Selective 4e-/4H+ O2 reduction by an iron(tetraferrocenyl)porphyrin complex: From proton transfer followed by electron transfer in organic solvent to proton coupled electron transfer in aqueous medium

Article abstract of DOI:10.1021/ic402297f

An iron porphyrin catalyst bearing four ferrocenes and a hydrogen bonding distal pocket is found to catalyze 4e^-/4H⁺ oxygen reduction reaction (ORR) in organic solvent under homogeneous conditions in the presence of 2-3 equiv of Trifluoromethanesulphonic acid. Absorption spectroscopy, electron paramagnetic resonance (EPR), and resonance Raman data along with H₂O₂ assay indicate that one out of the four electrons necessary to reduce O₂ to H₂O is donated by the ferrous porphyrin while three are donated by the distal ferrocene residues. The same catalyst shows 4e^-/4H⁺ reduction of O₂ in an aqueous medium, under heterogeneous conditions, over a wide range of pH. Both the selectivity and the rate of ORR are found to be pH independent in an aqueous medium. The ORR proceeds via a proton transfer followed by electron transfer (PET) step in an organic medium and while a 2e^-/1H ⁺ proton coupled electron transfer (PCET) step determines the electrochemical potential of ORR in an aqueous medium.

Full text of DOI:10.1021/ic402297f

Article
pubs.acs.org/IC
Selective 4e−/4H+ O2 Reduction by an Iron(tetraferrocenyl)Porphyrin
Complex: From Proton Transfer Followed by Electron Transfer in
Organic Solvent to Proton Coupled Electron Transfer in Aqueous
Medium
Kaustuv Mittra, Sudipta Chatterjee, Subhra Samanta, and Abhishek Dey*
Department of Inorganic Chemistry, Indian Association for Cultivation of Science, Kolkata, India 700032
S
* Supporting Information
ABSTRACT: An iron porphyrin catalyst bearing four ferrocenes
and a hydrogen bonding distal pocket is found to catalyze 4e−/
4H+ oxygen reduction reaction (ORR) in organic solvent under
homogeneous conditions in the presence of 2−3 equiv of
Trifluoromethanesulphonic acid. Absorption spectroscopy, elec-
tron paramagnetic resonance (EPR), and resonance Raman data
along with H2O2 assay indicate that one out of the four electrons
necessary to reduce O2 to H2O is donated by the ferrous
porphyrin while three are donated by the distal ferrocene residues.
The same catalyst shows 4e−/4H+ reduction of O2 in an aqueous
medium, under heterogeneous conditions, over a wide range of
pH. Both the selectivity and the rate of ORR are found to be pH
independent in an aqueous medium. The ORR proceeds via a
proton transfer followed by electron transfer (PET) step in an organic medium and while a 2e−/1H+ proton coupled electron
transfer (PCET) step determines the electrochemical potential of ORR in an aqueous medium.
particularly important for applications in H2/O2 fuel cells.
Ferrocene undergoes 1 electron oxidation to form ferrocenium
and has recently been used effectively as an external electron
donor in the presence of a Cu catalyst and strong acids in the
four electron reduction of oxygen to water.8 Recently an iron
porphyrin complex, containing 4 electron donating ferrocene
groups covalently attached to the porphyrin ring, was reported
(Figure1). This complex was found to catalyze reduction of O2
to O2− in an organic solvent and a selective 4e−/4H+ reduction
of oxygen to water at pH 7.19 Importantly the selectivity for
4e−/4H+ ORR in aqueous medium was retained under both
fast and slow electron fluxes from the electrode. This was due
to the availability of electron donating ferrocene groups in the
molecule. The crystal structure of an analogous Zn porphyrin
complex showed the presence of strong H-bonding interactions
in the distal pocket facilitated by the triazole rings and mediated
by water molecule. This H-bonding interaction forces axial
ligands like methanol and O2 to bind inside the sterically
congested distal pocket instead of the open proximal side.20 H-
bonding residues are known to act as proton translocator for
ORR catalysts. This attribute is known to aid proton coupled
electron transfer (PCET) steps often invoked to be involved in
ORR.14
1. INTRODUCTION
The oxygen reduction reaction (ORR) is important in
biological respiration as well as in H2/O2 fuel cell applications.1
H2/O2 fuel cells oxidize hydrogen at the anode and reduce
oxygen at the cathode generating an EMF of 1.23 V. A platinum
catalyst is used on both cathode and anode.
2H2 → 4H+ + 4e− (Anode Reaction)
O2 + 4H+ + 4e− → 2H2O (Cathode Reaction)
The ORR at the cathode is much slower than the hydrogen
oxidation reaction occurring at the anode.2 To speed up the
ORR kinetics, to make practical fuel cells, an efficient ORR
catalyst is required.3,4 Platinum, even though it is the most
efficient catalyst, is expensive.2−5 Hence, extensive research
over the past few decades has targeted the development of
efficient non-noble metal based ORR catalysts.6−9 This effort
has led to identification of several key attributes essential for a
good ORR catalyst.
It is desirable that an ORR catalyst should not only be stable
but also be able to reduce oxygen efficiently over a wide range
of pH. Most known metalloporphyrin based ORR electro-
catalysts show optimal activity (both selectivity and rate) at a
specific pH.10−15 There are rare examples of catalysts which
show comparable selectivity in acidic, neutral, and basic
pHs.16,17 Even rarer are examples of catalysts that show the
same catalytic rate of ORR over a large pH range.18 This is
Received: September 10, 2013
Published: December 4, 2013
© 2013 American Chemical Society
14317
dx.doi.org/10.1021/ic402297f | Inorg. Chem. 2013, 52, 14317−14325

Inorganic Chemistry
Article
100 μL of this 1 mM FeFc4 solution was added to an EPR tube (A).
To three other EPR tubes (B), (C), (D), 0.5 equiv (5 μL) of Na2S was
added. It has been recently shown that in a nonpolar organic solvent
Na2S can reduce FeIII porphyrin to FeII porphyrin and itself gets
oxidized to elemental sulfur in the process.21 While the sample in tube
A showed a high spin FeIII signal in EPR, the samples in tubes B, C, D
were EPR silent indicating that all the FeIII has been reduced to FeII.
This was also indicated by the shift of the Soret band from 426 nm in
the oxidized to 433 nm in the reduced form. Next, oxygen gas was
bubbled through the solution in the EPR tubes B, C, D at −80 °C.
This was followed by addition of 20−30 μL (2−3 equiv) of
trifluoromethanesulphonic acid (10 mM in THF) to C and addition
of about 50 μL (5 equiv) of the same acid to EPR tube D all at −80
°C. The EPR and Raman data of all tubes were recorded.
The UV experiments to monitor the oxidation of ferrocenes were
done similarly except that a 0.5 mM concentration of the catalyst was
used.
H2O2 Detection. A xylenol orange assay was used to detect H2O2
produced during O2 reduction under homogeneous conditions. A 4.9
mg portion of Mohr’s salt and 3.9 mg of xylenol orange were dissolved
in 5 mL of 250 mM H2SO4 and stirred for 10 min. A 200 μL portion
of this solution was taken in 1.8 mL of triple distilled water, and a
calibration curve for quantitative estimation of H2O2 was obtained by
adding 20 μL aliquots of H2O2 having different concentrations and
recording their absorbance at 560 nm. The concentrations of H2O2
used were 0.05 μM, 0.1 μM, 0.5 μM, 1 μM, 2.5 μM, 5 μM, 10 μM, and
100 μM.
A 200 μL portion of the xylenol orange H2SO4 mixture was added
to 1.8 mL of H2O in a cuvette, and the absorbance was recorded. A
100 μL portion of 1 mM reduced FeFc4 (1 mM oxidized α4-FeFc4 in
THF + Na2S (5 μL of 10 mM in THF)) in an organic solvent (THF)
was exposed to dry O2 gas at −80 °C. This solution was then extracted
with 200−400 μL of H2O. Twenty microliters of this aqueous extract
was added to the cuvette containing the xylenol orange and H2SO4
mixture. Absorbance for this was recorded. The absorbance of the
above solution (after subtracting the control) at 560 nm is fitted in the
calibration curve (obtained as described above) to get the
corresponding H2O2 concentrations.22 This concentration is scaled
accounting for dilution to get the concentration of H2O2 produced in
the original α4-FeFc4 solution. Addition of 20 μL of the catalyst
solution in THF to the xylenol orange and H2SO4 mixture did not
result in increase of absorbance at 560 nm. The same process was
repeated to acquire the data after addition of 2−3 equiv and 5 equiv
trifluoromethanesulphonic acid at −80 °C to the dioxygen adducts of
FeFc4.
Figure 1. Schematic diagram of the α4-FeFc4 complex.
In this paper we report that the α4-FeFc4 catalyst can, in fact,
reduce O2 to water under homogeneous conditions in an
organic solvent in the presence of acid. Further this complex
can catalyze ORR with almost equal efficiency in an aqueous
medium, under heterogeneous conditions, over a wide range of
pH (1 to 9). Not only the selectivity but also the catalytic rate
of ORR is also found to be independent of pH. This is likely
due to the H-bonding cavity consisting of triazoles (pKa ∼1) in
the catalyst which provides efficient proton translocation into
the catalytic center over the entire pH range.
2. EXPERIMENTAL DETAILS
2.1. Synthesis. The complexes are synthesized as previously
reported.19
2.2. Materials and Methods. Sodium sulphide was purchased
from Rankem, trifluoromethanesulphonic acid (triflic acid) and
methanol (MeOH) were purchased from Sigma Aldrich (U.S.A.). p-
Toluenesulfonic acid (PTSA) was purchased from Spectrochem.
Unless otherwise stated all chemicals were used as purchased, and
reactions were performed at room temperature. Edge Plane Graphite
(EPG) discs were purchased from Pine Instruments. All the electron
paramagnetic resonance (EPR) spectra were recorded on a JEOL
instrument. Resonance Raman (rR) data were collected using 413.1
nm excitation from a Kr+ ion source (Coherent Inc.) and a Trivista
555 triple spectrophotometer (gratings used in the three stages were
900, 900, and 1800 grooves/mm) fitted with an electronically cooled
Pixis CCD camera (Princeton Instruments). The irradiation power
was limited to 10 mW at the sample to avoid degradation. Data were
collected at room temperature and 77 K for 200 s.
2.3. Electrochemical Measurements. A 50 μL portion of a
dilute solution (1 mM) of the α4-FeFc4 catalyst (in CHCl3) was
uniformly distributed on the graphite disc. After the CHCl3 had
evaporated, the surface was sonicated in ethanol for 30 s and washed
with triple distilled water. The linear sweep voltammetry (LSV) were
recorded on electrochemical analyzer purchased from CH Instruments
using CHI720 D. A Pt wire was used as a counter electrode. The
measurements were made against an Ag/AgCl aqueous reference
electrode with scan rates varying from 50 mV/s to 500 mV/s. 100 mM
phosphate buffers were used to maintain pH from 0 to 9 (pH adjusted
by adding H3PO4 or KOH). 100 mM KPF6 was used as supporting
electrolyte in each case.
3. RESULTS
3.1. 4e−/4H+ ORR in Organic Solvent (Homogeneous).
The initial oxidized α4-FeFc4 in THF solvent is reduced by 0.5
equiv of Na2S (dissolved in dry degassed methanol). The
reaction of the reduced complex with O2 in an organic solvent
is followed by EPR and rR spectroscopy. EPR spectroscopy is a
useful tool to determine the spin state of iron porphyrin
complexes. rR can be used to determine the oxidation and spin
states of iron porphyrin complexes using ν4 and ν2 modes of
the porphyrin ring; also known as oxidation and spin state
marker bands.
a. EPR. The EPR spectrum of oxidized α4-FeFc4 is shown in
(Figure 2, blue). The oxidized complex exhibits a g ∼ 6 axial
EPR signal indicating that the Fe in the α4-FeFc4 complex is in
a high-spin S = 5/2 FeIII state. On reduction the complex
becomes EPR silent consistent with the formation of a reduced
FeII complex. Upon exposing the reduced sample to O2 at −80
°C no new EPR signal is observed (Figure 2, red) suggesting
the presence of a diamagnetic dioxygen adduct.23 On addition
of 2−3 equiv and excess (5 equiv) of trifluoromethanesul-
phonic acid (triflic acid) to the dioxygen adduct, a new rhombic
2.4. H2O2 Detection and Other Experimental Details. A 1.67
mg portion of α4-FeFc4 catalyst was dissolved in 1 mL of dry degassed
tetrahydrofuran (THF) solvent so that the final strength was 1 mM. A
2.4 mg portion of Na2S was dissolved in minimum volume of
methanol and diluted with dry degassed THF to make the final volume
1 mL, so that the final strength was 10 mM. THF was rigorously
degassed using the freeze−pump−thaw technique. Next, in a glovebox,
14318
dx.doi.org/10.1021/ic402297f | Inorg. Chem. 2013, 52, 14317−14325

Inorganic Chemistry
Article
and 1553 cm−1(Figure 3, sky blue), respectively. Addition of 5
equiv of triflic acid follows the same trend, and the bands shift
to 1364 cm−1 and 1553 cm−1 (Figure 3, violet). Thus, it is clear
that in both cases a high spin FeIII species is formed when the
diamagnetic FeII−O2 adduct of the α4-FeFc4 complex is treated
with triflic acid. This is consistent with the EPR data. Note that
both the rR (ν4 and ν2 bands are different) and the EPR data
(the product has more rhombic character) indicate that the
starting high-spin FeIII α4-FeFc4 complex is different from the
final product after addition of 3 and 5 equiv of acid.26 There is
however no change in the rR data in the presence of a weaker
acid, p-toluenesulphonic acid (PTSA, pKa = 8.5) compared to
triflic acid (pKa = 0.3) or in the presence of methanol (pKa =
27.9) which can also act as a proton source. Even addition of 1
equiv of triflic acid does not change the EPR or rR spectra.
c. H2O2 Assay. Hydrolysis of the FeII−O2 adduct prepared in
homogeneous organic medium yielded 48 3% H2O2 in the
resultant solution, that is, the α4-FeFc4 catalyst selectively
Figure 2. EPR spectra of FeFc4 showing the oxidized, dioxygen adduct
and the final product at room temperatue after addition of 3 equiv and
excess of triflic acid at −80 °C.
−
reduces O2 to O2 and not H2O in a nonpolar organic solvent
(O2− disproportionate to H2O2 and O2 yielding an overall 50%
H2O2).20,22 On addition of 2−3 equiv of triflic acid to the FeII−
O2 complex yielded only 1−5% H2O2. This is an interesting
observation as it indicates that addition of 2−3 equiv of triflic
acid to the FeII−O2 adduct leads to 4e− reduction of O2 to H2O
while the lack of acid produces 50% H2O2. This implies that the
presence of acid is necessary to drive 4e− reduction of O2, that
is, the ORR catalyzed by α4-FeFc4 catalyst likely involves a
proton assisted step. However 50% of H2O2 is detected when 1
equiv of triflic acid or a weaker acid like PTSA or when no acid
is used instead, that is, the reaction is stalled at the FeII−O2
stage under these conditions. Alternatively, when more than 5
equiv of triflic acid is added ∼50% H2O2 is observed, that is,
significant 2e−/2H+ ORR is observed when excess acid is used,
implying hydrolysis of the FeII−O2 or FeIII−OOH intermedi-
ates (Scheme 3) to yield H2O2 under strongly acidic conditions.
d. Quantification of Ferrocenium Produced during O2
Reduction. The UV spectrum (Figure 4) shows a new
signal at g ∼ 6 (Figure1, green) is obtained indicating that the
complex returns to a high spin FeIII oxidation state. The
rhombicity in the EPR spectra of the final product is due to the
replacement of the Br− counteranion present in the starting
complex by a π anisotropic hydroxo or methoxy ligand in the
final product.20
b. Resonance Raman. rR data of the oxidized sample shows
that the oxidation and spin state marker ν4 and the ν2 bands are
at 1361 cm−1 and 1552 cm−1, respectively (Figure 3, black),
Figure 3. rR data of the as isolated α4-FeFc4 complex (black), the fully
reduced α4-FeFc4 complex (red), the oxygen adduct (green, note some
of the reduced complex is still present as indicated by the presence of a
weak ν4 band at 1346 cm−1) and the final product (blue). 1 mM in
THF, 413.1 nm excitation, 77 K, 10 mW.
which are typical of S = 5/2 FeIII porphyrins, consistent with
the EPR data (Figure 2, black).24 Upon reduction these bands
shift to 1343 cm−1 and 1541 cm−1, indicating the formation of a
high-spin S = 2 FeII species (Figure 3, red). The diamagnetic O2
adduct shows a new set of bands at 1368 cm−1 and 1568 cm−1
(Figure 3, green).24 This species is an end-on S = 0 FeII−O2
adduct (Supporting Information, Figure S1).23 Dioxygen
adduct of a ferrous porphyrin can be described as FeII−O2 or
Figure 4. UV spectra in the region of 700−860 nm showing the FeFc4
catalyst in its initial oxidized state (blue) and after addition of 5 equiv
of CAN (green). The spectrum in red is obtained after adding 3 equiv
of triflic acid to the oxygenated catalyst at 298 K.
absorbance band at 788 nm after addition of ∼3 equiv of
triflic acid to the oxygenated α4-FeFc4 at 298 K. Neither the
oxidized nor the reduced α4-FeFc4 complex show a λmax at 788
nm and thus this new band at 788 nm indicates the 1 electron
oxidation of the ferrocene groups of the complex.27−29 Thus,
the ferrocene groups supply the electrons necessary for
reduction of oxygen to water in organic medium, in the
presence of 2−3 equiv of acid as they themselves get oxidized
to ferrocenium. Further in an effort to quantify the number of
−
FeIII−O2 . Recent X-ray absorption data and its analysis
indicate that none of the limiting descriptions may be
adequate.25 Thus this species is referred to as a FeII−O2
species in the rest of the paper. On addition of 2−3 equiv of
triflic acid to the diamagnetic dioxygen adduct at −80 °C
initially causes no change. However, warming up to about −40
°C results in shifting of the ν4 and the ν2 bands to 1363 cm−1
14319
dx.doi.org/10.1021/ic402297f | Inorg. Chem. 2013, 52, 14317−14325

Inorganic Chemistry
Article
Figure 5. LSV data of α4-FeFc4 deposited on EPG surface at multiple rotations in pH 0, pH 2, pH 4, pH 6, pH 7, pH 8 phosphate buffers, using 100
mM KPF6 as the supporting electrolyte and Pt and Ag/AgCl as counter and reference electrodes, respectively. (inset) K-L plot of the α4- FeFc4
catalyst (black bold line). The theoretical plots for 4e- and 2e- processes are indicated by dashed (red) and dotted lines (black), respectively. The
RDE data for pH 1, pH 3, pH 5, pH 9 are shown in Supporting Information, Figure S2.
ferrocenes getting oxidized, a methanol solution of cerric
ammonium nitrate (CAN) is added to the α4-FeFc4 solution,
where the iron in the porphyrin is in the oxidized ferric state
while the ferrocenes are reduced. CAN being a strong oxidizing
agent (E0 = 1.39 V vs Ag/AgCl) can oxidize the ferrocenes (E0
= 0.73 V vs Ag/AgCl) present in the complex. Since, each α4-
FeFc4 molecule contains 4 ferrocene groups, increasing
equivalents of CAN is added to oxidize the ferrocenes to
ferroceniums. The resultant spectrum (green) showed a peak at
788 nm indicating the oxidation of ferrocene groups.
Approximately 5 equiv of CAN is required to oxidize all the
ferrocene groups (Supporting Information, Figure S4). The
ratio of the intensities at 788 nm for the spectrum obtained
after reduction of the O2 in presence of ∼3 equiv of acid and
the spectrum obtained after the oxidation of the four ferrocene
groups in α4-FeFc4 indicate that only 80% of the ferrocenes
present in the catalyst are oxidized to ferrocenium during the
reduction of oxygen to water. Since, oxidation of three
ferrocenes out of four should yield a ratio of 0.75, the observed
ratio of 0.8 thus implies that only 3 electrons for the reduction
of oxygen to water are derived from the ferrocenes while the
fourth is supplied by Fe when it gets oxidized from +2 to +3 as
is indicated by EPR and rR data in sections 3.1.a and 3.1.b,
respectively.
3.2. 4e−/4H+ ORR in Aqueous Medium (Heteroge-
neous). a. Selectivity of O2 Reduction. Linear sweep
voltammetry (LSV) data of this catalyst immobilized on edge
plane graphite (EPG) electrodes and immersed in aqueous
solution show a substrate diffusion limited catalytic O2
reduction current below −0.1 V (Figure 5) at pH 0−pH 9 in
aqueous phosphate buffer solutions. The cathodic peak at +300
mV represents the reduction of the ferrocene groups, and it
serves as an internal reference to quantify the amount of
catalyst on the electrode as each iron porphyrin molecule
contains four ferrocene substituents. Its potential does not
change with pH. The onset of the catalytic current coincides
14320
dx.doi.org/10.1021/ic402297f | Inorg. Chem. 2013, 52, 14317−14325

Inorganic Chemistry
Article
with the reduction of the iron in the porphyrin from the
oxidized ferric state to the reduced ferrous state. The O2
reduction current increases with increasing rotation rates
the α4-FeFc4 catalyst reduce oxygen by 4 electrons in between
pH 1−9, but also has the similar catalytic kcat values indicating
that the efficiency of the catalyst to reduce oxygen to water
does not change in this pH range. Note that the kcat values are
corrected for differences in loading in every experiment
(Supporting Information, Table S1). Thus this is a rare
example of a catalyst where both the kcat and the selectivity are
retained in the pH range of 1−9, that is, across a proton
concentration gradient of 109.
following the Koutecky−Levich equation, I−1 = iK(E)−1
+
iL−1, where iK(E) is the potential dependent kinetic current and
iL is the Levich current. iL is expressed as 0.62nFA[O2]-
(DO2)
2/3ω1/2v‑1/6, where n is the number of electrons
transferred to the substrate, A is the macroscopic area of the
disc (0.125 cm2), [O2] is the concentration of O2 in an air
saturated buffer (0.26 mM) at 25 °C, DO2 is the diffusion
coefficient of O2 (1.8 × 10−5 cm2 s−1) at 25 °C, ω is the angular
velocity of the disc, and v is the kinematic viscosity of the
solution (0.009 cm2 s−1) at 25 °C. The plot of I−1 at multiple
rotation rates vs the inverse square root of the angular rotation
rate (ω−1/2) is linear. The slopes obtained from the
experimental data for the catalyst at pH 1−9 is almost identical
with the theoretical slope predicted for a 4e− process (Figure 5
and Supporting Information, Figure S2).30 The data at pH 0
indicates significant 2e−/2H+ reduction from O2 to H2O2. The
turn over number (TON) of oxygen reduction by α4-FeFc4
catalyst at pH 7 is estimated (from the amount of charge
consumed and the coverage of the catalyst) to be >103 at pH 7.
b. Variation of kcat with pH. The intercept of the K-L plot
can be used to obtain the kinetic current, ik, which can be used
to obtain the second order ORR rates for the catalyst (ik =
nFA[O2]kcatΓcatalyst, where Γcatalyst is the catalyst concentration
in moles cm−2). The data (Table 1) suggests that not only does
4. DISCUSSION
In an aqueous medium O2 is reduced to H2O over a wide range
of pH. It is evident that the potentials for ORR (EORR) shifts
toward positive values as the pH is lowered (Figure 6). A plot
of EORR with pH shows a slope of 28 mV per pH unit. A general
expression of the pH dependence of a Proton Coupled Electron
Transfer (PCET) pathway is −0.059f/n(pH), where n is the
number of electrons and f is the number of protons.31 In this
case the observed slope is consistent with n = 2 and f = 1, that
is, the ORR by the α4-FeFc4 catalyst involves a two electron and
1H+ proton step as the potential determining step.31,32 O2
reduction to water is a multistep reaction and several of them
can show a 2e−/1H+ PCET. The protonation of the FeII−O2
adduct, leading to the formation of a FeIII−O2H species,
followed by the cleavage of the O−O bond and simultaneous
protonation of the distal oxygen of the FeIII−O2H species can
be expected to show a 2e/1H+ PCET behavior (Proposal 1,
Scheme 1). A similar proposal was put forward by Nocera to
explain the mechanism of ORR by binuclear-cobalt pacman
porphyrin.33 Alternatively, initial protonation of a tentative
FeIII−OOH species followed by a O−O bond cleavage can also
be expected to show a 2e−/1H+ PCET pathway (Proposal 2,
Scheme 1). Another likely scenario is the reduction of the FeIII
to FeII followed by oxygen binding and transfer of an electron
and proton to the intermediate FeII−O2 species to generate a
FeIII−O22− species (Proposal 3, Scheme 1). Similarly a coupled
H+ and e− transfer to a FeII−O2 intermediate to generate a
FeIII−OOH species followed by its reduction to a FeII−OOH
species can also show a 2e−/H+ PCET(Proposal 4, Scheme 1).
Proposal 1 (Scheme 2, Pathway C) is less likely in the case of
mononuclear catalysts as the protonation of the FeII−O2
species is likely to increase its hydrolysis rate. Proposal 2 will
require a concerted electron and proton transfer to a FeIII−
OOH2 species. A protonated FeIII−OOH2 species can be
a
Table 1. ORR Activity of the α4- FeFc4 Catalyst
pH
kcat× 105 (M−1 s−1
)
no. of electrons
Tafel slopes (mV/dec)
0
1
2
3
4
5
6
7
8
9
3.9 0.5
1.0 0.1
1.1 0.2
1.2 0.3
2.0 0.6
0.7 0.3
2.3 0.3
1.2 0.2
1.0 0.1
4.5 0.5
3.0
3.9
4.0
3.9
4.0
4.0
3.9
3.7
3.8
3.9
ND
ND
ND
ND
ND
139
136
140
118
119
a
ND = Not determined.
Figure 6. Plot of catalytic half peak potential vs pH.
14321
dx.doi.org/10.1021/ic402297f | Inorg. Chem. 2013, 52, 14317−14325

Inorganic Chemistry
Article
a
Scheme 1. Possible Steps in ORR That Can Show a 2e−/1H+ PCET
a
From the top; Proposal 1: Protonation of the FeII−O2 adduct followed by O−O bond cleavage involving a 2e−/1H+ PCET step; Proposal 2:
Protonation of hydroperoxo species followed by bond cleavage involving a 2e−/1H+ step. The highlighted species (inside the box with brown dotted
border) is expected to be unstable and undergo facile O−O bond cleavage; Proposal 3: Reduction of the FeIII to FeII followed by oxygen binding
and 1 e- transfer to the oxygenated species to generate a FeIII−O22− species. For a 2e−/1H+ step the thermodynamic potential of the first step has to
be more negative than the last, that is, E1 < E2 and kO2 > kET; Proposal 4: The two steps, FeII−O2 to FeIII−OOH and reduction of FeIII−OOH to
FeII−OOH, can couple to give a 2e−/1H+ step provided the thermodynamic potential of the first step is more negative than the second, that is, E1 <
E2.
expected to undergo facile O−O bond cleavage, hence it being
the potential determining step (PDS) is slim. Proposal 3
requires simultaneous PCET reduction of a FeIII−OH species
to FeII−OH2 species, oxygen binding to result in a FeII−O2
uphill than reduction of a FeIII−OOH species (Scheme 2,
Pathway A) such that the reduction of the FeIII−OOH species
to a FeII−OOH species will proceeds spontaneously at a
potential where the FeII−O2 species gets reduced to a FeIII−
OOH species. This agrees with the observation that reduction
of the FeII−O2 species is indeed the most energetically
demanding step in an organic medium.
The protonation required in these cases can be facilitated by
the hydrogen bonding distal pocket which can also stabilize
FeII−O2 species. Note that the stabilization of dioxygen adduct
by hydrogen bonding in this distal pocket is already
demonstrated.23 The Tafel slope of ∼120−130 mV between
pH 6−9 (Supporting Information, Figure S3) suggests that the
rate determining step (rds) of ORR is a one electron transfer
step.3 This could be any of the 1 e− reduction steps depicted in
Scheme 2. However most of the reduction steps in Scheme 2
have a simultaneous protonation associated with them. Since
the kcat is pH independent (Table 1) as well as the electron
transfer to any of the Fen+−OHx (x = 0, 1 n = 4, 3) is not likely
to be the r.d.s. The reduction steps are known to involve PCET
species and its reduction resulting in a FeIII−O2 species. For
2−
these set of steps to show a 2e−/H+ PCET the O2 binding rate
has to be much faster that the electron transfer steps and the
most energy demanding step has to be reduction of the FeIII
species to FeII so that the subsequent reduction of a FeII−O2
2−
species to a FeIIIO2 is thermodynamically more feasible, that
is, when the potential of the second step is higher than the first,
these two steps can combine to show a 2e−/1H+ PCET.31
However, the results obtained in the organic solvent do not
support this pathway. This is because the reducing agent
(Na2S) can reduce the FeIII species to FeII and not the FeII−O2
species (stable in the presence of excess Na2S). This would
suggest that the reduction of a FeII−O2 species is
thermodynamically more uphill than reduction of a FeIII
species. Proposal 4, to be a 2e−/1H+ PCET step, will require
reduction of the FeII−O2 species to be thermodynamically
14322
dx.doi.org/10.1021/ic402297f | Inorg. Chem. 2013, 52, 14317−14325

Inorganic Chemistry
Article
a
Scheme 2. Oxygen Reduction Mechanism in Aqueous Medium
a
In the pH range 1-9 the cycle takes Pathway A resulting in 4 electron reduction of O2 to H20. At pH = 0, when the triazoles get protonated,
Pathway B takes its course resulting in increase in release of O2 which is detected as H2O2. Pathway C indicates an alternative 2e−/1H+ PCET
−
pathway involving superoxide protonation as seen in Pacman porphyrins. The abbreviation PDS stands for potential determining step in ORR
electrocatalysis.
a
and hence should, in principle, show pH dependence.
Alternatively, the reduction of FeIII−OOH species to FeII−
OOH, the second step of the overall 2e−/1H+ PCET process,
could be the r.d.s. In fact recent data on an analogous iron
porphyrin (ferrocene groups replaced by ester groups) show
accumulation of FeIII−OOH and FeIVO intermediates on the
electrode at steady state with the help of rR spectroscopy.34
Since the reduction of FeIVO is likely to be a PCET step, the
reduction of FeIII−OOH could be the r.d.s in ORR catalyzed by
these mononuclear iron porphyrin complexes.
Scheme 3. ORR in Organic Medium
−
Under neutral conditions O2 is reduced to O2 in organic
solvents.20 Alternatively, in the presence of 2−3 equiv of triflic
acid in organic medium and at pH > 1(in aqueous buffer), O2 is
reduced to H2O in both mediums. When the acidity of the
medium is further increased (5 equiv of triflic acid in organic
medium and pH < 1 in aqueous medium), enhanced 2e−/2H+
reduction is observed. Recently published results indicate that
the fully reduced α4-FeFc4 catalyst binds O2 and forms a stable
FeII−O2 adduct in an organic medium which hydrolyzes to
20
−
produce O2 . This is in spite of the presence of reducing
equivalents, in the form of reduced Ferrocene groups in the
distal pocket, and protic solvent (MeOH) in the medium. Only
in the presence of 2−3 equiv of a strong acid does the O−O
bond cleavage occur. This observation is very similar to recent
reports by Fukuzumi and Karlin where rapid reduction of CuII−
OOH species by ferrocene was observed in organic solvents
only in the presence of acid.27 Since the protonation of the
a
On addition of 2−3 equiv of triflic acid, the cycle follows Pathway A.
On addition of excess acid (5 equiv) the triazoles get protonated and
the cycle follows Pathway B. Weaker acids like MeOH and PTSA can
not reduce the FeII−O2 adduct. Triflic acid being a stronger acid is
required to drive the reaction forward. The species that have been
identified in solution are highlighted in the (boxes with dotted brown
border) scheme.
−
high-valent P+·FeIVO species. The FeIII−OOH or any high-
valent intermediates are not observed at −40 °C suggesting that
their decay via oxidation of the other Fc groups must be fast.27
Hence only the final hydroxide bound FeIII species is observed
with three Fc oxidized to Fc+, consistent with recent proposals
by Fukuzumi and Karlin.27 Further the reduction of the FeII−
bound O2 species would only lead to hydrolysis, the observe
O−O bond cleavage necessitates synchronous oxidation of an
electron donor species (Scheme 3).35 Under the reaction
conditions, the only available sources of this electron are the Fc
residues. The resultant FeIII−OOH intermediate likely under-
goes fast O−O bond cleavage and may be expected to produce
14323
dx.doi.org/10.1021/ic402297f | Inorg. Chem. 2013, 52, 14317−14325

Inorganic Chemistry
Article
Notes
O2 adduct proceeds only in the presence of a strong acid like
triflic acid and not in the presence of a weak acid like PTSA.
Thus in an organic medium, the FeII−O2 adduct likely
undergoes a proton transfer followed by electron transfer
(PET) (Scheme 3, Pathway A) during ORR in presence of 2−3
equiv of acid. Three out of the four electrons required for
complete reduction of oxygen to water are supplied by the
ferrocenes while one is donated by iron. However the addition
of excess equivalents of the acid causes the FeII−O2 adduct to
be hydrolyzed as superoxide which is detected as PROS
(Supporting Information, Table S2). This acid catalyzed
hydrolysis is very similar to the hydrolysis of the FeII−O2
case when no acid is added (Scheme 3, Pathway B).
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was funded by the Department of Science and
Technology, India, Grant # DST/SR/IC-35-2009 and by
Council of Scientific and Industrial Research (CSIR), India,
01(2412)10/EMr-II. K.M. and S.C. acknowledge CSIR-SRF
fellowship. S.S. acknowledges integrated PhD program of IACS.
REFERENCES
■
(1) Saveant, J.-M. Chem. Rev. 2008, 108, 2348.
(2) Jaouen, F.; Proietti, E.; Lefevre, M.; Chenitz, R.; Dodelet, J.-P.;
Wu, G.; Chung, H. T.; Johnston, C. M.; Zelenay, P. Energy Environ. Sci.
2011, 4, 114.
A close comparison of the results obtained in organic and
aqueous medium reveals certain contrasts and analogies. Under
−
neutral conditions O2 is reduced to O2 and H2O in organic
(3) SongC. Z. In PEM Fuel Cell Electrocatalysts and Catalyst Layers
Fundamentals and Applications; Zhang, J., Ed.; Springer: London, U.K.,
2008.
(4) Wu, G.; Zelenay, P. Acc. Chem. Res. 2013, 46, 1878.
(5) Shin, D.; Jeong, B.; Mun, B. S.; Jeon, H.; Shin, H.-J.; Baik, J.; Lee,
J. J. Phys. Chem. C 2013, 117, 11619.
(6) Liang, Y.; Wang, H.; Diao, P.; Chang, W.; Hong, G.; Li, Y.; Gong,
M.; Xie, L.; Zhou, J.; Wang, J.; Regier, T. Z.; Wei, F.; Dai, H. J. Am.
Chem. Soc. 2012, 134, 15849.
solvents and aqueous medium, respectively. At pH > 1 and in
the presence of 2−3 equiv of acid this catalyst reduced O2 to
H2O in both aqueous and nonaqueous conditions. However
when in an aqueous medium the O−O bond cleavage involves
a PCET pathway; in an organic medium, where O−O cleavage
does not proceed in the presence of protic solvent (MeOH) or
PTSA and 2−3 equiv of triflic acid is required to drive the
reaction, it likely follows a PET pathway. At pH < 1 in buffered
aqueous solutions and in the presence of >5 equiv of triflic acid
in organic solutions, the triazoles get protonated (pKa = 1.5)
which results in acid catalyzed hydrolysis of the iron-oxy adduct
leading to enhanced production of H2O2 (Scheme2, Pathway
B). Indeed addition of triflic acid to the resting oxidized FeFc4
complex in an organic solvent produced a large change in the
Soret (Supporting Information, Figure S6) but no change in
the EPR data (Supporting Information, Figure S7) likely
because of the protonation of the triazole group(s).
(7) Carver, C. T.; Matson, B. D.; Mayer, J. M. J. Am. Chem. Soc. 2012,
134, 5444.
(8) Fukuzumi, S.; Kotani, H.; Lucas, H. R.; Doi, K.; Suenobu, T.;
Peterson, R. L.; Karlin, K. D. J. Am. Chem. Soc. 2010, 132, 6874.
(9) Thorseth, M. A.; Letko, C. S.; Tse, E. C. M.; Rauchfuss, T. B.;
Gewirth, A. A. Inorg. Chem. 2013, 52, 628.
(10) Chang, C. J.; Loh, Z.-H.; Shi, C.; Anson, F. C.; Nocera, D. G. J.
Am. Chem. Soc. 2004, 126, 10013.
(11) Kadish, K. M.; Fremond, L.; Ou, Z.; Shao, J.; Shi, C.; Anson, F.
́
C.; Burdet, F.; Gros, C. P.; Barbe, J.-M.; Guilard, R. J. Am. Chem. Soc.
2005, 127, 5625.
5. CONCLUSION
(12) Ou, Z.; Lu, A.; Meng, D.; Huang, S.; Fang, Y.; Lu, G.; Kadish, K.
M. Inorg Chem. 2012, 51, 8890.
̈
In summary, the α4-FeFc4 complex can act both as a
homogeneous catalyst (in an organic solvent in the presence
of 2−3 equiv of acid) as well as a heterogeneous catalyst (in an
aqueous medium between pH 1−9) for ORR. The triazole
residues, which were bound by hydrogen bonds to water
molecules in the structure of the analogous Zn complex, likely
provide an efficient proton transfer pathway into the active site,
via hydrogen bonding, which is essential for ORR activity.20
Further more of these residues having pKa ∼ 1 act as a local
buffer in the vicinity of the O2 reducing center helping the
catalyst retain its selectivity for 4e−/4H+ ORR over a wide
range of pH. Only below pH 1 in aqueous medium or in the
presence of 5 equiv of acid in organic medium, when the
triazoles are likely protonated, does the selectivity of ORR
decrease, and it shows enhanced H2O2 production due to
hydrolysis of the FeII−O2.
(13) Schechter, A.; Stanevsky, M.; Mahammed, A.; Gross, Z. Inorg.
Chem. 2012, 51, 22.
(14) Matson, B. D.; Carver, C. T.; Von Ruden, A.; Yang, J. Y.; Raugei,
S.; Mayer, J. M. Chem. Commun. 2012, 48, 11100.
(15) Collman, J. P.; Boulatov, R.; Sunderland, C. J.; Fu, L. Chem. Rev.
2003, 104, 561.
(16) Forshey, P. A.; Kuwana, T. Inorg. Chem. 1983, 22, 699.
(17) Shigehara, K.; Anson, F. C. J. Phys. Chem. 1982, 86, 2776.
(18) Zhang, L.; Song, C.; Zhang, J.; Wang, H.; Wilkinson, D. P. J.
Electrochem. Soc. 2005, 152, A2421.
(19) Samanta, S.; Sengupta, K.; Mittra, K.; Bandyopadhyay, S.; Dey,
A. Chem. Commun. 2012, 48, 7631.
(20) Samanta, S.; Mittra, K.; Sengupta, K.; Chatterjee, S.; Dey, A.
Inorg. Chem. 2013, 52, 1443.
(21) Collman, J. P.; Ghosh, S.; Dey, A.; Decrea
Acad. Sci. U.S.A. 2009, 106, 22090.
́
u, R. A. Proc. Natl.
(22) Pramanik, D.; Dey, S. G. J. Am. Chem. Soc. 2011, 133, 81.
(23) Mittra, K.; Chatterjee, S.; Samanta, S.; Sengupta, K.;
Bhattacharjee, H.; Dey, A. Chem. Commun. 2012, 48, 10535.
(24) Burke, J. M.; Kincaid, J. R.; Peters, S.; Gagne, R. R.; Collman, J.
P.; Spiro, T. G. J. Am. Chem. Soc. 1978, 100, 6083.
ASSOCIATED CONTENT
■
S
* Supporting Information
Additional electrochemical data, tables of surface coverages
determined from cyclic voltammetry and Tafel plots for ORR.
This material is available free of charge via the Internet at
http://pubs.acs.org.
́
(25) Wilson, S. A.; Kroll, T.; Decreau, R. A.; Hocking, R. K.;
Lundberg, M.; Hedman, B.; Hodgson, K. O.; Solomon, E. I. J. Am.
Chem. Soc. 2013, 135, 1124.
(26) Simultaneous oxidation of the Fc centers could not be
confirmed from these data as the rR and absorption data is dominated
by the porphyrin and while Fc+ is paramagnetic, it does not generally
show an EPR signal at 77 K.
(27) Kakuda, S.; Peterson, R. L.; Ohkubo, K.; Karlin, K. D.;
Fukuzumi, S. J. Am. Chem. Soc. 2013, 135, 6513.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: icad@iacs.res.in.
14324
dx.doi.org/10.1021/ic402297f | Inorg. Chem. 2013, 52, 14317−14325

Inorganic Chemistry
Article
(28) Mase, K.; Ohkubo, K.; Fukuzumi, S. J. Am. Chem. Soc. 2013,
135, 2800.
(29) Fukuzumi, S.; Mandal, S.; Mase, K.; Ohkubo, K.; Park, H.;
Benet-Buchholz, J.; Nam, W.; Llobet, A. J. Am. Chem. Soc. 2012, 134,
9906.
(30) Bard, A. J. Faulkner, L. R. Electrochemical Methods; Wiley: New
York, 2001.
(31) Smith, E. T.; Davis, C. A.; Barber, M. J. Anal. Biochem. 2003,
323, 114.
(32) Stine, K. J.; Andrauskas, D. M.; Khan, A. R.; Forgo, P.; D’Souza,
V. T. J. Electroanal. Chem. 1999, 465, 209.
(33) Rosenthal, J.; Nocera, D. G. Acc. Chem. Res. 2007, 40, 543.
(34) Sengupta, K.; Chatterjee, S.; Samanta, S.; Dey, A. Proc. Natl.
Acad. Sci. U.S.A. 2013, 110, 8431.
(35) Sengupta, K.; Chatterjee, S.; Samanta, S.; Bandyopadhyay, S.;
Dey, A. Inorg. Chem. 2013, 52, 2000.
14325
dx.doi.org/10.1021/ic402297f | Inorg. Chem. 2013, 52, 14317−14325