Small Reorganization Energy for Ligand-Centered Electron-TransferReduction of Compound I to Compound II in a Heme Model StudyNami Fukui,† Xiao-Xi Li,‡ Wonwoo Nam,*,‡ Shunichi Fukuzumi,*,‡,§ and Hiroshi Fujii*,†

†Department of Chemistry, Graduate School of Humanities and Sciences, Nara Women’s University, Nara 650-8506, Japan‡Department of Chemistry and Nano Science, Ewha Womans University, Seoul 03760, Korea§Faculty of Science and Engineering, Meijo University, SENTAN, Japan Science and Technology Agency, Nagoya, Aichi 468-0073,Japan

*S Supporting Information

ABSTRACT: The electron-transfer (ET) processes fromelectron-donor substrates to oxoiron(IV) porphyrin π-cation-radical species (Cpd I) are key steps in theiroxygenation reactions. Here, we have evaluated the rateconstants of the outer-sphere ET reduction of Cpd Imodel complexes of meso-tetramesitylporphyrin (1) and2,7,12,17-tetramesityl-3,8,13,18-tetramethylporphyrin (2)in light of the Marcus theory of ET to determine the ETreorganization energies (λ). The λ values of the ligand-centered ET reduction of Cpd I model complexes aremuch smaller than those of the metal-centered ETreduction of various oxoiron(IV) complexes. Moreover,the λ value of 1 is larger than that of 2, resulting from thedifference in the nature of the a1u/a2u porphyrin π-cation-radical orbitals.

Oxoiron(IV) porphyrin π-cation-radical species, referredto as compound I (Cpd I), are pivotal reaction

intermediates in many biological reactions catalyzed by hemeenzymes such as peroxidases, catalases, and cytochromesP450.1−4 Because Cpd I is reduced to an oxoiron(IV)porphyrin species, referred to as compound II (Cpd II), or aferric heme complex in their biological reactions, the chemicalreactions of Cpd I accompany electron-transfer (ET) processesfrom substrates to Cpd I. The reaction mechanisms of Cpd Iwith various substrates have so far been extensively studied byusing heme enzymes and their synthetic model com-pounds.5−11 The ET processes are key to gaining anunderstanding of the reactivity and selectivity of Cpd I.11

The N-demethylation reactions of Cpd I of horseradishperoxidase (HRP) with N,N-dimethylanilines were studiedpreviously to discuss the ET pathways in hydrogen-atomtransfer from organic substrates to Cpd I.11,12 However, outer-sphere ET processes from one-electron donor to Cpd I andanalogues have yet to be scrutinized in light of the Marcustheory of ET to determine the reorganization energies (λ) ofouter-sphere ET.12

We report herein the ligand-centered outer-sphere ET fromvarious one-electron donors to Cpd I model complexes,evaluating the ET rate constants in light of the Marcus theoryof outer-sphere ET to determine the λ values,13 which can becompared with metal-centered outer-sphere ET reactions ofoxoiron(IV) species. The difference in the λ values depending

on the types of orbitals of the porphyrin π-cation radicals hasalso been clarified. To the best of our knowledge, this is thefirst time for clarification on how the ligand-centered versusmetal-centered ET controls the ET reactivity of Cpds I and II.We prepared structurally different Cpd I model complexes

(Figure 1) of meso-tetramesitylporphyrin (1) and 2,7,12,17-

tetramesityl-3,8,13,18-tetramethylporphyrin (2).14,15 1 and 2have the bulky mesityl group at the meso position and thepyrrole β position, respectively. The overall structure of 2resembles that of a protoheme (Figure 1), the prostheticgroups of most heme enzymes, rather than that of 1. The redoxpotentials of Cpd I of 1 and 2 were reported to be almost thesame, and the reported E1/2 value is calibrated to be 0.974 Vversus saturated calomel electrode (SCE).8 We examined thereactions of Cpd I of 1 and 2 with one-electron donors, such astriphenylamine derivatives and 1,1′-diacetylferrocene, by usinga low-temperature rapid-mixing stopped-flow technique. Cpd Imodel complexes of 1 and 2 were prepared by premixing ferricnitrate complexes of 1 and 2 with 5 equiv of m-chloroperoxybenzoic acid in dichloromethane at −20 °C,respectively. After Cpd I (∼10 s) was generated, the reactionmixture was rapidly mixed with the solution of the reductant.Figure 2 shows the absorption spectral change after rapid

mixing of 1 with tris-p-tolylamine in dichloromethane at −20°C. The peaks at 504, 549, and 672 nm appear after rapidmixing. A comparison with authentic spectra indicates that thepeaks at 504, 549, and 672 nm result from the ferric porphyrin

Received: April 11, 2019Published: June 11, 2019

Figure 1. Structures of Cpd I model complexes of protoheme, 1, and2.

Communication

pubs.acs.org/ICCite This: Inorg. Chem. 2019, 58, 8263−8266

© 2019 American Chemical Society 8263 DOI: 10.1021/acs.inorgchem.9b01051Inorg. Chem. 2019, 58, 8263−8266

Dow

nloa

ded

by E

WH

A W

OM

AN

S U

NIV

at 2

0:55

:23:

276

on J

uly

01, 2

019

from

http

s://p

ubs.

acs.

org/

doi/1

0.10

21/a

cs.in

orgc

hem

.9b0

1051

.

complex, Cpd II and tris-p-tolylamine π-cation radical,respectively (Figure S1). The intensity of the peak at 549nm increased for 0.090 s after mixing but then decreased(Figure 2a, inset). On the other hand, the intensities of thepeaks at 504 and 672 nm increased in the reaction. Finally, theabsorption spectrum changed to that of a mixture of ferricporphyrin complex and tris-p-tolylamine π-cation radical. Theobserved spectral changes can be interpreted by the reactionwhere Cpd I is reduced to Cpd II, followed by the furtherreduction of Cpd II to the ferric porphyrin complex, with theformation of tris-p-tolylamine π-cation radical. In addition,clear isosbestic points were observed at the initial and laststages of the reaction. The time course of the absorbancechange can be simulated with a double-exponential function,providing apparent reduction rate constants for Cpds I and II.The reduction rate constant for Cpd I linearly depends on theconcentration of tris-p-tolylamine, providing the second-orderrate constant (Figure S2), as listed in Table 1.Similar absorption spectral changes were observed for the

reaction of Cpd I of 2 with tris-p-tolylamine (Figure S3). Thesecond-order rate constant for 2 was also determined from thedependence of the rate constant on the concentration of tris-p-tolylamine (Figure S4 and Table 1). Interestingly, the rateconstant of the reduction from Cpd I to Cpd II for 2 is about

20-fold faster than that for 1, in spite of the same redoxpotential, suggesting that the ET reorganization energy (λ) of2 is smaller than that of 1.To investigate the dependence of the ET rate constants on

the ET driving force, we determined ET rate constants usingvarious triphenylamine derivatives having different redoxpotentials and 1,1′-diacetylferrocene. The redox potentials ofthe triphenylamine derivatives used here were determined bycyclic and differential pulse voltammetry (Figure S5 and Table1). The absorption spectral changes for the reactions withtriphenylamine derivatives other than tris-p-bromophenyl-amine were similar to those with tris-p-tolylamine (FiguresS6−S8). The reactions of Cpd I with tris-p-bromophenylaminestopped in the middle of the reaction because of theendergonic ET process, even when a large excess (∼500equiv) of tris-p-bromophenylamine was present (Figure S9). Inaddition, the time courses for the reactions with tris-phenylamine, tris-p-bromophenylamine, and 1,1′-diacetylferro-cene could be simulated with single-exponential functionsbecause the reduction reactions from Cpd I to Cpd II becameslower than those from Cpd II to ferric porphyrin complexes(Figures S9−S11). The second-order rate constants of ET(ket) determined are summarized in Table 1. The ket values of2 are also much larger than those of 1 for all other reductants.On the basis of these results, we analyzed rate constants of

ET from one-electron reductants to Cpd I of 1 and 2 in light ofthe Marcus theory of outer-sphere ET.13 Figure 3 shows thedependence of log ket on the driving force of ET from one-electron donors to Cpd I of 1 and 2 at −20 °C. The log ketvalues of ET from one-electron donors to Cpd I of 1 (redcircles) and 2 (blue circles) are well fitted by the Marcusequation for adiabatic outer-sphere ET (eq 1):

λ λ= [− + Δ ]k Z G k Texp ( /4)(1 / ) /et et2

B (1)

Z is the collision frequency, taken as 1 × 1011 M−1 s−1, λ is thereorganization energy of ET, kB is the Boltzmann constant, andT is the absolute temperature.13 From the fitting of the rateconstants, the reorganization energy (λ) values of ET fromtriphenylamines to Cpd I model complexes of 1 and 2 wereestimated to be 1.44 and 1.21 eV, respectively.

Figure 2. Absorption spectral change for the reaction of Cpd I of 1(12.5 μM) with tris-p-tolylamine (250 μM) in dichloromethane at−20 °C: (a) 0−0.091 s; (b) 0.091−0.500 s. Green line: immediatelyafter rapid mixing. Red line: after 0.091 s. Blue line: after 0.500 s. Theinset shows the time course (0−0.500 s) of the absorbance at 549 nm:red circle, experimental data; black line, simulation line obtained froma least-squares curve fit with a double-exponential function.

Table 1. Redox Potentials and ET Rate Constants ofTriphenylamine Derivatives and 1,1′-Diacetylferrocene

ket (M−1 s−1)

E1/2a V vs

SCE inCH2Cl2 1 2

Para-Substituted TriphenylamineOCH3 OCH3 Br 0.800 6.11 × 105 1.93 × 106

CH3 CH3 CH3 0.829 1.54 × 105 1.76 × 106

OCH3 H H 0.860 6.43 × 104 1.15 × 106

CH3 CH3 Br 0.964 1.43 × 103 9.63 × 104

OCH3 Br Br 0.980 8.45 × 102 5.93 × 104

CH3 H H 0.988 5.07 × 103 n.d.H H H 1.020b 3.34 × 102 9.43 × 103

Br Br Br 1.172 4.01 × 10 2.63 × 102

Ferrocene1,1′-diacetyl

0.974 4.94 × 103 5.56 × 104

aE1/2 of ferrocene was 0.484 V versus SCE. bReference 16. Thereported E1/2 value (0.92 V) was calibrated using the reported E1/2(0.76 V) for 4-methoxyltriphenylamine.

Inorganic Chemistry Communication

DOI: 10.1021/acs.inorgchem.9b01051Inorg. Chem. 2019, 58, 8263−8266

8264

The estimated λ values of Cpd I model complexes of 1 and 2are smaller than those (1.6−1.8 eV) of one-electron-reductionprocesses of Cpd II model complexes of myoglobin andHRP.12,17−19 These λ values are also much smaller than those(2.05−2.74 eV) of ET of nonheme oxoiron(IV) com-plexes.12,20−22 The small λ values for Cpd I may result fromthe site of the ET reaction, which is on the porphyrin ligandmoiety. Previous studies showed that the λ values for metal-centered redox processes are significantly larger than those forporphyrin ligand-centered redox processes because the metal-centered ET process induces the change of the oxidation stateof the metal ion, which results in changes of the charge andionic radius of the metal ion and enforces the structural changearound the metal ion with the redox process.22

The λ value for 2 is smaller than that for 1. The difference inthe λ values can be interpreted with the nature of theporphyrin π-cation-radical orbitals of these Cpd I species.Previously, we showed that the porphyrin π-radical electron ofCpd I of 1 is in the a2u orbital, whereas that of 2 is in the a1uorbital.19,20 The porphyrin π-radical electron in the a2u orbitalof 1 delocalized not only in the porphyrin ligand but also in theiron oxo moiety (∼5%) because of the orbital interactionbetween the a2u orbital and the Fe 4pz−O 2pz antibondingorbital (Figure 4). On the other hand, the porphyrin π-radicalelectron in the a1u orbital of 2 mainly delocalized in the

porphyrin ligand because of the absence of suitable orbitalinteraction with oxoiron orbitals (i.e., mismatch of the orbitalsymmetry). Therefore, the ET process of Cpd I of 2 shows theporphyrin ligand-centered character with no metal-centeredcharacter. In contrast, 1 contains some metal-centered redoxcharacter. Consequently, as discussed above, the λ value of theCpd I of 1 becomes larger than that of 2. In addition, theflexibility of the porphyrin ligand also affects the λ values. Theabsence of the substituent at the meso position in 2 makes theporphyrin ligand flexible, giving the smaller λ value. Thesimilarity of the λ values suggests that the contribution of theprotein moiety of HRP to the reorganization energy for the ETreduction process of Cpd I would be small becauseparticipation of the protein moiety is expected to increasethe λ value.The reorganization energies for the ET exchange processes

of Cpd I model complexes of 1 and 2 were also estimated fromdensity functional theory (DFT) calculations. Details of theDFT calculations are described in the Supporting Information.The calculated λ value (1.15 eV) for the ET reduction processof 1 is larger than that (1.05 eV) of 2 (Table S1), which agreeswell with the experimental results in Figure 3. The geometricanalysis, which can disclose the flexibility difference betweentwo porphyrin systems, indicates that the ET reduction of theCpd I of 1 resulted in a larger structural change, showing ahigher root-mean-square-deviation (RMSD) value of 0.262 Å,whereas that of 2 presented a smaller structural change with alower RMSD value of 0.107 Å (Table S2). The larger is thestructural change during the ET process, the higher thereorganization energy (vide supra). Thus, the results from bothelectronic and geometric analyses support a higher reorganiza-tion energy (i.e., a lower ET reactivity) for Cpd I of 1compared to that of 2.In conclusion, we report the λ values for the outer-sphere

ET reductions of Cpd I to Cpd II in heme model complexes.The estimated λ values for Cpd I model complexes are muchsmaller than those for nonheme oxoiron(IV) complexes. Thisis due to the fact that the site of the reduction is the porphyrinligand in the heme system, and thus the λ value is modulatedby the nature of the a1u/a2u porphyrin π-cation-radical orbital.Understanding the λ values for the ligand-centered ET of CpdI provides quantitative insight into the difference in the redoxreactivity between Cpds I and II.

■ ASSOCIATED CONTENT*S Supporting InformationThe Supporting Information is available free of charge on theACS Publications website at DOI: 10.1021/acs.inorg-chem.9b01051.

Experimental and computational details, Scheme S1,Figures S1−S11, and Tables S1 and S2 (PDF)

■ AUTHOR INFORMATIONCorresponding Authors*E-mail: wwnam@ewha.ac.kr.*E-mail: fukuzumi@chem.eng.osaka-u.ac.jp.*E-mail: fujii@cc.nara-wu.ac.jp.ORCIDWonwoo Nam: 0000-0001-8592-4867Shunichi Fukuzumi: 0000-0002-3559-4107Hiroshi Fujii: 0000-0003-4611-2983

Figure 3. Plots of log ket for the ET reduction of Cpd I modelcomplexes of 1 and 2 by one-electron donors versus their drivingforces, −ΔGet = −e(Eox − Ered). Ered (1 and 2) = 0.974 V. Red andblue circles show plots for 1 and 2, respectively. Red and blue linesshow simulation lines using eq 1 for 1 and 2, respectively.

Figure 4. Electron-accepting orbitals for 1 (the a2u orbital) and 2 (thea1u orbital).

Inorganic Chemistry Communication

DOI: 10.1021/acs.inorgchem.9b01051Inorg. Chem. 2019, 58, 8263−8266

8265

NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThis work was supported by grants from JSPS (Grants25288032 and 17H03032 to H.F. and Grant 16H02268 toS.F.) and CREST (to H.F.) and the NRF of Korea throughCRI (Grant NRF-2012R1A3A2048842 to W.N.) and GRL(Grant NRF-2010-00353 to W.N.).

■ REFERENCES(1) Liu, W.; Groves, J. T. Manganese Catalyzed C-H Halogenation.Acc. Chem. Res. 2015, 48, 1727−1735.(2) Fujii, H. Model Complexes of Heme Peroxidases. In HemePeroxidases; RSC Metallobiology Series; Raven, E., Dunford, B., Eds.;Royal Society of Chemistry: London, 2016; pp 183−217.(3) Rittle, J.; Green, M. T. Cytochrome P450 Compound I: Capture,Characterization, and C-H Bond Activation Kinetics. Science 2010,330, 933−937.(4) Nam, W. High-Valent Iron(IV)-Oxo Complexes of Heme andNon-Heme Ligands in Oxygenation Reactions. Acc. Chem. Res. 2007,40, 522−531.(5) Huang, X.; Groves, J. T. Oxygen Activation and RadicalTransformations in Heme Proteins and Metalloporphyrins. Chem.Rev. 2018, 118, 2491−2553.(6) Zaragoza, J. P.; Yosca, T. H.; Siegler, M. A.; Moen̈ne-Loccoz, P.;Green, M. T.; Goldberg, D. P. Direct Observation of OxygenRebound with an Iron-Hydroxide Complex. J. Am. Chem. Soc. 2017,139, 13640−13643.(7) Asaka, M.; Fujii, H. Participation of Electron Transfer Process inRate-Limiting Step of Aromatic Hydroxylation Reactions byCompound I Models of Heme Enzymes. J. Am. Chem. Soc. 2016,138, 8048−8051.(8) Takahashi, A.; Kurahashi, T.; Fujii, H. Redox Potentials ofOxoiron(IV) Porphyrin π-Cation Radical Complexes: Participation ofElectron Transfer Process in Oxygenation Reactions. Inorg. Chem.2011, 50, 6922−6928.(9) Guengerich, F. P. Common and Uncommon Cytochrome P450Reactions Related to Metabolism and Chemical Toxicity. Chem. Res.Toxicol. 2001, 14, 611−650.(10) Goto, Y.; Matsui, T.; Ozaki, S.-I.; Watanabe, Y.; Fukuzumi, S.Mechanism of Sulfoxidation Catalyzed by High-Valent Intermediatesof Heme Enzymes: Electron-Transfer vs Oxygen-Transfer Mecha-nism. J. Am. Chem. Soc. 1999, 121, 9497−9502.(11) Goto, Y.; Watanabe, Y.; Fukuzumi, S.; Jones, J. P.; Dinnocenzo,J. P. Mechanisms of N-Demethylations Catalyzed by High-ValentSpecies of Heme Enzymes: Novel Use of Isotope Effects and DirectObservation of Intermediates. J. Am. Chem. Soc. 1998, 120, 10762−10763.(12) Fukuzumi, S. Electron-Transfer Properties of High-ValentMetal-Oxo Complexes. Coord. Chem. Rev. 2013, 257, 1564−1575.(13) Marcus, R. A. Chemical and Electrochemical Electron-TransferTheory. Annu. Rev. Phys. Chem. 1964, 15, 155−196. (b) Marcus, R. A.Electron Transfer Reactions in Chemistry: Theory and Experiment(Nobel Lecture). Angew. Chem., Int. Ed. Engl. 1993, 32, 1111−1121.(14) Fujii, H. Effects of the Electron-Withdrawing Power ofSubstituents on the Electronic Structure and Reactivity in Oxoiron-(IV) Porphyrin π-Cation Radical Complexes. J. Am. Chem. Soc. 1993,115, 4641−4648.(15) Takahashi, A.; Kurahashi, T.; Fujii, H. Effect of Imidazole andPhenolate Axial Ligands on the Electronic Structure and Reactivity ofOxoiron(IV) Porphyrin π-Cation Radical Complexes: DrasticIncrease in Oxo-Transfer and Hydrogen Abstraction Reactivities.Inorg. Chem. 2009, 48, 2614−2625.(16) Seo, E. T.; Nelson, R. F.; Fritsch, J. M.; Marcoux, L. S.; Leedy,D. W.; Adams, R. N. Anodic Oxidation Pathways of Aromatic Amines.Electrochemical and Electron Paramagnetic Resonance Studies. J. Am.Chem. Soc. 1966, 88, 3498−3503.

(17) Fukuzumi, S.; Nakanishi, I.; Tanaka, K.; Suenobu, T.; Tabard,A.; Guilard, R.; Van Caemelbecke, E.; Kadish, K. M. Electron-Transfer Kinetics for Generation of Organoiron(IV) Porphyrins andIron(IV) Porphyrin π-Radical Cations. J. Am. Chem. Soc. 1999, 121,785−790.(18) (a) Park, J.; Morimoto, Y.; Lee, Y.-M.; Nam, W.; Fukuzumi, S.Unified View of Oxidative C-H Bond Cleavage and Sulfoxidation byNonheme Iron(IV)-Oxo Complex via Lewis Acid-Promoted ElectronTransfer. Inorg. Chem. 2014, 53, 3618−3628.(19) Fenwick, C. W.; English, A. M.; Wishart, J. F. pH and DrivingForce Dependence of Intramolecular Oxyferryl Heme Reduction inMyoglobin. J. Am. Chem. Soc. 1997, 119, 4758−4764.(20) Fukuzumi, S.; Kotani, H.; Prokop, K. A.; Goldberg, D. P.Electron- and Hydride-Transfer Reactivity of an Isolable Manganese-(V)-Oxo Complex. J. Am. Chem. Soc. 2011, 133, 1859−1869.(21) Fukuzumi, S.; Fujioka, N.; Kotani, H.; Ohkubo, K.; Lee, Y.-M.;Nam, W. Mechanistic Insights into Hydride-Transfer and Electron-Transfer Reactions by a Manganese(IV)-Oxo Porphyrin Complex. J.Am. Chem. Soc. 2009, 131, 17127−17134.(22) Lee, Y.-M.; Kotani, H.; Suenobu, T.; Nam, W.; Fukuzumi, S.Fundamental Electron-Transfer Properties of Non-heme Oxoiron(IV)Complexes. J. Am. Chem. Soc. 2008, 130, 434−435.

Inorganic Chemistry Communication

DOI: 10.1021/acs.inorgchem.9b01051Inorg. Chem. 2019, 58, 8263−8266

8266