, N. Furutachi, T. Matsumoto, S. Nagamoto, T. Tosha,
(b) M.B. Davies, Polyhedron 11 (1992) 285–321.
[21] R.C. Weast (Ed.), Handbook of Chemistry and Physics, 56th ed., CRC Press, Boca
Raton, FL, 1976, p. D150.
S. Terada, S. Fujinami, M. Suzuki, T. Kitagawa, J. Am. Chem. Soc. 127 (2005)
212–5223;
d) S. Thyagarajan, N.N. Murthy, A.A. Narducci Sarjeant, K.D. Karlin, S.E. Rokita,
[22] (a) A. Rigo, M. Scarpa, E. Argese, P. Ugo, P. Viglino, in: W. Bors, M. Saran, D. Tait
(Eds.), Oxygen Radicals Chemistry and Biology, Walter de Gruyter, Berlin and
New York, 1984, p. 170;
5
(
J. Am. Chem. Soc. 128 (2006) 7003–7008.
14] Y. Yano, S. Takano, Y. Kato, W. Tagaki, J. Chem. Soc. [Perkin II] (1979) 1227–1229.
15] P. Stevens, J.M. Waldeck, J. Strohl, R. Nakon, J. Am. Chem. Soc. 100 (1978)
(b) I.B. Afanas’ev, V.V. Grabovetskii, N.S. Kuprianova, J. Chem. Soc. [Perkin II]
(1987) 281–285.
[23] (a) K. Pfeilsticker, F. Marx, M. Bockisch, Carbohydr. Res. 45 (1975)
269–274;
[
[
3
632–3634.
[
[
16] R.F. Jameson, N.J. Blackburn, J. Chem. Soc. Dalton (1976) 1596–1602.
17] (a) I.O. Fritsky, H. Kozłowski, P.J. Sadler, O.P. Yefetova, J. S´ wi a˛ tek-Kozłowska,
(b) E.K. Koliou, P.V. Ioannou, Carbohydr. Res. 340 (2005) 315–318;
(c) R.C. Kerber, J. Chem. Educ. 85 (2008) 1237–1242.
V.A. Kalibabchuk, T. Głowiak, J. Chem. Soc. Dalton Trans. (1998) 3269–3274;