Meso-Substitution Activates Oxoiron(IV) Porphyrin π-Cation Radical Complex More Than Pyrrole-β-Substitution for Atom Transfer Reaction

Nami Fukui, Kanako Ueno, Masahiko Hada, and Hiroshi Fujii*

Article

meso-Substitution Activates Oxoiron(IV) Porphyrin π‑Cation Radical

Complex More Than Pyrrole-β-Substitution for Atom Transfer

Reaction

Cite This: Inorg. Chem. 2021, 60, 3207 −3217

Read Online

ACCESS

Metrics & More

Article Recommendations

sı Supporting Information

ABSTRACT: There have been two known categories of porphyrins: a

meso-substituted porphyrin like meso-tetramesitylporphyrin (TMP)

and a pyrrole-β-substituted porphyrin like native porphyrins and

,7,12,17-tetramethyl-3,8,13,18-tetramesitylporphyrin (TMTMP). To

reveal the chemical and biological function of native hemes, we

compare the reactivity of the oxoiron(IV) porphyrin π-cation radical

complex (Compound I) of TMP (TMP-I) with that of TMTMP

(TMTMP-I) for epoxidation, hydrogen abstraction, hydroxylation,

sulfoxidation, and demethylation reactions. Kinetic analysis of these

reactions indicated that TMP-I is much more reactive than TMTMP-I

when the substrate is not sterically bulky. However, as the substrate is

sterically bulkier, the diﬀerence of the reactivity between TMP-I and

TMTMP-I becomes smaller, and the reactivity of TMP-I is

comparable to that of TMTMP-I for a sterically hindered substrate. Since the redox potential of TMP-I is almost the same as

that of TMTMP-I, we conclude that TMP-I is intrinsically more reactive than TMTMP-I for these atom transfer reactions, but the

steric eﬀect of TMP-I is stronger than that of TMTMP-I. This is contrary to the previous result for the single electron transfer

reaction: TMTMP-I is faster than TMP-I. DFT calculations indicate that the orbital energies of the FeO moiety for TMTMP-I

are higher than those for TMP-I. The diﬀerence in steric eﬀect between TMP-I and TMTMP-I is explained by the distance from the

mesityl group to the oxo ligand of Compound I. Signiﬁcance of the pyrrole-β-substituted structure of the hemes in native enzymes is

also discussed on the basis of this study.

INTRODUCTION

iron oxo species and to develop a ligand system that can make

2,3

■

the iron oxo species much more reactive.

High-valent iron oxo species have been identiﬁed as the key

intermediates in the reactions of iron-containing catalysts, as

well as metalloenzymes.

The ligand eﬀect on the reactivity can be seen in heme

enzymes. Many heme enzymes have the same heme (iron

protoporphyrin IX) but various heme proximal (axial) ligands

−6

The oxoiron(IV) porphyrin π-

cation radical species called Compound I and oxoiron(IV)

porphyrin species called Compound II have been identiﬁed in

the catalytic cycles of peroxidases, catalases, and cytochrome

(

e.g., histidine in peroxidases, tyrosine in catalases, and

cysteine in cytochrome P450s). The axial ligand is highly

conserved in each family having the same biological function.

−6

P450s.

Synthetic model complexes of Compound I and

It has been believed that the axial ligand tunes the reactivity of

Compound II have also been prepared to study the electronic

states and reactivity of these complexes. Nonheme oxoiron-

V) and oxoiron(IV) complexes have also been synthesized by

Compound I. Green et al. reported a high pK value of the oxo

ligand in Compound II of cytochrome P450, and they

proposed that it is due to the strong donor eﬀect from the

thiolate axial ligand and the origin of the strong oxidizing

power of Compound I for hydroxylation of unactivated C−H

(

using various synthetic ligand systems. All of these synthetic

and enzymatic studies revealed that the reactivity of the high-

valent oxoiron species is changed by the structural and

electronic eﬀects of the ligands. For example, oxoiron(IV)

complexes of 1,4,8,11-tetramethyl-1,4,8,11-tetraazacyclotetra-

decane have been reported to be stable even at room

bonds. We reported a pronounced axial ligand eﬀect of

Received: December 4, 2020

Published: February 15, 2021

temperature, but those of tris((N-methylbenzimidazol-2-

yl)methyl)amine react with cyclohexane even at −40 °C

(

structures of ligands: see the Appendix in the Supporting

Information). Recently, much attention has been directed to

reveal how the ligand controls the reactivity of the high-valent

2021 American Chemical Society

207

Inorg. Chem. 2021, 60, 3207−3217

Inorganic Chemistry

and TMTMP-I have been reported previously.

Article

imidazole and phenolate ligands on the reactivity of synthetic

Compound I model complexes; the reactivity increased about

To clarify whether the reactivity of Compound I is altered by

the position of the substituent, the meso-position and the

pyrrole-β-position, we study here the reactivity of TMP-I and

TMTMP-I toward epoxidation, hydrogen abstraction, hydrox-

ylation, sulfoxidation, and demethylation reactions. The

present kinetic study clearly shows that the intrinsic reactivity

of TMP-I is much higher than that of TMTMP-I, but the

steric eﬀect of TMP-I for these reactions is greater than that of

TMTMP-I. We conclude that the TMP-I is more reactive than

TMTMP-I for the oxygen- or hydrogen-atom transfer reaction,

but TMTMP-I is more reactive than TMP-I for the single

electron transfer reaction. The factors controlling the reactivity

and the steric eﬀect for TMP-I and TMTMP-I are studied

from the activation parameters and the DFT calculations.

00-fold with the coordination of imidazoles and phenols.

The axial ligand eﬀect on Compound I could be explained by

the stability of the ground spin state of the product state.

similar axial ligand eﬀect has also been observed for the

corrolazine (5,10,15-triazacorrole) ligand system.

A signiﬁcant ligand eﬀect has been reported not only for the

,3,14−32

axial ligand but also for the equatorial ligand.

The eﬀect

of the equatorial ligand has been comprehensively studied by

,3,14−32

using synthetic porphyrin ligands.

These studies

revealed that introduction of electron-withdrawing substituents

into porphyrin ligands increases reactivity of high-valent oxo

4−17,19,21−28

species.

For example, reactivity of Compound I

becomes higher as the number of halogen atoms in the meso-

14−16,19,22,23

phenyl group is larger.

The introduction of the

RESULTS

electron-withdrawing substituents increased the redox poten-

■

22,25−29

tials of Compound I.

Linear correlations could be

Epoxidation Reactions. TMP-I and TMTMP-I were

prepared by the oxidation of iron(III) nitrate complexes of

TMP and TMTMP with ozone gas at low temperature,

found between the activation energy and the redox potential of

7,28,32

Compound I.

The steric and electronic eﬀect of the

11,12

substituent changed the selectivity of the oxygen transfer

respectively.

The spectroscopic characterizations of TMP-I

4,30,31

11,16

reactions.

These porphyrin ligand eﬀects have been

Figure 2

studied by using meso-tetraarylporphyrins, in which the

sterically hindered aryl groups bind at the meso-position of a

porphyrin. Groves et al. reported the ﬁrst example of the

synthetic Compound I model complex (TMP-I) from meso-

tetramesitylporphyrin (TMP) shown in Figure 1 . However,

Figure 1. Structures of Compound I of TMP (TMP-I), native heme

(protoheme), and TMTMP (TMTMP-I).

this substitution pattern is diﬀerent from those of iron

porphyrins in native enzymes, in which the substituents bind

at the pyrrole-β position of a porphyrin (Figure 1). The

synthetic Compound I model complex from a nativelike

porphyrin was ﬁrst synthesized by using 2,7,12,17-tetramethyl-

,8,13,18-tetramesitylporphyrin (TMTMP) shown in Figure

The spectroscopic property of Compound I of TMTMP

(TMTMP-I) was similar to those of Compound I in native

enzymes. The redox potential of TMTMP-I is very close to

that of TMP-I; thus, reactivity of TMTMP-I was expected to

be close to that of TMP-I. A previous result of the competitive

epoxidation reaction of TMP-I and TMTMP-I with cyclo-

hexene suggested that reactivity of TMP-I is close to that of

TMTMP-I. However, a recent study on the rate of single

electron transfer (ET) reactions showed a signiﬁcant diﬀerence

between TMP-I and TMTMP-I. The ET rates of TMTMP-I

are 1 to 2 orders magnitude faster than those of TMP-I, and

the estimated reorganization energy (λ) for TMP-I is about 0.2

eV larger than that of TMTMP-I. These results suggest that a

further study is needed to reveal biological signiﬁcance of the

pyrrole-β-substituted structure of the native hemes in many

enzymes.

Figure 2. Absorption spectral change for the reactions of (a) TMP-I

and (b) TMTMP-I with norbornene in dichloromethane at −50 °C.

Red line, immediately after the addition of norbornene; and blue line,

after the reaction. Gray broken line, after the addition of norbornene

at constant time interval. (a) TMP-I (0.1 mM), norbornene (19.6

mM), and time intervals of 100 s. (b) TMTMP-I (0.1 mM),

norbornene (19.9 mM), and time intervals of 585 s. Inset: time course

of the absorbance in the reactions. (a) Black circle, at 507 nm; red

circle, at 667 nm. (b) Black circle, at 502 nm; red circle, at 632 nm.

Solid line: simulation line obtained from a least-squares curve ﬁt with

a single exponential function.

208

Inorg. Chem. 2021, 60, 3207−3217

Inorganic Chemistry

Article

Table 1. Second-Order Rate Constants for the Reactions of TMP-I and TMTMP-I with Oleﬁns in Dichloromethane at −50 °C

TMP-I k × 10 /M s⁻¹(yield/%)

−1

TMTMP-I k × 10 /M s⁻¹(yield/%)

−1

substrate

norbornene

styrene

22.0 (86)

2.23

1.97 (28)

0.538

p-methoxystyrene

cyclopentene

cyclohexene

cyclooctene

86.2 (52)

13.5 (32)

3.89 (68)

2.98 (64)

19.6 (7)

4.59 (18)

1.85 (54)

3.00 (48)

shows the absorption spectral change for the reactions of

TMP-I and TMTMP-I with 200 equiv of norbornene in

dichloromethane at −50 °C. With the addition of norbornene,

the absorption spectra of TMP-I and TMTMP-I change to

those of their respective initial iron(III) porphyrin complexes.

The time courses of the absorbance for the reactions are

simulated well with a single-exponential function, providing

apparent rate constants. The second-order rate constants were

Table 2. Second-Order Rate Constants for the Hydrogen

Abstraction and Hydroxylation Reactions of TMP-I and

TMTMP-I in Dichloromethane at −50 °C

TMP-I k /M s⁻¹

−

TMTMP-I k /M s⁻¹

−1

substrate

(yield/%)

,4-cyclohexadiene

-methyl-1,4-

cyclohexadiene

5.60 (49)

7.54 (64)

0.489 (18)

1.45 (22)

−

−1 −1

estimated to be 2.20 × 10

for TMP-I and 1.97 ×

1,4-

dihydronaphthalene

3.42

4.64

−

−1 −1

for TMTMP-I from the dependence of the rate

,10-

0.522 (28)

0.461 (<1)

constants on the concentration of norbornene (Table 1 and

Figures S1 and S2). Product analyses of the reaction for TMP-

I and TMTMP-I showed the formation of norbornene oxide in

dihydroanthracene

xanthene

tetraline

1.04 (75)

1.35

−2

2.16 × 10 (16)

2.77 × 10 (1)

6% and 28% yields, respectively. The rate constant for TMP-I

At −30 °C. Ref 36.

is 1 order magnitude greater than that for TMTMP-I. To

investigate whether TMP-I is more reactive than TMTMP-I

for other oleﬁns, we examined the epoxidation reactions of

styrene, 4-methoxystyrene, cyclopentene, cyclohexene, and

cyclooctene (Figures S3 and S4 ). The second-order rate

constants were estimated from the dependence of the apparent

rate constants on the concentration of oleﬁns (Figures S5 −

S14 ) and are summarized in Table 1. As observed for

norbornene, the second-order rate constants of styrene, 4-

methoxystyrene, cyclopentene, and cyclohexene for TMP-I are

larger than those for TMTMP-I, but that of cyclooctene for

TMP-I is almost identical to that for TMTMP-I. For cyclic

oleﬁns, the reaction rate constant of TMP-I is decreased with

an increase in the ring size, but that of TMTMP-I is

unchanged. These results can be explained by the diﬀerence

of the steric eﬀect between TMP-I and TMTMP-I. Since the

meso-position is nearer to the FeO moiety than the pyrrole-β

position, the reaction space around the oxo ligand of TMP-I is

smaller than that of TMTMP-I (see the Discussion section).

The product analyses of these reactions also aﬀorded the

corresponding epoxides in moderate yields (Table 1). The

yields of the epoxides for TMP-I are higher than those for

TMTMP-I. In addition, TMP-I aﬀorded cyclohexene oxide, 2-

cyclohexen-1-ol, and 2-cyclohexen-1-one from cyclohexene,

but TMTMP-I formed only cyclohexene oxide.

order rate constants are estimated from the dependence of the

rate constant on the concentration of these substrates (Figures

S16 −S19 ). The second-order rate constants are summarized in

Table 2. The reaction rate constant decreases with an increase

in the molecular size of the substrate: 1,4-cyclohexadiene <

,4-dihydronapthalene < 9,10-dihydroanthracene.

The absorption spectrum of TMTMP-I also changes with

clear isosbestic points upon the addition of these substrates,

but the absorption spectra of the ﬁnal reaction products are not

identical with that of the initial iron(III) porphyrin complex of

TMTMP (Figure S20). The absorption spectra of the ﬁnal

reaction solutions show a new peak around 780 nm, and the

new peak becomes signiﬁcant for 1,4-dihydronaphthalene and

9,10-dihydroanthracene. This peak was not unchanged when

tetra-n-butylammonium iodide is added to the ﬁnal reaction

solution at −50 °C but changed to the spectrum of an iron(III)

porphyrin complex when the ﬁnal solution was allowed to

warm to room temperature. Product analyses indicate that the

yields of the corresponding aromatic compounds from the

reactions of TMTMP-I with 1,4-cyclohexadiene and 1-methyl-

,4-cyclohexadiene are lower than those of TMP-I, and no

product is obtained from the reaction with 9,10-dihydroan-

thracene (Table 2). The reaction rate constants for the

reactions of TMTMP-I with these substrates were also

estimated using absorption spectroscopy (Figures S21 −S24

and Table 2). The rate constants for TMTMP-I with 1,4-

cyclohexadiene and 1-methyl-1,4-cyclohexadiene are much

smaller than those for TMP-I, but the reaction rates of

TMTMP-I are close to those of TMP-I for more sterically

hindered substrates such as 1,4-dihydronaphthalene and 9,10-

dihydroanthracene.

Hydrogen Abstraction Reactions. We further compared

the reactivity of TMP-I with TMTMP-I for hydrogen

abstraction reactions by using 1,4-cyclohexadiene, 1-methyl-

,4-cyclohexadiene, 1,4-dihydronaphthalene, and 9,10-dihy-

droanthracene at −50 °C. The absorption spectral changes

for the reactions of TMP-I with these substrates are shown in

Figure S15 . As observed for the epoxidation reactions, the

absorption spectra of TMP-I change to those of the initial

iron(III) nitrate complex of TMP with clear isosbestic points

upon the addition of these substrates. Product analyses of these

reactions showed the formation of the corresponding aromatic

compounds in moderate yields (Table 2). The reaction rate

constants for the hydrogen abstraction reactions were also

estimated from time course of the absorbance, and the second-

Hydroxylation Reactions. Previously, we showed that

TMP-I reacts with xanthene and tetraline to aﬀord xanthydrol

and 1,2,3,4-tetrahydro-1-naphthol, respectively. To compare

the reactivity of TMP-I and TMTMP-I for the hydroxylation

reaction, we examined the reactions with xanthene and

tetraline. With the addition of xanthene, the absorption

209

Inorg. Chem. 2021, 60, 3207−3217

Inorganic Chemistry

Article

spectrum of TMP-I changes to that of the initial iron(III)

porphyrin complex with clear isosbestic points, but the

absorption spectrum of TMTMP-I changes to that of the

iron(III) porphyrin complex, concomitant with the formation

of the new peak at around 780 nm at −30 °C (Figure S25).

These spectral changes were close to those observed for the

hydrogen abstraction reactions of 1,4-dihyrdonaphthalene and

Demethylation Reactions of N,N-Dimethylanilines.

We also examined the reactions of TMP-I and TMTMP-I with

N,N-dimethylanilines. The formation of N-methylanilines and

their yields from the reactions were conﬁrmed by HPLC

analysis of the reaction products. The yields of N-methyl-p-

nitroaniline and N-methylaniline are shown in Table 3.

For N,N-dimethyl-p-nitroaniline, whose redox potential is

higher than those of TMP-I and TMTMP-I, the reactions of

TMP-I and TMTMP-I could be followed with an absorption

spectrometer at −50 °C. As observed for other substrates, the

absorption spectra of TMP-I and TMTMP-I change to those

of the corresponding iron(III) porphyrin complexes with clear

isosbestic points (Figure S37). The second-order rate

constants, estimated from the absorption spectral change

(Figures S38 and S39), are shown in Table 3. The reaction rate

of TMP-I is larger than that of TMTMP-I.

,10-dihyrdoanthracene (Figure S20). The second-order rate

constants, estimated from the absorption spectroscopy

Figures S26 −S29 ), are summarized in Table 2. The reaction

rates of TMTMP-I for xanthene and tetraline are close to those

of TMP-I. Product analysis showed that the hydroxylation

products were produced from the reactions of TMP-I,

xanthydrol (75%) from xanthene and 1,2,3,4-tetrahydro-1-

naphthol (16%) from tetraline, but the hydroxylation products

hardly formed from the reactions of TMTMP-I. These results

are the same as the hydrogen abstraction reactions.

For the reactions with N,N-dimethylaniline and N,N-

dimethyl-p-bromoaniline, whose redox potentials are lower

than those of TMP-I and TMTMP-I, we followed the

reactions using a rapid-mixing stopped ﬂow system at −20

°C because the reactions were too fast to follow with an

ordinary absorption spectrometer. The absorption spectral

changes for the reactions of TMP-I and TMTMP-I with N,N-

dimethyl-p -bromoaniline and N,N-dimethylaniline are shown

in Figure S37. The absorption spectral changes show the

formation of the iron(III) porphyrin complex with clear

isosbestic points within 100 ms after mixing of TMP-I with

N,N-dimethyl-p-bromoaniline and N,N-dimethylaniline, and

mixing of TMTMP-I with N,N-dimethyl-p-bromoaniline. No

intermediate is detected in the reactions, indicating that the

initial step, the step from Compound I to Compound II, is the

rate-limiting step for these demethylation reactions. The

second-order rate constants estimated from the absorption

spectral changes for these reactions (Figures S40 −S42) are

summarized in Table 3. However, a reliable second-order rate

constant cannot be estimated for the reaction of TMTMP-I

with N,N-dimethylaniline. The reaction was too fast to follow

even with a stopped ﬂow system, and only the last process of

the reaction could be observed (Figure S37). Moreover, the

absorption spectral change for the reaction of TMTMP-I with

N,N-dimethylaniline is diﬀerent from those of other

demethylation reactions; the spectral change for TMTMP-I

results from the reaction from Compound II to iron(III)

porphyrin, but those for other reactions result from the

reaction from Compound I to iron(III) porphyrin. The rate-

limiting step for the reaction of TMTMP-I with N,N-

dimethylaniline changes to the step from Compound II to

iron(III) porphyrin. Therefore, the estimated reaction rate for

the reaction of TMTMP-I with N,N-dimethylaniline cannot

compare directly with the rates for other reactions. The

reaction rate from Compound I to Compound II for the

reaction of TMTMP-I with N,N-dimethylaniline would be

Previously, we reported a signiﬁcant deuterium kinetic

isotope eﬀect (KIE) for the reaction of TMP-I with

xanthene. The KIE value for xanthene was reported to be

4. To compare the proton transfer process, we further

determined the KIE value for TMTMP-I . The KIE value for

xanthene-d was estimated to be 9.9 (Figure S27), smaller than

that of TMP-I. These KIE values suggest the participation of

H-transfer processes in the rate-limiting steps. The smaller KIE

value for TMTMP-I would be related to the formation of the

complex having a peak at around 780 nm.

Sulfoxidation Reactions. We compared the reactivity of

TMP-I with TMTMP-I for the sulfoxidation reactions of

thioanisole and p-chlorothioanisole at −80 °C. The reactions

were too fast to follow with an absorption spectrometer at −50

C. TMP-I and TMTMP-I react with these sulﬁdes to produce

the corresponding initial iron(III) porphyrin complexes

Figure S30 ). The reaction rate constants of these sulfox-

idation reactions, estimated from the time courses of the

absorbance (Figures S31 −S36), are summarized in Table 3.

Table 3. Second-Order Rate Constants for the Reactions of

TMP-I and TMTMP-I with Sulﬁdes and N,N-

Dimethylanilines in Dichloromethane

TMP-I

TMTMP-I

−1 −1

k₂/M _s−1

−

k /M

substrate

vs SCE

(yield/%)

thioanisole

1.34

90.7

55.6

3.21 × 10

22.3

12.9

1.75 × 10³

p-chlorothioanisole

1.37

tert-butyl

methylsulﬁde

N,N-dimethyl-p-

nitroaniline

N,N-dimethyl-p-

bromoaniline

N,N-dimethylaniline

0.96

4.14 (66)

0.919 (32)

9.97 × 10⁵

very fast (45)

7.33 × 10⁴

0.51

0.57

1.60 × 10 (63)

At −80 °C. At −20 °C. At −50 °C. Ref 37. Ref 38.

−1 −1

much larger than the reaction rate (∼3 × 10 M s )

estimated from the observed spectral change from Compound

II to iron(III) porphyrin.

The sulfoxidation reactions of TMP-I with thioanisole and p-

chlorothioanisole are about 4-fold faster than those of

TMTMP-I. This is consistent with the results of the

epoxidation and hydrogen abstraction reactions for sterically

less-hindered substrates. Furthermore, we examined the

reactions of TMP-I and TMTMP-I with tert-butyl methyl

sulﬁde to investigate the steric eﬀect. As observed for the other

reactions, the reaction of TMTMP-I is faster than that of

TMP-I for tert-butyl methyl sulﬁde (Table 3).

To study the reaction mechanism, we studied the deuterium

KIE for the demethylation reactions. The kinetic data for the

reactions of TMP-I and TMTMP-I with deuterium labeled

N ,N -dimethyl-p-nitroaniline are summarized in Figures S43

and S44 . The estimated KIE values, k /k , for the reactions of

TMP-I and TMTMP-I with N,N-dimethyl-p-nitroaniline were

4.4 and 2.1, respectively. These KIE values suggest the

participation of H-transfer processes in the rate-limiting steps.

210

Inorg. Chem. 2021, 60, 3207−3217

Inorganic Chemistry

Article

As observed for the hydroxylation reaction, the KIE value for

TMTMP-I was smaller than that for TMP-I and can be related

to the formation of the complex having the peak at around 780

nm.

Activation Parameters. In order to obtain information on

the transition states of the reactions of TMP-I and TMTMP-I,

we estimated the activation parameters for the hydrogen

abstraction reaction of 1,4-cyclohexadiene from the reaction

rates in the temperature range from −50 to −80 °C. Eyring

plots for these reactions are shown in Figure 3 , and the

Figure 4. Comparison of orbital energies of TMP-I and TMTMP-I

estimated by the DFT calculation. The average of the orbital energies

for the α-spin and β-spin are used for the orbital energies of occupied

orbitals.

and porphyrin π orbitals. The unpaired electrons of porphyrin

ligands in TMP-I and TMTMP-I are calculated to occupy the

a and a orbitals, respectively (Figure S46 ). These results are

Figure 3. Eyring plots for the reactions of TMP-I (blue) and

TMTMP-I (red) with 1,4-cyclohexadiene in dichloromethane. The

solid lines indicate the simulation of the least-squares ﬁts with a linear

function.

consistent with the previous spectroscopic studies, indicating

16,39

the accuracy of the present calculations.

The energy of the

^a2_uorbital for TMP-I is just slightly lower than that of the a₁_u

orbital for TMTMP-I. This is consistent with the previous

result that the redox potential of TMP-I is almost the same as

that of TMTMP-I. On the other hand, the calculated energies

of the iron d orbitals for TMP-I are eventually lower than

those of TMTMP-I (Figure 4 and Table S2). Since the

energies of the iron d orbitals are modulated by the energies of

the occupied porphyrin orbitals interacting with these iron d

orbitals, these results suggest that the energies of the occupied

porphyrin orbitals for TMP are lower than those for TMTMP.

This is conﬁrmed by the DFT calculations of free bases of

TMP and TMTMP (Table S3). The calculated energies of the

occupied orbitals, interacting with the iron d orbitals, for the

free base of TMTMP are higher than those for TMP. The

diﬀerence of the energies of these porphyrin orbitals would

result from the diﬀerence of the electron densities between the

meso-position and the pyrrole-β-position. When signiﬁcant

electron density is present at the substituted position, the

porphyrin orbital is destabilized due to the electronic repulsion

between the porphyrin and the substituent. This is conﬁrmed

by the electron distribution of these porphyrin orbitals shown

in Table S3.

⧧

enthalpy of activation (ΔH ) and the entropy of activation

⧧

−1

(

ΔS ) calculated from the Eyring plots are 17.7 kJ mol and

−

−1

−

148.8 J K mol for TMP-I and 21.2 kJ mol and −154.4 J

1 −1

−

⧧

mol for TMTMP-I, respectively. The ΔH value for

−1

TMP-I is 3.5 kJ mol smaller than that for TMTMP-I, and

⧧

−1

the |ΔS | value for TMP-I is 6.0 J K mol smaller than for

⧧

TMTMP-I. The negative ΔS values indicate the associative

rate-limiting step for the hydrogen abstraction reaction;

Compound I and the substrate tightly bind in the transition

⧧

state. The ΔS values for the hydrogen abstraction reactions

are larger than those for the epoxidation reactions but close to

⧧

those of hydroxylation reactions. The −T ΔS values at a

⧧

temperature over 140 K are larger than the ΔH values,

indicating the entropy control for the hydrogen abstraction

⧧

reaction. The free energy of activation (ΔG ) for TMP-I is

smaller than that for TMTMP-I, indicating that TMP-I is

more reactive than TMTMP-I at any reaction temperature.

DFT Calculations. To explain the diﬀerence of the

reactivity between TMP-I and TMTMP-I, we performed

density functional theory (DFT) calculations. Details of the

DFT calculations are described in the Experimental Section.

Optimized structures of the model complexes and their bond

lengths and bond angles are shown in Figure S45 and Table S1 ,

respectively. Figure 4 shows energies of molecular orbitals near

the HOMO and iron d orbitals. The energies of these orbitals

are summarized in Tables S2 and S3 . These model complexes

have lower symmetry than the D₄_hbecause of the substitution

of the porphyrin ring and the coordination of the axial ligand,

but we here denote the molecular orbitals with symmetry

labels of the D₄_hsymmetry. The ground states of the model

complexes for TMP-I and TMTMP-I are calculated to be the

quartet (S = 3/2) state having unpaired electrons in d , d ,

DISCUSSION

■

Comparison of Reactivity between TMP-I and

TMTMP-I. To compare the reactivity between TMP-I and

TMTMP-I for various substrates studied here, we calculated

logarithms of the values for the second-order rate constants of

TMP-I over those of TMTMP-I, log(k

). The

TMP‑I TMTMP‑I

calculated results are summarized in Figures 5, 7, and 8. The

log(k /k ) value becomes positive when TMP-I is

TMP‑I TMTMP‑I

more reactive than TMTMP-I, but the value becomes negative

when TMTMP-I is more reactive than TMP-I. As shown in

211

Inorg. Chem. 2021, 60, 3207−3217

Inorganic Chemistry

abstraction and sulfoxidation reactions. As shown in Figures 7

Article

The above discussion can also be applicable to the hydrogen

Figure 5. Comparison of the reactivity of TMP-I with TMTMP-I for

the epoxidation reaction. The numbers in the ﬁgure are log(k

TMP‑I

k ) values.

TMTMP‑I

Figure 7. Comparison of the reactivity of TMP-I with TMTMP-I for

the hydrogen abstraction reaction and hydroxylation reaction. The

Figure 5, the log(k

reactions are positive except for cyclooctene (almost 0). As the

size of the cyclic oleﬁn is larger, the log(k /k ) value

) values for the epoxidation

TMP‑I TMTMP‑I

numbers in the ﬁgure are log(k

) values.

TMP‑I TMTMP‑I

for cyclic oleﬁn becomes smaller. These results would be

explained by the diﬀerence of the steric eﬀect around the oxo

ligand between TMP-I and TMTMP-I (Figure 6 ). Since the o-

Figure 6. CPK models of the optimized structures of TMP-I (left)

and TMTMP-I (right) by the DFT calculations. The distances

between the o-methyl carbon atoms facing each other in TMP-I and

TMTMP-I are 9.75 and 11.53 Å, respectively.

Figure 8. Comparison of the reactivity of TMP-I with TMTMP-I for

the sulfoxidation reaction (yellow bars) and the demethylation

reaction (blue bars). The numbers in the ﬁgure are log(k

TMP‑I

) values.

TMTMP‑I

methyl group of the mesityl group at the meso-position in

TMP-I is closer to the oxo ligand than that at the pyrrole β-

position in TMTMP-I, the reaction space around the oxo

ligand of TMP-I is smaller than that of TMTMP-I. When the

alkene moiety of the substrate is not sterically bulky, the

diﬀerence of the steric eﬀect hardly aﬀects the reaction rate.

and 8, the log(k

abstraction and sulfoxidation reactions indicate positive values

when the substrates are less hindered. However, the log-

) values for the hydrogen

TMP‑I TMTMP‑I

) values for the sterically hindered substrates

TMP‑I TMTMP‑I

such as 1,4-hydronapthalene and 9,10-dihydroanthracene

indicate values close to 0, and the large negative value is

obtained for tert-butyl methyl sulﬁde. These results clearly

indicate that TMP-I is intrinsically more reactive than

TMTMP-I even for the hydrogen abstraction and sulfoxidation

reactions although the steric eﬀect of TMP-I is stronger than

that of TMTMP-I. This is also true for the hydroxylation

Therefore, for such a substrate, the log(k

) value

TMP‑I TMTMP‑I

reﬂects the diﬀerence of the intrinsic reactivity between TMP-I

and TMTMP-I; TMP-I is more reactive than TMTMP-I for

the epoxidation reaction. As the substrate is bulkier, the steric

eﬀect resulting from the o-methyl group of TMP-I becomes

more signiﬁcant than that of TMTMP-I. This is conﬁrmed by

the reaction rates in Table 1. The reaction rate of TMP-I for

cyclic oleﬁns drastically decreases with an increase in the ring

size, while those of TMTMP-I do not. All of these results

indicate that TMP-I is intrinsically more reactive than

TMTMP-I for the epoxidation reaction although the steric

eﬀect of TMP-I is stronger than that of TMTMP-I.

reactions. The log(k

) values are close to 0 for

TMP‑I TMTMP‑I

these substrates. Since the structures of tetraline and xanthene

are close to those of 1,4-hydronapthalene and 9,10-

dihydroanthracene, respectively, these values mainly reﬂect

the diﬀerence of the steric eﬀect, rather than the intrinsic

reactivity, between TMP-I and TMTMP-I.

212

Inorg. Chem. 2021, 60, 3207−3217

Inorganic Chemistry

Article

For the demethylation reactions of N,N-dimethylanilines, a

Compound I in many groups. For example, Groves et al.

proposed the rebound mechanism for the hydroxylation

reaction, in which the hydrogen atom is abstracted from a

substrate by the oxo ligand of Compound I, followed by the

rebound between the formed iron-bound OH group and the

positive log(k

dimethyl-p-nitroaniline, while a negative log(k

) value is observed for N,N-

TMP‑I TMTMP‑I

)

TMP‑I TMTMP‑I

value is observed for N,N-dimethyl-p-bromoaniline (Figure 8).

The value for N,N-dimethylaniline would also be negative

because the reaction of TMTMP-I is faster than that of TMP-

I. Since these substrates are not sterically bulky, these results

cannot be explained by the steric eﬀect as discussed above.

These values can be rationalized with the reaction mechanism

of the demethylation reactions of these N,N-dimethylanilines.

Previous studies of the demethylation reactions of N,N-

dimethylanilines proposed that the reaction proceeds via the

single electron transfer process followed by the H-atom

40,41

substrate radical.

The large hydrogen/deuterium kinetic

isotope eﬀects for the hydroxylation reactions indicate the

participation of the hydrogen abstraction process in the rate-

limiting step; the hydrogen atom (one electron−one proton)

transfer process is involved in the rate-limiting step of the

36,42−49

hydroxylation reaction.

This rate-limiting step can be

applicable to the hydrogen abstraction reaction. For the

epoxidation reaction, the rate-limiting steps of these mecha-

nisms also involve the electron transfer process and the O−C

bond formation process in some manner although many

29,38

(

hydrogen or proton) transfer process.

When the redox

potential of a N,N-dimethylaniline is lower than that of

Compound I (TMP-I, 0.88 V; TMTMP-I, 0.88 V), the rate-

limiting step of the reaction is the electron transfer process

from a N,N-dimethylaniline to Compound I. However, when

the redox potential of a N,N-dimethylaniline is higher than that

of Compound I, the rate-limiting step involves the H-atom

transfer process. Therefore, the rate-limiting step would be a

single electron transfer reaction for N,N-dimethylaniline and

N,N-dimethyl-p-bromoaniline, but it would be the H-atom

transfer reaction for N,N-dimethyl-p-nitroaniline. The present

KIE values for N,N-dimethyl-p-nitroaniline support this

7,32,45,50−52

reaction mechanisms have been proposed.

previous study for the sulfoxidation reaction with TMP-I

proposed the direct oxygen atom transfer mechanism.

Therefore, the O−S bond formation process and the electron

transfer process must be involved in the rate-limiting step. As

mentioned in a previous paragraph, the demethylation reaction

of N,N-dimethylaniline proceeds via the single electron transfer

process followed by the H-atom (hydrogen or proton) transfer

process, and the nature of the rate-limiting step is changed by

29,38

the redox potential of the aniline.

Considering a recent

reaction mechanism. The negative log(k

for N,N-dimethyl-p-bromoaniline and N,N-dimethylaniline are

consistent with our recent result that TMTMP-I is superior to

TMP-I for the single electron transfer reaction. The positive

value for N,N-dimethyl-p-nitroaniline is also reasonable with

the present result that TMP-I is superior to TMTMP-I for the

hydrogen abstraction reaction.

) values

study on the single electron transfer reaction and the reaction

mechanisms discussed above, we conclude that TMP-I is

intrinsically much more reactive than TMTMP-I when the

bond formation process with the oxo ligand is involved in the

rate-limiting step while TMTMP-I is intrinsically more reactive

than TMP-I when only the single electron transfer process is

involved in the rate-limiting step (Scheme 1).

TMP‑I TMTMP‑I

The absorption spectral change for the reaction of TMTMP-

I with N,N-dimethylaniline suggests the change of the rate-

limiting step of the reaction from the electron transfer process

to the H-atom transfer process because the redox potential of

N,N-dimethylaniline is lower than that of TMTMP-I. The rate

constant for the electron transfer reaction of TMTMP-I with

Scheme 1. Reaction Mechanisms Proposed for the Single

Electron Transfer Reaction and the Atom Transfer

Reactions

−1 −1

N,N-dimethylaniline can be estimated to be 1 × 10 M

from the redox potential of N,N-dimethylaniline and a Marcus

plot for TMTMP-I. The estimated rate supports the change

of the rate-limiting step because this rate is much larger than

the rate of the demethylation reaction.

All of the present results indicate that TMP-I is intrinsically

much more reactive than TMTMP-I for the epoxidation,

hydrogen abstraction, hydroxylation, sulfoxidation, and deme-

thylation reactions while the steric eﬀect of the mesityl group

in TMP-I is intrinsically more signiﬁcant than that in

TMTMP-I. This conclusion seems to conﬂict with our

previous proposal that the reactivity of TMP-I is comparative

to that of TMTMP-I. We deduced this proposal from the

absorption spectral change, which showed that both TMP-I

and TMTMP-I react with cyclohexene under the competitive

reaction in the TMP-I and TMTMP-I mixture. This result is

also supported by the present individual kinetic analysis (Table

The hydrogen abstraction reaction and hydroxylation

reactions of TMTMP-I produced a new compound having a

peak at around 780 nm. This can be seen in Figures S4, S20,

and S25 . This compound would be an intermediate that kills or

inhibits the hydrogen atom transfer and hydroxylation

reactions because the product yields from the reactions are

always low when it is observed. The identity of this compound

is not clear at present, and further spectroscopic study is

required to clarify.

), where the reaction rate constant of TMP-I is close to that

Factors Controlling the Reactivity of Compound I.

Numerous previous studies revealed that the redox potential of

Compound I is an essential factor for determining the

of TMTMP-I for sterically hindered substrates such as

cyclohexene and cyclooctene. As discussed in the above

paragraphs, the apparent reactivity of TMP-I for cyclohexene

accidentally coincides with that of TMTMP-I due to a

decrease in the reactivity of TMP-I by the steric eﬀect.

The reaction mechanisms of Compound I with various

substrates have been studied to understand the reactivity of

,3,27,28,32

reactivity.

This is quite reasonable because the

electron transfer process is involved in the rate-limiting step

of most reactions of Compound I. Linear relationships

between the activation energy and the redox potential of

27,28,32

high-valent oxo porphyrin species have been reported.

213

Inorg. Chem. 2021, 60, 3207−3217

Inorganic Chemistry

Article

However, the redox potential is not the factor for

discriminating the intrinsic reactivity of TMP-I and

TMTMP-I because the redox potentials of these complexes

π orbitals, which further changes the energies of the iron d

orbitals, and ﬁnally the reactivity of Compound I.

Chemical and Biological Relevance. This study clearly

shows that the meso-substituted Compound I is more reactive

than the pyrrole-β-substituted Compound I for the O-atom

and H-atom transfer reactions, but this reactivity is reversed for

the single electron transfer reaction. Although further study is

needed to support the present results, this study suggests that

the meso-substituted heme is more suitable than the pyrrole β-

substituted heme to develop a superior heme catalyst. The high

reactivity of the meso-substituted heme would make the

oxidation of a more inactive substrate possible. Since the meso-

position of the heme tends to be more susceptible than the

pyrrole β-position for the substitution reaction, as can be seen

in the heme oxygenase reaction and the nitration reaction of

are approximately the same.

One of the signiﬁcant diﬀerences between TMP-I and

TMTMP-I is the position of the substituent: the meso-

substituted porphyrin or the pyrrole β-substituted porphyrin.

This structural diﬀerence changes the steric eﬀect around the

oxo ligand, as discussed in the previous paragraph. However,

the steric eﬀect is not the factor for determining the intrinsic

reactivity because of the opposite trend to the reactivity. This

structural diﬀerence also changes the ﬂexibility of the

porphyrin ligand. TMTMP is more ﬂexible than TMP because

of the absence of the substituent at the meso-position. Since the

meso-position is nearer to the iron center of heme than the

pyrrole-β position, the o-methyl group of the meso-phenyl

group at the meso-position in TMP induces steric repulsion

with the porphyrin plane when the porphyrin plane is

deformed, but that of the pyrrole-β position does not. We

previously proposed that the ﬂexibility of the porphyrin ring of

TMTMP makes the structural change in the single electron

transfer reaction easy, decreasing the reorganization energy (λ)

3,54

heme,

the meso-substituted heme would be more

sustainable than the pyrrole-β-substituted heme for the heme

degradation reaction in the catalytic reactions. Moreover, the

meso-substituted heme is easier to synthesize than the pyrrole

β-substituted heme, facilitating the modiﬁcation and/or

functionalization of heme.

Most heme enzymes in nature have protoporphyrin IX or its

derivatives, which can be classiﬁed into the pyrrole-β-

substituted porphyrin, in their active sites. Therefore, the

intrinsic reactivity of Compound I in native heme enzymes

would be close to that of TMTMP-I, rather than TMP-I. The

pyrrole-β-substituted structures of the hemes are favorable for

peroxidases because the Compounds I in peroxidases catalyze

the single electron transfer reaction from the substrate to

and increasing the electron transfer rate. The diﬀerence of

the reorganization energy between TMP-I and TMTMP-I is

also conﬁrmed by the DFT calculations. The ﬂexibility of the

porphyrin ligand would aﬀect the reactivity of Compound I for

the O-atom and H-atom transfer reactions. As expected from

⧧

the large negative value of ΔS for the reaction of Compound

I, TMP-I and TMTMP-I tightly interact with a substrate in the

transition state. The ﬂexible porphyrin ring of TMTMP would

be unfavorable to have the tight transition state, and the

substrate should come closer to the oxo ligand of TMTMP-I

than TMP-I in the transition state. This is also supported by

Compound I. On the other hand, the pyrrole-β-substituted

structures of the hemes in cytochromes P450 and catalases are

unfavorable because compounds I of these enzymes catalyze

5−57

the O-atom and H-atom transfer reactions.

This

⧧

the larger ΔH value for TMTMP-I. We propose that the

disadvantage would be compensated by other inherent factors

in their heme enzymes. The ﬁrst one is the axial ligand;

cytochrome P450 and catalase have a thiolate axial ligand from

cysteine and a phenolate axial ligand from tyrosine,

respectively. The σ-donor ability of these axial ligands is

stronger than the axial ligand of peroxidase (histidine), because

diﬀerence of the ﬂexibility of the porphyrin plane makes TMP-

I more reactive for the atom transfer reaction and TMTMP-I

more reactive for the single electron transfer reaction.

The diﬀerence of the energies of the porphyrin σ and π

orbitals would also be another possible factor for changing the

reactivity. The present DFT calculations show a signiﬁcant

diﬀerence in the energies of the molecular orbitals around the

iron-oxo moiety between TMP-I and TMTMP-I (see Figure

the pK

imidazole. Green et al. proposed that the strong binding of the

thiolate axial ligand in cytochrome P450 increases the pK of

values of thiol and phenol are larger than that of

). The energies of the iron d orbitals for TMP-I are lower

the oxo ligand of Compound II, enhancing the H-atom

abstraction ability. We revealed in a previous study using

synthetic Compound I model complexes that, as the binding of

the axial ligand is stronger, Compound I changes to being

more reactive for the O-atom and H-atom transfer reactions.

Moreover, the interactions of the heme with amino acid

residues in the heme pocket would also assist in keeping the

reactivity very high. The ﬂexibility of the heme is decreased by

the interactions of the heme in the heme pockets, such as

hydrogen bonds and steric interaction. The binding site of a

substrate, which ﬁxes the substrate not to ﬂuctuate and to be

placed near the oxo ligand of Compound I, also assists in

stabilizing the transition state.

than those for TMTMP-I. As discussed in a previous

paragraph, the atom transfer reaction involves the electron

transfer process and the bond formation process in the rate-

limiting step. Since TMTMP-I is faster than TMP-I for the

single electron transfer process, the reactivity for the atom

transfer reaction should be controlled by the bond formation

process. The bond formation occurs at the oxo ligand of

Compound I, whose orbital is linked to the iron d orbital.

Therefore, the stabilization of the iron d orbital would facilitate

the bond formation process in the reaction of TMP-I,

decreasing the activation energy. Since the energies of the

iron d orbitals are determined by the orbital energies of the

ligands coordinating to the iron, the porphyrin σ and π orbitals

interacting with the iron d orbitals should be related to the

diﬀerence of the reactivity between TMP-I and TMTMP-I.

This is conﬁrmed by the DFT calculations (Table S3 ). The

energies of the porphyrin σ and π orbitals, interacting with the

iron d orbitals, for TMP are more stable than those for

TMTMP. After all, the diﬀerence in the substituted position of

the porphyrin ring changes the energies of the porphyrin σ and

EXPERIMENTAL SECTION

■

Instrumentation. UV−vis absorption spectra were recorded on

an Agilent 8454 spectrometer (Agilent Technologies) equipped with a

USP-203 low-temperature chamber (UNISOKU). H NMR spectra

were measured on a Lambda-400 spectrometer (JEOL). The chemical

shifts were referenced to the residual peaks of the deuterated solvents

chloroform (7.24 ppm) and toluene−CH₃(2.09 ppm). The

214

Inorg. Chem. 2021, 60, 3207−3217

Inorganic Chemistry

The Supporting Information is available free of charge at

Article

concentrations of the NMR samples were 1−3 mM. The reaction

product analyses were performed using a GC-MS QP-2010 SE gas

chromatograph mass spectrometer (Shimadzu) or high-performance

liquid chromatography (HPLC) LC-10AT system with a multidiode

array detector (Shimadzu), attached to an octadodecylsilyl (ODS)

column, 4.6 mm × 250 mm (SHISEIDO). Ozone gas was generated

by the UV irradiation of oxygen gas (99.999%) with a PR-1300 ozone

generator (Clear Water) and used without further puriﬁcation. The

kinetic analyses for fast reactions were performed using a stopped-ﬂow

rapid mixing system USP-R100 (UNISOKU) with a low-temperature

double mixer USP-SFM-CRD10 (UNISOKU).

DFT Calculations. The DFT calculations were performed using

the Gaussion09 program package. All basis sets were taken from the

Gaussian09 program. We used two combinations of the DFT

functional and basis set. LC-BLYP was used with cc-pVTZ for the

OFeCl and the C and N atoms in porphyrin ring and sto-6g for

the porphyrin substituents (mesityl group and methyl group). All

compounds were fully optimized in the ground state of each spin

multiplicity.

ASSOCIATED CONTENT

sı Supporting Information

https://pubs.acs.org/doi/10.1021/acs.inorgchem.0c03548.

■

Materials. Anhydrous organic solvents were obtained commer-

cially and stored in a glovebox. Dichloromethane was puriﬁed by

passing through an alumina column just before use in the glovebox.

Other chemicals were purchased commercially and used without

further puriﬁcation. TMP and TMTMP were synthesized by reported

Kinetic analyses and data from DFT calculations and

additional data and ﬁgures including absorption spectral

changes, optimized structures, spin distribution, bond

lengths and angles, and molecular orbital energies

methods. Syntheses of iron(III) complexes of TMP and TMTMP

were described in a previous report. Substrates for the reactions are

puriﬁed by passing through active alumina just before use. Deuterium

labeled N,N-dimethyl-p-nitroaniline was prepared from the reaction of

(PDF )

p-nitroaniline with deuterium labeled methyl iodide (99.5% D).

Kinetic Analysis: General Procedure. The ferric porphyrin

complex (100 μM) in dichloromethane was prepared in a 1 cm quartz

cuvette in a low-temperature chamber set on a UV−vis absorption

spectrometer. After stabilization of the temperature of the solution at

AUTHOR INFORMATION

Corresponding Author

■

Hiroshi Fujii − Department of Chemistry, Graduate School of

Humanities and Sciences, Nara Women’s University,

Kitauoyanishi, Nara 630-8506, Japan; orcid.org/0000-

0003-4611-2983 ; Email: fujii@cc.nara-wu.ac.jp

−

50 °C, O gas was slowly bubbled into the solution with a gastight

syringe. The oxidation process was monitored using the absorption

spectrometer. After conﬁrming the generation of the oxoiron(IV)

porphyrin π-cation radical complex, the excess O gas in the cell was

Authors

removed by bubbling argon gas with a gastight syringe. An excess of

substrate (more than 10 equiv) was then added to the solution with

stirring, and the progress of the reaction was followed at a constant

time interval using absorption spectroscopy. The reaction rate

constants were determined by simulation of the time course of the

absorbance of the peaks with the commercial program Igor Pro 7

Nami Fukui − Department of Chemistry, Graduate School of

Humanities and Sciences, Nara Women’s University,

Kitauoyanishi, Nara 630-8506, Japan

Kanako Ueno − Department of Chemistry, Graduate School of

Humanities and Sciences, Nara Women’s University,

Kitauoyanishi, Nara 630-8506, Japan

(

WaveMetrics). The absorption spectral change followed the ﬁrst-

order kinetics in the presence of a large excess of the substrate, and

Masahiko Hada − Department of Chemistry, Graduate School

of Science, Tokyo Metropolitan University, Hachioji 192-

the second-order reaction rate constants (k ) were determined from

the linear dependence of the pseudo-ﬁrst-order rate constant on the

concentration of the substrate.

0397, Japan; orcid.org/0000-0003-2752-2442

Product Analysis: General/Typical Procedure. The ferric

porphyrin complex (1.0 mM) in dichloromethane (0.22 mL) was

prepared in a Schlenk ﬂask and cooled to −50 °C in a temperature-

controlled cooling bath. Ozone gas was slowly bubbled into the

solution until the formation of the Compound I model complexes.

The excess ozone gas was removed by the freeze−thaw method.

Excess substrate (10 equiv) dissolved in dichloromethane was added

to the solution at −50 °C. After completion of the reactions (∼30

min), the reaction products and their yields were analyzed with GC-

MS. The assignments of the products were performed on the basis of

the retention times and the mass numbers of the authentic samples.

The product yields of N-methylaniline and N-methyl-p-nitroaniline

from the demethylation reactions of N,N-dimethylaniline and N,N-

dimethyl-p-nitroaniline were analyzed by HPLC with an ODS column

Complete contact information is available at:

https://pubs.acs.org/10.1021/acs.inorgchem.0c03548

Notes

The authors declare no competing ﬁnancial interest.

ACKNOWLEDGMENTS

This study was supported by grants from JSPS (Grant

7H03032 and 19H04581) and CREST.

■

REFERENCES

■

1) Huang, X.; Groves, J. T. Oxygen Activation and Radical

Transformations in Heme Proteins and Metalloporphyrins. Chem.

(

4.6 mm × 250 mm) using an acetonitrile−water (8:2) mixture as an

eluent.

Stopped-Flow Rapid Mixing Experiments: General Proce-

Rev. 2018, 118, 2491−2553.

(2) Baglia, R. A.; Zaragoza, J. P. T.; Goldberg, D. P. Biomimetic

dure. The iron(III) complex of TMP or TMTMP (∼50 μM) was

prepared in dichloromethane. The solution of iron(III) porphyrin and

m-chloroperoxybenzoic acid (5 equiv) was placed in sample holders

of a stopped ﬂow and cooled at −20 °C. After cooling for 1 min, the

solutions were rapidly mixed. After aging the mixture until the

formation of Compound I (10−60 s), a solution of substrate (N,N-

dimethylanilines or tert-butyl methyl sulﬁde) was further rapidly

mixed to initiate the reaction. The reactions were monitored by the

absorption spectral change at constant time intervals. The reaction

rate constants were determined by simulation of the time course of

the absorbance of the peaks. The second-order reaction rate constant

Reactivity of Oxygen-Derived Manganese and Iron Porphyrinoid

Complexes. Chem. Rev. 2017 , 117 (21), 13320 −13352.

(3) Nam, W. High-Valent Iron(IV)−Oxo Complexes of Heme and

Non-Heme Ligands in Oxygenation Reactions. Acc. Chem. Res. 2007,

40, 522−531.

(4) Rittle, J.; Green, M. T. Cytochrome P450 compound I: capture,

characterization, and C-H bond activation kinetics. Science 2010, 330,

933−937.

(5) Fujii, H. Electronic structure and reactivity of high-valent oxo

iron porphyrins. Coord. Chem. Rev. 2002, 226, 51−60.

(6) Fujii, H., Model Complexes of Heme Peroxidases. In Heme

Peroxidases; Raven, E.; Dunford, B., Eds.; The Royal Society of

Chemistry: London, 2016; pp 183−217.

(

k₂) were determined from the linear dependence of the pseudo-ﬁrst-

order rate constant on the concentration of the substrate.

215

Inorg. Chem. 2021, 60, 3207−3217

Inorganic Chemistry

Tsui, W. M.; Zhou, X. G.; Zhou, Z. Y.; Zhu, N.; Chang, C. K.

Article

(

7) Rohde, J. U.; In, J. H.; Lim, M. H.; Brennessel, W. W.; Bukowski,

(25) Che, C. M.; Zhang, J. L.; Zhang, R.; Huang, J. S.; Lai, T. S.;

Hydrocarbon oxidation by beta-halogenated dioxoruthenium(VI)

porphyrin complexes: effect of reduction potential (RuVI/V) and

C-H bond-dissociation energy on rate constants. Chem. - Eur. J. 2005,

11 (23), 7040−53.

M. R.; Stubna, A.; Munck, E.; Nam, W.; Que, L., Jr. Crystallographic

and spectroscopic characterization of a nonheme Fe(IV)-O complex.

Science 2003, 299 (5609), 1037−9.

(

8) Seo, M. S.; Kim, N. H.; Cho, K.-B.; So, J. E.; Park, S. K.;

Clémancey, M.; Garcia-Serres, R.; Latour, J.-M.; Shaik, S.; Nam, W. A

mononuclear nonheme iron(iv)-oxo complex which is more reactive

than cytochrome P450 model compound I. Chem. Sci. 2011, 2 (6),

(26) Porhiel, E.; Bondon, A.; Leroy, J. (β -Octafluoro-meso-

tetraarylporphyrin)manganese Complexes: Synthesis, Characteriza-

tion and Catalytic Behaviour in Monooxygenation Reactions. Eur. J.

Inorg. Chem. 2000, 2000 (5), 1097−1105.

039.

9) Sono, M.; Roach, M. P.; Coulter, E. D.; Dawson, J. H. Heme-

Containing Oxygenases. Chem. Rev. 1996 , 96 (7), 2841 −2888.

10) Rittle, J.; Green, M. T. Cytochrome P450 compound I: capture,

characterization, and C-H bond activation kinetics. Science 2010, 330

6006), 933−7.

11) Takahashi, A.; Kurahashi, T.; Fujii, H. Effect of Imidazole and

(27) Garrison, J. M.; Ostovic, D.; Bruice, T. C. Is a linear

relationship between the free energies of activation and one-electron

oxidation potential evidence for one-electron transfer being rate

determining? Intermediates in the epoxidation of alkenes by

cytochrome P-450 models. 4. Epoxidation of a series of alkenes by

oxo(meso-tetrakis(2,6-dibromophenyl)porphinato)chromium(V). J.

Am. Chem. Soc. 1989, 111 (13), 4960 −4966.

Phenolate Axial Ligands on the Electronic Structure and Reactivity of

Oxoiron(IV) Porphyrin π -Cation Radical Complexes: Drastic

Increase in Oxo-Transfer and Hydrogen Abstraction Reactivities.

Inorg. Chem. 2009, 48, 2614−2625.

(28) Asaka, M.; Fujii, H. Participation of Electron Transfer Process

in Rate-Limiting Step of Aromatic Hydroxylation Reactions by

Compound I Models of Heme Enzymes. J. Am. Chem. Soc. 2016, 138

(12) Takahashi, A.; Yamaki, D.; Ikemura, K.; Kurahashi, T.; Ogura,

(

26), 8048−51.

T.; Hada, M.; Fujii, H. Effect of the axial ligand on the reactivity of the

oxoiron(IV) porphyrin π -cation radical complex: higher stabilization

of the product state relative to the reactant state. Inorg. Chem. 2012,

29) Takahashi, A.; Kurahashi, T.; Fujii, H. Redox potentials of

oxoiron(IV) porphyrin π -cation radical complexes: participation of

electron transfer process in oxygenation reactions. Inorg. Chem. 2011,

1 (13), 7296−305.

13) Prokop, K. A.; de Visser, S. P.; Goldberg, D. P. Unprecedented

0, 6922−8.

30) Bhyrappa, P.; Young, J. K.; Moore, J. S.; Suslick, K. S.

rate enhancements of hydrogen-atom transfer to a manganese(V)-oxo

corrolazine complex. Angew. Chem., Int. Ed. 2010 , 49 (30), 5091 −5.

hemins with electronegative substituents: efficient catalysts for

hydroxylation and epoxidation. J. Chem. Soc., Chem. Commun. 1984,

Dendrimer-Metalloporphyrins: Synthesis and Catalysis. J. Am. Chem.

Soc. 1996, 118 (24), 5708−5711.

14) Traylor, P. S.; Dolphin, D.; Traylor, T. G. Sterically protected

(31) Groves, J. T.; Viski, P. Asymmetric hydroxylation by a chiral

iron porphyrin. J. Am. Chem. Soc. 1989, 111 (22), 8537−8538.

(32) Sainna, M. A.; Kumar, S.; Kumar, D.; Fornarini, S.; Crestoni,

(

79−280.

15) Gold, A.; Jayaraj, K.; Doppelt, P.; Weiss, R.; Chottard, G.; Bill,

E.; Ding, X.; Trautwein, A. X. Oxoferryl complexes of the halogenated

porphinato)iron catalyst [tetrakis(2,6-dichlorophenyl)porphinato]-

iron. J. Am. Chem. Soc. 1988 , 110 (17), 5756 −5761.

16) Fujii, H. Effects of the electron-withdrawing power of

M. E.; de Visser, S. P. A comprehensive test set of epoxidation rate

constants for iron(iv)-oxo porphyrin cation radical complexes. Chem.

Sci. 2015, 6 (2), 1516−1529.

(33) Groves, J. T.; Haushalter, R. C.; Nakamura, M.; Nemo, T. E.;

Evans, B. J. High-valent iron-porphyrin complexes related to

peroxidase and cytochrome P-450. J. Am. Chem. Soc. 1981, 103

substituents on the electronic structure and reactivity in oxoiron(IV)

porphyrin π -cation radical complexes. J. Am. Chem. Soc. 1993, 115

11), 4641−4648.

34) Fujii, H.; Ichikawa, K. Preparation and characterization of an

10), 2884−2886.

A1u Oxoiron(IV) porphyrin π -cation-radical complex. Inorg. Chem.

992, 31 (6), 1110−1112.

35) Fukui, N.; Li, X. X.; Nam, W.; Fukuzumi, S.; Fujii, H. Small

17) Goh, Y. M.; Nam, W. Significant Electronic Effect of Porphyrin

Ligand on the Reactivities of High-Valent Iron(IV) Oxo Porphyrin

Reorganization Energy for Ligand-Centered Electron-Transfer Re-

Cation Radical Complexes. Inorg. Chem. 1999 , 38 (5), 914 −920.

duction of Compound I to Compound II in a Heme Model Study.

Inorg. Chem. 2019, 58 (13), 8263−8266.

(18) Collins, T. J.; Ryabov, A. D. Targeting of High-Valent Iron-

TAML Activators at Hydrocarbons and Beyond. Chem. Rev. 2017,

17 (13), 9140−9162.

19) Ishimizu, Y.; Ma, Z.; Hada, M.; Fujii, H. Experimental and

(36) Cong, Z.; Kinemuchi, H.; Kurahashi, T.; Fujii, H. Factors

Affecting Hydrogen-Tunneling Contribution in Hydroxylation

Reactions Promoted by Oxoiron(IV) Porphyrin π -Cation Radical

Complexes. Inorg. Chem. 2014, 53 (19), 10632−41.

theoretical studies of the porphyrin ligand effect on the electronic

structure and reactivity of oxoiron(iv) porphyrin π -cation-radical

complexes. JBIC, J. Biol. Inorg. Chem. 2019 , 24 (4), 483 −494.

(37) Goto, Y.; Matsui, T.; Ozaki, S.; Watanabe, W.; Fukuzumi, S.

(20) Afanasiev, P.; Sorokin, A. B. μ -Nitrido Diiron Macrocyclic

Mechanisms of Sulfoxidation Catalyzed by High-Valent Intermediates

of Heme Enzymes: Electron-Transfer vs Oxygen-Transfer Mecha-

nism. J. Am. Chem. Soc. 1999, 121, 9497−9502.

Platform: Particular Structure for Particular Catalysis. Acc. Chem. Res.

016, 49 (4), 583−93.

21) Kumar, D.; Tahsini, L.; de Visser, S. P.; Kang, H. Y.; Kim, S. J.;

(

(38) Goto, Y.; Watanabe, Y.; Fukuzumi, S.; Jones, J. P.; Dinnocenzo,

Nam, W. Effect of porphyrin ligands on the regioselective

dehydrogenation versus epoxidation of olefins by oxoiron(IV) mimics

of cytochrome P450. J. Phys. Chem. A 2009, 113 (43), 11713 −22.

J. P. Mechanisms of N-Demethylations Catalyzed by High-Valent

Species of Heme Enzymes: Novel Use of Isotope Effects and Direct

Observation of Intermediates. J. Am. Chem. Soc. 1998, 120, 10762−

(22) Sugimoto, H.; Tung, H. C.; Sawyer, D. T. The formation,

10763.

characterization, and reactivity of the oxene adduct of [tetrakis(2,6-

dichlorophenyl)porphinato]iron(III) perchlorate in acetonitrile.

Model for the reactive intermediate of cytochrome P-450. J. Am.

Chem. Soc. 1988 , 110 (8), 2465 −2470.

39) Fujii, H.; Yoshimura, T.; Kamada, H. ESR Studies of A1u and

A2u Oxoiron(IV) Porphyrin π -Cation Radical Complexes. Spin

Coupling between Ferryl Iron and A1u/A2u Orbitals. Inorg. Chem.

1996, 35 (8), 2373−2377.

(40) Groves, J. T.; McClusky, G. A. Aliphatic Hydroxylation via

Oxygen Rebound. Oxygen Transfer Catalyzed by Iron. J. Am. Chem.

Soc. 1976, 98, 859−861.

(41) Zaragoza, J. P. T.; Yosca, T. H.; Siegler, M. A.; Moenne-Loccoz,

P.; Green, M. T.; Goldberg, D. P. Direct Observation of Oxygen

Rebound with an Iron-Hydroxide Complex. J. Am. Chem. Soc. 2017,

139 (39), 13640−13643.

23) Fujii, H. Characterization of High Valent Iron Porphyrin in

Catalytic Reaction by Iron(III) Tetrapentafluorophenylporphyrin.

Chem. Lett. 1994, 23, 1491−1494.

(24) Song, W. J.; Ryu, Y. O.; Song, R.; Nam, W. Oxoiron(IV)

porphyrin π -cation radical complexes with a chameleon behavior in

cytochrome P450 model reactions. JBIC, J. Biol. Inorg. Chem. 2005,

10, 294−304.

216

Inorg. Chem. 2021, 60, 3207−3217

Inorganic Chemistry

Article

42) Hjelmeland, L. M.; Aronow, L.; Trudell, J. R. Intramolecular

[18F]FPPA-c(RGDfK). Bioorg. Med. Chem. Lett. 2013, 23 (22),

6068−6072.

determination of primary kinetic isotope effects in hydroxylations

catalyzed by cytochrome P-450. Biochem. Biophys. Res. Commun.

43) Jones, J. P.; Trager, W. F. The separation of the intramolecular

977, 76 (2), 541−549.

isotope effect for the cytochrome P-450 catalyzed hydroxylation of n-

octane into its primary and secondary components. J. Am. Chem. Soc.

(

987, 109 (7), 2171−2173.

44) Kadkhodayan, S.; Coulter, E. D.; Maryniak, D. M.; Bryson, T.

A.; Dawson, J. H. Uncoupling oxygen transfer and electron transfer in

the oxygenation of camphor analogues by cytochrome P450-CAM.

Direct observation of an intermolecular isotope effect for substrate C-

H activation. J. Biol. Chem. 1995, 270 (47), 28042 −8.

45) Groves, J. T.; Adhyam, D. V. Hydroxylation by cytochrome P-

50 and metalloporphyrin models. Evidence for allylic rearrangement.

J. Am. Chem. Soc. 1984, 106 (7), 2177 −2181.

46) Groves, J. T.; Nemo, T. E. Aliphatic hydroxylation catalyzed by

iron porphyrin complexes. J. Am. Chem. Soc. 1983, 105 (20), 6243−

248.

47) Jeong, Y. J.; Kang, Y.; Han, A. R.; Lee, Y. M.; Kotani, H.;

Fukuzumi, S.; Nam, W. Hydrogen atom abstraction and hydride

transfer reactions by iron(IV)-oxo porphyrins. Angew. Chem., Int. Ed.

48) Sorokin, A.; Robert, A.; Meunier, B. Intramolecular kinetic

008, 47, 7321−4.

isotope effects in alkane hydroxylations catalyzed by manganese and

iron porphyrin complexes. J. Am. Chem. Soc. 1993, 115 (16), 7293−

49) Pan, Z.; Horner, J. H.; Newcomb, M. Tunneling in C-H

299.

oxidation reactions by an oxoiron(IV) porphyrin radical cation: direct

measurements of very large H/D kinetic isotope effects. J. Am. Chem.

Soc. 2008, 130 (25), 7776−7.

(50) Groves, J. T.; Nemo, T. E. Epoxidation reactions catalyzed by

iron porphyrins. Oxygen transfer from iodosylbenzene. J. Am. Chem.

Soc. 1983, 105 (18), 5786−5791.

(51) Traylor, T. G.; Nakano, T.; Dunlap, B. E.; Traylor, P. S.;

Dolphin, D. Mechanisms of hemin-catalyzed alkene epoxidation. The

effect of catalyst on the regiochemistry of epoxidation. J. Am. Chem.

Soc. 1986, 108 (10), 2782−2784.

(52) Collman, J. P.; Brauman, J. I.; Meunier, B.; Raybuck, S. A.;

Kodadek, T. Epoxidation of olefins by cytochrome P-450 model

compounds: mechanism of oxygen atom transfer. Proc. Natl. Acad. Sci.

U. S. A. 1984, 81 (10), 3245−8.

(53) Fujii, H.; Zhang, X.; Tomita, T.; Ikeda-Saito, M.; Yoshida, T. A

Role for Highly Conserved Carboxylate, Aspartate-140, in Oxygen

Activation and Heme Degradation by Heme Oxygenase-1. J. Am.

Chem. Soc. 2001, 123, 6475−6484.

(54) Bonnett, R.; Stephenson, G. F. The meso Reactivity of

Porphyrins and Related Compounds. I. Nitration. J. Org. Chem. 1965,

0 (8), 2791−2798.

55) Dunford, H. B. Historical: Poineering work on Horseradish and

(

Yeast Cyochrome c Peroxidases. In Peroxidases and Catalases;

Dubford, H. B., Ed.; John Wiley & Sons: New Jersey, 2010; pp 1−8.

(56) Munro, A. W.; McLean, K. J.; Grant, J. L.; Makris, T. M.

Structure and function of the cytochrome P450 peroxygenase

enzymes. Biochem. Soc. Trans. 2018 , 46 (1), 183 −196.

(57) Ortiz de Montellano, P. R. Mechanism and role of covalent

heme binding in the CYP4 family of P450 enzymes and the

mammalian peroxidases. Drug Metab. Rev. 2008, 40 (3), 405−26.

(58) Ono, N.; Kawamura, H.; Bougauchi, M.; Maruyama, K.

Porphyrin synthesis from nitrocompounds. Tetrahedron 1990, 46

21), 7483−7496.

59) Yokota, S.; Fujii, H. Critical Factors in Determining the

(

Heterolytic versus Homolytic Bond Cleavage of Terminal Oxidants

by Iron(III) Porphyrin Complexes. J. Am. Chem. Soc. 2018, 140 (15),

(

127−5137.

60) Monaco, A.; Zoete, V.; Alghisi, G. C.; Ruegg, C.; Michelin, O.;

Prior, J.; Scapozza, L.; Seimbille, Y. Synthesis and in vitro evaluation

of a novel radioligand for alphavbeta3 integrin receptor imaging:

217