Oxygen Reduction by Iron Porphyrins with Covalently Attached Pendent Phenol and Quinol

Article abstract of DOI:10.1021/jacs.0c10385

Phenols and quinols participate in both proton transfer and electron transfer processes in nature either in distinct elementary steps or in a concerted fashion. Recent investigations using synthetic heme/Cu models and iron porphyrins have indicated that p

Full text of DOI:10.1021/jacs.0c10385

pubs.acs.org/JACS
Article
Oxygen Reduction by Iron Porphyrins with Covalently Attached
Pendent Phenol and Quinol
Asmita Singha, Arnab Mondal, Abhijit Nayek, Somdatta Ghosh Dey, and Abhishek Dey*
Cite This: https://dx.doi.org/10.1021/jacs.0c10385
Read Online
ACCESS
Metrics & More
Article Recommendations
*
sı Supporting Information
ABSTRACT: Phenols and quinols participate in both proton
transfer and electron transfer processes in nature either in distinct
elementary steps or in a concerted fashion. Recent investigations
using synthetic heme/Cu models and iron porphyrins have
indicated that phenols/quinols can react with both ferric
superoxide and ferric peroxide intermediates formed during O2
reduction through a proton coupled electron transfer (PCET)
process as well as via hydrogen atom transfer (HAT). Oxygen
reduction by iron porphyrins bearing covalently attached pendant
phenol and quinol groups is investigated. The data show that both
−
+
of these can electrochemically reduce O selectively by 4e /4H to H O with very similar rates. However, the mechanism of the
2
2
reaction, investigated both using heterogeneous electrochemistry and by trapping intermediates in organic solutions, can be either
PCET or HAT and is governed by the thermodynamics of these intermediates involved. The results suggest that, while the reduction
III
−
III
̇
of the Fe −O species to Fe −OOH proceeds via PCET when a pendant phenol is present, it follows a HAT pathway with a
2
pendant quinol. In the absence of the hydroxyl group the O reduction proceeds via an electron transfer followed by proton transfer
2
III
−
III
−
III
̇
̇
to the Fe −O species. The hydrogen bonding from the pendant phenol group to Fe −O and Fe −OOH species provides a
2
2
unique advantage to the PCET process by lowering the inner-sphere reorganization energy by limiting the elongation of the O−O
bond upon reduction.
8
,9
1
. INTRODUCTION
medium. The unique post-translational modification has
been emulated in synthetic heme/Cu systems (Scheme 1b,c)
as well. It has been observed that incorporation of Cu metal in
the distal site of Fe-porphyrin not only produces a very stable
ferric superoxide intermediate but also weakens the O−O
bond enhancing the rate of its scission.
distance between Fe and Cu metal centers are flexible, addition
of O to the fully reduced heme/Cu complex results in the
formation of a stable bridging η :η Fe -peroxo-Cu species
Factors that can affect the rate and selectivity of the reduction
of molecular oxygen are of great contemporary interest. In
aerobic eukaryotes and higher animals, selective 4e /4H
oxygen reduction is performed by the membrane-bound
cytochrome C oxidase (CcO), which is a terminal oxidase in
the respiratory redox chain in mitochondria. The active site
of CcO (Scheme 1a) is comprised of a histidine-bound heme
a and a histidine-ligated Cu center. Of the three histidine
3
3
−
+
10,11
However, when the
1
,2
2
3
2
2
III
II
B
residues, one is cross-linked with a tyrosine residue, which acts
as an electron and proton donor during the reduction of
(Scheme 1b), where the peroxo ligand coordinates with Fe and
2
2
2
1
Cu in either η :η or η :η fashion depending on the
12−16
oxygen in the case of a catalytically relevant mixed-valent
coordination number of the Cu center.
Inclusion of a
4
(
MV) state of heme-Cu oxidases. The reduction of oxygen to
phenol, either covalently attached to an imidazole or added
externally, results in the formation of a phenoxyl radical akin to
−
+
water requires 4e and 4H , and in the catalytic cycle of CcO
out of the four electrons two are supplied by reduced heme a
3
the Tyr244 radical formed in the active site of CcO (P state
M
II
IV
17,18
(
Fe ), which is oxidized to ferryl (Fe O), one is supplied by
of CcO).
When these complexes are attached to an
I
II
Cu (Cu ), which is oxidized to Cu , and the last electron is
B
electrode and electrochemical oxygen reduction is investigated,
these can selectively reduce oxygen to water even when the
derived from the tyrosine residue, which is covalently attached
to the His240 residue (Scheme 1a) through a post-transla-
tional modification, when the system is electron-deficient, that
is, MV state, which avoids the formation of partially reduced
Received: September 29, 2020
−
oxygen species (PROS) like O 22 and OH·, which are
,4−65−7
detrimental to biological systems.
Several synthetic
Fe-porphyrins and heme/Cu systems (Scheme 1b−d) have
been reported, which can catalyze an electrochemical oxygen
reduction both in an organic solvent as well as in an aqueous
©
XXXX American Chemical Society
https://dx.doi.org/10.1021/jacs.0c10385
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
A

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
a
Scheme 1. Structures Investigated in This Work
a
(A) Active site of CcO (truncated from pdb id: 2OCC). Schematic drawings of (B) proposed peroxide intermediate heme-Cu complex with
external phenol. (C) Heme-Cu complex with a covalently attached phenol. (D) Fe porphyrin with a basic amine residue in the distal site and (E)
Fe porphyrins having hydroquinone (FeQH ), phenol (FePh), and 2,5-dimethoxy phenyl ring (FeQMe ).
2
2
2
7
supply of electrons from the electrode is retarded (utilizing a
self-assembled monolayer of thiol on Au electrodes), where the
phenol donates an electron to oxygen to complete its 4e
hexadecanethiol-coated Au electrodes. The same peroxide
intermediate is found to accumulate in iron porphyrins with a
different axial ligand and in heme/Cu systems (both synthetic
and biosynthetic) indicating that the cleavage of the O−O
bond is the rate-determining step in ORR by iron porphyrins
−
reduction under these conditions mimicking the reactivity of
19
the MV state of CcO. Several recent investigations have
highlighted the role of axial ligand and hydrogen bonding in
these heme/Cu systems, with a particular focus on the factors
that enable efficient scission of the peroxide bond, which is the
last step of the reduction of oxygen to water, that is, cleaving
30−32
as well as heme/Cu systems under these conditions.
pK
The
III
of the proximal oxygen (bound to the iron) of the Fe −
a
OOH intermediate is between 5.5 and 7.0 for neutral axial
ligand and that of the distal oxygen (not bound to the iron) is
Accordingly, spatial control of proton
transfer directing it specifically to the distal oxygen of this
Fe −OOH to affect a heterolytic O−O bond cleavage was
envisaged to enhance both the selectivity and rate of O
reduction. Recently, mononuclear Fe-porphyrins having
pendant residues like aliphatic amine, pyridine and guanidine
etc. in the distal site, as envisaged above, (Scheme 1d) have
demonstrated a 100 fold increase relative to previously
reported heme and heme/Cu systems in the rate of oxygen
2
0−24
33,34
the O−O bond.
about ∼1 unit lower.
Electrochemical oxygen reduction being a multielectron
III
multiproton process often results in the formation of PROS
−
like O2 and H O2 due to the incomplete reduction of
2
2
2
5−28
oxygen.
spectroscopy coupled to rotating disc electrochemistry
SERRS-RDE) on mononuclear Fe porphyrins having no
discrete distal superstructure such as Fe-tetraphenyl porphyrin
In situ surface enhanced resonance Raman
(
III
(
FeTPP) show that the hydrolysis of an Fe −OOH
intermediate is responsible for the release of H O (Scheme
). The hydrolysis is accentuated when the electron transfer
−
+
reduction with >95% 4e /4H selectivity even under slow
2
2
35
2
9
electron fluxes from the electrode. Kinetic and spectroscopic
investigations in organic solutions indicated that the
protonated amines can rapidly cleave the O−O bond
heterolytically forming compound I with activation barriers
of ∼12.5 kcal/mol only explaining the rate acceleration and
high selectivity exhibited by these mononuclear iron
porphyrins during oxygen reduction reaction (ORR) in
2
to this species is slow offering more time for this species to be
hydrolyzed, and as much as 100% H O can be detected when
2
2
the electron transfer rate from the electrode to the immobilized
−
1
iron porphyrin catalyst is slowed down to ∼10 s using
Scheme 2. A General Mechanism of O Reduction by Iron
2
III
III
aqueous medium. Thus, the reduction of Fe −OOH during
Porphyrins Indicating the Fate of the Key Fe −OOH
electrochemical ORR in an aqueous medium likely involves a
proton transfer step (PT) followed by electron transfer (ET)
to the ensuing compound I species; that is, the reaction
Species As Observed using SERRS-RDE under
Heterogeneous Aqueous Conditions
36
proceeds via PTET. The active site of CcO, however, does
not have such pendant bases; rather, it has the tyrosine residue
with a phenol headgroup and can still exhibit a selective
reduction of O to H O. In a recent report on a heme/Cu
2
2
system, a combination of kinetic experiments and theoretical
III
calculations suggests that the conversion of end-on Fe -
II
IV
II
̇
peroxo-Cu to Fe -oxo-Cu −OH-PhO (intermediate analo-
B
https://dx.doi.org/10.1021/jacs.0c10385
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
1
Figure 1. (A) X-ray structure of FePh, (B) schematic representation of FeQMe indicating the -OMe H resonances, (C) schematic representation
2
of FeBz.
gous to P in CcO) proceeds through a transition state (TS),
the methylated quinol can only act as a H-bond acceptor. An
iron porphyrin with a pendant phenyl group and FeTPP act as
the control samples. The electrochemical data show that, while
M
where a H-bonding interaction from PhOH cleaves the O−O
bond homolytically prior to the proton coupled electron
transfer (PCET) from the PhOH residue as indicated by
experimental and theoretical values of the kinetic isotope effect
−
+
all of these complexes can selectively reduce O by 4e /4H
2
with more than 95% selectivity and with similar rates, the
solution characterization of the intermediates involved shows
1
8
(
KIE) of the reaction. Very recently, H-bond assisted O−O
bond homolysis has been reported for a ferric hydroperoxide
that the mechanism of the O reduction varies from PCET,
2
complex during the reduction of O in an organic medium by a
HAT, and ETPT depending on the chemical nature of the
pendant group. The experimental data, along with density
functional theory (DFT) calculations, suggest that, while a
preorganization of the pendant group is key, intrinsic
thermodynamic parameters of the pendant hydroxyl groups
and a lowering of the inner-sphere reorganization (λinner‑sphere)
2
mononuclear Fe porphyrin having a hydroquinone ring
covalently attached in the distal superstructure (Scheme 1e,
3
7
FeQH2). The complex utilizes two consecutive hydrogen
atom transfer (HAT) steps to reduce oxygen to water
completely via ferric superoxide, ferric hydroperoxide, and
ferryl oxo intermediate species. A similar HAT by heme
superoxide complexes has recently been observed in other
III
−
III
̇
for the PCET from Fe −O to Fe −OOH species by
2
hydrogen bonding from these groups play a decisive role in
determining the reactivity and mechanism of action.
3
8−40
heme systems.
Thus, the crucial O−O bond cleavage of
an intermediate ferric peroxide species is reported to be aided
by PTET (heme/distal base, Scheme 1d), PCET (heme/Cu/
Phenol, Scheme 1c), as well as HAT (heme/Quinol, Scheme
2. METHODS AND MATERIALS
37
All reagents and solvents are purchased from commercial sources.
Sodium sulphide is purchased from Merck. 2,5-Dimethoxy benzoic
acid, 2-hydroxy benzoic acid, FeBr , 2,4,6-collidine, KPF , and CDCl
3
1
e), albeit in very different systems. Both phenol and
hydroquinone are known to participate in PCET reactions in
nature as well. Phenol is involved in oxygen reduction reaction
by CcO in its mixed-valent form, while hydroquinone is found
to donate a proton and electron to cytochrome C in Q-
cytochrome c oxidoreductase, known as complex III, during
2
6
1
8
are purchased from Sigma-Aldrich. O is purchased from Sigma-
2
Aldrich and Icon Isotopes. The sample preparations for resonance
Raman (rR) and electron paramagnetic resonance (EPR) experiments
and the handling of air-sensitive materials are performed under an
18,41,42
1
electron transport in mitochondria.
Phenols and quinols
inert atmosphere inside the glovebox. Room-temperature H NMR
are active participants in the reactivity of several Cu enzymes
spectra are collected on a Bruker DPX-500 spectrometer. The EPR
spectra are recorded on a JEOL instrument using either an Oxford
43,44
and related synthetic complexes.
However, it is still
unknown if these two residues can directly participate in
oxygen reduction reaction when they are attached with a
mononuclear Fe porphyrin macrocycle. Apart from its obvious
relation to the active site of CcO, how these residues tune the
mechanism (PCET, HAT, etc.) of the reaction is of
fundamental interest. Thus, a systematic investigation into
the role of groups like phenol and quinol, which are capable of
participating in proton and electron transfer using concerted or
independent steps, in a closely related molecular framework,
into the rates and selectivity of ORR (either in solution or
under heterogeneous conditions) is desirable.
flow cryostat (4 K) or liquid N finger dewar (77 K). Resonance
2
+
Raman data are collected using 413.1 nm excitation from a Kr ion
source (Sabre, Coherent Inc.) and a Trivista 555 triple spectropho-
tometer (gratings used in the three stages are 900, 900, and 1800/
2400 grooves/mm) fitted with an electronically cooled Pixis charge-
coupled device (CCD) camera (Princeton Instruments). The first two
stages of the spectrometer are used as a tunable bandpass. X-ray
single-crystal data are collected at 100 K on a Bruker D8VENTURE
Microfocus diffractometer equipped with PHOTON II Detector, with
Mo Kα radiation (λ = 0.710 73 Å), controlled by the APEX3
(v2017.3-0) software package. Raw data are integrated and corrected
for Lorentz and polarization effects using the Bruker APEX II95/
APEX III program suite.
In this manuscript, O reduction is investigated by three
2
porphyrins having a pendant hydroquinone (Scheme 1e,
FeQH ), phenol (Scheme 1e, FePh), and 2,5-dimethoxy
3. EXPERIMENTAL PROCEDURE
2
phenyl ring (Scheme 1e, FeQMe ) covalently attached to the
2
All the porphyrins were synthesized starting from o-aminophenyl-
tris(phenyl)-porphyrin (monoamine, 1a), which, in turn, was
synthesized by condensation of 1 equiv of o-nitrobenzaldehyde, 3
equiv of benzaldehyde, and 4 equiv of pyrrole followed by a reduction
of the nitro group by adding SnCl2 in dilute HCl followed by
purification via a silica gel column.
porphyrin ring (Figure 1) both in organic solution as well as
under heterogeneous electrochemical aqueous conditions.
Quinol and phenol pendant groups can act as both H-bond
donor and acceptor and can serve as both a proton and
electron donor in distinct as well as concerted pathways, while
C
https://dx.doi.org/10.1021/jacs.0c10385
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
1
3
.1. FePh. One equivalent of 2-methoxy benzoic acid is dissolved
protons in its H NMR consistent with a high-spin ferric porphyrin
(Figure S8).
in dry tetrahydrofuran (THF) solvent. It is then reacted with 4 equiv
of oxalyl chloride added dropwise under refluxing condition and kept
under Ar atmosphere overnight. Excess oxalyl chloride is evaporated
to get the acid chloride (quantitative yield) as a yellowish oil. It is
then treated with o-aminophenyl-tris(phenyl)-porphyrin dissolved in
dry THF in the presence of 4 equiv of dry triethylamine and kept
overnight at RT. The reaction mixture is then evaporated, dissolved in
dichloromethane (DCM) solvent, and washed with water in a
separatory funnel. The organic part is extracted, dried over anhydrous
Na SO , and evaporated via a rotary evaporator. The solid crude
3.3. FeBz. Benzoic acid was dissolved in dry DCM. 10 equiv of
thionyl chloride was added to it under refluxing conditions and stirred
for 4 h under an Ar atmosphere. The excess thionyl chloride was
evaporated, and a solution of o-aminophenyl-tris(phenyl)-porphyrin
in dry DCM was added to it. The green reaction mixture was refluxed
for 2−3 h under an Ar atmosphere (Scheme S1). The resulting
reaction mixture was neutralized with triethylamine, whereby it
changed color to purple, and washed with water and extracted with
DCM. The organic phase was separated, dried over Na SO ,
evaporated, and purified by column chromatography using silica gel
60−120 mesh). The desired product was obtained by using an 80%
DCM-hexane solvent mixture as the eluent. The violet colored
compound was isolated when the solvent was evaporated. Yield: 80%;
H NMR (400 mHz, CDCl ) δ ppm = 8.93 (m, 8H), 8.21 (m, 7H),
.93 (m, 1H), 7.86 (m, 10H), 7.59 (m,1H), 6.86 (t, 1H),
2
4
2
4
product is then purified with column chromatography with silica gel
60−120 mesh) and an 80% DCM-hexane solvent mixture as the
eluent. The final product is a violet powder. Yield: (>90%).
The resulting porphyrin is dissolved in dry DCM and allowed to
react with 50 equiv of BBr for 20 h at 0 °C. It is then neutralized with
a saturated aqueous solution of sodium bicarbonate. The reaction
mixture is then evaporated, dissolved in DCM, and washed with water
in a separatory funnel. The organic part is extracted and then dried
over anhydrous Na SO and evaporated via a rotary evaporator. The
crude reaction mixture is then purified using column chromatography
with silica gel (60−120 mesh) and a 60% DCM-hexane mixture. The
final product is a reddish-brown powder. Yield: (80%). H NMR
400mHz, CDCl ) δ ppm = 11.51 (s, 1H), 8.90 (m, 8H), 8.27 (m,
H), 8.15 (m, 8H), 7.82 (m, 4H), 7.81(m,1H), 7.78 (m,1H), 7.61 (d,
H), 7.26 (s,1H), 6.5 (d, 1H), −2.61 (s, 2H) (Figure S2).
Electrospray ionization mass spectrometry (ESI-MS) (positive ion
mode in acetonitrile (ACN)): m/z (%) = 749 (100) (Figure S1).
The porphyrin ligand is then metalated using FeBr in dry THF in
(
(
1
3
3
7
6
.56(m,4H), −2.60 (s, 2H) (Figure S9), ESI-MS (positive ion
mode in ACN): m/z (%) = 734 (100) (Figure S10).
The resulting porphyrin ligand was then metalated with FeBr in
THF in the presence of 2 equiv of 2,4,6-collidine under an Ar
2
4
2
atmosphere. The excess FeBr was then removed by working up the
1
2
reaction mixture with dilute HCl (1:4 concentration HCl/water) and
DCM. The organic part was then separated, dried over anhydrous
Na SO , and evaporated using a rotary evaporator. It was then
purified by column chromatography on silica gel (60−120 mesh) and
using a 1% DCM-methanol solvent mixture as the eluent. A deep
purple colored compound was isolated. It exhibits paramagnetic shifts
of the meso-phenyl protons in its H NMR spectrum consistent with a
high-spin ferric porphyrin (Figure S11). Yield: 80%. ESI-MS (positive
ion mode in ACN): m/z (%) = 787 (100) (Figure S12).
(
7
1
3
2
4
2
1
the presence of 2 equiv of 2,4,6-collidine under Ar atmosphere. The
excess FeBr is removed by working up the reaction mixture with
dilute HCl, and the complex is extracted using DCM. The organic
part is then dried over anhydrous Na SO and evaporated using a
2
3
.4. FeQH . FeQH is synthesized according to the previously
2 2
37
2
4
reported procedure.
.5. Resonance Raman. An Fe-porphyrin is dissolved in dry and
rotary evaporator. The crude product is then purified using column
chromatography with silica gel (60−120 mesh) and a 1% methanol-
DCM solvent mixture as the eluent. The final product is a deep purple
powder. Yield: (70−75%) ESI-MS (positive ion mode in ACN): m/z
3
degassed THF so that the concentration of the solution is 1 mM in
iron porphyrin. It is then reduced to its ferrous form by adding 0.5
equiv of Na S (20 mM stock solution in MeOH) under an Ar
atmosphere. A 200 μL portion of the stock solution of the reduced
porphyrin is then taken in a sample tube. The reduced sample tubes
are then oxygenated at −80 °C and frozen in liquid N at specified
times. Resonance Raman data of these sample are collected at 77 K
liquid N dewar) using a 413 nm laser source.
2
(
%) = 804 (100) (Figure S3). It exhibits paramagnetic shifts of the
1
meso-phenyl protons in its H NMR consistent with a high-spin (HS)
ferric porphyrin (Figure S4). The structure of the molecule is further
confirmed using single-crystal X-ray diffraction (Figure 1A).
2
3.2. FeQMe . 2,5-Dimethoxy benzoic acid is dissolved in dry
2
(
2
DCM, and 10 equiv of thionyl chloride is added dropwise under a
refluxing condition and kept under an Ar atmosphere for 4 h. Excess
thionyl chloride along with solvent are then evaporated to get the
corresponding acid chloride as a yellow oil in quantitative yields. It is
then treated with o-aminophenyl-tris(phenyl)-porphyrin dissolved in
dry DCM and stirred for another 3−4 h at 45 °C under Ar
atmosphere. The reaction mixture is then neutralized with a saturated
aqueous solution of sodium bicarbonate and extracted with DCM and
water; it was dried over anhydrous Na SO and evaporated using a
rotary evaporator and purified by column chromatography on silica
gel (60−120 mesh), where the desired product was obtained using an
0% DCM-hexane solvent mixture as the eluent. A violet colored
compound was isolated. Yield: (80−85%); H NMR (400mHz,
CDCl ) δ ppm = 8.93 (m, 8H), 8.22 (m, 6H), 8.02 (m, 1H), 7.91 (m,
1H), 7.81 (m, 2H), 7.66 (d, 1H), 7.51(s,1H), 5.5 (d, 1H), 3.65 (s,
H), 1.62 (s, 1H), −0.12 (s, 2H), −2.60 (s, 2H) (Figure S6). ESI-MS
positive ion mode in ACN): m/z (%) = 794 (100) (Figure S5).
The resulting porphyrin ligand is then metalated with FeBr in
THF in the presence of 2 equiv of 2,4,6-collidine under an Ar
atmosphere. The excess FeBr is then removed when the reaction
3
.6. Electron Paramagnetic Resonance. The sample prepara-
tion for the EPR experiment is very similar to that for the resonance
Raman experiments. The only difference is that, for collecting the
EPR data, the sample solution is taken in an EPR tube instead of a
Raman tube. The EPR data are recorded using an X-band EPR
instrument at 4 K in the case of FePh and at 77 K in the case of
FeQMe2.
3
.7. Electrochemical Experiments. 3.7.1. Cyclic Voltammetry.
The cyclic voltammograms (CV) are recorded using a CH instrument
potentiostat model (710D). A very dilute solution (∼1 mM) of Fe-
porphyrins dissolved in CHCl is physiadsorbed on an edge-plane
graphite (EPG) electrode. The loosely bound iron porphyrins are
removed by sonicating the electrode in MeOH for 30 s. The resultant
EPG bearing physiadsorbed complexes are cleaned with deionized
water and used as the working electrode. A Pt wire is used as a
counter electrode, and a Ag/AgCl (saturated KCl) standard electrode
is used as the reference electrode. Phosphate buffer solutions (100
mM in phosphate) are used to maintain the pH of an electrolyte
2
4
8
3
1
3
1
3
(
2
solution containing 100 mM KPF .
2
6
mixture is washed with dilute HCl (1:4 concentration HCl/water).
The organic part is extracted with DCM, separated, dried over
anhydrous Na SO , and evaporated using a rotary evaporator. It is
3.7.2. Rotating Disk Electrochemistry (RDE). The RDE measure-
ments are performed using a CHI 710D bipotentiostat along with a
Pine Instruments modulated speed rotor fitted with an E6 series
change-disc tip. The complex is physiadsorbed on an edge plane
graphite electrode as described above. The RDE experiment is
performed by measuring linear sweep voltammetry (LSV) at a scan
rate of 10−100 mV/s at different rotation rates using Ag/AgCl
(saturated KCl) reference and Pt counter electrodes.
2
4
then purified using column chromatography on silica gel (60−120
mesh), and the desired product is obtained using a 1% DCM-
methanol solvent mixture as eluent. A deep purple colored compound
is isolated. Yield: (∼80%). ESI-MS (positive ion mode in ACN): m/z
=
847 (Figure S7). It exhibits paramagnetic shifts of the meso-phenyl
D
https://dx.doi.org/10.1021/jacs.0c10385
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Scheme 3. Synthetic Scheme of Preparation of FePh (1c) and FeQMe (2b)
2
3
.7.3. Rotation Ring Disk Electrochemistry (RRDE): Reactive
4. RESULTS
.1. Synthesis. The porphyrin ligand with a pendant
phenol is obtained via a multistep synthesis by first converting
-methoxy benzoic acid to the corresponding acid chloride
Oxygen Species (ROS) Detection and Calculation. For RRDE
experiments a Pt ring-disc assembly is used, where the Pt ring is
placed encircling the disc electrode, and both Pt ring and the disc are
used as working electrodes. The Pt ring is first cleaned by polishing it
with alumina powder (grit sizes: 1, 0.3, and 0.05 μ) and then
electrochemically using a cylindrical glass cell equipped with Ag/AgCl
4
2
(
(
Scheme 3). It is then refluxed with o-aminophenyl-tris-
phenyl)-porphyrin in dry DCM for 4−5 h to yield the
corresponding amide (1a). Similarly, for the synthesis of the
dimethylquinone substituent, 2,5-dimethoxy benzoic acid is
converted to its acid chloride, which is then refluxed with o-
aminophenyl-tris(phenyl)-porphyrin in the presence of dry
DCM to obtain the corresponding amide (2a). The methoxy
group in the distal phenyl ring of 1a is then demethylated to
yield the corresponding phenol by treating the porphyrin with
(
0
saturated KCl) reference and Pt counter electrodes immersed into
.5 M H SO solution. In the RRDE technique, oxygen is reduced
2
4
catalytically on the disk, whose potential is swept from positive to
negative, and the Pt ring is held at a constant potential (0.7 V at
pH7), where the partially reduced oxygen species such as superoxide
peroxide, etc., produced during oxygen reduction on the disc gets
oxidized resulting in an oxidation current. The PROS is the ratio of
the 2e /2H ring current (corrected for collection efficiency) and the
catalytic disc current and is normalized to 100%. The collection
efficiency (CE) is evaluated at a 10 mV/s scan rate using 2 mM
K Fe(CN) and 0.1 M KNO solutions at a rotation speed of 300
−
+
BBr at 0 °C to result in ligand 1b. The ligands 1b and 2a are
3
metalated with Fe using standard protocols to yield FePh and
FeOMe (Scheme 3). These porphyrins and iron complexes
2
1
3
6
3
are characterized using ESI-MS and H NMR. The FePh
rpm. The CE is measured as 18 ± 1% during these experiments.
The RRDE was performed on the Fe porphyrins, which were
physiadsorbed on edge plane graphite electrode as well as alkane thiol
complex is further characterized using single-crystal X-ray
diffraction, where the single crystals are obtained by a slow
diffusion of acetonitrile into a chloroform solution of FePh
(
C8-SH and C16-SH) self-assembled monolayers (SAM) modified
(
Figure 1A). The structure shows that the phenyl ring is atop
gold electrode. The thiols are adsorbed on a freshly cleaned gold
electrode by dipping freshly cleaned Au electrodes (disc or wafers) in
a 0.1 mM ethanolic solution of the desired thiol solutions for 24 h to
form uniform SAM. Dilute solutions of iron porphyrins in CHCl are
deposited on these SAM following the protocol used for the EPG
electrodes.
the porphyrin ring. The chloride counterion is derived during
purification, which involved a workup in dilute HCl. While the
1
FeQMe complex could not be crystallized, the H NMR
2
3
spectrum of the precursor complex 2a shows one of the two
−
−
OMe singlets at −0.12 ppm. Such an upfield shift of the
OMe group (3.65 ppm) is due to the aromatic current of the
3.8. Computational Details. DFT calculations are performed
porphyrin ring implying that the −OMe group is oriented
above the porphyrin plane (Figure 1B). The control FeBz
complex, which does not possess any of the pendant groups, is
synthesized by the condensation of o-aminophenyl-tris-
using the Gaussian 03 software package. Geometry optimizations are
done with the B3LYP functional within an unrestricted formalism.
A mixed basis set consisting of 6-311G* on Fe and 6-31G* on other
atoms is used for geometry optimization and frequency calculation.
The structures of the ferric superoxide complexes have been
45,46
47
(
phenyl)-porphyrin with benzoyl chloride followed by
48−51
subsequent metalation of the porphyrin with FeBr2.
optimized using a broken symmetry approach.
structures yielded no imaginary frequency. Tight self-consistent field
SCF) convergence and a 6-311+G* basis set in a polarizable
All the optimized
4.2. Heterogeneous Electrochemistry. The complexes
(
FePh, FeQMe , and FeQH are adsorbed on an EPG electrode,
2
2
continuum model were used on all atoms for single-point energy
calculations. For the calculation of the reduction potential water is
used as the solvent.
and their CV data, in degassed pH 7 buffer solution, show the
III/II
0
Fe
E (Figure 2A−C) at −310, −170, and −250 mV versus
Ag/AgCl (saturated KCl) reference, respectively. In addition
E
https://dx.doi.org/10.1021/jacs.0c10385
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
ORR on EPG electrodes can be assessed by a Koutechy-Levich
52
(
K-L) analysis of RDE data. In this experiment, the linear
sweep voltammetry data of oxygen reduction by an Fe-
porphyrin complex, physiadsorbed on an edge plane graphite
electrode, are collected at different rates of rotation. The
catalytic current of oxygen reduction increases with an increase
in rotation rate following eqs i and (ii). The number of
electroactive species on the electrode is estimated by
performing CV in the absence of O before conducting the
2
ORR experiments. The inverse of current in a mass-transfer
limited region varies linearly with the inverse of the square root
of the angular rate of rotation, and the plot yields both the
second-order rate constant of ORR (eq i, calculated from the
intercept of the K-L plot) as well as the number of electrons
provided to the substrate, O (n in eq ii, calculated from the
2
Figure 2. Anaerobic CV of FeQMe (orange), FeQH (blue), and
2
2
−1/2
−1
slope of the K-L plot). The plot of ω
potential for FePh (Figure 4A), FeQH (Figure 4C), and
with I at −0.4 V
FePh (green) physiadsorbed on an EPG electrode immersed in pH 7
buffer solution in the presence of Ag/AgCl (saturated KCl) reference
electrode and Pt counter electrode. The data were recorded at a scan
rate 500 mV/s at 25 °C.
2
FeQMe (Figure 4E) yields second-order rate constants of
2
6
−1 −1
6
−1 −1
(7.09 ± 1) × 10 M s , (1.97 ± 1) × 10 M s , and 5.31
7
−1 −1
±
2) × 10 M s , respectively, and the slope (inset) of all of
+
−
to the FeIII/II process at −250 mV, the quinol (QH ) to
these complexes indicates that they reduce oxygen by 4e /4H
to H O. Thus, these mononuclear iron porphyrins reduce O
2
2
2
semiquinone (SQ) CV is observed at +200 mV versus Ag/
to H O at very similar rates in pH 7 buffer solution when
2
AgCl (saturated KCl) for the FeQH complex (Figure 2b).
2
III/II
0
adsorbed on graphite electrodes. Alternatively, the control
FeBz complex decays during oxygen reduction such that
reproducible K-L analysis could not be achieved. Such a decay
is generally associated with the generation of PROS, that is,
The plot of both the Fe
A, simulation to the data points−brown and red lines) shows
a slope of 60 mV/pH suggesting that both these processes
and SQ/QH E versus pH (Figure
2
3
−
+
III/II
involve 1e /1H PCET. Similarly, the Fe
FePh is also a 1e /1H PCET as indicated by the slope (60
mV/pH) of E versus pH in the plot (Figure 3B, simulation to
data points−red line) of FePh. In all these cases the Br
counterion bound to the resting ferric complex synthesized is
exchanged with OH in the aqueous medium. The reduction
process for the
−
+
−
+
lower selectivity for 4e /4H reduction. The selectivity for
ORR, however, is much more accurately determined using
rotating ring disc electrochemistry (RRDE) and is described
below.
0
−
−
−
1
= ik−1 + il−1
I
(i)
III
of this Fe −OH center is associated with the protonation of
the hydroxide ligand to water resulting in the PCET behavior
2/3
1/2ν−1/6
i = 0.62nFA[O ](D )
ω
(ii)
III/II
III
−
+
II
l
2
O2
of the Fe
OH2.
process, that is, Fe −OH + e + H → Fe −
At pH 7, the onset potential of the ORR currents obtained
III/II
0
4
.2.1. Electrocatalytic Oxygen Reduction. The electrode
using the FePh and FeQH complexes align with the Fe
E
2
III
II
bearing these complexes is immersed in an air-saturated pH 7
buffer, and on sweeping the potential below the E of FeIII/II,
the CV is replaced by an electrocatalytic oxygen reduction
current. Electrochemical ORR is characterized by rate, onset
(Figure 4B,D), indicating that the reduction of the Fe to Fe
0
is the most thermodynamically uphill process in the ORR
catalytic cycle. However, the onset potential of the ORR
current for the FeQMe complex is cathodically shifted by 130
2
−
+
−
+
III/II
potential, and selectivity (2e /2H vs 4e /4H ). The rate and
selectivity of the ORR by these iron porphyrins catalyzing
mV with respect to the Fe
process, indicating that the ORR
involves at least one step that requires more driving force than
III
II
Figure 3. (A) CV data of FeQH physiadsorbed on EPG are collected in a buffer solution having different pH. Plot of the potential of the Fe /Fe
2
redox couple and a semiquinone/hydroquinone redox couple vs pH of the deoxygenated buffer solution (blue triangles and green diamonds,
respectively) and the simulation of the pH dependence using a PCET formalism (brown and red lines, respectively). (B) CV data of FePh
III
II
physiadsorbed on EPG are collected in buffer solutions having different pH values. Plot of the potential of the Fe /Fe redox couple vs pH of the
III II
deoxygenated buffer solution shows a 60 mV shift of the Fe / redox potential per unit change in pH (blue diamonds) and the simulation of the
pH dependence using a PCET formalism (red line).
F
https://dx.doi.org/10.1021/jacs.0c10385
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 4. RDE and LSV data of FePh (A, B), FeQH (C, D), and FeQMe (E, F) deposited on an EPG surface in pH 7 phosphate buffer using 100
2
2
mM KPF as the supporting electrolyte and Ag/AgCl (saturated KCl) and Pt as reference and counter electrodes, respectively. The RDE data were
6
−
−
collected at multiple rotations. (inset) K-L plot of Fe porphyrins is represented by a solid line. The theoretical plots for 4e and 2e processes are
indicated by dashed lines in red and blue, respectively. LSV and RDE data were recorded at 100 mV/s scan rate, while the CV were recorded at 500
mV/s.
the reduction of FeIII to Fe (Figure 4F). Note that, in the
II
the large catalytic wave, as the latter is almost 2 orders of
magnitude higher than the former. Indeed, the reduction
III/II
presence of oxygen, the Fe
observed in FeQMe , as O binding to ferrous iron porphyrins
EC step) are of the order of 10 s (the second-order rate
constant is ∼10 mol s , and the oxygen concentration is 0.2
mM in water ), and the resulting 1e prewave is buried under
reduction wave was not
III
−
III
potential of Fe −O is measured to be lower relative to Fe
2
2
2
3
−1
53
(
in organic solvents in the absence of protons. The pH
dependence of ORR (defined as the peak potential of the
electrocatalytic ORR current) of FeQH and FeQMe show
7
−1 −1
8
2
2
G
https://dx.doi.org/10.1021/jacs.0c10385
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 5. Plot of EORR vs pH for (A) FePh, (B) FeQH , and (C) FeQMe molecules physiadsorbed on EPG immersed in buffer solutions having
2
2
different pH (data are in blue diamonds, and the corresponding PCET simulation is shown in the red line). The data were recorded in the presence
of a Ag/AgCl reference electrode and Pt counter electrode at 25 °C. All the cyclic voltammograms were recorded at a 500 mV/s scan rate.
−
+
the same 1e /1H slope similar to the pH dependence of the
Fe
Scheme 4. Schematic Representation of a General
Mechanism of Oxygen Reduction by Synthetic Fe
Porphyrins
III/II
process for these complexes (Figure 5B,C). The pH
dependence of the EORR of the FePh complex (Figure 5A) has
a slope of 30 mV/pH, indicative of a 2e /1H PCET step as
−
+
the potential defining step of the ORR. This is different from
0
−
+
that of the pH dependence of the E , which is 1e /1H PCET.
−
+
The 2e /1H dependence can be due to a second ET step
having a higher or the same E as the Fe
ORR catalysis. The differences in onset potential and pH
dependencies clearly suggest that, despite similarities in the
catalytic ORR rates, the mechanism of the ORR is different for
these different porphyrins, which vary only marginally in their
distal substituents.
0
III/II
involved in the
5
4
4
.2.2. Measurement of Amount of Partially Reduced
−
+
Oxygen Species during Reduction of Oxygen. A 4e /4H
ORR by mononuclear iron porphyrins with selectivity as high
as 95% is very rare. Most mononuclear iron porphyrins show a
−
+
25−28
substantial (30−100%) 2e /2H ORR forming H O .
In
2
2
situ resonance Raman spectroscopy has revealed that the fate
III
of an Fe −OOH intermediate, which accumulates under a
steady state and is detected using resonance Raman, formed
during ORR is mostly responsible for the generation of H O
2
2
2
9,55
(Scheme 4).
Depending on the axial ligand on the iron
3
6
porphyrin, this intermediate species is either protonated or
using rapid kinetics, EPR, and reactivity. Hydrolysis of the
−
+
III
reduced for O−O bond cleavage (i.e., the next step of 4e /4H
Fe −OOH intermediate formed during ORR is accentuated
3
4
ORR). However, if protons and electrons are not provided
by slowing down electron transfer (ET) from the electrode to
the catalyst, which enhances the extent of competitive
hydrolysis of this intermediate. Slowing down of the ET
from the electrode can be achieved by immobilizing these
catalysts on SAM of thiols on Au electrodes, where the rate of
III
rapidly, the Fe −OOH hydrolyzes to produce H O , that is,
2
2
−
+
2
e /2H ORR. In iron porphyrin complexes with pendant
amines that are protonated at neutral pH values and deliver
III
protons to the unbound distal oxygen atom of the Fe −OOH
5
species, the O−O bond heterolysis is very facile, which
ET across these electrodes can be attenuated between 10 and
10 s by extending the chain length of the thiol.
complexes are immobilized on octanethiol SAM (C SAM) and
hexadecanethiol (C SAM), where the rates of ET are ∼10
and ∼10 s , respectively, and the extent of H O released is
−
1
19,56
rescinds the possibility of PROS production via a hydrolysis of
These
III
35
the Fe −OOH intermediate. The heterolysis of the O−O
bond of a ferric peroxide of these porphyrins to result in
compound I has recently been established in organic solutions
8
3
1
6
−1
2
2
H
https://dx.doi.org/10.1021/jacs.0c10385
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
III/II
E0 at pH 7, and the selectivity of ORR stays at
assessed using RRDE. In RRDE any H O released on the
that of Fe
2
2
−
+
working electrode is diffused to and can be reoxidized at a Pt
ring electrode, encircling the working Au disc electrode by
poising a constant potential on it, while the ORR current is
4e /4H even under a slow ET flux. The FeQH complex, with
2
6
−1
a quinol, shows an ORR rate constant of (1.97 ± 1) × 10 M
−
1
−
+
s , a 1e /1H PCET onset potential that overlays with the
52
III/II
0
−
+
investigated at the working electrode. Thus, an assessment of
Fe
E at pH 7, and the selectivity stays at 4e /4H even
the extent of H O released under very slow ET fluxes allows a
under a slow ET flux. The FeQMe complex, where the
2
2
2
III
direct investigation of the inherent reactivity of the Fe −OOH
intermediate; that is, larger PROS reflect a lower stabilization
and/or slower rate of O−O bond cleavage. The data (Table 1,
hydroxyl groups of FeQH are methylated, show an ORR rate
2
7
−1 −1
−
+
of (5.31 ± 2) × 10 M
s
on EPG, an 1e /1H PCET onset
III/II
0
potential that is shifted cathodically relative to the Fe
E at
−
+
pH 7, and the selectivity for the 4e /4H ORR is gradually
reduced at slow ET fluxes. To gain insight into these
differences, the reaction of the reduced ferrous FePh,
FeQH , and FeQMe with O is monitored in an organic
Table 1. % of PROS Produced by Fe Porphyrins on
Different Electrodes
2
2
2
C SAM
1
6
medium.
C SAM
modified Au
8
EPG
modified Au
(k = 6−10
4.3. Reaction with O in Organic Solution. The FePh,
2
ET
complex (kET = 10 s−1) (kET = 10 s
6
3
−1
)
s
−1
)
refs
FeQH , and FeQMe complexes are reduced to their ferrous
2
2
FeQH2
3.3 ± 0.2
4.0 ± 0.5
1.8 ± 0.5
11 ± 0.5
10
5.5 ± 0.6
3.3 ± 0.4
3.6 ± 0.4
10.2 ± 0.5
27 ± 2
this work
forms, and their reaction with O is followed at −80 °C (dry
2
(
Figure
ice/MeOH bath) to trap and characterize the reaction
S14A)
intermediates and elucidate the mechanism of O activation
2
FePh
3.2 ± 0.3
4.6 ± 0.2
18 ± 1
29
this work
by these complexes. The reduced FeQMe complex when
(
Figure
2
S14B)
exposed to dry O gas forms an EPR-silent species. The
2
FeQMe2
FeBz
this work
resonance Raman data of this species exhibit the oxidation and
(
Figure
−1
spin-state marker bands ν and ν at 1368 and 1568 cm ,
4
2
S14C)
respectively, consistent with the formation of a low-spin (LS)
this work
(
Figure
Fe(III) species in solution (Figure S16A). The Fe−O vibration
III
−
−1
S15)
of this diamagnetic Fe −O ·
̇
species is observed at 585 cm
2
−
1
−1
18
FeTPP
rapid
degradation
35
and is confirmed by a 28 cm shift to 557 cm on using O-
labeled O gas (Figure 6). This Fe −O · species does not
57
III
−
̇
2
2
Figure S13 blue) show that FePh, FeQH , and FeQMe
2
2
produce only 4.0 ± 0.5, 3.3 ± 0.2, and 1.8 ± 0.5% PROS on
EPG electrodes, respectively. The data on the EPG electrode
are consistent with the RDE data, which showed a selective
−
+
4
e /4H ORR for all of these complexes as well. The FePh as
−
+
well as the FeQH complexes show a very selective 4e /4H
2
ORR even under very slow ET rates, with a maximum of 4%
PROS (Table 1, rows 1 and 2, Figure S13), suggesting that
either these complexes have an inherent mechanism for
promoting the O−O bond cleavage or these complexes can
III
stabilize the Fe −OOH intermediate, substantially avoiding
their hydrolysis. In contrast, the FeQMe complex shows a
2
gradual increase in the PROS from 1.8% to 10% (i.e., a fivefold
increase, Table 1 row 3, Figure S13) as the ET rate is slowed.
Figure 6. Resonance Raman data of the superoxide complex of
1
6
FeQMe . The blue spectra represent the O adduct of FeQMe , and
2
2
2
1
8
The behavior of FeQMe is similar to iron porphyrins without
the red one represents the O adduct of FeQMe .
2 2
2
distal architectures, where as much as 100% PROS has been
reported on C SAM, albeit the effect of slow ET is quite
1
6
demure in FeQMe relative to the iron porphyrins reported
react any further and, with time, hydrolyzes to regenerate the
2
previously (Table 1, rows 5 and 6) for reasons discussed later.
RRDE data on control samples FeBz (Figure 1C) and
previously reported FeTPP show higher PROS during
electrochemical ORR (Figure S15, Table 1). Thus, the three
synthetic model complexes (FeQH , FePh, and FeQMe ) used
herein impress on the profound role of distal residues on
enhancing the selectivity of ORR by mononuclear iron
porphyrins despite having potentially very different mecha-
nisms for achieving the same.
ferric porphyrin (Scheme 5C). Note that the FeQMe complex
2
needs an additional thermodynamic driving force for catalyzing
an ORR under heterogeneous conditions in an aqueous
solvent, indicting that the oxy adduct produced in either
aqueous or nonaqueous medium is not capable of taking the
reaction forward toward ORR. The reaction of the reduced
2
2
3
7
FeQH complex has been recently reported in detail. The
2
III
−
FeQH complex forms a Fe −O · species that performs an
2
2
III
HAT from the pendant quinol generating a [Fe −OOH]SQ
Despite similar rates, the onset potential and its pH
dependence and selectivity under different ET rates of ORR
by these three complexes in buffered aqueous solutions
indicate that very different ORR mechanisms are at play for
these complexes. In the FePh complex, where there is a
species (Scheme 5A). This species is characterized by the EPR
III
18
2
and rR of the Fe −OOH unit with the help of O and H
III
substitution. The [Fe −OOH]SQ species then undergoes
IV
another HAT to form a [Fe O]Q species, which is
−
1
characterized by an Fe−O vibration at 750 cm and the
6
−1
pendant phenol, the ORR rate constant is (7.09 ± 1) × 10
symmetric quinone mode at 1520 cm . The kinetics of
−
1
−1
−
+
M
s , a 2e /1H PCET onset potential that overlays with
formation and decay of these intermediates are investigated
I
https://dx.doi.org/10.1021/jacs.0c10385
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
a
Scheme 5
a
(
A) Schematic representation of a mechanism of an ORR by FeQH involving a double HAT pathway. (B) Schematic representation of a
2
mechanism of an ORR by FePh involving a disproportionation reaction of ferric superoxide complex. (C) Schematic representation of an ORR by
FeQMe in organic and aqueous mediums. The red and blue colors are used to denote features observed in organic and aqueous mediums,
2
respectively. The features colored with both red and blue are proposed to occur in both the mediums.
2
marker band ν 1568 cm characteristic of low-spin FeIII
−1
using EPR, and H isotope effects are consistent with two
2
consecutive HATs (Scheme 5A). Thus, once reduced, the
ground state (Figures 7B and S16). The Fe−O vibration of this
III
−
III
−
−1
−1
Fe −O · species formed upon oxygen binding can complete
Fe −O ·
̇
species is at 582 cm , which shifts to 554 cm on
2
2
1
8
ORR without any additional electron from an external source,
O substitution (Figure 7A). Note that the Fe−O stretching
2
which is consistent with the fact that ORR proceeds at the
frequency is lower than that of FeQMe and similar to that of
2
III/II
Fe
potential under heterogeneous conditions in aqueous
the previously characterized ferric superoxide intermediate of
solvents.
FeQH and other Fe-porphyrins having H-bonded triazole
2
37,57,62
The reduced FePh complex reacts with O gas to form an
residues in the distal pocket.
A H-bonding interaction
2
III
−
III
−
Fe −O ·
̇
species characterized by a diamagnetic ground state
between the hydrogen-bond donor and Fe −O · leads to a
2
2
̇
62
III
−
̇
reduction in the Fe−O stretching frequency. The Fe −O ·
with no EPR signal. The rR data of this species show the
2
−
1
oxidation-state marker band ν at 1369 cm and the spin-state
species evolves into a mixture of high-spin and low-spin species
4
J
https://dx.doi.org/10.1021/jacs.0c10385
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
1
6
Figure 7. (A) Resonance Raman data of a superoxide complex of FePh. The blue spectrum represents the O adduct of FePh, and the red one
2
represents the 18O adduct of FePh. (B) Oxidation-state marker (ν ) of ferric superoxide and ferric hydroperoxide species of FePh. When the
2
4
III
−1
reduced FePh is oxygenated at −80 °C and allowed to react with O for 8 min it shows a maximum Fe low-spin signal (ν at 1367 cm )
corresponding to a ferric superoxide complex (green spectra). When the sample is kept at −80 °C for 10 min ferric superoxide begins to form ferric
hydroperoxide (red spectra), and after 12 min some amount of low-spin signal (corresponding to ferric superoxide) significantly decreases, and the
2
4
−
1
oxidation-state marker (1361 cm ) for the high-spin ferric complex increases (blue spectra), which confirms the disproportionation reaction of
ferric superoxide to ferric hydroperoxide and HS ferric complex. (C) EPR data of ferric hydroperoxide complex of FePh recorded at 4 K. The
III
III
intensity of the EPR signal of both Fe HS and Fe LS species increases simultaneously with time. The blue spectrum represents the EPR signal of
a ferric hydroperoxide complex incubated at −80 °C for 10 min, and the red spectrum represents the EPR signal of the same intermediate species
with an incubation period of 12 min in a −80 °C bath.
−
1
with the ν and ν values at 1361 and 1555 cm and 1369 and
1
Similar g-values are also obtained for the same intermediate of
FeQH in a previous report using the FeQH complex, which
are confirmed by an O / O and H/D isotope shift of Fe−O
4
1
2
−
568 cm , respectively. Consistently, the corresponding EPR
2
2
1
6
18
spectra shows the presence of a signal at g = 6 corresponding
to the high-spin ferric species and a low-spin ferric species with
g-values at 2.35, 2.26, and 1.90 (Figure 7C). These g-values are
consistent with those of 6C low-spin ferric hydroperoxide
Fe −OOH) complexes (Table 2) reported in the past.
2
2
37
and O−O vibrations in its resonance Raman spectra. While
for the ferric hydroperoxide intermediate of FePh the Fe−O
and O−O vibrations could not be identified in the rR data, the
−1
ν and the ν values are at 1369 and 1568 cm (Figures 7B
4
2
III
(
and S16), respectively, consistent with those of low-spin iron
porphyrin as observed in the EPR data. These observations
here are very similar to those reported for iron porphyrins
Table 2. g Values of Previously Reported Ferric Heme
III
−
Hydroperoxide Complexes
bearing pendant −COOH groups, where an Fe −O ·
̇
adduct
2
III
−
+
is reduced to an Fe −OOH via an intramolecular 1e /1H
PCET process, where the proton is delivered from the pendant
−COOH group and the electron is derived from unreacted Fe
species in solution. Considering the data obtained here and
name of ferric heme hydroperoxide complex
g values
ref
III
II
(
(
TMPIm)Fe −OOH
2.32, 2.19, 1.95
2.31, 2.19, 1.95
2.26, 2.14, 1.96
2.26, 2.16, 1.96
2.32, 2.19, 1.94
2.27, 2.16, 1.96
2.29, 2.17, 1.95
2.38, 2.20, 1.90
2.35, 2.26, 1.90
58
59
53
60
60
61
61
37
III
63
TMPIm)Fe −OOH
III
Fe −F TPP−OOH
HOO-met-hemoCD−OH
HOO-met-hemoCD-Im
Fe -TPP−OH-OOH
Fe OEP−OH-OOH
FeQH -MeOH-OOH
previous literature precedence, the FePh reacts with O to
20
2
III
−
̇
form a Fe −O · species initially, which is then transformed
2
III
to an Fe −OOH species via an intramolecular PCET, where
III
the proton is delivered from the pendant phenol and the
III
III
−
̇
electron is derived from another Fe −O complex present in
solution (Scheme 5B). The Fe high-spin species thus
generated by an oxidation of the Fe −O species results in
2
III
2
FePh-MeOH-OOH
this work
III
̇
−
2
K
https://dx.doi.org/10.1021/jacs.0c10385
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
III
an increase in the g = 6 signal (Figure 7C) in the EPR and new
an Fe −OOH species (Scheme 5B) even at −80 °C. The fact
III −
−
1
̇
2
ν and ν vibrations at 1361 and 1555 cm , respectively. The
that an Fe −O · can provide the electron needed for the
4
2
III
−
̇
rise of these high-spin ferric ν and ν is associated with the
reduction of another Fe −O · implies that, electrochemi-
4
2
2
concomitant decrease of the ν (Figure 7B) and ν (Figure
cally, this ET can be achieved at the same potential where the
4
2
−
1
III
̇
Fe −O · is formed, that is, at the potential where Fe is
− + III
−
II
S17) vibrations at 1369 and 1567 cm originating from the
2
III
−
̇
Fe −O · species. Curiously, the electron required to reduce
formed. Hence the 1e /1H PCET reduction of Fe −OH to
2
III
−
III
−
II
̇
̇
the Fe −O · is derived from another Fe −O · species.
Fe −OH under heterogeneous conditions will be rapidly
2
2
2
III
−
̇
Thus, one may expect that the ORR should occur at the same
followed by O binding and an ET to reduce Fe −O · to
2
2
III
II
III
−
+
potential where Fe is reduced to Fe , which is exactly what
happens under heterogeneous aqueous conditions. Eventually,
this intermediate decays to form a single high-spin species with
Fe −OOH resulting in a 2e /1H PCET as observed here.
Thus, the same mechanisms are operative under heteroge-
neous ORR in aqueous solutions and homogeneous O
reduction in an organic solution. The selectivity for 4e /4H
2
−
1
−
+
ν and ν vibrations at 1361 and 1555 cm , respectively
4
2
(
Figure S16B).
ORR is retained by the FePh complex even under slow ET
rates suggesting that there must be a mechanism to stabilize
the Fe −OOH species against competitive hydrolysis. This is
investigated using DFT calculations below.
4
.4. Reconciliation of the Reactivities in Aqueous and
III
Organic Solvents. The FeQMe complex with the methy-
lated quinol can bind O in an organic solution in its ferrous
2
2
state but cannot take the reaction forward. This is consistent
with the onset potential of an ORR in an aqueous medium
4.5. DFT Calculations. Geometry-optimized DFT calcu-
lations are used to address the remarkable selectivity of the
being lower than the E0 Fe . The Fe −O ·
III/II
III
−
species
−
+
FePh for 4e /4H even at slow ET rates from the electrode. As
discussed earlier, recent SERRS-RDE data show that the PROS
2
̇
immediately formed upon reduction of the iron to its ferrous
form is incapable of reacting further, unless an additional
electron is provided from the electrode at a higher
thermodynamic potential in an aqueous medium under a
III
are released via the hydrolysis of the Fe −OOH intermediate
produced during ORR. Stabilization of this intermediate or
activation of the O−O bond for cleavage results in an
−
+
heterogeneous condition (Scheme 5C). Anxolabeh
́
er
̀
e-Mallart
enhanced selectivity for 4e /4H ORR. Before the DFT
III
−
̇
calculations are used to understand these effects, it is used to
and Banse have reported that the reduction of the Fe −O ·
2
0
III/II
species requires ∼500 mV greater driving force than that of
calculate the relative E for the Fe
process to evaluate its
III
Fe porphyrin in organic solvents for both heme and nonheme
accuracy in predicting the properties of these systems.
5
3,64
III/II
−
+
ligands.
Hence, the onset potential of ORR for FeQMe2,
The Fe
process is 1e /1H PCET. Thus, the ΔG for the
which does not have an intrinsic proton or electron source, is
more cathodic than the Fe
intermediate species, which will be subsequently formed, must
be stabilized against hydrolysis to result in a lowering of the
PROS formation relative to unfunctionalized porphyrins.
Further insight into this is obtained using DFT calculations,
vide infra.
following reaction is calculated.
III/II
0
III
E . Note that the Fe −OOH
III
−
+
II
Fe −OH + e + H
→ Fe −OH
solv
2
A value of 4.43 eV is used as the value of the free electron,
and the free energy of solvation of 271 kcal/mol is assumed for
+
a H . Note that both these values are subject to the
experimental conditions like the reference electrode used.
0
Recently published results show that the FeQH reduces O
Hence the relative E values are used normalizing both the
2
2
0
all the way to water using two electrons from the iron and two
electrons and two protons from the pendant quinol group
DFT-calculated and experimental E of FeQMe to be 0 V.
2
The results show a qualitative agreement between the
IV
0
resulting in the formation of an Fe O quinone species in an
experimental and computed ΔE values (Table 3) justifying
37
organic solvent (Scheme 5A). Here, during a heterogeneous
ORR, the same mechanism can be operative, which should
Table 3. Experimental and Theoretical Relative Energies of
Ferric and Ferrous Porphyrins
III/II
0
result in an onset potential the same as the Fe
E as is
IV
observed in this investigation. The Fe O and quinone
formed during the electrochemical ORR can be reduced back
to the ferrous quinol state on the electrode at the potential
where ORR is observed (EORR at −200 mV; the QH to SQ
process is observed in the CV at +200 mV). Automatically,
such a mechanism will make the ORR intrinsically selective for
the 4e /4H reduction of O to water, and this selectivity
should be independent of the rate of ET from the electrode as
is evidenced from the lack of PROS even at very slow ET rates
III/II
0
III/II
ΔE Fe
(mV)
ΔE Fe
(mV)
complexes (H bond)
(calculated)
(observed)
FePh (−OH as H-bond
−520
−140
2
acceptor)
FeQMe (−OH as H-bond
0
+30
0
2
donor)
−
+
FeQH (−OH as H-bond
2
2
donor)
FeQH (−OH as H-bond
−500
−80
2
acceptor)
(
Table 1). Because of an alternate mechanism for the O−O
III
bond cleavage, stabilization of an Fe −OOH may not be
important here. Furthermore, the experimental pseudo-first-
order rate of ORR observed here is 394 s− (kcat from K-L
the use of the method and basis set for these calculations.
1
These calculations also offer an opportunity to investigate the
0
analysis × [O ]) for ORR, which, at RT, indicates a barrier of
∼150 mV difference in the E between FeQMe and FePh
2
2
1
3.9 kcal/mol, which is very close to the barrier calculated
complexes; a difference that is qualitatively reproduced in the
calculations as well.
from the previously reported experimentally measured rate-
III
determining second HAT (13.7 kcal/mol) by an Fe −OOH
species from an SQ in an organic solvent.
The optimized structures of the oxidized ferric state show
3
7
−
that the axial OH ligand to the iron is hydrogen-bonded to
III
The ferrous FePh complex reacts with O forming a Fe −
the pendant phenolic groups. While the Fe−OH is a hydrogen-
2
−
̇
O · species, which rapidly picks up a proton from the pendant
phenol and an electron from another ferric superoxide to form
bond acceptor in the FePh and FeQH complexes (Figure
2
2
8A,E) and the ortho hydroxy substituent on the pendant
L
https://dx.doi.org/10.1021/jacs.0c10385
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 8. Optimized structure of the ferric hydroxide and ferrous aquo complexes of all the three iron porphyrins. (A) Optimized structure of
III
Fe QH −OH with metal-bound hydroxide ligand acting as H-bond acceptor, (B) the optimized structure of the corresponding aquo complex, (C)
2
III
the optimized structure of Fe QH −OH with metal-bound hydroxide complex acting as H-bond donor, (D) the optimized structure of
2
III
corresponding ferrous aquo complex, (E) the optimized structure of Fe Ph−OH where the bound hydroxide is a H-bond acceptor, (F) the
corresponding ferrous aquo complex, (G) the optimized structure of FeQMe −OH where the bound hydroxide is a H-bond donor, and (H) the
2
optimized structure of its ferrous aquo complex. The relevant atoms are represented using ball and sticks (color codes are as follows, Fe = purple, O
=
red, N = blue, C = gray, and H = white), and the porphyrin backbone is represented using thin bonds for clarity.
phenyl ring is the donor (the other geometry for QH higher in
in all three complexes. The reason for the lower potential in
2
energy, Figure 8B), it is a hydrogen-bonding donor in FeQMe2
FePh and FeQH is the greater hydrogen-bonding stabilization
2
(
Figure 8G) where the meta methoxy substituent in the
of the Fe−OH acceptor ortho hydroxy donor orientation over
the Fe−OH donor meta hydroxy acceptor orientation. The
strong hydrogen bonding stabilizes the ferric state of the FePh
hydrogen-bond acceptor. In the reduced state where the axial
ligand is H O, the axial H O acts as a hydrogen-bonding donor
2
2
M
https://dx.doi.org/10.1021/jacs.0c10385
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
and FeQH complexes, where this can be achieved due to
ORR catalyzed by FeQMe proceeds at −130 mV more
III/II
2
2
ability of the ortho-phenolic −OH group to act as a H-bond
cathodic potential than its Fe
potential. This is because the
III
−
̇
donor but not in FeQMe . This observation is further
reduction of the Fe −O · species to produce water requires
2
2
0
supported by the fact that the calculated E for the model of
electrons to be supplied by the electrode to it, and the reaction
follows an ETPT pathway to form ferric hydroperoxide in an
FeQH , where the Fe−OH donates H-bond to a pendant
2
hydroquinone (Figure 8C), is ∼530 mV more positive than
aqueous medium (Scheme 5C). The FeQH complex follows a
2
III
that of the isomer, where the Fe −OH acts as a H-bond
consecutive HAT pathway during an oxygen reduction
III
−
acceptor (Figure 8A).
̇
reaction in organic medium involving an Fe −O · -QH
2
2
III
IV
Extending these calculations to obtain the hypothetical
species, Fe −OOH-SQ species, and Fe -oxo-Q intermediates
Scheme 5A), which have been characterized using UV−vis,
EPR, and resonance Raman spectroscopy previously. Thus,
of the four electrons required for the reduction of oxygen to
water, two are supplied by ferrous porphyrin, which is oxidized
to a ferryl species, and the rest are supplied by a distal
hydroquinone group, which is oxidized to quinone at the end
III
structures of an Fe −OOH species, likely to be involved in
(
−
3
7
ORR, one reaches very similar conclusions. The axial OOH
ligand is hydrogen-bonded in all these complexes (Figure
9
A,B). Abiding by the mechanism arrived at experimentally,
of the reaction. Here a 60 mV/pH shift in the E
which overlays the E measured at these pH values, implies
potential,
ORR
1
/2
+
−
that the potential determining step of ORR involves a 1H /1e
III
II
PCET reduction of Fe −OH to Fe OH as is expected based
2
on the previous results obtained in an organic medium.
Furthermore, the pseudo-first-order rate constant of an oxygen
reduction reaction in an aqueous medium is found to be 394
s , which is similar to the rate of the slow second HAT step
found in an organic medium. Thus, the consecutive HAT
−
1
Figure 9. DFT-optimized structure of hydroperoxide-bound (A)
pathway is likely operative during ORR by FeQH both in an
2
FePh and (B) FeQMe . The atoms are represented using ball and
aqueous medium as well as in an organic medium. When
2
sticks (color codes are as follows, Fe = purple, O = red, N = blue, C =
gray, and H = white).
III
̇
−
hydroquinone is replaced with phenol in FePh, the Fe −O ·
2
species formed in solution does not abstract a H atom from a
pendant phenol residue unlike FeQH . Rather in the case of
2
FePh a ferric superoxide complex is reduced to a low-spin
ferric hydroperoxide intermediate via a PCET mechanism,
where a proton comes from a pendant phenol, and the electron
III
the Fe −OOH in FePh is hydrogen-bonded to the pendant
III
phenolate. Similarly, the Fe −OOH is hydrogen-bonded to
the pendant SQ in FeQH and to the 2,5-dimethoxy phenyl
group in FeQMe . In all these cases the hydrogen bonding is
strong, making the species Fe −OOH resistant toward
hydrolysis-reducing PROS. This is consistent with the
2
III
−
̇
is donated by another Fe −O · species (Scheme 5B). During
2
2
III
a heterogeneous oxygen reduction in aqueous medium FePh
+
−
follows a 1H /2e PCET pathway, where the potential
III
−
+
determining step for ORR is the reduction of Fe −OH to
experimental data, where the selectivity for 4e /4H ORR is
retained even under very slow ET in sharp contrast to iron
porphyrins without such a preorganized hydrogen-bonding
network, where as much as 50−100% H O has been
II
Fe −H O followed by rapid oxygen binding to ferrous
2
porphyrin and a second electron transfer resulting in the
reduction of a ferric superoxide species to form a ferric
peroxide species. The fact that the ferric superoxide formed on
the electrode during ORR can be rapidly reduced to the ferric
hydroperoxide at the same potential is consistent with the fact
that an analogous ferric superoxide formed in solution can be
reduced by another ferric superoxide (ΔG ≈ 0) via PCET.
Thus, the FePh complex follows a PCET pathway to reduce
oxygen to water both in organic solvent as well as in aqueous
medium (Scheme 5B).
2
2
2
5,28
reported.
The DFT calculations further show that the H-
bonding distance between the bound −OOH ligand and
pendant phenolate oxygen is 1.63 Å, while in the case of
FeQMe the H-bonding distance between Fe bound −OOH
2
and phenyl −OMe group is 1.93 Å (Figure 9). Thus, the
hydrogen bonding of the bound hydroperoxide is much
stronger with the pendant phenolate than it is with the
methoxy group, which explains why the FeQMe complex
produces more PROS than the FePh complex when the
electron transfer from the electrode is slow. Note that the
2
The DFT calculations reveal that the HAT process for FePh
(
ΔG = +12.2 kcal/mol) is more energy demanding than that
III
FeQH complex is primed for a facile HAT by the Fe −OOH
for FeQH (ΔG = +3.9 kcal/mol). The bond dissociation
2
2
−
moiety, which would naturally bias the selectivity toward 4e /
energy (BDE) of the O−H bond of phenol and hydroquinone
+
4
H .
calculated using equation iii shows that the BDE of phenol
(
BDEO−H = 87 kcal/mol) is ∼6−7 kcal/mol higher compared
65
5
. DISCUSSION
to that of the hydroquinone (BDE
= 80 kcal/mol). This
O−H
is probably because of the higher redox potential of the RO·/
Three Fe porphyrins bearing pendant phenol, hydroquinone,
and a 2,5-dimethoxyphenyl ring are investigated. Under a
heterogeneous condition in aqueous medium all three
−
RO redox couple and lower pK of phenol compared to
a
hydroquinone. In aqueous solution in neutral pH the reduction
mononuclear Fe porphyrins reduce O electrochemically to
potential of phenol (∼0.8 V) is much higher than that of
2
65
H O with very high selectivity and with almost the same
hydroquinone (∼0.46 V). Furthermore, the pK of the
a
2
second-order rate constant when the catalysts are physiad-
sorbed on edge plane graphite electrode, where the ET to the
catalytic center is facile. Under heterogeneous conditions, the
phenol in an aqueous medium is 9.98, while that of
66
hydroquinone is 11.4. Please note that, while the absolute
0
values of pK and E will be different in these substituted
a
N
https://dx.doi.org/10.1021/jacs.0c10385
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
phenols and quinols relative to free phenols and quinol, the
relative values are likely to be the same.
0
BDEOH = C + 1.37pk + 23.06 ×E
H
(iii)
a
where C represents the enthalpic constant and includes the
H
free energy of formation of a hydrogen radical, free energy of
solvation of hydrogen radical in the solvent, and an entropy
term.
Thus the higher BDE of the O−H bond of phenol compared
to that of hydroquinone prevents the ferric superoxide of FePh
from abstracting a H atom from the pendant phenol group;
rather, it chooses a different mechanism involving a concerted
proton-coupled electron transfer pathway. The high endother-
micity of an HAT process in FePh relative to FeQH in the
2
superoxide adduct of FePh should increase the energy of the
corresponding transition state ∼12 kcal/mol higher than ∼13.4
kcal/mol calculated for the FeQH complex (Figure 10)
2
Figure 11. DFT-optimized structure of ferric superoxide and ferric
hydroperoxide complexes, which are proposed to form after PCET to
III
̇
−
the precursor Fe −O complexes of FeTPP (A, B), FeQH (C, D),
2
2
and FePh (E, F). The relevant atoms are represented using ball and
sticks (color codes are as follows, Fe = purple, O = red, N = blue, C =
gray and H = white), and the porphyrin backbone is represented as
wire for clarity.
18
phenol. The results presented here show that a suitably
placed phenol with a proton with a low pK can enable a facile
a
III
−
̇
PCET in an Fe −O · species as well.
2
Figure 10. Energy profile diagram for an HAT and PCET reaction
−
Finally, the FeTPP-O complex does not show PCET when
2
involved in the conversion from ferric heme superoxide to ferric heme
reacted with external PhOH despite having the same ΔG
PCET
hydroperoxide complex for FeQH and FePh.
2
−
as the FePh-O species. The optimized structures of the
2
III
−
III
̇
reactant Fe −O · and product Fe −OOH of FeTPP, FePh,
2
III
−
III
−
̇
̇
and FeQH show that the O−O bond length of Fe −O · is
III
Alternatively, for the Fe −O · adduct of the FePh
2
2
2
complex, which has a high barrier for HAT owing to the
high BDEOH of PhOH, PCET is a more favored reaction. A
shorter, while that for Fe −OOH is longer for FeTPP relative
III −
̇
2
to both FePh and FeQH (Table 4). The Fe −O · adducts
2
III
−
̇
PCET to an Fe −O · can be viewed as a reduction of this
of both FePh and FeQH have pendant hydroxyl groups, which
2
2
species followed by a proton transfer from the pendant phenol
or quinol groups. The overall energy of this process can be
expressed as
act as hydrogen-bond donor to the bound superoxide. The
polarization of the electron density on the superoxide by the
hydroxyl groups in FePh and FeQH lead to elongation of the
2
O−O bond (Table 4). Similarly, the deprotonated phenolate
ΔGPCET = ΔGET + ΔGPT
and quinolate oxygen act as hydrogen-bond acceptor to the
III
where ΔG represents the free energy of reduction of the
bound hydroperoxide of the Fe −OOH species. This
ET
III
−
̇
Fe −O · species by the reductant in solution, and ΔG
represents the free energy of proton transfer from the pendant
phenol/quinol to the Fe −O species formed. The ΔG is
almost 0 for both the complexes, and the optimized structure
hydrogen bonding leads to an increased electron density on
2
PT
the bound hydroperoxide leading to a shorter O−O distance in
III
2−
III
these hydrogen-bonded Fe −OOH species (Table 4). Overall,
2
ET
the hydrogen bonding between the pendant phenol/quinol in
III
III
−
̇
(
Figure 11) of the resultant Fe −OOH is almost identical for
the reactant Fe −O · species makes the O−O bond longer,
2
both FePh and FeQH . The only difference is in the heterolytic
whereas hydrogen bonding between the phenolate/quinolate
2
III
bond dissociation energies of the phenolic and the quinolic
OH groups. This is reflected in their pK values. The pK of
quinol is ∼1.42 unit greater than that of phenol, which
translates to 1.94 kcal/mol higher proton transfer energy for
with Fe −OOH in the product makes the O−O bond shorter,
−
which is responsible for the smaller displacement in the O−O
a
a
bond during the PCET reaction for FePh/FeQH compared to
2
FeTPP.
the FeQH complex relative to the FePh complex. Thus, the
The inner-sphere reorganization energy (λinner‑sphere) for an
2
III
−
2
̇
PCET to an Fe −O · species should be favorable in the
ET process can be expressed as Σk (x ) , where k is calculated
2
i
i
i
FePh complex by 1.94 kcal/mol relative to the FeQH
from the individual force constant of the bonds involved in the
2
complex. Recent results on synthetic heme/Cu systems suggest
that the H-bonding interaction with the acidic proton of a
phenol can help to polarize the O−O bond of a bridging
peroxide, where a proton and electron are transferred from the
ith normal mode both for the reactant and product molecules
using eqs va, vb, and vc, and x is the change of equilibrium
i
bond length between reactant and product along the ith
normal mode. The individual force constant (f) of any
O
https://dx.doi.org/10.1021/jacs.0c10385
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
−
1
Table 4. Calculated Bond Lengths (Å) and Stretching Frequencies (cm ) of the Reactant and Products That Are Supposed to
Be Involved in the PCET Process
−
ferric-OOH-RO−
ferric-O · -ROH
2
Fe−O bond
O−O bond
O−H bond
Fe−O bond
O−O bond
O−H bond
length
(Å)
ν
length
(Å)
ν
length
(Å)
ν
length
(Å)
ν
length
(Å)
ν
length
(Å)
ν
Fe−O
O−O
O−H
−1
Fe−O
O−O
O−H
−
1
−1
−1
−1
−1
complex
(cm
)
(cm
)
(cm
)
(cm
)
(cm
)
(cm
)
FeTPP
FeQH2
FePh
1.85
1.83
1.83
1.84
502
559
562
545
1.28
1.30
1.30
1.31
1244
0.97
0.98
0.98
0.98
3731
1.77
1.76
1.76
1.79
655
1.44
1.42
1.42
1.44
907
0.97
1.00
1.01
1.0
3672
1246
1244
1241
3476
3463
3527
661
674
622
948
944
902
3129
2987
3090
FePh-
3
H O
2
particular bond is calculated from its stretching frequency (ν)
and reduced mass (μ) using eq iv.
FeQH2
= kFe‐O(dFe‐Ohyd − dFe‐O
sup 2
λinner‐sphere
)
67,68
hyd
sup 2
+
k
(d
− dO‐O
)
O‐O O‐O
1
/2
hyd
ROH 2
ν = (1/(2πc)) × (f /μ)
+ kO‐H(dO‐H − dO‐H
)
hyd
2
hyd 2
=
=
3941.89 + 24 586.85 + 722.63
29.2 kJ/mol = 6.98 kcal/mol
f
= (2πc) × μ × ν
)
Fe‐O
(iv)
(va)
(vb)
(vc)
Fe‐O
hyd
sup
hyd
sup
k
k
k
= (fFe‐O × fFe‐O )/(f
+ fFe‐O
)
Fe‐O
Fe‐O
hyd 2
FePh
hyd
sup 2
λinner‐sphere
= kFe‐O(dFe‐O − dFe‐O )
2
sup 2
=
((2πc) × μ × ((ν
) × (νFe‐O ) )
hyd
sup 2
Fe‐O
+ kO‐O(dO‐O − dO‐O )
hyd 2
sup 2
/
((νFe‐O ) + (νFe‐O ) )
hyd
ROH 2
+ kO‐H(dO‐H − dO‐H
)
=
4390.92 + 25 483.93 + 1538.558
31.4 kJ/mol = 7.5 kcal/mol
hyd
sup
hyd
sup
= (fO‐O × fO‐O )/(f
+ fO‐O
)
O‐O
O‐O
hyd 2
=
2
sup 2
=
((2πc) × μ × ((ν
) × (νO‐O ) )
O‐O
sup 2
hyd 2
FeTPP
hyd
sup 2
/
((νO‐O ) + (νO‐O ) )
λinner‐sphere
= kFe‐O(dFe‐O
− dFe‐O
hyd
)
sup 2
+
k
(d
− dO‐O
)
O‐O O‐O
hyd
ROH
hyd
ROH
hyd
ROH 2
= (fO‐H
× fO‐H
)/(f
+ fO‐H
)
+ kO‐H(dO‐H − dO‐H
)
O‐H
O‐H
hyd 2
2
ROH 2
= 4485.61 + 39 536.21 + 54.70
=
((2πc) × μ × ((ν
) × (ν
) )
O‐H
O‐H
hyd 2
) + (νO‐H
ROH 2
=
44.07 kJ/mol = 10.53 kcal/mol
/
((νO‐H
) )
λinner‐sphereFePh‐H2O = kFe‐O(d
hyd
− dFe‐O )
sup 2
2
Fe‐O
λinner‐sphere = ∑ kixi
hyd
hyd
sup 2
+
k
(d
− dO‐O
)
O‐O O‐O
hyd
sup 2
hyd
sup 2
ROH 2
=
k
+
(d
− dFe‐O ) + k (d
− dO‐O
)
Fe‐O Fe‐O
O‐O O‐O
+ kO‐H(dO‐H − dO‐H
)
hyd
ROH 2
k
(d
− dO‐H
)
O‐H O‐H
=
=
1859.48 + 29 238.16 + 721.96
31.8 kJ/mol = 7.6 kcal/mol
(vi)
The computed λinner‑sphere for a PCET process suggests that
both FeQH and FePh have a comparable λ for the
−
The λinner‑sphere in FeTPP-O is thus 3 kcal/mol higher than
2
inner‑sphere
2
−
−
̇ ̇
in FePh-O · and FeQH −O · making the PCET
inner‑sphere
PCET process and that the values are sufficiently lower than
that of the FeTPP. The O−O bond is strong with a force
constant of 7.29 mdyne/Å for a superoxide and 3.87 mdyne/Å
for a hydroperoxide of FeTPP. As a result, the λinner‑sphere is
dominated by the contribution from the elongation of the O−
λ
2
2
2
substantially unfavorable in the former. These estimates
indicate a clear advantage of having hydrogen bonding in
III
−
III
̇
lowering the λinner‑sphere for the PCET from Fe −O · to Fe −
2
OOH. The H bonding between the distal superoxide and
hydroperoxide with distal residues may or may not be retained
in an aqueous medium, where these experiments are
performed. To validate this DFT optimization was performed
on ferric superoxide and hydroperoxide complexes of FePh
with three water molecules in the distal pocket (Figure S18).
The optimized geometries and calculated frequencies are very
similar to the parent FePh complex (Table 4, rows 3 and 4). In
particular, the critical differences in the O−O bond lengths
upon reduction are retained even when three explicit water
molecules are added. As a result, the inner-sphere reorganiza-
tion energy for the conversion of ferric-hydroperoxide
complexes from the ferric-superoxide complexes calculated
from the optimized structure and frequencies is found to be
III
−
III
̇
O both from Fe −O · to Fe −OOH. While the O−H bond
has a higher force constant, its contribution to the λ
2
is
inner‑sphere
negligible due to a smaller change in the O−H bond length
III
between the product Fe −OOH and the reactant PhOH/
QH (Table 4). Given that the λ
varies with the square
2
inner‑sphere
of the displacement, a lower displacement of the O−O bond
results in a lower λ in the FePh/FeQH complex
compared to FeTPP. Thus, the strong interaction with the
pendant phenol/hydroquinone is quintessential for lowering
the λinner‑sphere for PCET in FePh/FeQH relative to FeTPP by
3 kcal/mol making the process more favorable in the
complexes bearing hydrogen bonding in the distal site
promoted by the pendant groups.
inner‑sphere
2
2
∼
P
https://dx.doi.org/10.1021/jacs.0c10385
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
very similar to that of the FePh without external water
molecules in the distal pocket. This suggests that the effect of a
distal residue on the mechanism of conversion of ferric
hydroperoxide from a ferric superoxide complex may not alter
substantially with a change in the solvent from organic to
aqueous.
Abhijit Nayek − School of Chemical Science, Indian
Association for the Cultivation of Science, Kolkata 700032,
India; orcid.org/0000-0001-9981-4271
Somdatta Ghosh Dey − School of Chemical Science, Indian
Association for the Cultivation of Science, Kolkata 700032,
India; orcid.org/0000-0002-6142-2202
Complete contact information is available at:
https://pubs.acs.org/10.1021/jacs.0c10385
6
. SUMMARY
The electrocatalytic ORR by three structurally analogous iron
porphyrin complexes bearing phenol, quinol, and methylqui-
Notes
The authors declare no competing financial interest.
none groups and their reactivities in solution with O show
2
that the mechanism of O reduction can vary dramatically,
2
ACKNOWLEDGMENTS
The research is funded by Department of Science and
Technology (SERB) Grant No. EMR-0008063.
even though their apparent rates and selectivity for ORR are
similar under certain conditions. Probing the electrochemical
ORR under different pH values and electron transfer rates
reveals very different mechanisms of ORR at play. Simulta-
neous investigations of an O2 reaction in homogeneous
conditions help to further confirm the differences in the
mechanism. A pendant methylquinone group can restrain the
■
REFERENCES
■
(
1) Ferguson-Miller, S.; Babcock, G. T. Heme/Copper Terminal
Oxidases. Chem. Rev. 1996, 96, 2889−2908.
III
hydrolysis of an Fe −OOH group offering good selectivity for
(2) Proshlyakov, D. A.; Pressler, M. A.; Babcock, G. T. Dioxygen
activation and bond cleavage by mixed-valence cytochrome c oxidase.
Proc. Natl. Acad. Sci. U. S. A. 1998, 95, 8020.
−
+
4
e /4H ORR via an ETPT pathway only when the ET from
the electrode is fast. The low reduction potential of a quinol
allows the iron porphyrin complex with a pendant quinol to
(
3) Yoshikawa, S.; Shinzawa-Itoh, K.; Nakashima, R.; Yaono, R.;
III
−
Yamashita, E.; Inoue, N.; Yao, M.; Fei, M. J.; Libeu, C. P.; Mizushima,
T.; Yamaguchi, H.; Tomizaki, T.; Tsukihara, T. Redox-Coupled
Crystal Structural Changes in Bovine Heart Cytochrome c Oxidase.
Science 1998, 280, 1723.
(4) Proshlyakov, D. A.; Pressler, M. A.; DeMaso, C.; Leykam, J. F.;
DeWitt, D. L.; Babcock, G. T. Oxygen Activation and Reduction in
Respiration: Involvement of Redox-Active Tyrosine 244. Science 2000,
290, 1588.
̇
reduce O via two consecutive HATs to an Fe −O · adduct.
2
2
This pathway is inaccessible when the quinol is replaced by a
phenol due to the high reduction potential of the phenol.
Rather the low pK (greater acidity) of the pendant phenol
a
III
−
̇
allows ORR via a favorable PCET to this Fe −O · species.
The advantages of a pendant phenol or quinol are (a)
prearranged favorable orientation of the Fe −O · and the H
atom/proton source, (b) lowering of λ for the PCET from
Fe −O · to Fe −OOH, and (c) stabilization the Fe −
OOH intermediate by hydrogen bonding (evidenced by its
detection in organic solutions), which helps these complexes
2
III
−
̇
2
(5) Morgan, J. E.; Verkhovsky, M. I.; Wikstrom, M. Observation and
̈
Assignment of Peroxy and Ferryl Intermediates in the Reduction of
Dioxygen to Water by Cytochrome c Oxidase. Biochemistry 1996, 35,
III
−
III
III
̇
2
1
2235−12240.
−
+
(6) Yoshikawa, S.; Shimada, A. Reaction Mechanism of Cytochrome
retain more than 95% selectivity for 4e /4H ORR even under
c Oxidase. Chem. Rev. 2015, 115, 1936−1989.
7) Babcock, G. T. How oxygen is activated and reduced in
very slow ET flux from the electrode.
(
respiration. Proc. Natl. Acad. Sci. U. S. A. 1999, 96, 12971.
(8) Collman, J. P.; Boulatov, R.; Sunderland, C. J.; Fu, L. Functional
Analogues of Cytochrome c Oxidase, Myoglobin, and Hemoglobin.
Chem. Rev. 2004, 104, 561−588.
ASSOCIATED CONTENT
sı Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/jacs.0c10385.
■
*
(
9) Amanullah, S.; Singha, A.; Dey, A. Tailor made iron porphyrins
for investigating axial ligand and distal environment contributions to
1H NMR data of the ligands, ESI-MS of the ligands and
the complexes, additional spectroscopic data, and
optimized coordinates (PDF)
electronic structure and reactivity. Coord. Chem. Rev. 2019, 386, 183−
208.
(
10) Collman, J. P.; Rapta, M.; Broring, M.; Raptova, L.;
̈
Schwenninger, R.; Boitrel, B.; Fu, L.; L’Her, M. Close Structural
Analogues of the Cytochrome c Oxidase Fea3/CuB Center Show
Clean 4e- Electroreduction of O2 to H2O at Physiological pH. J. Am.
Chem. Soc. 1999, 121, 1387−1388.
X-ray crystallographic information (CIF)
(11) Collman, J. P.; Sunderland, C. J.; Berg, K. E.; Vance, M. A.;
AUTHOR INFORMATION
■
Solomon, E. I. Spectroscopic Evidence for a Heme−Superoxide/
Cu(I) Intermediate in a Functional Model of Cytochrome c Oxidase.
J. Am. Chem. Soc. 2003, 125, 6648−6649.
Corresponding Author
Abhishek Dey − School of Chemical Science, Indian
Association for the Cultivation of Science, Kolkata 700032,
India; orcid.org/0000-0002-9166-3349; Email: icad@
iacs.res.in
(
12) Sasaki, T.; Nakamura, N.; Naruta, Y. Formation and
Spectroscopic Characterization of the PeroxoFeIII-CuII Complex. A
Modeling Reaction of the Heme-Cu Site in Cytochrome c Oxidase.
Chem. Lett. 1998, 27, 351−352.
(
13) Chufan, E. E.; Karlin, K. D. An Iron−Peroxo Porphyrin
́
Authors
Complex: New Synthesis and Reactivity Toward a Cu(II) Complex
Asmita Singha − School of Chemical Science, Indian
Association for the Cultivation of Science, Kolkata 700032,
India; orcid.org/0000-0001-5512-253X
Arnab Mondal − School of Chemical Science, Indian
Association for the Cultivation of Science, Kolkata 700032,
India
Giving a Heme−Peroxo−Copper Adduct. J. Am. Chem. Soc. 2003,
1
(
25, 16160−16161.
14) Chishiro, T.; Shimazaki, Y.; Tani, F.; Tachi, Y.; Naruta, Y.;
Karasawa, S.; Hayami, S.; Maeda, Y. Isolation and Crystal Structure of
a Peroxo-Bridged Heme−Copper Complex. Angew. Chem., Int. Ed.
2003, 42, 2788−2791.
Q
https://dx.doi.org/10.1021/jacs.0c10385
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(
15) Ghiladi, R. A.; Hatwell, K. R.; Karlin, K. D.; Huang, H.-w.;
Moenne-Loccoz, P.; Krebs, C.; Huynh, B. H.; Marzilli, L. A.; Cotter,
R. J.; Kaderli, S.; Zuberbuhler, A. D. Dioxygen Reactivity of
c oxidase as an electrocatalyst for oxygen reduction. Nat. Commun.
2015, 6, 8467.
̈
̈
(31) Chatterjee, S.; Sengupta, K.; Hematian, S.; Karlin, K. D.; Dey,
A. Electrocatalytic O2-Reduction by Synthetic Cytochrome c Oxidase
Mimics: Identification of a “Bridging Peroxo” Intermediate Involved
in Facile 4e−/4H+ O2-Reduction. J. Am. Chem. Soc. 2015, 137,
12897−12905.
Mononuclear Heme and Copper Components Yielding A High-
Spin Heme−Peroxo−Cu Complex. J. Am. Chem. Soc. 2001, 123,
6
(
183−6184.
16) Ghiladi, R. A.; Chufan, E. E.; del Río, D.; Solomon, E. I.; Krebs,
́
C.; Huynh, B. H.; et al. Further Insights into the Spectroscopic
Properties, Electronic Structure, and Kinetics of Formation of the
Heme−Peroxo−Copper Complex [(F8TPP)FeIII−(O22-)−CuII-
TMPA)]+. Inorg. Chem. 2007, 46, 3889−3902.
(
(32) Sengupta, K.; Chatterjee, S.; Dey, A. Catalytic H2O2
Disproportionation and Electrocatalytic O2 Reduction by a Func-
tional Mimic of Heme Catalase: Direct Observation of Compound 0
and Compound I in Situ. ACS Catal. 2016, 6, 1382−1388.
(
17) Collman, J. P.; Decreau, R. A.; Yan, Y.; Yoon, J.; Solomon, E. I.
́
(33) Mukherjee, S.; Bandyopadhyay, S.; Chatterjee, S.; Dey, A.
Intramolecular Single-Turnover Reaction in a Cytochrome c Oxidase
Electrocatalytic O2 reduction by a monolayer of hemin: the role of
pKa of distal and proximal oxygen of a FeIII−OOH species in
determining reactivity. Chem. Commun. 2014, 50, 12304−12307.
Model Bearing a Tyr244 Mimic. J. Am. Chem. Soc. 2007, 129, 5794−
5
(
795.
18) Schaefer, A. W.; Kieber-Emmons, M. T.; Adam, S. M.; Karlin,
(34) Chatterjee, S.; Sengupta, K.; Samanta, S.; Das, P. K.; Dey, A.
K. D.; Solomon, E. I. Phenol-Induced O−O Bond Cleavage in a Low-
Spin Heme−Peroxo−Copper Complex: Implications for O2 Reduc-
tion in Heme−Copper Oxidases. J. Am. Chem. Soc. 2017, 139, 7958−
Concerted Proton−Electron Transfer in Electrocatalytic O Reduc-
2
tion by Iron Porphyrin Complexes: Axial Ligands Tuning H/D
Isotope Effect. Inorg. Chem. 2015, 54, 2383−2392.
7
(
973.
(35) Bhunia, S.; Rana, A.; Roy, P.; Martin, D. J.; Pegis, M. L.; Roy,
19) Collman, J. P.; Devaraj, N. K.; Decreau, R. A.; Yang, Y.; Yan, Y.-
́
B.; Dey, A. Rational Design of Mononuclear Iron Porphyrins for
L.; Ebina, W.; Eberspacher, T. A.; Chidsey, C. E. D. A Cytochrome c
Oxidase Model Catalyzes Oxygen to Water Reduction Under Rate-
Limiting Electron Flux. Science 2007, 315, 1565.
−
+
Facile and Selective 4e /4H O Reduction: Activation of O−O Bond
2
by 2nd Sphere Hydrogen Bonding. J. Am. Chem. Soc. 2018, 140,
9
(
444−9457.
(20) Halime, Z.; Kieber-Emmons, M. T.; Qayyum, M. F.; Mondal,
36) Bhunia, S.; Rana, A.; Dey, S. G.; Ivancich, A.; Dey, A. A
B.; Gandhi, T.; Puiu, S. C.; Chufan, E. E.; Sarjeant, A. A. N.;
́
designed second-sphere hydrogen-bond interaction that critically
influences the O−O bond activation for heterolytic cleavage in ferric
iron−porphyrin complexes. Chem. Sci. 2020, 11, 2681.
Hodgson, K. O.; Hedman, B.; Solomon, E. I.; Karlin, K. D. Heme−
Copper−Dioxygen Complexes: Toward Understanding Ligand-
Environmental Effects on the Coordination Geometry, Electronic
Structure, and Reactivity. Inorg. Chem. 2010, 49, 3629−3645.
(37) Singha, A.; Dey, A. Hydrogen atom abstraction by synthetic
heme ferric superoxide and hydroperoxide species. Chem. Commun.
019, 55, 5591−5594.
38) Kim, H.; Rogler, P. J.; Sharma, S. K.; Schaefer, A. W.; Solomon,
(21) Kieber-Emmons, M. T.; Li, Y.; Halime, Z.; Karlin, K. D.;
2
(
Solomon, E. I. Electronic Structure of a Low-Spin Heme/Cu Peroxide
Complex: Spin-State and Spin-Topology Contributions to Reactivity.
Inorg. Chem. 2011, 50, 11777−11786.
E. I.; Karlin, K. D. Heme-FeIII Superoxide, Peroxide and Hydro-
peroxide Thermodynamic Relationships: FeIII-O2•− Complex H-
Atom Abstraction Reactivity. J. Am. Chem. Soc. 2020, 142, 3104−
(
22) Kieber-Emmons, M. T.; Qayyum, M. F.; Li, Y.; Halime, Z.;
Hodgson, K. O.; Hedman, B.; Karlin, K. D.; Solomon, E. I.
Spectroscopic Elucidation of a New Heme/Copper Dioxygen
Structure Type: Implications for O···O Bond Rupture in Cytochrome
c Oxidase. Angew. Chem., Int. Ed. 2012, 51, 168−172.
3
116.
39) Mukherjee, M.; Dey, A. Catalytic C−H Bond Oxidation Using
Dioxygen by Analogues of Heme Superoxide. Inorg. Chem. 2020, 59,
415−7425.
40) Mondal, P.; Wijeratne, G. B. Modeling Tryptophan/Indole-
(
7
(
(
23) Kim, H.; Sharma, S. K.; Schaefer, A. W.; Solomon, E. I.; Karlin,
K. D. Heme−Cu Binucleating Ligand Supports Heme/O2 and FeII−
CuI/O2 Reactivity Providing High- and Low-Spin FeIII−Peroxo−
CuII Complexes. Inorg. Chem. 2019, 58, 15423−15432.
amine 2,3-Dioxygenase with Heme Superoxide Mimics: Is Ferryl the
Key Intermediate? J. Am. Chem. Soc. 2020, 142, 1846−1856.
(41) Sarewicz, M.; Osyczka, A. Electronic Connection Between the
(
24) Schaefer, A. W.; Ehudin, M. A.; Quist, D. A.; Tang, J. A.; Karlin,
Quinone and Cytochrome c Redox Pools and Its Role in Regulation
of Mitochondrial Electron Transport and Redox Signaling. Physiol.
Rev. 2015, 95, 219−243.
K. D.; Solomon, E. I. Spin Interconversion of Heme-Peroxo-Copper
Complexes Facilitated by Intramolecular Hydrogen-Bonding Inter-
actions. J. Am. Chem. Soc. 2019, 141, 4936−4951.
(42) Cramer, W. A.; Hasan, S. S.; Yamashita, E. The Q cycle of
(
25) Rigsby, M. L.; Wasylenko, D. J.; Pegis, M. L.; Mayer, J. M.
cytochrome bc complexes: A structure perspective. Biochim. Biophys.
Acta, Bioenerg. 2011, 1807, 788−802.
Medium Effects Are as Important as Catalyst Design for Selectivity in
Electrocatalytic Oxygen Reduction by Iron−Porphyrin Complexes. J.
Am. Chem. Soc. 2015, 137, 4296−4299.
(43) Solomon, E. I.; Heppner, D. E.; Johnston, E. M.; Ginsbach, J.
W.; Cirera, J.; Qayyum, M.; Kieber-Emmons, M. T.; Kjaergaard, C.
H.; Hadt, R. G.; Tian, L. Copper Active Sites in Biology. Chem. Rev.
(
26) McGuire, R., Jr; Dogutan, D. K.; Teets, T. S.; Suntivich, J.;
Shao-Horn, Y.; Nocera, D. G. Oxygen reduction reactivity of
cobalt(ii) hangman porphyrins. Chemical Science 2010, 1, 411−414.
(
2
(
014, 114, 3659−3853.
44) Anson, C. W.; Ghosh, S.; Hammes-Schiffer, S.; Stahl, S. S.
27) Samanta, S.; Mittra, K.; Sengupta, K.; Chatterjee, S.; Dey, A.
Co(salophen)-Catalyzed Aerobic Oxidation of p-Hydroquinone:
Mechanism and Implications for Aerobic Oxidation Catalysis. J. Am.
Chem. Soc. 2016, 138, 4186−4193.
(45) Becke, A. D. Density-functional exchange-energy approxima-
tion with correct asymptotic behavior. Phys. Rev. A: At., Mol., Opt.
Phys. 1988, 38, 3098−3100.
(46) Becke, A. D. Density-functional thermochemistry. III. The role
of exact exchange. J. Chem. Phys. 1993, 98, 5648−5652.
(47) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. Self-
consistent molecular orbital methods. XX. A basis set for correlated
wave functions. J. Chem. Phys. 1980, 72, 650−654.
(48) Yan, J. J.; Kroll, T.; Baker, M. L.; Wilson, S. A.; Decreau, R.;
́
Lundberg, M.; Sokaras, D.; Glatzel, P.; Hedman, B.; Hodgson, K. O.;
Solomon, E. I. Resonant inelastic X-ray scattering determination of
Second Sphere Control of Redox Catalysis: Selective Reduction of O2
to O2− or H2O by an Iron Porphyrin Catalyst. Inorg. Chem. 2013, 52,
1
(
443−1453.
28) Collman, J. P.; Decrea
́
u, R. A.; Lin, H.; Hosseini, A.; Yang, Y.;
Dey, A.; Eberspacher, T. A. Role of a distal pocket in the catalytic O2
reduction by cytochrome c oxidase models immobilized on
interdigitated array electrodes. Proc. Natl. Acad. Sci. U. S. A. 2009,
106, 7320.
(
29) Sengupta, K.; Chatterjee, S.; Samanta, S.; Dey, A. Direct
observation of intermediates formed during steady-state electro-
catalytic O2 reduction by iron porphyrins. Proc. Natl. Acad. Sci. U. S.
A. 2013, 110, 8431.
(
30) Mukherjee, S.; Mukherjee, A.; Bhagi-Damodaran, A.;
Mukherjee, M.; Lu, Y.; Dey, A. A biosynthetic model of cytochrome
R
https://dx.doi.org/10.1021/jacs.0c10385
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
the electronic structure of oxyhemoglobin and its model complex.
Proc. Natl. Acad. Sci. U. S. A. 2019, 116, 2854.
(65) Warren, J. J.; Tronic, T. A.; Mayer, J. M. Thermochemistry of
Proton-Coupled Electron Transfer Reagents and its Implications.
Chem. Rev. 2010, 110, 6961−7001.
(49) Han Du, W.-G.; Noodleman, L. Broken Symmetry DFT
Calculations/Analysis for Oxidized and Reduced Dinuclear Center in
Cytochrome c Oxidase: Relating Structures, Protonation States,
(66) Li, C.; Hoffman, M. Z. One-Electron Redox Potentials of
Phenols in Aqueous Solution. J. Phys. Chem. B 1999, 103, 6653−6656.
(67) Hatcher, E.; Soudackov, A. V.; Hammes-Schiffer, S. Proton-
Coupled Electron Transfer in Soybean Lipoxygenase. J. Am. Chem.
Soc. 2004, 126, 5763−5775.
Energies, and Mo
Inorg. Chem. 2015, 54, 7272−7290.
50) Du, W.-G. H.; Gotz, A. W.; Noodleman, L. DFT Fea3−O/O−
̈
̈
ssbauer Properties in ba3 Thermus thermophilus.
(
(
68) Zhou, Z.; Khan, S. U. M. Inner-sphere reorganization energy of
ions in solution: a molecular orbital calculation. J. Phys. Chem. 1989,
3, 5292−5295.
O Vibrational Frequency Calculations over Catalytic Reaction Cycle
States in the Dinuclear Center of Cytochrome c Oxidase. Inorg. Chem.
9
2
(
019, 58, 13933−13944.
51) Noodleman, L.; Case, D. A. Density-Functional Theory of Spin
Polarization and Spin Coupling in IronSulfur Clusters. In Advances
in Inorganic Chemistry; Academic Press, 1992; Vol. 38; pp 423−470.
DOI: 10.1016/S0898-8838(08)60070-7
(
52) Heinze, J.; Bard, A. J.; Faulkner, L. F. Electrochemical
Methods−Fundamentals and Applications. Ber. Bunsen-Ges. 1981,
5, 1085−1086.
53) Oliveira, R.; Zouari, W.; Herrero, C.; Banse, F.; Scho
Fave, C.; Anxolabehere-Mallart, E. Characterization and Subsequent
8
(
̈
llhorn, B.;
́
̀
Reactivity of an Fe-Peroxo Porphyrin Generated by Electrochemical
Reductive Activation of O2. Inorg. Chem. 2016, 55, 12204−12210.
(
54) Mittra, K.; Chatterjee, S.; Samanta, S.; Dey, A. 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.
Inorg. Chem. 2013, 52, 14317−14325.
(55) Chatterjee, S.; Sengupta, K.; Mondal, B.; Dey, S.; Dey, A.
Factors Determining the Rate and Selectivity of 4e−/4H+ Electro-
catalytic Reduction of Dioxygen by Iron Porphyrin Complexes. Acc.
Chem. Res. 2017, 50, 1744−1753.
(56) Chidsey, C. E. D. Free Energy and Temperature Dependence
of Electron Transfer at the Metal-Electrolyte Interface. Science 1991,
51, 919.
57) Mittra, K.; Chatterjee, S.; Samanta, S.; Sengupta, K.;
2
(
Bhattacharjee, H.; Dey, A. A hydrogen bond scaffold supported
synthetic heme FeIII−O2− adduct. Chem. Commun. 2012, 48,
1
(
0535−10537.
58) Liu, J.-G.; Shimizu, Y.; Ohta, T.; Naruta, Y. Formation of an
End-On Ferric Peroxo Intermediate upon One-Electron Reduction of
a Ferric Superoxo Heme. J. Am. Chem. Soc. 2010, 132, 3672−3673.
(59) Liu, J.-G.; Ohta, T.; Yamaguchi, S.; Ogura, T.; Sakamoto, S.;
Maeda, Y.; Naruta, Y. Spectroscopic Characterization of a Hydro-
peroxo−Heme Intermediate: Conversion of a Side-On Peroxo to an
End-On Hydroperoxo Complex. Angew. Chem., Int. Ed. 2009, 48,
9
(
262−9267.
60) Kitagishi, H.; Tamaki, M.; Ueda, T.; Hirota, S.; Ohta, T.;
Naruta, Y.; Kano, K. Oxoferryl Porphyrin/Hydrogen Peroxide System
Whose Behavior is Equivalent to Hydroperoxoferric Porphyrin. J. Am.
Chem. Soc. 2010, 132, 16730−16732.
(61) Tajima, K.; Shigematsu, M.; Jinno, J.; Ishizu, K.; Ohya-
Nishiguchi, H. Generation of FeOEP-hydrogen peroxide complex
OEP = octaethylporphyrinato) by reduction of FeOEP−O2 with
ascorbic acid sodium salt. J. Chem. Soc., Chem. Commun. 1990, 144−
45.
62) Singha, A.; Mittra, K.; Dey, A. Effect of hydrogen bonding on
(
1
(
innocent and non-innocent axial ligands bound to iron porphyrins.
Dalton Transactions 2019, 48, 7179−7186.
(63) Nagaraju, P.; Ohta, T.; Liu, J.-G.; Ogura, T.; Naruta, Y. The
secondary coordination sphere controlled reactivity of a ferric-
superoxo heme: unexpected conversion to a ferric hydroperoxo
intermediate by reaction with a high-spin ferrous heme. Chem.
Commun. 2016, 52, 7213−7216.
(64) Segaud, N.; Anxolabehere-Mallart, E.; Senechal-David, K.;
́ ́ ̀ ́ ́
Acosta-Rueda, L.; Robert, M.; Banse, F. Electrochemical study of a
nonheme Fe(ii) complex in the presence of dioxygen. Insights into
the reductive activation of O2 at Fe(ii) centers. Chemical Science
2015, 6, 639−647.
S
https://dx.doi.org/10.1021/jacs.0c10385
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX