Metal corroles as electrocatalysts for oxygen reduction

Article abstract of DOI:10.1039/b512982f

The rotating ring disk electrode method has been used to study O ₂ electroreduction with metal corroles. Catalysis begins at potentials that are 0.5-0.7 V more positive than the expected potential of the M^III/II couple based on studies in non-aqueous solutions. The path of O₂ reduction depends on the nature of the metal ion. Cobalt corroles promote O₂ reduction to H₂O₂. Iron corroles catalyse O₂ reduction via parallel two- and four-electron pathways, with a predominate four-electron reaction. The rate constants for the individual O₂ reduction paths are given at pH 7. Mechanisms are proposed on the basis of pH dependence, inhibition studies, and Tafel slopes. An imidazole-tailed iron corrole catalyses H₂O₂ disproportionation analogous to catalase. The Royal Society of Chemistry 2006.

Full text of DOI:10.1039/b512982f

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Metal corroles as electrocatalysts for oxygen reduction†
James P. Collman,* Marina Kaplun and Richard A. Decre´au
Received 13th September 2005, Accepted 14th October 2005
First published as an Advance Article on the web 2nd November 2005
DOI: 10.1039/b512982f
The rotating ring disk electrode method has been used to study O2 electroreduction with metal corroles.
Catalysis begins at potentials that are 0.5–0.7 V more positive than the expected potential of the MIII/II
couple based on studies in non-aqueous solutions. The path of O2 reduction depends on the nature of
the metal ion. Cobalt corroles promote O2 reduction to H2O2. Iron corroles catalyse O2 reduction via
parallel two- and four-electron pathways, with a predominate four-electron reaction. The rate constants
for the individual O2 reduction paths are given at pH 7. Mechanisms are proposed on the basis of pH
dependence, inhibition studies, and Tafel slopes. An imidazole-tailed iron corrole catalyses H2O2
disproportionation analogous to catalase.
Introduction
Metal macrocycles are of interest for electrocatalysis of O2
reduction because they mimic the active sites of enzymes (e.g.,
quinol and cytochrome c oxidases),1 and might also be used as
electrode materials in technological devices (fuel cells, metal–air
batteries, oxygen sensors, etc.).2 Porphyrins and phthalocyanines
have been widely studied in a continuing search for catalysts
that might promote direct 4e− O2 reduction.3–6 Herein, we report
catalysis of O2 reduction using several metal complexes of a
different lesser studied macrocycle–corrole. Corroles are closely
related to porphyrins except they lack one meso-carbon and
are trianionic ligands stabilizing higher metal oxidation states
than porphyrins. Several efficient syntheses had been described
Fig. 1 Syntheses of compounds 5 and 6, and structure of the studied
lately that made corroles accessible,7 and have led to studies
of their metallo-derivatives as catalysts for oxidation reactions8
and electroreduction of CO2.9 Recently, a series of Co corroles
have been reported as effective catalysts for O2 reduction10 that
promote parallel 2e− and 4e− pathways similar to related cofacial
Co diporphyrins and porphyrins which can form dimers when
adsorbed on an electrode surface.3d,6 Our interest has been to
identify and compare pathways of O2 reduction with monomeric
corroles that differ in the nature of the central metal ion (M = 3H,
Mn, Fe, Co). A highly electron withdrawing corrole with three
bulky meso-substituents was thought necessary in order to have
catalysts that do not form dimers and to raise the potential of
the MIII/II transition,9 since the potentials where metal complexes
catalyse O2 reduction are usually close to that of the MIII/II
transition of the catalyst, and the MIII/II potential in corroles is
more negative than in the corresponding porphyrins.11 Therefore
we synthesized metal complexes of tris(pentafluorophenyl)corrole
2–4, and of bis(pentafluorophenyl)corrole 5–6 bearing an imi-
dazole axial ligand, which is covalently attached to the corrole
macrocycle7n (Fig. 1).
complexes. Reagents and conditions: (a) (i) FeBr2, THF–CH3OH, AcONa,
RT, N2; (ii) O2; (iii) HCl, CH2Cl2, 52%; (b) Co(AcO)2·4H2O, C2H5OH,
AcONa, RT, 2 h, N2, 79%. The preparation of compounds 1–4 is described
in ref. 12.
Experimental
Preparation of the catalysts
5,10,15-Tris(pentafluorophenyl)corrole [H3(tpfc)] (1), 5,10,15-tris-
(pentafluorophenyl)corrole manganese(III) [Mn(tpfc)] (2), 5,10,15-
tris(pentafluorophenyl)corrole iron(IV) chloride [Fe(tpfc)Cl] (3),
and 5,10,15-tris(pentafluorophenyl)corrole cobalt(III) triphenyl
phosphine [Co(tpfc)PPh3] (4) were synthesized according to
literature procedures.12
5,15-Bis(pentafluorophenyl)-10-(2-3-(1-imidazolylmethyl)-benza-
midophenyl)-corrole iron(III) [Fe(mapc-t)] (5). A solution of free
base tailed corrole, H3(mapc-t), (50 mg, 53.2 lmol) was stirred
with FeBr2 (36 mg, 165.6 lmol) in a THF/CH3OH mixture (4 :
1 vol.) in the presence of sodium acetate (44.0 mg, 532 lmol)
at room temperature for 16 h under inert atmosphere. CH2Cl2
was added (100 mL), air was bubbled through the solution for
1 h, and the solution was washed with 7% HCl (2 × 50 mL).
After evaporation of the solvent, the residue was subjected to
chromatography (SiO2, elution hexane/Et2O 2 : 1 vol.). 5 crushed
out after an excess of hexanes was added to a solution of 5 in
EtOH/acetone (1 : 2 vol.). 5 was obtained in 52% yield (17 mg,
Department of Chemistry, Stanford University, Stanford, CA 94305, USA.
E-mail: jpc@stanford.edu; Fax: +1 650 7250259; Tel: +1 650 7250283
† Electronic supplementary information (ESI) available: Analysis of elec-
trochemical data, and characterization data for compounds 5 and 6. See
DOI: 10.1039/b512982f
554 | Dalton Trans., 2006, 554–559
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
purple; pink solution in Et2O). 1H-NMR (400 MHz, CDCl3)
dH(ppm): 128.26 (2H), 83.28 (2H), 12.53 (2H), 10.02 (2H), 7.69
(2H), 6.96 (2H), 6.47 (2H), 5.12 (2H), −0.08 (1H), −60.22
(2H), −63.42 (2H). 19F-NMR (400 MHz, C6D6) dF(ppm): −93.2
(brs, 2F), −110.8 (brs, 2F), −150.78 (s, 2F), −156.05 (s, 2F),
−157.67 (s, 2F). UV-Vis. (CH2Cl2) k (Abs.) 10−3 × e (M−1 cm−1):
328 (10.7), 360 (11.4), 410 (19.3), 552 (5.5), 626 (1.4 sh),
754 (1.6) nm. MS (ESI+, m/z) = 958.3 [M]+, HR-MS (ESI+,
m/z) = 958.1078 (Calcd for C48H22F10FeN7O). Anal. Calcd for
C48H22F10FeN7O·H2O·3EtOH: C, 58.18; H, 3.80; N, 8.79; Cl,
0.00; Found: C, 58.56; H, 3.41; N, 8.23; Cl, 0.40. TLC (silica,
Et2O 100% vol.) Rf 0.72.
and Pt ring (Pine Instruments, N0 = 22.5%, r10 = 2.5 mm,
r20 = 3.75 mm, and r30 = 4.25 mm). The collection efficiency
of the Pt ring towards H2O2 was determined as a function of
the rotation speed13 by carrying out the O2 reduction at a bare
EPG disk of the EPG-disk/Pt-ring rotating electrode. A BAS CV-
50 W potentiostat (Bioanalytical Systems) and a Pine AFCBP1
bipotentiostat (Pine Instruments) with an ASR speed controller
(Pine Instruments) were used for rotating disk and ring-disk
experiments, respectively. The auxiliary electrode was a Pt mesh
and the reference electrode was a low flow rate saturated calomel
electrode. All potentials are given with respect to the normal
hydrogen electrode (ESCE = 0.243 V vs. NHE).
The Pt ring was cleaned with 0.05 micron c-alumina paste for
1 min followed by sonication in methanol. The EPG disk was
cleaned with a 600 grit SiC paper and sonicated for 1 min in
methanol immediately prior to depositing a catalytic film. The
catalysts were deposited on the electrode surface by syringing
a 0.6–2.0 lL aliquot of a fresh 0.25–0.5 mM solution in a
dimethoxyethane/water mixture (5.6 : 1 vol.) or acetone on a
vertically positioned rotating electrode (300 rpm). The stability
of the catalysts was verified in every experiment by measuring
the voltammetric response at 200 rpm before and after the set of
catalytic waves at different rotations was collected. The limiting
current value varied by only 2–4%. The presented data are
representative of at least 10 separate measurements. Measurements
were conducted at ambient temperature, 22 2 ◦C, in N2 deaerated
solutions when necessary.
5,15-Bis(pentafluorophenyl)-10-(2-3-(1-imidazolylmethyl)-benza-
midophenyl)-corrole cobalt(III) [Co(mapc-t)] (6). Corrole
6
was prepared from the adaptation of a reported method12c by
dissolving H3(mapc-t) (22.6 mg, 25 lmol), Co(OAc)2·4H2O
(30 mg, 120 lmol) and NaOAc (30 mg, 366 lmol) in ethanol
(10 mL) and stirring the resulting solution for 2 h at 25 ◦C. After
evaporation of the solvent, the residue was subjected to
chromatography (SiO2, 17 × 2 cm, loading CH2Cl2, eluent
CH2Cl2). 6 was obtained in 78.8% yield (19 mg, dark brown;
brown solution in CH2Cl2). Limited amount of dimer was
1
obtained. H-NMR (400 MHz, CDCl3) dH(ppm): 9.02 (d, 2H,
J = 4.4 Hz), 8.95 (d, 1H, J = 6.8 Hz), 8.37 (d, 2H, J = 4.8 Hz),
8.18 (d, 2H, J = 4.8 Hz), 8.12 (d, 2H, J = 4.4 Hz), 7.95 (d, 1H,
J = 6.0 Hz), 7.88 (s, 1H), 7.86 (s, 1H), 7.77 (t, 1H, J = 7.2 Hz),
7.40 (t, 1H, J = 6.0 Hz), 7.19 (t, 1H, J = 6.0 Hz), 6.77 (d, 1H,
J = 5.6 Hz), 5.02 (s, 1H), 4.06 (s, 1H), 3.89 (s, 2H), 3.28 (s, 1H),
2.58 (s, 1H). 19F-NMR (400 MHz, CDCl3) dF(ppm): −137.52
(dd, J = 25.2 Hz, J = 7.6 Hz, 2F), −139.45 (dd, J = 26.0 Hz,
J = 7.6 Hz, 2F), −153.62 (t, J = 19.6 Hz, 2F), −161.35 (dt, J =
25.2 Hz, 2F), −162.26 (dt, J = 26.0 Hz, 2F). 13C-NMR (75 MHz,
CDCl3) dC(ppm): 163.25, 150.14, 147.26, 145.97, 144.89, 143.00,
138.62, 136.81, 135.57, 135.47, 135.23, 132.64, 130.98, 130.53,
129.66, 129.62, 129.14, 128.57, 128.31, 127.91, 127.57, 125.83,
123.75, 121.98, 120.74, 120.41, 119.58, 118.71, 116.95, 106.80,
50.17, 29.93. UV-Vis. (CH2Cl2) k (Abs.) 10−3 × e (M−1 cm−1): 380
(47.7), 412 (26.2), 546 (7.4), 586 (4.3), 604 (4.2) nm. MS (ESI+,
m/z) = 961.5 [M]+, HR-MS (ESI+, m/z) = 961.1091, MS (ESI−,
m/z) = 960.5 [M − H]− (Calcd for C48H22CoF10N7O). TLC (silica,
CH2Cl2 100% vol.) Rf 0.53 (silica, hexane/ethyl acetate 1 : 1 vol.)
Rf 0.70.
Kinetic analysis
The following simplified general scheme of O2 reduction was used
to analyse the data:
O2 is cathodically reduced either to H2O (with the constant k1)
or to H2O2 (k2). The resulting H2O2 can be further reduced to
H2O (k3), catalytically decomposed on the electrode surface (k4),
or removed into the bulk of the solution.
For such a scheme, the expressions for distinguishing between
various reaction paths from the disk (Id) and ring (Ir) currents
were developed by Bagotskii:14
Materials
NId
Ir
2k1
k2
(k3 + k4) (1 + 2k1/k2) + k3
= 1 +
+
x−1/2
(1)
KNO3 (≥ 99%, Acros) was recrystallized. NaCN (≥ 95%, Baker &
Adamson) was used as supplied. WARNING: Cyanide is highly
toxic, the experiments must be performed in a properly ventilated
area. All discarded electrolytes were neutralized with bleach before
disposal. The buffers employed were (0.1 M): KC8H5O4, pH = 4;
KH2PO4, pH = 4.4; Na2HPO4/KH2PO4, pH = 5.1, 6.2, 7, 8;
Na2CO3/NaHCO3, pH = 9; Na3PO4/NaHCO3, pH = 10, 11.
ZH
O
2
2
N(I1 − Id )
2ZO k3 + k4
2ZO
k2
2
= 1 +
+
2 x1/2
(2)
Ir
ZH
k2
O
2
2
where Id and Ir (A) are the disk and the ring currents, Il (A) is the
limiting diffusion current, N is the collection efficiency of the ring
electrode, Zi = 0.62Di2/3v−1/6, D (cm2 s−1) is a diffusion coefficient,
v (cm2 s−1) is the kinematic viscosity of the solution, x (rad s−1)
is the speed of electrode rotation, ki (cm s−1) are the first-order
heterogeneous rate constants for individual paths of O2 reduction.
The values of the parameters are: D = 1.71 × 10−5 (H2O2) and
2.1 × 10−5 cm2 s−1 (O2), v = 0.01 cm2 s−1, [O2] = 0.24 mM, Il =
84 lA for a 4e− process.
Procedures and instrumentation
Experiments were carried out in a one compartment home-made
glass cell (100 mL) with a tightly fitting lid using an edge plane
graphite (EPG) rotating disk working electrode (Pine Instruments,
0.195 cm2) or a ring-disk electrode with a removable EPG disk
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 554–559 | 555
©

View Article Online
Mn(tpfc) shows redox couples at 0.88 V and 0.73 V at pH 7.
Both peaks shift by −60 mV pH−1 and are sensitive to the presence
of imidazole. Therefore, the redox processes are proposed to be
MnV/IV and MnIV/III transitions (Fig. 3). Cobalt corroles do not
exhibit well-defined redox peaks when adsorbed on EPG. This
behavior has previously been observed for Co complexes, and was
suggested to arise from different coordination preferences of CoIII
and CoII.3d In our discussion, we have used the CoIII/II reduction
potentials in acetonitrile (ca. −0.16 V vs. NHE).9
Results and discussion
Redox properties of the catalysts
The redox properties of the catalysts were studied by cyclic
voltammetry. The iron corroles adsorbed on edge-plane graphite
(EPG) disk electrode show two redox couples at 0.66 V and
−0.50 V vs. NHE for Fe(tpfc)Cl and 0.71 V and −0.63 V for
Fe(mapc-t) (Fig. 2). Comparison with spectroelectrochemical data
in acetonitrile9 suggests that these redox processes should be FeIV/III
and FeIII/II transitions. To verify the proposed assignments, we
examined the effects of pH and of the addition of ligands to the
solution. The potential of the higher couple in Fe(tpfc)Cl shifts
by −60 mV pH−1 (Fig. 3), in accord with the FeIV/III assignment.
The lower transition does not depend on pH and is shifted to
more negative potentials in the presence of imidazole, as expected
for the FeIII/II process. Surprisingly, this couple is not sensitive to
CN−, and is not linked to O2 reduction, as the redox peaks persist
in air-saturated solution (Fig. 2). Thus, it seems likely that this
redox process is ligand-centered (vide infra). The process can not
be ascribed to FeII/I transition because the electrode potential of
FeII/I in acetonitrile is −1.60 V vs. SCE.9
Electrocatalysis of O2 reduction
Electrocatalysis of O2 reduction was studied by rotating ring-
disk electrode voltammetry16 in air-saturated aqueous solutions.
The currents for O2 electroreduction using selected metal corroles
adsorbed on the EPG disk electrode, as well as the corresponding
limiting currents of H2O2 oxidation on the Pt ring electrode at
pH 7 are shown in Fig. 4A. The presence of Co and Fe corroles
on the disk electrode causes a positive shift in the half-wave
potential of O2 reduction by 0.4–0.2 V, while the activity of
H3(tpfc) (not presented) and Mn(tpfc) is comparable to that of
bare EPG.
Free base, Mn, and Co corroles17 catalyse only 2e− reduction
to H2O2 as evidenced by the correspondence of the ring and disk
currents, and by the number of electrons found from the slope of
the Koutecky–Levich plots (1/Id vs. x−1/2).16 Additional support
comes from a separate study of the catalysts in 0.25 mM H2O2
solution which showed no significant H2O2 reduction activity
(Fig. 4B, curves 3–5).
With Fe corroles as precatalysts, the larger catalytic currents
and the lower ring currents (Fig. 4A) show that much less H2O2
is produced and are indicative of involvement of a 4e− pathway.
For Fe(tpfc)Cl, two waves involving O2 reduction are observed in
the disk current with corresponding waves of H2O2 oxidation in
the ring current. The number of electrons on the plateaux of the
waves is 3.6 and 4, respectively, indicating that the two waves are
not a result of (2 + 2)e− reduction but originate from catalysis
by two active forms of the catalyst. Further analysis of the data
using eqn (1) and (2) (by plotting the data at desired potential
Fig. 2 Cyclic voltammograms for the Fe and Mn corroles in N2 deaerated
and air-saturated 0.1 M KNO3, pH 7, scan rate: 0.05 V s−1, coverage:
2.56 nmol cm−2
.
Fig. 3 The redox potentials of Fe(tpfc)Cl, Fe(mapc-t), and Mn(tpfc) as a function of pH. pKa values are obtained from the curve-fit of the appropriate
Nernst–Clark equation (ref. 15) to the experimental data. The potential of the invoked intramolecular electron transfer FeIII(cor3−)/FeII(cor2−•) is not
presently clear (see text for details).
556 | Dalton Trans., 2006, 554–559
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
on potential, which indicates that the number of the active sites
(supposedly, FeIII–O2H intermediate or/and FeIII form of corrole,
vide infra) decreases with reducing potential. The value of 4.0 ×
103 M−1 s−1 estimated at 0.2 V is ca. 4 orders of magnitude less than
that for dismutation by catalase18a but among the largest reported
for synthetic FeIII complexes (ca. 1.5 × 103 M−1 s−1).5e,18b
Catalysis of O2 reduction by Co and Fe corroles begins at
∼
potentials (E 0.3 V vs. NHE) that are 0.5–0.7 V more positive
=
than the expected potential of the MIII/II couple based on studies
in non-aqueous solutions, as recently observed in Kadish et al.10
where the onset of catalytic waves correlates with the CoIV/III
potential of the catalysts. This is unusual considering that MIII
is not expected to bind O2 and hence should not be catalytically
active. To probe the nature of the active sites (metal ion vs.
ligand), we performed inhibition experiments in air-saturated
20 mM NaCN solutions. The results indicate that O2 activation
occurs at the metal center, because the reduction waves in the
potential range of the MIII form of the catalysts are strongly
suppressed in the presence of CN− (Fig. 4A, dashed lines). Thus,
it seems reasonable to propose the presence of the MII form,
which may arise from an internal corrole-to-metal(III) electron
•
transfer, MIII(cor3−) ꢀ MII(cor2− ),19 that is inhibited by CN−
due to stabilization of the MIII oxidation state. The waves at
higher reducing potentials are not sensitive to CN−, suggesting
•
a MII(cor2− ) + e− ꢀ MII(cor3−) redox transition in the catalysts
Fig. 4 (A) Rotating ring-disk electrode data for the reduction of O2 in
the absence (solid lines) and presence of 20 mM NaCN (dashed lines) at:
(1,1ꢀ) Fe(tpfc)Cl; (2,2ꢀ) Fe(mapc-t); (3,3ꢀ) Co(mapc-t); (4,4ꢀ) Mn(tpfc) in
0.1 M KNO3 at pH 7, Ering = 0.5V vs. SCE, scan rate: 0.02 V s−1, rotation
speed: 200 rpm, coverage: 1.5 nmol cm−2. (B) Rotating disk electrode data
for the reduction of 0.25 mM H2O2 at: (1) Fe(tpfc)Cl; (2) Fe(mapc-t);
(3) Co(tpfc)PPh3; (4) Co(mapc-t); (5) Mn(tpfc) in 0.1 M KNO3 at pH 7,
based on a low affinity of CN− for MII.20
Two plausible mechanisms of O2 reduction are suggested to
account for the observed results. One possibility is that the MII
form reacts relatively fast with O2 if we assume that the onset
of O2 reduction correlates with the potential of the proposed
electron transfer (the mechanism typical of O2 catalysis with iron
porphyrins), and thus MII(cor2−•) form becomes favored at E ≤
0.3 V.21 Alternatively, the MII(cor2−•) form might arise directly
from the MIV/III transition, suggesting a slow formation of an
O2 adduct which is reduced at more negative potentials.3e Tafel
plots (applied potential vs. the log. of the kinetic current or rate
constants) constructed for O2 reduction on H3(tpfc), Mn(tpfc),
and Fe(tpfc)Cl, show slopes of −120 mV dec−1, indicating that
the rate of the overall turnover is determined by a first electron
scan rate: 0.02 V s−1, rotation speed: 200 rpm, coverage: 1.5 nmol cm−2
.
as NId/Ir vs. x−1/2 and N(Il–Id)/Ir vs. x1/2) shows that for all
potentials where H2O2 is detected, O2 is reduced by parallel 2e−
and 4e− pathways. The k1/k2 ratio varies from 2 to 12 (Fig. 5), and
thus O2 is predominantly reduced to water.
transfer to adsorbed O2:4
•
MIII(cor3−) ꢀ MII(cor2−
)
•
MII(cor2− ) + O2 ꢀ MIII(cor2−•)–(O2
)
−
MIII(cor2− •)–(O2−) + e− → (slow)
Catalysis on Fe(mapc-t) and Co corroles shows Tafel slopes of
−60 mV dec−1, suggesting the slow chemical step (e.g., O2 binding
or further protonation of the adduct).4
The pH dependence of O2 reduction was examined with Co and
Fe corroles. On Co corroles, the Tafel slopes are pH dependent
and decrease gradually from −86 mV dec−1 to −42 mV dec−1
over the pH range 4–11 (Fig. 6). The pH dependence of potential
has a slope of −40 mV pH−1 (Fig. 7A), giving pH dependent
order in protons, p(H+) = (∂E/∂pH)/(∂E/∂logk). The data
indicate a different mechanism in alkaline solutions with the rate
determining first electron transfer. On Fe(tpfc)Cl, the mechanism
does not change with pH as the Tafel slopes are −120 mV dec−1 at
all studied pHs, and the potential shows a weak pH dependence.
On Fe(mapc-t), the rate of dismutation goes through a maximum
at pH 9 (Fig. 7B). Kinetic parameters of individual paths were not
determined because of insufficiently long linear parts of the Tafel
Fig. 5 Potential dependence of the heterogeneous rate constants for the
2e− (k2) and 4e− (k1) reduction of O2 and for catalytic dismutation of H2O2
(k4) at: (1) Co(mapc-t); (2) Fe(tpfc)Cl; (3) Fe(mapc-t).
Similar reduction paths are drawn for catalysis on Fe(mapc-t)
at negative potentials. An additional wave above 0 V is explained
as dismutation of intermediate H2O2 based on the treatment of
the data giving (k3 + k4) = 0 and on the fact that the catalyst
is not active toward H2O2 reduction (k3 = 0) in this potential
range (Fig. 4B). The calculated rate constant (k4, Fig. 5) depends
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 554–559 | 557
©

View Article Online
alysts for O2 reduction. The activity series is similar to tetraphenyl-
porphyrins and dibenzo-tetraazaannulenes: Co > Fe ꢂ Mn,
3H. The behavior of Co, Mn and H3 corroles parallels findings
on related porphyrins, whereas the results from the Fe corroles
show remarkable differences. First, the Fe corroles promote O2
reduction to H2O directly, rather than by the (2 + 2)e− pathway
typical of Fe porphyrins. Second, the activity for H2O2 reduction
is less than that for O2 reduction, whereas for Fe porphyrins the
activities are similar. Reduction of H2O2 is observed only at E ≤
4
−0.3V (k /coverage 2 × 10 M−1 s−1 at −0.6 V), which might be
∼
=
3
explained by inhibition of a supposed intramolecular transition in
the catalysts due to H2O2 reacting with FeIII through a homolytic
process.
Fig. 6 Tafel plots for the reduction of O2 at Co(mapc-t) in 0.1 M KNO3
solutions of different pH.
Acknowledgements
Work was supported by NIH (Grant No. 5R01 GM-17880-35)
and NSF (Grant No. CHE-0131206). R. A. D. thanks the French
Foreign Ministry for a Lavoisier fellowship.
References and notes
1 J. Abramson, S. Riistama, G. Larsson, A. Jasaitis, M. Svensson-Ek,
L. Laakkonen, A. Puustinen, S. Iwata and M. Wikstro¨m, Nat. Struct.
Biol., 2000, 7, 910; D. Zaslavsky and R. B. Gennis, Biochim. Biophys.
Acta, 2000, 1458, 164; S. Ferguson-Miller and G. T. Babcock, Chem.
Rev., 1996, 96, 2889.
2 A. Bettelheim, L. Soifer and E. Korin, J. Power Sources, 2004, 130, 158;
Y. Kiros, O. Lindstro¨m and T. Kaimakis, J. Power Sources, 1993, 45,
219; B. Shentu, K. Oyaizu and H. Nishide, J. Mater. Chem., 2004, 14,
3308.
3 (a) J. P. Collman, R. Boulatov, C. J. Sunderland and L. Fu, Chem. Rev.,
2004, 104, 561; (b) M. R. Tarasevich and K. A. Radyushkina, Russ.
Chem. Rev., 1980, 49, 718; (c) J. H. Zagal, Coord. Chem. Rev., 1992,
119, 89; (d) E. Song, C. Shi and F. C. Anson, Langmuir, 1998, 14,
4315; (e) C.-L. Ni and F. C. Anson, Inorg. Chem., 1985, 24, 4754; (f) H.
Jahnke, M. Scho¨nborn and G. Zimmermann, Top. Curr. Chem., 1976,
61, 133.
4 F. Van den Brink, W. Visscher and E. Barendrecht, J. Electroanal.
Chem., 1983, 157, 305; J. Zagal, P. Bindra and E. Yeager, J. Electrochem.
Soc., 1980, 127, 1506.
5 (a) H. Behret, W. Clauberg and G. Sandstede, Ber. Bunsenges. Phys.
Chem., 1979, 83, 139; (b) H. Behret, W. Clauberg and G. Sandstede,
Z. Phys. Chem., 1978, 113, 97; (c) T. Sawaguchi, T. Matsue, K. Itaya
and I. Uchida, Electrochim. Acta, 1991, 36, 703; (d) F. Van den Brink,
W. Visscher and E. Barendrecht, J. Electroanal. Chem., 1984, 172, 301;
(e) R. Boulatov, J. P. Collman, I. M. Shiryaeva and C. J. Sunderland,
J. Am. Chem. Soc., 2002, 124, 11923.
6 C.-L. Ni, I. Abdalamuhdi, C. K. Chang and F. C. Anson, J. Phys.
Chem., 1987, 91, 1158; J. P. Collman, N. H. Hendricks, K. Kim
and C. S. Bencosme, J. Chem. Soc., Chem. Commun., 1987, 1537; F.
D’Souza, Y. Y. Hsieh and G. R. Deviprasad, Chem. Commun., 1998,
1027.
Fig. 7 (A) Plots of potential at constant current (10 lA) vs. pH for the
reduction of O2 on cobalt and iron corroles derived from corresponding
Tafel plots. (B) The currents of O2 electroreduction at Fe(mapc-t) at
different pH, scan rate: 0.04 V s−1, rotation speed: 200 rpm.
7 (a) E. Rose, A. Kossanyi, M. Quelquejeu, M. Soleilhavoup, F.
Duwavran, N. Bernard and A. Lecas, J. Am. Chem. Soc., 1996, 118,
1567; (b) Z. Gross, N. Galili and I. Saltsman, Angew. Chem., Int. Ed.,
1999, 38, 1427; (c) R. Paolesse, L. Jaquinod, D. J. Nurco, S. Mini,
F. Sagone, T. Boschi and K. M. Smith, Chem. Commun., 1999, 1307;
(d) Z. Gross, N. Galili, L. Simkhovich, I. Saltsman, M. Botoshansky,
D. Blaser, R. Boese and I. Goldberg, Org.Lett., 1999, 4, 599; (e) R.
Paolesse, S. Nardis, F. Sagone and R. G. Khoury, J. Org. Chem., 2001,
66, 550; (f) D. T. Gryko and K. Jadach, J. Org. Chem., 2001, 66, 4267;
(g) B. Ramdhanie, C. L. Stern and D. P. Goldberg, J. Am. Chem.
Soc., 2001, 123, 9447; (h) R. Guilard, D. T. Gryko, G. Canard, J. M.
Barbe, B. Koszarna, S. Brandes and M. Tasior, Org. Lett., 2002, 4,
4491; (i) J. Sankar, V. G. Anand, S. Venkatraman, H. Rath and T. K.
Chandrashekar, Org. Lett., 2002, 4, 4233; (j) D. T. Gryko, Eur. J. Org.
Chem., 2002, 11, 1735; (k) R. Paolesse, A. Marini, S. Nardis, A. Froiio,
F. Mandoj, D. J. Nurco, L. Prodi, M. Montalti and K. M. Smith,
plots. These kinetic data correspond to the first reduction wave.
The mechanism at more negative potentials was not analysed due
to overlap of the second reduction wave (ascribed to catalysis
by MII(cor3−) form) with residual catalysis by the first active form
of the catalysts, MII(cor2−•), and with possible contribution from
catalysis by EPG due to the porous nature of the catalytic layer.
Conclusions
In summary, we have demonstrated that metal corroles adsorbed
on EPG electrode operate at potentials comparable with those
observed for catalysis with metal porphyrins and are effective cat-
558 | Dalton Trans., 2006, 554–559
This journal is
The Royal Society of Chemistry 2006
©

View Article Online
J. Porphyrins Phthalocyanines, 2003, 7, 25; (l) R. A. Decre´au and J. P.
Collman, Tetrahedron Lett., 2003, 44, 3323; (m) G. R. Geier, J. F. B.
Chick, J. B. Callinan, C. G. Reid and W. P. Auguscinski, J. Org. Chem.,
2004, 69, 4159; (n) J. P. Collman and R. A. Decre´au, Org. Lett., 2005,
7, 975.
8 Z. Gross and H. B. Gray, Adv. Synth. Catal., 2004, 346, 165.
9 J. Grodkowski, P. Neta, E. Fujita, A. Mahammed, L. Simkhovich and
Z. Gross, J. Phys. Chem. A, 2002, 106, 4772.
10 K. M. Kadish, L. Fre´mond, Z. Ou, J. Shao, C. Shi, F. C. Anson, F.
Burdet, C. P. Gros, J.-M. Barbe and R. Guilard, J. Am. Chem. Soc.,
2005, 127, 5625.
11 V. A. Adamian, F. D’Souza, S. Licoccia, M. L. Di Vona, E. Tassoni,
R. Paolesse, T. Boschi and K. M. Kadish, Inorg. Chem., 1995, 34, 532;
E. Van Caemelbecke, S. Will, M. Autret, V. A. Adamian, J. Lex, J.-P.
Gisselbrecht, M. Gross, E. Vogel and K. M. Kadish, Inorg. Chem.,
1996, 35, 184.
12 (a) Z. Gross, G. Golubkov and L. Simkhovich, Angew. Chem., Int. Ed.,
2000, 39, 4045; (b) L. Simkhovich, A. Mahammed, I. Goldberg and
Z. Gross, Chem. Eur. J., 2001, 7, 1041; (c) A. Mahammed, I. Giladi, I.
Goldberg and Z. Gross, Chem. Eur. J., 2001, 7, 4259; (d) J. P. Collman
and R. A. Decre´au, Tetrahedron Lett., 2003, 44, 1207.
13 T. Geiger and F. C. Anson, Anal. Chem., 1980, 52, 2448.
14 V. S. Bagotskii, V. Yu. Filinovskii and N. A. Shumilova, Russ. J. Elec-
trochem., 1968, 4, 1129; K.-L. Hsueh and D.-T. Chin, J. Electroanal.
Chem., 1983, 153, 79.
17 The behavior of Co(tpfc)PPh3 and Co(mapc-t) is the same; Co(tpfc)
without an axial base is unstable and could not be isolated, as previously
reported in ref. 12c.
18 (a) G. R. Schonbaum and B. Chance, In The Enzymes, 3rd edn., ed.
P. D. Boyer, Academic Press, New York, 1976, 13, 363; (b) J. Paschke,
M. Kirsch, H.-G. Korth, H. de Groot and R. Sustmann, J. Am. Chem.
Soc., 2001, 123, 11099.
19 (a) A similar electron transfer was reported for Fe catecholate (ref. 19a),
Cu corroles (ref. 19b,c) and Rh porphyrins (ref. 19d): D. D. Cox and
L. Que, Jr., J. Am. Chem. Soc., 1988, 110, 8085; (b) C. Bru¨ckner, R. P.
Brin˜as and J. A. Krause Bauer, Inorg. Chem., 2003, 42, 4495; (c) R.
Guilard, C. P. Gros, J.-M. Barbe, E. Espinosa, F. Je´roˆme, A. Tabard,
J.-M. Latour, J. Shao, Z. Ou and K. M. Kadish, Inorg. Chem., 2004, 43,
7441; (d) J. P. Collman and R. Boulatov, J. Am. Chem. Soc., 2000, 122,
11812.
20 A possible reason is electronic repulsion of CN− due to the negative
charge of MII(cor3−) as suggested in: B. Ramdhanie, J. Telser, A.
Caneschi, L. N. Zakharov, A. L. Rheingold and D. P. Goldberg, J. Am.
Chem. Soc., 2004, 126, 2515; N. S. Hush and I. S. Woolsey, J. Chem.
Soc., Dalton Trans., 1974, 24.
21 The reason for and the potential of the invoked internal electron
transfer are not presently clear. A possible explanation is that at
reducing potentials the electrode becomes negatively charged and can
favour the electronic state of the adsorbed species which is avoided
when the potential is maintained at positive values; by analogy with
the well-known phenomenon of the reorientation of adsorbed aromatic
substances (flat orientation vs. vertical one): B. B. Damaskin, O. A.
Petrii and V. B. Batrakov, Adsorption of Organic Compounds on
Electrodes, New York, Plenum Press, 1971.
15 W. M. Clark, Oxidation-Reduction Potentials of Organic Systems,
Baltimore, Williams & Wilkins, 1960.
16 A. J. Bard and L. R. Faulkner, Electrochemical Methods, Wiley, New
York, 2001.
This journal is
The Royal Society of Chemistry 2006
Dalton Trans., 2006, 554–559 | 559
©