A Minimalist Iron Porphyrin Which Can Catalyze Both Peroxidation and Oxygen Reduction Reaction

Abstract

Heme enzymes catalyze several key reactions in nature. O₂ reduction and peroxidation are two such reactions crucial for respiration and oxidation of organic and inorganic substrates in nature, respectively. Over the last several decades, there has been a focused effort to generate small-molecule analogues which mimic these reactions. Small-molecule mimics of these metallo-enzymes may find potential use as an oxygen reduction reaction (ORR) catalyst in a fuel cell and as a catalyst for decontamination of wastewater in addition to providing deeper insights into the reactivity of the enzyme they mimic. An iron porphyrin with a pendant imidazole group appears to catalyze both these reactions quite efficiently. On the one hand, the pendant imidazole group provides a binding site for both H₂O₂ and the substrate, which results in efficient peroxidase catalysis that shows an enzyme-like “ping-pong” mechanism which is extremely rare in a small molecule. On the other hand, the pendant imidazole aids stabilization of intermediates during the ORR and facilitates O–O bond cleavage, resulting in fast catalysis with a high selectivity for 4e^–/4H⁺ ORR under homogeneous as well as heterogeneous conditions. The reactive species produced during the ORR can oxidize organic substrates, acting like an oxygenase.

Introduction

Oxygen and hydrogen peroxide activation are key reactions in nature involved in respiration and biosynthesis across different species. Oxygen reduction reaction (ORR) is the final step of respiration where O₂ is selectively reduced by 4e^–/4H⁺ to H₂O and is coupled to the synthesis of ATP in a process referred to as oxidative phosphorylation. , Cytochrome c oxidase has a heme/Cu active site (Figure A) where heme a is tethered to the protein via an iron histidine bond and the Cu_B is bound to three histidine residues, one of which is covalently bound to a tyrosine residue by a post-translational modification. − The tyrosine residue donates an electron to the ORR during catalysis and also participates in vectorial proton transfer. ,, Peroxidases (Figure B) are heme-based enzymes where heme b is tethered to the protein via an iron histidine bond. In addition, there are highly conserved residues like histidine and arginine residues in the distal side which help in peroxide activation and substrate binding. − In both these proteins, distal second sphere residues play an important role in catalysis and proton transfer, which is key to both these reactions and seem to be managed efficiently in these active sites.

1.

(A) Active site of cytochrome c oxidase (pdb id: 2Y69) and the reaction it catalyzes and (B) active site of peroxidase (pdb id: 1DCC) and the reaction it catalyzes.

Over the last several decades, the scientific community has been trying to create artificial mimics of these metallo-enzymes to access these reactivities outside of a protein matrix. − Synthetic iron porphyrins and biochemical models have been envisaged to catalyze the 4e^–/4H⁺ reduction of O₂ to H₂O which not only is crucial for generating ORR catalysts for fuel cells but also aids the understanding of the roles of the different elements present in the active site of CcO. − Nocera and co-workers, in particular, developed synthetic systems to investigate proton-coupled electron transfer in O₂ reduction with hangman porphyrins. − Remarkable developments were achieved, including rates above that of the natural enzyme, and even catalytic oxidation of cytochrome c by a synthetic iron porphyrin could be achieved. − Similarly, peroxidase activity has been realized in synthetic and biochemical systems. − The design of these catalysts is rather cumbersome, and realizing these reactions in a minimal iron porphyrin will be ideal.

Mechanistic investigations are crucial for distilling the necessary functional elements of a catalytic active site to envisage a simple, yet effective, analogue that can catalyze these reactions. Efficient peroxidase activity requires facile heterolytic cleavage of the O–O bond of a Fe^III–OOH species, formed upon the reaction of H₂O₂ with ferric heme, to generate a highly oxidizing species Compound I which is a Fe^IVO unit bound to a porphyrin cation radical unit (P⁺Fe^IVO). This protonation is facilitated by the second sphere histidine residue in the enzyme and biochemical models. ,,− In both synthetic and biochemical models of cytochrome c oxidase, in situ surface-enhanced resonance Raman spectroscopy indicated that the rate-limiting step is the cleavage of the O–O bond of a heme peroxide species, and the site of protonation of this Fe^III–OOH determined the selectivity of the process. ,− These discoveries allowed the realizations of small iron porphyrins with proton transfer residues in the second sphere which were shown to result in facile O–O bond cleavage of Fe^III–OOH species and rapid and selective 4e^–/4H⁺ reduction. − This approach was subsequently extended successfully to Cu and nonheme iron systems involved in O₂ activation. − Given the dominating role of proton transfer to both peroxidase activity and selective oxygen reduction reaction by iron porphyrins, it is possible to conceive a simple minimalistic iron porphyrin catalyst that can facilitate both these reactions efficiently.

In this article, a simple iron porphyrin with a single pendant imidazole group in the second sphere is reported to catalyze both peroxidase activity and ORR efficiently. This molecular catalyst exhibits enzyme-like kinetics implying a “ping-pong” mechanism at play with catalytic parameters comparable to the best peptide-based biochemical catalysts. This catalyst is found to rapidly facilitate the heterolysis of the O–O bond of H₂O₂ to generate a Compound I like oxidant. The rapid cleavage of the peroxide bond also allows this compound to catalyze selective 4e^–/4H⁺ ORR under physiological conditions as well, with rates of orders of magnitude higher than simple iron porphyrins.

Experimental Details

Methods and Materials

All reagents and solvents were obtained from Merck and used without further purification unless otherwise noted. Tetrahydrofuran (THF) was distilled from Na/benzophenone under an Ar atmosphere. Dimethylformamide (DMF) was distilled under an Ar atmosphere. The other degassed solvents were also treated by the freeze–thaw method. Preparation and handling of air sensitive materials were carried out under Ar in an MBraun glovebox. UV–vis experiments were carried out using an Agilent Cary 3500 spectrophotometer. A 1 cm path length quartz cell cuvette was used to perform all of the UV–vis experiments. ¹H spectra were recorded on Bruker DRX 400 and 300 MHz NMR spectrometers. Elemental analyses were performed on an Elementar Unicube CHN analyzer. CW electron paramagnetic resonance (EPR) spectra were collected using an X-band Bruker E580 spectrometer. Electrospray ionization (ESI) mass spectra were recorded with a Waters QTOF Micro YA263 instrument.

Heterogeneous Electrochemical Experiment

Cyclic voltammograms are recorded using a CH instrument potentiostat model 710E. Ag/AgCl (saturated KCl) was purchased from CH Instruments. Pt wire was used as the counter electrode. All cyclic voltammetry (CV) experiments were done in pH 7 buffer containing 100 mM Na₂HPO₄·2H₂O and 100 mM KPF₆ (supporting electrolyte).

Rotating Disk Electrochemistry

Rotating disk electrochemistry (RDE) measurements were performed on a CHI 710E bipotentiostat with a Pine Instruments modulated speed rotor fitted with an E6 series change-disc tip. The graphite surface was cleaned by uniformly polishing it on silicon carbide grinding paper. The complex was physiadsorbed on the disk. The RDE experiment was carried out by measuring LSV at 100 mV/s scan rate at different rotation rates using Ag/AgCl (saturated KCl) reference and Pt counter electrodes. The complex was physiadsorbed by the following procedureA 50 μL portion of catalyst from a 1 mM solution of the respective complex in chloroform (CHCl₃) is deposited on a freshly cleaned edge plane graphite (EPG) electrode mounted on an RDE setup. After evaporation of the solvent, the surface was rinsed with CHCl₃ and sonicated in ethanol and thoroughly dried with N₂ gas. Finally, the modified electrodes were washed with Milli-Q water before being used for electrochemical experiments. Kinetics of ORR is calculated using Koutecky–Levich (K–L) analysis. In the RDE experiment, the ORR current increases with increasing rotation following the K–L equation:

$$i^{-1}=i_{\mathrm{K}}(E{)}^{-1}+i_{\mathrm{L}}^{-1}$$

where i _K(E) is the potential dependent kinetic current, i _K(E)= nFA[O₂]k _catΓ_catalyst, i _L is the Levich current, i _L= 0.62 nFA[O₂]­(D _O2 )^2/3ω^1/2υ^–1/6, n = the number of electrons transferred to the substrate, F = Faraday constant, A = macroscopic area of the disc (0.096 cm²), [O₂] = concentration of O₂ in an air-saturated buffer (0.26 mM) at 25 °C, k _cat = second-order rate of catalytic O₂ reduction, Γ_catalyst = catalyst concentration in mol/cm², D _O2 = diffusion coefficient of O₂ (1.8 × 10^–5 cm² s ^–1) at 25 °C, ω = angular velocity of the disc, and υ = kinematic viscosity of the solution (0.009 cm² s ^–1) at 25 °C.

The plot of I ^–1 at multiple rotation rates with the inverse square root of the angular rotation rate (ω^–1/2) is linear. The number of electrons (n) involved in the O₂ reduction reaction by a catalyst can be calculated from the slope and rate of catalysis (k _cat) from the intercept of this linear plot.

Rotating Ring Disk Electrochemistry

Partially reduced oxygen species detection and calculation. The Pt ring and the Au disc were both polished by alumina powder (grit sizes: 1, 0.3, and 0.05 μm) and electrochemically cleaned and inserted into the rotating ring disk electrochemistry (RRDE) tip, which was then mounted on the rotor and immersed into a cylindrical glass cell equipped with Ag/AgCl reference and Pt counter electrodes. The ratio of the 2e^–/2H⁺ current (corrected for collection efficiency) at the ring and the catalytic current at the disk is expressed as reactive oxygen species, and it provides an in situ measure of the 2e^–/2H⁺ reduction side reaction. A 20 ± 2% CE was generally recorded during these experiments. The potential at which the ring was held during the collection experiments at pH 7 for detecting H₂O₂ was obtained from the literature.

Homogeneous Electrochemical Measurements

CV measurements of acetonitrile solutions containing FeTPPMIm and FeTPP were collected on CH Instruments (CHI) model 710E/700E potentiostats using a three-electrode configuration. Glassy carbon (2 mm, CHI), platinum wire, and silver wire pseudoreference electrodes were used as the working, auxiliary, and reference electrodes, respectively. The nonaqueous Ag/AgCl pseudoreference electrode was prepared by sanding a silver wire and immersing it in a capillary containing 0.1 M [nBu₄N]­[ClO₄] in acetonitrile solution, which was separated from the bulk solution by a Vycor tip. An internal standard ferrocene was added in the electrolytic solution, and the potentials were reported with respect to the Fc⁺/Fc⁰ couple. All CVs were corrected for uncompensated resistance (< 2 mV shift in the current–potential response under catalytic conditions) prior to analysis. Anaerobic measurements were performed inside a glovebox under a N₂ atmosphere, and the ORR data were recorded under aerobic conditions in an O₂ saturated acetonitrile medium. The overpotentials were estimated from the catalytic waves at the potentials having a catalytic current of 15 μA for all of the complexes. The homogeneous CV experiments were carried out in an acetonitrile solution of 0.5 mM catalyst with 100 mM TBAP (supporting electrolyte) in an electrochemical cell. The scan rate was 100 mV s^–1 in most cases, unless mentioned otherwise. One stock solution of acid TsOH (p-toluenesulfonic acid) (500 mM) was prepared, which was gradually added in the resulting solution to perform the catalytic ORR at different acid concentrations. As previously reported, a foot-of-the-wave analysis (FOWA) was used to analyze the data and obtain values for the turnover frequency (TOF) at a 5 mM TsOH concentration.

Substrate Oxidation

The substrate used here was toluene. Phosphate buffer solutions (pH 7) were saturated with toluene by adding two drops of ethanol to the suspension of these organic molecules in water, followed by vigorous shaking of the previously mentioned suspension. The use of ethanol increases the solubility of the organic substrates in the aqueous buffer. The mixture was allowed to settle and was separated using a separating flask. The aqueous layer was extracted, and electrocatalysis was done in heterogeneous condition. Electrolysis was performed at a constant potential of −500 mV [vs Ag/AgCl (saturated KCl)] with this solution for 15 min. Product analysis represents the average of three to five individual experiments. After the electrochemical experiments, the resultant aqueous electrolytes containing reactants and products were extracted with chloroform (CHCl₃). The CHCl₃ layer was dried and evaporated, and the product left behind was subjected to GC-MS analysis.

Synthesis

Methyl Imidazole-Substituted Monoamino Porphyrin (H₂TPPMIm)

100 mg (0.15 mmol) of meso-mono­(o-aminophenyl)­triphenylporphyrin (Scheme , 1) was first dissolved in 20 mL of dry DMF. A solution of 2-imidazolecarboxaldehyde at 4 equiv (60.67 mg, 0.632 mmol) was prepared in dry DMF and added dropwise to the reaction vessel, followed by the addition of 57 μL (0.742 mmol) of trifluoroacetic acid (TFA). The vessel was flushed with Ar for 10 min to maintain an inert atmosphere, and the reaction was allowed to proceed at room temperature for 3 h followed by the addition of 5 equiv of sodium borohydride (NaBH₄) (29.88 mg, 0.79 mmol), and the reaction mixture was stirred overnight under the same inert conditions. The reaction was then worked up by extracting it with dichloromethane (DCM) and water, and the resulting product was purified via column chromatography using 60–120 mesh neutral silica gel. The column was initially packed with a 20:80 DCM–hexane solvent mixture, with the polarity gradually increased until the final product was eluted with a 1:90 methanol–DCM mixture. ¹H NMR (CDCl₃): δ (ppm) 8.7–8.6 (m, 8H), 8.1–7.4 (m, 19H), 6.3 (s,2H), 3.8 (d, 2H), 3.5 (t, 1H), −2.8 (s, 2H). ESI-MS (Positive ion mode, CH₂Cl₂): 710.2305 ([M + H]⁺); Yield: (56.78 mg, ∼56%). UV–vis (THF): λ/nm = 417, 515, 549, 594, 652.

1. Synthetic Scheme of Methyl Imidazole-Substituted Monoamino Iron Porphyrin (FeTPPMIm).

Methyl Imidazole-Substituted Monoamino Iron Porphyrin (FeTPPMIm)

The ligand methyl imidazole-substituted monoamino porphyrin (H₂TPPMIm) (Scheme , 2) (100 mg, 0.14 mmol) was dissolved in 15 mL of dry and degassed THF. Two equivalents of 2,4,6-collidine (36 μL, 0.28 mmol) was added to this solution followed by the addition of 4 equiv of FeBr₂ (116.48 mg, 0.56 mmol). The solution was stirred overnight in glovebox under an inert atmosphere in the dark. The reaction mixture was worked up with dilute HCl, and the metal complex was extracted using DCM. The organic layer in DCM was collected and dried over anhydrous Na₂SO₄. The solvent was removed on a rotary evaporator, and the residue was purified by column chromatography using 60–120 mesh silica gel. The column was initially packed with 60:40 DCM–hexane, the polarity of the solvent mixture was gradually increased, and the final product was eluted with 4:96 MeOH–DCM mixture. Anal. calcd for C₄₈H₃₃ClFeN₇.C₆H₁₄·2H₂O: C, 70.40; H, 5.58; N, 10.64. Found: C, 70.75; H, 5.69; N, 9.62. ¹H NMR (CDCl₃): δ (ppm) 75–81 (β pyrrole proton). ESI-MS (Positive ion mode, CH₂Cl₂): 763.1347 ([M⁺]. Yield: 76 mg (71%). UV–vis (THF): λ/nm = 416, 509, 575.

Results and Discussion

Synthesis

The ligand H₂TPPMIm was synthesized by Schiff base condensation between aromatic amine and imidazole carboxaldehyde using TFA to form a carbon–nitrogen imine bond at room temperature. Further, the imine bond was reduced by using NaBH₄. Iron was introduced by metalating the porphyrin with FeBr₂ using 2,4,6-trimethylpyridine as the base in THF solvent, and post aerobic workup with dilute HCl, the chloride-bound ferric porphyrin with the pendant imidazole group was obtained (Scheme ).

Peroxidase Activity

Peroxidase activity is evaluated by catalytic oxidation of substrates like 3,3′,5,5′-tetramethylbenzidine (TMB) by two electrons by H₂O₂ or organic peroxides. Kinetic traces of the two-electron oxidized product of TMB which is a diamine product were recorded by monitoring 461 nm band (ε ≈ 30,250 M^–1 cm^–1 in a 5% v/v H₂O–acetonitrile mixture) (Scheme and Figure A). TMB is rapidly and catalytically oxidized by FeTPPMIm, and the rate is much faster than that of FeTPP (Figure F). Imidazole group in the second sphere clearly plays an important role to accelerate the catalysis. FeTPPMIm also shows peroxidase activity with o-Phenylenediamine (OPD), and the evolution kinetic trace of the two-electron oxidation product at 444 nm was monitored (Figure S3). FeTPPMIm also shows peroxidase activity with organic hydroperoxide (Figure S2) with these substrates, which establishes the generality of the catalysis.

2. Schematic Representation of Peroxidase-like Activity of FeTPPMIm, Illustrating Its Ability to Facilitate the Oxidation of TMB in the Presence of H₂O₂ .

2.

(A) Evolution of UV–vis absorption spectrum resulting from the reaction of TMB (150 μM) with FeTPPMIm (2.5 μM) and H₂O₂ (50 μM) in a 5% H₂O-acetonitrile mixture. Appearance of 461 nm band corresponds to the di-imine formation with time. (B) First-order rate constant of TMB oxidation plotted against varying catalyst concentrations (0.05 to 1.25 μM) keeping the H₂O₂ and TMB concentrations fixed. (C) Pseudo-first-order rate constant of TMB oxidation plotted against varying H₂O₂ concentrations (1–15 mM) keeping catalyst (0.05 μM) and TMB concentrations (90 μM) fixed. (D) Saturation kinetics of TMB oxidation plotted against different TMB concentrations (25 to 200 μM) keeping catalyst (0.05 μM) and H₂O₂ (100 μM) concentrations fixed. (E) Evolution of UV–vis absorption spectrum resulting from the reaction of TMB (150 μM) with FeTPP (2.5 μM) and H₂O₂ (50 μM) in 5% H₂O–acetonitrile mixture. Appearance of 461 nm band corresponds to the di-imine formation with time. (F) Kinetic traces (461 nm band) of TMB oxidation by FeTPPMIm (k = 0.033 s^–1) and FeTPP (k = 0.0011 s^–1).

Enzyme-like Catalysis

As the oxidized product of TMB is a chromophore, it is commonly used as a substrate for peroxidases which allows convenient determination of kinetic parameters. Kinetics of TMB oxidation by H₂O₂ catalyzed by FeTPPMIm was determined by independently varying the concentrations of the catalyst, H₂O₂, and TMB. The rate of TMB oxidation increases linearly with increasing catalyst concentration, suggesting that the reaction is first order with respect to the catalyst (Figure B). The oxidation rate displays a linear relationship with H₂O₂ at its low concentration, characteristic of first-order kinetics. At high concentrations of H₂O₂ (Figure C), the rate shows saturation implying binding of H₂O₂ to the catalyst prior to the reaction. This behavior is reminiscent of enzyme–substrate interactions where substrate binding precedes an irreversible rate-determining step consistent with Michaelis–Menten kinetics. The oxidation of TMB using FeTPPMIm as the catalyst, as it turns out, follows the saturation kinetics (Figure D) with respect to the TMB concentration as well. Very similar catalytic behavior of FeTPPIm was recorded with another peroxidase substrate OPD (Figure S3).

To investigate the enzyme-like catalysis and assess the kinetic parameters, the Michaelis–Menten kinetic behavior of FeTPPMIm was investigated. The Michaelis–Menten type saturation is observed for FeTPPMIm in the presence of TMB (Figure D) and H₂O₂ (Figure C) concentration ([TMB] < K _M , [H₂O₂] < K _M . Note that, Michaelis–Menten like saturation kinetics profile observed here is unusual for the molecular catalyst as it generally does not form a catalyst-substrate complex unlike its enzymatic counterparts, and only in very rare cases have such behavior been observed in molecular catalysts. , In contrast, FeTPP shows a very slow rate of oxidation of TMB (Figure E,F), and most importantly, the catalysis does not show saturation with either TMB or H₂O₂ implying there is no enzyme-like catalysis involved in FeTPP. The double reciprocal plot is used to investigate enzyme kinetics which shows a “ping-pong” mechanism (equation below).

$$\frac{1}{V_{\mathrm{TMB}}}=\frac{K_{\mathrm{M}}^{\mathrm{TMB}}}{V_{\max }}\frac{1}{C_{\mathrm{TMB}}}+\frac{K_{\mathrm{M}}^{{\mathrm{H}}_2{\mathrm{O}}_2}}{V_{\max }}\frac{1}{C_{{\mathrm{H}}_2{\mathrm{O}}_2}}+\frac{1}{V_{\max }}$$

A series of experiments were performed varying one substrate and keeping another constant, and the corresponding kinetic parameters are calculated from the double reciprocal plot which shows near parallel lines (Figure ) suggesting that TMB oxidation catalyzed by FeTPPMIm in the presence of H₂O₂ follows a “ping-pong” mechanism normally associated with enzyme catalysis. Secondary plot provides V _max = 5.9 × 10^–5 M s^–1, K _M = 5.5 ± 0.3 mM, K _M = 1.6 ± 0.5 mM, K _cat = 29 ± 3 s^–1, catalytic efficiency η_H2O2 = K _cat/K _M = 5.3 × 10³ M^–1 s^–1, and η_TMB = (K _cat/ K _M ) = 18.44 × 10³ M^–1 s^–1. The most relevant kinetic parameters for peroxidase are compared in the Table .

3.

Double reciprocal plot of peroxidase activity of FeTPPMIm with respect to the concentration of TMB (C _TMB), (H₂O₂ varied from 1.5 to 3 mM H₂O₂).

1. Comparison of Kinetic Parameters of Peroxidase Activity Using H₂O₂ and Organic Substrates.

catalyst	k cat (s–1)	K M (mM)	ηH2O2  (M–1 s–1) = K cat/K M	organic substrates	ref
Biot2–FePP·Sav S112 E K121H	3 × 103	4.37 ± 0.2	1.81 × 106	TMB
heme–peptide metalloenzyme	371 ± 14	44 ± 2	8.4 × 103	ABTS
artificial mimochrome VI	5.8 ± 3 × 103	440 ± 50	0.013 ± 0.002 × 103	ABTS
MP-8	22.9 ± 1.3		680 ± 10	tyramine
antibody-heme complex	6.56	24	274
C 45	1200	94	1.3 × 104	ABTS
chimeric peptide-DNAzymes	784 ± 67	127 ± 18	6.17 ± 1.6	ABTS
peroxidase (HRP,MPO)	0.01–2100	3.7	4.6 × 106	broad range of substrates	,,
(H42L)HRPC*		2
(R38L)HRPC*		12
FeMARG	25 ± 5	4.76 ± 0.5	5.2 × 103	TMB
FeTPPMIm	29 ± 3	5.5 ± 0.3	5.3 × 103	TMB	this work

K _M for H₂O₂ and TMB for FeTPPMIm are in the same range as that of native HRP and is similar to the best peptide-based catalysts reported (Table ). k _cat is not as high as those of peptide-based mimics of HRP and, of course, HRP itself. However, the catalytic rate of TMB oxidation by FeTPPMIm and η_H2O2 is substantially greater than that of any synthetic complexes and is comparable with the FeMARG complex reported earlier where a guanidinium group is present in the second sphere.

HRP has two important amino acid residues histidine and arginine in the distal side, and both play a significant role in the peroxidase activity. − , Mutation of His42 compromises the catalytic cycle of peroxidase. The rate of formation of Compound I decreases by ∼10⁶. Distal site histidine plays an important role in peroxide activation. Similarly, mutation of arginine shows that Arg 38 has a significant role in heterolytic O–O bond cleavage and consequently formation of Compound I. , FeTPPMIm is a synthetic complex that has only an imidazole pendant group, and FeMARG is a synthetic complex that has only a guanidium pendant group. So, a combination of these two complexes gives insight into the individual role of the two amino acids toward peroxidase activity, i.e., FeTPPMIm is analogous to the arginine mutant and FeMARG is analogous to the histidine mutant. The K _M for both the complexes are comparable indicating that both imidazole and guanidine groups can assist H₂O₂ binding to the iron center which is not very surprising, considering that both of these distal residues can form a strong hydrogen bond. Interestingly, the catalytic efficiency of H₂O₂ (Table , column 4) is comparable for the two complexes as well. In HRP, histidine acts as a general base to accelerate the binding of H₂O₂ and also acts as a general acid to facilitate the O–O bond heterolytic cleavage of Fe^III–OOH intermediate and eventually formation of Compound I. The pendant imidazole group of FeTPPMIm seems to be capable of acting as a general acid–base, and the K _M value is similar to HRP and the catalytic efficiency (η_H2O2 ) is higher than that of any other synthetic complex reported until date. The K _M for FeTPPMIm is lower than that of FeMARG. It suggests that TMB has more binding affinity toward FeTPPMIm where the pendant imidazole group is present than FeMARG where one of the four meso-aryl groups of TPP has an ortho-guanidine group. Catalytic efficiency or enzyme specificity toward TMB is much higher in the case of FeTPPMIm (K _cat/K _M = 18.44 × 10³ M^–1 s^–1) than FeMARG (K _cat/K _M = 7.35 × 10³ M^–1 s^–1). The higher binding affinity of TMB toward FeTPPMIm, the higher catalytic efficiency, and the higher catalytic rate of TMB oxidation may be attributed to a prearranged reaction site (Figure ).

4.

Plausible mechanistic pathway of electron and proton relay through Fe^III-hydroperoxide, imidazole moiety, and TMB.

Inhibition of enzymatic catalysis is a well-documented and extensively characterized phenomenon. Inhibitors are generally low-molecular-weight ligands which bind to the enzyme and lead to a decrease or complete loss of substrate conversion. Azide acts as a reversible or irreversible inhibitor of heme enzymes such as peroxidase, catalase, and cytochrome oxidase. , Azide shows a high binding affinity for ferric porphyrin, and it leads to competitive inhibition of HRP. The catalytic peroxidase activity of FeTPPMIm toward TMB oxidation is inhibited in the presence of azide. For competitive inhibition, the double reciprocal equation is

$$\frac{1}{V}=\frac{K_{\mathrm{M}}}{V_{\max }}\left(1+\frac{[I]}{K_I}\right)\frac{1}{[S]}+\frac{1}{V_{\max }}$$

The plot (Figure ) shows that V _max is not affected, and the lines intercept on $\frac{1}{V_{\mathrm{TMB}}}$ axis. The double reciprocal plot at different azide concentrations provides inhibition constant K _I. This type of kinetics is consistent with competitive inhibition, as reported for HRP. K _I for FeTPPMIm is calculated as 0.032 ± 0.02 mM, and for HRP, it is 1.47 mM. So, HRP has lower affinity toward azide binding than FeTPPMIm.

5.

Double reciprocal plot for competitive inhibition of catalytic TMB oxidation by FeTPPMIm as the catalyst in the presence of H₂O₂. Inhibitor (azide) concentrations are 12.5 and 25 μM.

In HRP oxidation, processes are initiated by protonation where the proton transfer is relayed to and from the distal histidine residue. The Michaelis–Menten kinetics and the catalytic efficiency of FeTPPMIm indicate a similar activation of ferric hydroperoxide to form the active oxidant, and substrate orientation is possible in this system as well, and this architecture facilitates a higher catalytic rate, resulting in a rare molecular catalyst which exhibits enzyme-like kinetics.

Reactive Intermediate

An iron porphyrin complex that exhibits efficient peroxidase activity is anticipated to generate ferric hydroperoxide (Fe^III–OOH) and compound I or an iron­(IV) oxo π cation radical as the reactive intermediate. Thus, the feasibility of such intermediates in FeTPPMIm, Fe^III–OOH and Compound I were generated by reacting Fe­(III)­TPPMIm with H₂O₂ and mCPBA, respectively.

To characterize the ferric hydroperoxide (Fe^III–OOH) intermediate, we utilized EPR spectroscopy at 77 K. Distinct new S = 1/2 low-spin EPR signals were observed with the g-values at 2.22, 2.12, and 1.95 (Figure A) when Fe­(III)­TPPMIm is reacted with H₂O₂ and rapidly frozen, which are very similar to those reported for heme-based Fe^III–OOH species previously reported (Table S1). To characterize Compound I or an iron­(IV) oxo π-cation radical intermediate, we utilized EPR spectroscopy at 10 K as this signal saturates at higher temperatures. A new EPR signal was observed at g = 3.99, 3.71, and 2.00 (Figure B) in addition to residual high-spin Fe­(III) porphyrin at 5.79 with its g = 2 signal when mCPBA is added to Fe­(III)­TPPMIm at −80 °C and rapidly frozen in liquid N₂. The new EPR signal is characteristic of exchange-coupled oxoferryl-porphyrin radical present in Compound I. ,

6.

(A) EPR spectral features of Fe–OOH species generated by the reaction of 2 mM solution of Fe^IIITPPMIm and H₂O₂ in THF. Data collected at 77 K. (B) EPR spectral features of Compound I generated by the reaction of 2 mM solution of Fe^IIITPPMIm and mCPBA in DCM. Data were collected at 10 K. (*) is the signal from decayed products.

ORR

Homogeneous

Electrochemical ORR of Fe-porphyrin complexes were performed in O₂-saturated CH₃CN solution using TsOH (p-Toluene sulfonic acid) as a proton source. The reversible Fe^III/II redox wave of FeTPPMIm, observed at a potential of −544 mV vs Fc^+/0 (Figure , light green) in an inert atmosphere is replaced by an irreversible electrocatalytic ORR current under aerobic condition (Figure , deep green). The irreversible catalytic current increases with increasing acid concentration which implies first-order dependence on acid concentration with TOF_max (Figure S7.1) consistent with previous reports of ORR by iron porphyrins. − The second-order rate constant (k _obs) of ORR by FeTPPMIm is calculated by FOWA analysis (Figure S7.2) to be 16 × 10² M^–1 s^–1 (Table ). The TOF values are obtained from FOWA to be 13 ± 1 s^–1 (Table ). The effective thermodynamic overpotential (at 15 μA current) determined for FeTPPMIm is 1.10 V and for FeTPP is 1.22 V. In the case of ORR catalyzed by FeTPP, where the second sphere residue is not present, the corresponding TOF_max and k _obs are 6 times lower (Table ) than FeTPPMIm (having an imidazole group in second sphere) with > 100 mV higher overpotential.

7.

CV data of 0.5 mM FeTPPMIm complex in dry and degassed acetonitrile solution under N₂ atmosphere and electrochemical ORR data in O₂-saturated acetonitrile solution at different TsOH (acid) concentrations using a GC working electrode, a Pt counter electrode, and a nonaqueous Ag/AgCl reference electrode with 100 mM [Bu₄N]⁺[ClO₄]⁻ as the supporting electrolyte at a scan rate of 100 mV s^–1 in the presence of a ferrocene internal standard.

2. Homogeneous Electrochemical Parameters.

catalyst	k (×102) (M–1 s–1)	TOF (s–1)	overpotential (V)
FeTPPMIm	16 ± 1	13 ± 1	1.10
FeTPP	2.7 ± 0.2	2.1 ± 0.1	1.22
FeL2	7.8	6.3	1.05

The mechanism of 4e^–/4H⁺ reduction of O₂ to H₂O in an organic solvent is crucial to understand the observed variations of rate and overpotential. An initial report found the protonation of a Fe­(III)–O₂ ^– species to be the rate-determining step of ORR in an organic solvent. The mechanism of O₂ reduction by iron porphyrins in organic solution has since been investigated by several groups using additional experimental and theoretical methods, and the same conclusion was reached, i.e., protonation of the Fe­(III)-O₂ ^– species is the rate- determining step in ORR in the presence of acid. ,,− Simultaneously, a linear correlation between log­(TOF) and overpotential of ORR by iron porphyrin under homogeneous condition in the presence of a strong acid has been described by Mayer and co-workers. The linear relationship between log­(TOF) and overpotential represents that faster TOF values originate from the consequence of higher overpotential. The scaling relationship proposed by Mayer is based on the fact that iron porphyrin with electron- donating substituents results in the enhanced pK _a of Fe­(III)–O₂ ^·– species which favors the protonation but at the same time lowers the Fe^III/II potential. However, deviation from this relationship is observed here, i.e., higher rate observed at lower overpotential (Figure ). This deviation of enhanced rate originates from the fact that other than primary coordination sphere, factors like second sphere interactions play crucial role in the rate-determining step by facilitating the proton transfer which has now been demonstrated in iron porphyrins and nonheme iron catalysts as well. − ,, The expected TOF value of the ORR of the FeTPPMIm complex based on the thermodynamic reduction potential can be estimated from the rate of FeTPP and using the slope from Mayer^’s scaling relationship to be 0.019 s^–1. However, the experimentally found TOF value of ORR by FeTPPMIm is 13 ± 1 s^–1 (Figure and Table ). Advantage of a higher ORR rate at a lower overpotential is clear in the Tafel plot (Figure S7.4). The rate of ORR of FeTPPMIm is almost twice that of an analogous complex with a pendant pyridine residue (FeL₂) (Table ) which breaks away from the scaling relationship as well. Thus, the pendant imidazole appears to accelerate the ORR in an organic solvent more than a pendant pyridine.

8.

log­(TOF) as a function of the catalyst-specific effective overpotential η_ORR with 5 mM TsOH under 1 atm O₂ using 0.1 M [nBu₄N]­[ClO₄] as the supporting electrolyte in acetonitrile solution. The compounds (FeTPPMIm, FeL₂, and FeTPP) studied here are represented by solid blue diamonds, and the empty circles indicate the expected log­(TOF) values based on their overpotential using the linear scaling relation developed by Mayer. The slope used is 21 ± 7 dec/V as PTSA is a strong acid.

Heterogeneous ORR

Heterogeneous ORR by iron porphyrin in an aqueous environment at neutral pH values proceeds via Fe­(III)–OOH species, and generally, the O–O bond cleavage is the rate-determining step of ORR. , Mechanistic investigations using in situ SERRS-RDE, a home-built setup to investigate heterogeneous catalysts in-operando, have revealed that the selectivity of the process depends on the site of protonation of an intermediate Fe^III–OOH species where the protonation of the proximal O-atom (bound to Fe) leads to 2e^–/2H⁺ ORR releasing H₂O₂, while protonation of the distal O-atom (not bound to the Fe) leads to O–O bond cleavage and 4e^–/4H⁺ ORR resulting in H₂O. ,, The ability of FeTPPMIm to activate the heterolytic O–O bond cleavage of a ferric peroxide species thus raises the possibility that this complex could facilitate selective 4e^–/4H⁺ ORR under aqueous conditions as well. In particular, the pK _a of the distal imidazole should be around that of 4-methylimidazole which is 7.9, implying that the pendant imidazole should be partly protonated at neutral pH values.

Oxygen reduction by FeTPPMIm was investigated under heterogeneous electrochemical conditions. FeTPPMIm was physiadsorbed over EPG electrodes, and electrochemical data are collected in pH 7 phosphate buffer solution against Pt counter electrode and Ag/AgCl in saturated KCl as the reference electrode. FeTPPMIm shows a well-defined Fe^III/Fe^II redox couple E _1/2 at −284 mV vs Ag/AgCl (−574 mV vs Fc⁰/Fc⁺) in the absence of O₂ in pH 7 (Figure A). The surface coverage was found to be 1.91 ± 0.5 × 10^–12 mol cm^–2 by integrating the CV current. In air-saturated pH 7 buffer solution, a large irreversible current was observed instead of a reversible couple, which was observed in the absence of O₂. This is the indication of electrocatalytic ORR by FeTPPMIm. LSV data show that O₂ reduces at potential −240 mV vs Ag/AgCl (−530 mV vs Fc⁰/Fc⁺) at the EPG electrode (Figure A). RDE data suggest a normal mass transfer limited current for FeTPPMIm below −240 mV (−530 mV vs Fc⁰/Fc⁺) (Figure B). The catalytic rate and the number of electrons transferred to the substrate by a heterogeneous catalyst can be obtained via K–L analysis using RDE. The K–L plot (Figure C) of 1/I _cat vs 1/ω^1/2 is linear, and the slope is close to the theoretical value for the 4e^– reduction process of O₂. From the K–L analysis (see experimental details section), the k _cat value obtained is 8.34 × 10⁶ M^–1 s^–1 which suggests that FeTPPMIm is a substantially faster catalyst for ORR than synthetic heme/Cu functional models of cytochrome c oxidases reported which have second-order rate constants of the order of 10⁵–10⁴ M^–1 s^–1. ,,

9.

(A) CV of FeTPPMIm over EPG in deoxygenated pH 7 buffer solution at a scan rate of 500 mV/s and LSV data of FeTPPMIm in air-saturated pH 7 buffer solution over EPG. (B) RDE data of FeTPPMIm over EPG for multiple rotations at the scan rate of 100 mV/s. (C) K–L plot for FeTPPMIm at the potential of −0.4 V. (D) RRDE data of FeTPPMIm deposited on the EPG electrode at a scan rate of 10 mV/s and at 300 rpm using Pt as the counter electrode and Ag/AgCl (saturated KCl) as the reference electrode.

Selectivity of ORR can be determined precisely by using RRDE where any partially reduced oxygen species (PROS) produced by the catalyst immobilized on the working electrode is diffused outward by the hydrodynamics created by the rotation of the ring and oxidized at a Pt ring electrode encircling the working electrode. RRDE data (Figure D) show detection of PROS at potentials where ORR occurs at the working electrode. The quantification of the PROS in situ reveals that the FeTPPMIm catalyst produces barely 2.2% PROS. Thus, FeTPPMIm is very selective for 4e^–/4H⁺ ORR to H₂O, producing very little H₂O₂ under aqueous conditions. Thus, the second sphere imidazole group can enhance both the rate and selectivity for the 4e^–/4H⁺ ORR at physiological pHs.

While O₂ reduction of a similar iron porphyrin with a pendant imidazole has not been reported in the past, iron porphyrin with three imidazole groups in the distal site were reported (Figure ). The three imidazole groups were installed to bind a Cu in the distal site to create mimics of cytochrome c oxidase. Nonetheless, the electrochemical ORR properties of this complex under experimental conditions similar to FeTPPMIm have been reported. The rate of ORR at pH 7 of this Fe-only complex is 10⁴ M^–1 s^–1 which is 2 orders of magnitude slower than FeTPPMIm under same experimental conditions. The imidazole group in both FeTPPMIm and Fe-only complexes can act as a H-bond acceptor, but the imidazole group in the Fe-only complex is attached via an electron-withdrawing amide linkage. This will result in lowering the pK _a of this imidazole by several units precluding its protonation at pH 7. The O₂ reduction by an analogous iron porphyrin with a pendant pyridine (FeL₂), which has 1 unit lower pK _a relative to imidazole, proceeds with the same rate as FeTPPMIm indicating that a lower pK _a is likely not responsible for the 2 orders of magnitude lower rate of the Fe-only complex. Similarly, the axial imidazole in the Fe-only complex would not affect the rate of ORR. Alternatively, both FeTPPMIm and FeL₂ share a similar geometric orientation of the pendant nitrogen relative to the aryl amine substituent on the porphyrin ring (Figure , highlighted in red) in that the H-bond acceptor nitrogen is separated by two carbon atoms from the aryl amine nitrogen. Spectroscopic data and electronic structure calculations on FeL₂ had revealed that this H-bonding interaction is very effective in activating the O–O bond of ferric peroxide species for cleavage. The Fe-only complex bears the H-bond acceptor nitrogen separated by three carbon atoms (Figure , highlighted in purple), and this subtle difference appears to have a pronounced effect on the kinetics of ORR.

10.

Schematic representation of ferric hydroperoxide intermediates of three different iron porphyrins FeTPPMIm, FeL₂, and Fe-only.

Oxidase Activity

During the catalytic ORR by iron porphyrins under heterogeneous conditions on SAM-covered Au electrodes, oxygen-derived reactive species accumulate on electrodes that have been used for the oxidation of organic substrates. The FeTPPMIm complex has the ability to support intermediates like Fe^III–OOH and Compound I species in organic solutions. To study the reactivity of the reactive high-valent intermediate species produced during ORR under heterogeneous conditions, toluene is chosen as a substrate (BDFE_CH of 89 kcal mol^–1). Oxygen reduction reaction was performed with a constant electrode potential at −500 mV vs Ag/AgCl using toluene as the substrate in pH 7 buffer solution (Figure ). Four electron oxidized product of toluene, benzaldehyde (Figure S8.1) is detected. The quantification of the benzaldehyde produced indicates a TON of 11404.56 and a TOF of 12.67­(s^–1) at room temperature. Note that without an axial thiolate, the ability of these high-valent species to oxidize stronger C–H bonds is limited.

11.

Schematic representation of toluene-to-benzaldehyde oxidation by O₂ catalyzed by FeTPPMIm physiadsorbed on the electrode in pH 7 buffer solution.

Conclusions

FeTPPMIm can thus catalyze both peroxidation (5% H₂O in CH₃CN) and O₂ reduction (both in organic and aqueous medium) quite efficiently. Most importantly, this dual catalysis is made possible by the inclusion of a single imidazole group in the distal environment of the iron porphyrin. In fact, in simple porphyrins, the O–O bond cleavage of ferric peroxide is homolytic and not heterolytic, and this step is the rate-determining step in heterogeneous O₂ reduction. The single imidazole group in the distal environment, analogous to the histidine group in peroxidases, facilitates O–O bond heterolysis of ferric peroxide intermediate which is a critical step in both peroxidation and the rate-determining step in oxygen reduction.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/jacsau.5c01252.

• Additional ¹H NMR, cyclic voltammetry, gas chromatography, and kinetics data (PDF)

§.

S.D. and T.R. contributed equally in this work.

The research is funded by the Department of Science and Technology, India, grant CRG/2021/000154.

The authors declare no competing financial interest.