Effect of imidazole and 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

Article abstract of DOI:10.1021/ic802123m

To study the effect of axial ligands on the electronic structure and reactivity of compound I of peroxidases and catalases, oxoiron(IV) porphyrin π-cation radical complexes with imidazole, 2-methylimidazole, 4(5)-methylimidazole, and 3-fluoro-4-nitropheno

Full text of DOI:10.1021/ic802123m

Inorg. Chem. 2009, 48, 2614-2625
Effect of Imidazole and 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
Akihiro Takahashi, Takuya Kurahashi, and Hiroshi Fujii*
Institute for Molecular Science and Okazaki Institute for IntegratiVe Bioscience, National Institutes
of Natural Sciences, and Department of Functional Molecular Science, The Graduate UniVersity
for AdVanced Studies, Myodaiji, Okazaki 444-8787, Japan
Received November 5, 2008
To study the effect of axial ligands on the electronic structure and reactivity of compound I of peroxidases and
catalases, oxoiron(IV) porphyrin π-cation radical complexes with imidazole, 2-methylimidazole, 4(5)-methylimidazole,
and 3-fluoro-4-nitrophenolate as the axial ligands were prepared by ozone oxidation of iron(III) complexes of
5,10,15,20-tetramesitylporphyrin (TMP) and 2,7,12,17-tetramethyl-3,8,13,18-tetramesitylporphyrin (TMTMP). These
complexes were fully characterized by absorption, 1H, 2H, and 19F NMR, electron paramagnetic resonance (EPR),
and electrospray ionization mass spectrometry (ESI-MS) spectroscopy. The characteristic absorption peak of
compound I at approximately 650 nm was found to be a good marker for estimation of the electron donor effect
from the axial ligand. The axial ligand effect did not change the porphyrin π-cation radical state, the a2u state of
the TMP complexes, or the a1u radical state of both the TMTMP complexes and compound I. The ferryl iron and
porphyrin π-cation radical spins were effectively transferred into the axial ligands for the a2u complexes but not for
the a1u complexes. Most importantly, the reactivity of the oxoiron(IV) porphyrin π-cation radical complex was drastically
increased by the imidazole and phenolate axial ligands. The reaction rate for cyclooctene epoxidation was increased
100- to 400-fold with axial coordination of imidazoles and phenolate. A similar increase was also observed for the
oxidation of 1,4-cyclohexadiene,N,N-dimethyl-p-nitroaniline and hydrogen peroxide. These results suggest extreme
enhancement of the reactivity of compound I by the axial ligand in heme enzymes. The functional role of axial
ligands on the compound I in heme enzymes is discussed.
Introduction
catalytic cycles of cytochrome P450.3 Although these
enzymes share a common reactive intermediate, the reactivity
of compound I differs from enzyme to enzyme. The
compound I in cytochrome P450 directly transfers a single
oxygen atom to a variety of substrates, while those in
peroxidases and catalases carry out oxidation of organic
substrates, such as amines and phenols, and oxidation of
hydrogen peroxide to water and oxygen gas, respectively.1-3
It is generally accepted that the diverse functions of
Oxoiron(IV) porphyrin π-cation radical species in peroxi-
dases and catalases, called compound I, is known as a
reactive intermediate of these enzymes.1,2 It is also believed
that compound I is formed as a reactive intermediate in the
* To whom correspondence should be addressed. E-mail: hiro@ims.ac.jp.
(1) (a) Hersleth, H.-P.; Ryde, U.; Rydberg, P.; Go¨rbitz, C. H.; Andersson,
K. K. J. Inorg. Biochem. 2006, 100, 460–476. (b) Watanabe, Y.; Fujii,
H. In Metal-Oxo and Metal-Peroxo Species in Catalytic Oxidations;
Meunier, B., Ed.; Springer: New York, 2000; Vol. 97, pp 61-89. (c)
Dunford, H. B. Heme Peroxidases; Wiley-VCH: New York, 1999.
(2) (a) Kirkman, H. N.; Gaetani, G. F. Trends Biochem. Sci. 2007, 32,
44–50. (b) Putnam, C. D.; Arvai, A. S.; Bourne, Y.; Tainer, J. A. J.
Mol. Biol. 2000, 296, 295–309. (c) Schonbaum, G. R.; Chance, B. In
The Enzymes; Boyer, P. D., Ed.; Academic Press: New York, 1976;
Vol. 13, pp 363-408.
(3) (a) Denisov, I. G.; Makris, T. M.; Sliger, S. G.; Schlichting, I. Chem.
ReV. 2005, 105, 2253–2278. (b) Oritz de Montellano, P. R. Cytochrome
P450. Structure, Mechanism, and Biochemistry, 2nd ed.; Plenum
Publishing Corporation: New York, 1995. (c) Watanabe, Y.; Groves,
J. T. In The Enzymes; Sigman, D. S., Ed.; Academic Press: New York,
1992; Vol. 20, pp 405-452.
2614 Inorganic Chemistry, Vol. 48, No. 6, 2009
10.1021/ic802123m CCC: $40.75  2009 American Chemical Society
Published on Web 02/13/2009

Oxoiron(IV) Porphyrin π-Cation Radical Complexes
Figure 1. Active site structures of (a) peroxidase (HRP, PDB file 1DZ9), (b) catalase (BL-CAT, PDB file 2A9E), and (c) cytochrome P450 (Cytochrome
P450cam, PDB file 1HCH).
compound I are controlled by heme environmental structures,
such as porphyrin peripheral structures, heme axial ligands,
and protein structures around heme.
synthetic iron(III) porphyrin complexes with imidazoles,
phenolates, and thiolates as axial ligands suggested that the
axial ligands have a significant effect on the reactivity and
selectivity of compound I.12-17 For example, in the catalytic
competitive oxidation of cyclooctane/cyclooctene, the imi-
dazole axial ligand produced greater product yields than the
thiolate axial ligand. On the other hand, the thiolate axial
ligand favored hydroxylation, whereas the imidazole axial
ligand preferentially yielded epoxidation.17 Recently, a
pronounced axial ligand effect on the rate of styrene
epoxidation was found by synthetic compound I model
complexes with various axial anion ligands, such as chloride,
fluoride, acetate, and so on.18 More recently, the axial ligand
effect was also studied in detail with other synthetic
compound I model systems, as well as with density-
functional theory (DFT) calculations.19,20 Although these
results clearly show the axial ligand effect on the electronic
structure and reactivity of compound I, it is unclear how the
enzyme axial ligands (imidazole, phenolate, and thiolate)
change the electronic state and the reactivity of compound
I. This problem is due to different protein structures making
it difficult to compare the reactivity of compound I in heme
enzymes and to the absence of stable compound I model
complexes with imidazole, phenolate, and thiolate axial
ligands because of the rapid reduction of compound I model
complexes by these axial ligands.
One significant structural difference in these heme en-
zymes is the heme axial ligand.1-3 As shown in Figure 1,
histidine and tyrosine residues (imidazole and phenolate) are
coordinated to the heme iron as axial ligands in peroxidases,
other than chloroperoxidase (CPO), and catalases, respec-
tively. In cytochrome P450 and CPO, a cysteine residue
(thiolate) is the heme axial ligand. The amino acids that serve
as the axial ligands are highly conserved in these heme
enzymes, suggesting some functional roles in these enzyme
functions. In the reaction cycles of these enzymes, the axial
ligand can affect both the formation of compound I and the
reaction of compound I with the substrate. The effect of the
axial ligand on the rate of compound I formation has been
studied by mutation of the aspartic acids that interact with
the proximal histidine residues in peroxidases; the mutations
weakened the electron donor effect of the proximal histidines,
drastically decreasing the rates of compound I formation.1,4
A similar axial ligand effect was also observed for synthetic
heme enzyme model complexes. The rate of heterolytic
cleavage of the O-O bond of iron bound m-chloroperben-
zoate, which leads to formation of the compound I model
complex, increased as the electron donor effect of the axial
imidazole ligand increased.5 These mutation and synthetic
enzyme model studies demonstrate that an increase in the
electron donor effect of the axial ligand enhances the rate
of compound I formation.
In previous studies, we reported the first preparation of
oxoiron(IV) porphyrin π-cation radical complexes with
(12) Traylor, T. G.; Lee, W. A.; Stynes, D. V. J. Am. Chem. Soc. 1984,
106, 755–764.
(13) (a) Dicken, C. M.; Woon, T. C.; Bruice, T. C. J. Am. Chem. Soc.
1986, 108, 1636–1643. (b) Beck, M. J.; Gopinath, E.; Bruice, T. C.
J. Am. Chem. Soc. 1993, 115, 21–29.
(14) Battioni, P.; Renaud, J. P.; Bartoli, J. F.; Reina-Artiles, M.; Fort, M.;
Mansuy, D. J. Am. Chem. Soc. 1988, 110, 8462–8470.
(15) Robert, A.; Loock, B.; Momenteau, M.; Meunier, B. Inorg. Chem.
1991, 30, 706–711.
The axial ligand also modulates the electronic structure
and reactivity of compound I. The electron paramagnetic
resonance (EPR) studies of compound I in peroxidases and
catalases indicated that the axial ligand changes the intramo-
lecular spin coupling contant between the ferryl iron and the
porphyrin radical spins.6-11 Catalytic oxidation studies using
(16) (a) Higuchi, T.; Uzu, S.; Hirobe, M. J. Am. Chem. Soc. 1990, 112,
7051–7052. (b) Suzuki, N.; Higuchi, T.; Urano, Y.; Kikuchi, K.;
Uekusa, H.; Ohashi, Y.; Uchida, T.; Kitagawa, T.; Nagano, T. J. Am.
Chem. Soc. 1999, 121, 11571–11572. (c) Urano, Y.; Higuchi, T.;
Hirobe, M.; Nagano, T. J. Am. Chem. Soc. 1997, 119, 12008–12009.
(17) Ohno, T.; Suzuki, N.; Dokoh, T.; Urano, Y.; Kikuchi, K.; Hirobe,
M.; Higuchi, T.; Nagano, T. J. Inorg. Biochem. 2000, 82, 123–125.
(18) (a) Gross, Z.; Nimri, S. Inorg. Chem. 1994, 33, 1731–1732. (b) Gross,
Z.; Nimri, S.; Barzilay, C. M.; Simkhovich, L. J. Biol. Inorg. Chem.
1997, 2, 492–506.
(4) (a) Smulevich, G.; Mauro, J. M.; Fishel, L. A.; English, A. M.; Kraut,
J.; Spiro, T. G. Biochemistry 1988, 27, 5477–5485. (b) Goodin, D. B.;
McRee, D. E. Biochemistry 1993, 32, 3313–3324.
(5) Yamaguchi, K.; Watanabe, Y.; Morishima, I. J. Am. Chem. Soc. 1993,
115, 4058–4065.
(6) Schluz, C. E.; Rutter, R.; Sage, J. T.; Debrunner, P. G.; Hager, L. P.
Biochemistry 1984, 23, 4743–4754.
(7) (a) Rutter, R.; Hager, L. P. J. Biol. Chem. 1982, 257, 7958–7961. (b)
Rutter, R.; Hager, L. P.; Dhonau, H.; Hendrich, M.; Valentine, M.;
Debrunner, P. Biochemistry 1984, 23, 6809–6816.
(8) Benecky, M. J.; Frew, J. E.; Scowen, N.; Jones, P.; Hoffman, B. M.
Biochemistry 1993, 32, 11929–11933.
(9) Patterson, W.; Polous, T. L.; Goodin, D. B. Biochemistry 1995, 34,
4342–4345.
(19) Song, W. J.; Ryu, Y. O.; Song, R.; Nam, W. J. Biol. Inorg. Chem.
2005, 10, 294–304.
(20) (a) Kuramochi, H.; Noodleman, L.; Case, D. A. J. Am. Chem. Soc.
1997, 119, 11442–11451. (b) de Visser, S. P.; Ogliaro, F.; Sharma,
P. K.; Shaik, S. Angew. Chem., Int. Ed. 2002, 41, 1947–1951. (c)
Kamachi, T.; Kouno, T.; Nam, W.; Yoshizawa, K. J. Inorg. Biochem.
2006, 101, 1464–1472. (d) Wang, R.; de Visser, S. P. J. Inorg.
Biochem. 2007, 101, 1464–1472.
(10) Ivancich, A.; Jouve, H. M.; Sartor, B.; Gaillard, J. Biochemistry 1997,
36, 9356–9364.
(11) Khindaria, A.; Aust, S. Biochemistry 1996, 35, 13107–13111.
Inorganic Chemistry, Vol. 48, No. 6, 2009 2615

Takahashi et al.
Scheme 1
two alternative routes, as shown in Scheme 1. [(TMP+•)-
FeIVO(L)](ClO4) and [(TMTMP+•)FeIVO(L)](ClO4), where
L is Im, 2-MeIm, and 5-MeIm, were prepared from (TMP)-
FeIII(ClO4) and (TMTMP)FeIII(ClO4) using route B, respec-
tively. [(TMTMP+•)FeIVO(L)](ClO4) also was prepared from
(TMTMP)FeIII(ClO4) using route A. However, pure [(TMP+•)-
FeIVO(L)](ClO4) could not be prepared using route A because
oxidation of (TMP)FeIII(ClO4) with ozone did not produce
pure (TMP+•)FeIVO(ClO4). To explore the effects of counter
anions, we examined the preparation from nitrate iron(III)
porphyrin complexes. In contrast to the perchlorate com-
plexes, [(TMP+•)FeIVO(L)](NO3) and [(TMTMP+•)FeIVO(L)]-
(NO3) were prepared from (TMP)FeIII(NO3) and (TMT-
MP)FeIII(NO3) using route A, respectively. On the other hand,
route B was not effective, except for the case of TMT-
MPFeIII(NO3) and 2-methylimidazole, because the addition
of 1 equiv of the imidazole derivative to the nitrate complex
did not afford its monoadduct complex but rather a mixture
of the parent and bis-adduct complexes (Supporting Informa-
tion, Figure S1). The oxoiron(IV) porphyrin π-cation radical
complexes with imidazole derivatives, prepared from nitrate
complexes using route A, were as stable as those prepared
from perchlorate complexes at temperatures below 233 K
in dichloromethane.
The preparation of oxoiron(IV) porphyrin π-cation radical
complexes with phenolate derivatives (4-nitrophenolate,
pentafluorophenolate, 3,4,5-trifluorophenolate, and 3-fluoro-
4-nitrophenolate) as axial ligands was examined using route
B for TMP and TMTMP complexes. The oxoiron(IV)
porphyrin π-cation radical complexes with 4-nitrophenolate,
3,4,5-trifluorophenolate, and pentafluorophenolate were very
unstable even at 183 K and quickly reduced to iron(III)
porphyrin complexes with dissociation and decomposition
of the axial phenol. However, the oxoiron(IV) porphyrin
π-cation radical complex with 3-fluoro-4-nitrophenolate was
stable enough for spectroscopic characterization at temper-
atures below 213 K in dichloromethane. In route A, the
addition of 1 equiv of phenol derivatives to oxoiron(IV)
porphyrin π-cation radical complex resulted in rapid reduc-
tion of the complex.
Figure 2. Structures of compound I model complexes prepared in this study.
imidazole and 4-nitrophenolate as models for compound-I
of peroxidases and catalases.21,22 Although the imidazole
1
complex could be characterized with absorption, H NMR,
and resonance Raman spectroscopy, the 4-nitrophenolate
complex was too unstable to characterize, except for its
absorption spectrum. To fully characterize and compare
reactivities, we comprehensively prepared oxoiron(IV) por-
phyrin π-cation radical complexes with imidazole (Im),
2-methylimidazole (2-MeIm), 4(5)-methylimidazole (5-
MeIm), and 3-fluoro-4-nitrophenolate (3-F-4-NO2-PhO) as
axial ligands by ozone oxidation of iron(III) complexes of
5,10,15,20-tetramesitylporphyrin (TMP) and 2,7,12,17-tet-
ramethyl-3,8,13,18-tetramesitylporphyrin (TMTMP), as shown
in Figure 2. A stable catalase compound I model complex
was synthesized with the 3-fluoro-4-nitrophenolate axial
ligand. These imidazole and phenolate complexes were fully
characterized using UV-visible absorption, 1H, 2H, and 19
F
NMR, EPR, and electrospray ionization mass spectrometry
(ESI-MS) spectroscopy. The TMTMP complexes more
closely mimic the spectroscopic properties of compound I
in peroxidases and catalases than the TMP complexes. The
most important findings in this study were as follows: (1)
that the porphyrin π-cation radical spin was significantly
transferred in the axial imidazole (phenolate) and oxo ligands
for the a2u radical complex, but not for the a1u radical
complex, and (2) that the oxidation reactivity of oxoiron(IV)
porphyrin π-cation radical complex was drastically increased,
more than 400 times in the most significant case, with the
coordination of imidazole and phenolate axial ligands,
suggesting extremely reactive compound I in heme enzymes.
On the basis of the present spectroscopic and kinetic studies,
the functional roles of axial ligands on compound I in heme
enzymes and catalytic oxidation reactions are discussed in
the final section.
Results
Synthesis of Compound I Models. The oxoiron(IV)
porphyrin π-cation radical complexes with imidazole and
phenolate derivatives as axial ligands were prepared using
The preparation of oxoiron(IV) porphyrin π-cation radical
complexes with thiolate axial ligands, as models for com-
pound I of cytochrome P450 and CPO, using route A and B
may be ineffective because of the sensitivity of iron(III)
porphyrin thiolate complexes to moisture and dioxygen and
(21) Fujii, H.; Yoshimura, T.; Kamada, H. Inorg. Chem. 1997, 36, 6142–
6143.
(22) Czarnecki, K.; Kincaid, J. R.; Fujii, H. J. Am. Chem. Soc. 1999, 121,
7953–7954.
2616 Inorganic Chemistry, Vol. 48, No. 6, 2009

Oxoiron(IV) Porphyrin π-Cation Radical Complexes
Table 2. Absorption Spectra and EPR Data for Compound I of
Peroxidases and Catalases
enzyme
axial ligand
λ/nm
EPR
ref
HRP
ASP
LiP
MnP
CIP
CPO
BL-CAT
Eo-CAT
Pm-CAT
Ml-CAT
His
His
His
His
His
Cys
Tyr
Tyr
Tyr
Tyr
400, 651
404, 650
408, 650
406, 650
402, 658
367, 688
400, 656
402, 657
655
∼ 2
6, 25
9, 26
11, 27
28
3.27, 1.99
3.42, 2.00
29
2.00, 1.73, 1.64
1.96 (1.98)
7, 30
10, 23
24
10
8
∼2
3.32, ∼2
The UV-visible absorption spectra of [(TMP+•)FeIVO-
(L)](NO3) and [(TMTMP+•)FeIVO(L)](NO3), where L is Im,
2-MeIm, and 5-MeIm, are shown in Supporting Information,
Figures S2 and S3. As summarized in Table 1, the axial
coordination of imidazole derivatives did not change the peak
positions for the TMP complexes but did change them for the
TMTMP complexes. The absorption spectral features of
[(TMTMP+•)FeIVO(L)](NO3) resembled those of compound I
of
peroxidases
more
closely
than
those
of
[(TMP+•)FeIVO(L)](NO3) (Table 2).25-30 Similar complexes
with perchlorate anions were also prepared, and the peak
positions were the same as those observed with nitrate anions.
Figure 4 shows the 1H NMR spectra of (TMP+•)FeIVO(3-
F-4-NO2-PhO) and (TMTMP+•)FeIVO(3-F-4-NO2-PhO) in
Figure 3. UV-visible absorption spectra of (P)FeIII(3-F-4-NO2-PhO), red
lines, and (P+•)FeIVO(3-F-4-NO2-PhO), blue lines, in dichloromethane
(concentration 1.0 × 10-5 M, 1 cm path length cell) at -80 °C: (a) P )
TMP, (b) P ) TMTMP.
1
Table 1. UV-visible Absorption Spectral Data of [(P+•)FeIVO(L)]X, in
CH2Cl2 at -80°C
dichloromethane-d2 at -80 °C. The H NMR signals were
assigned using deuterium-labeled compounds, signal intensi-
ties, and comparisons with those of previously characterized
oxoiron(IV) porphyrin π-cation radical complexes.31 (TMP+•)-
FeIVO(3-F-4-NO2-PhO) showed a pyrrole proton signal at
-13.7 ppm, the m-proton of the meso mesityl group at 71.7
ppm, the o-methyl protons at 25.4 and 26.9 ppm, and the
p-methyl protons at 11.3 ppm. The separation of the o-methyl
signals reflected the coordination of two different axial
ligands: oxo and phenolate. The large downfield shift of the
meso mesityl protons indicated an unpaired electron in the
a2u porphyrin π-orbital, a2u porphyrin π-cation radical state.
On the other hand, (TMTMP+•)FeIVO(3-F-4-NO2-PhO) showed
a large downfield shift of the pyrrole methyl signal at 101.9
ppm and a small shift of the meso proton signal at -1.0
ppm. The small paramagnetic shift of the meso proton signal
for (TMTMP+•)FeIVO(3-F-4-NO2-PhO) clearly indicated an
unpaired electron in the a1u porphyrin π-orbital, a1u porphyrin
π-cation radical state. These 1H NMR shifts are summarized
in Tables 3 and 4.
porphyrin
L
X
λ/nm
ref
-
TMP
ClO4
NO3
Im
403, 660
401, 666
403, 667
403, 667
404, 667
406, 666
387, 621
391, 630
396, 645
394, 643
389, 646
394, 643
388, 644
392, 655
-
-
ClO4-, NO3-
ClO4-, NO3-
ClO4-, NO3
2-MeIm
5-MeIm
3-F-4-NO2-PhO
-
TMTMP
ClO4
NO3
-
-
C6H5CO2
-
Im
ClO4-, NO3-
2-MeIm
5-MeIm
3-F-4-NO2-PhO
4-NO2-PhO
ClO4-, NO3-
ClO4-, NO3
21
also because of the reducing ability of thiolate. Further
modification will be required to prepare these complexes.
Characterization of Compound I Models. Figure 3
shows the UV-visible absorption spectra of (TMP+•)FeIVO(3-
F-4-NO2-PhO) and (TMTMP+•)FeIVO(3-F-4-NO2-PhO) pre-
pared from route B. The absorption maxima of the spectra
are listed in Table 1. The absorption spectral features of
(TMTMP+•)FeIVO(3-F-4-NO2-PhO) were close to those of
compound I of catalases (Table 2).10,23,24 While (TMP+•)FeIV-
O(3-F-4-NO2-PhO) slowly decomposed to (TMP)FeIII(3-F-
4-NO2-PhO) in dichloromethane at -80 °C (kdecomp ) 3.3
× 10-4 s-1), (TMTMP+•)FeIVO(3-F-4-NO2-PhO) was stable
for a few hours under the same conditions.
1
Figure 5 shows the H NMR spectra of [(TMP+•)FeIVO-
(L)](ClO4), where L is Im, 2-MeIm, and 5-MeIm, in
dichloromethane-d2 at -60 °C. The 1H NMR shifts of
[(TMP+•)FeIVO(L)](ClO4) are summarized in Table 3. The
(26) Lad, L.; Mewies, M.; Raven, E. L. Biochemistry 2002, 41, 13774–
13781.
(27) Renganathan, V.; Gold, M. H. Biochemistry 1986, 25, 1626–1631.
(28) Wariishi, H.; Akileswaran, L.; Gold, M. H. Biochemistry 1988, 27,
5365–5370.
(29) Anderson, M. B.; Hsuanyu, K.; Welinder, K. G.; Scheider, P.; Dunford,
H. P. Acta Chem. Scand. 1991, 45, 1080–1086.
(30) Palcic, M. M.; Rutter, R.; Araiso, T.; Hager, L. P.; Dunford, H. B.
Biochem. Biophys. Res. Commun. 1980, 94, 1123–1127.
(31) (a) Fujii, H. J. Am. Chem. Soc. 1993, 115, 4641–4648. (b) Fujii, H.;
Ichikawa, K. Inorg. Chem. 1992, 31, 1110–1112.
(23) Browett, W. R.; Stillman, M. J. Biochim. Biophys. Acta 1980, 623,
21–31.
(24) Hara, I.; Ichise, N.; Kojima, K.; Kondo, H.; Ohgiya, S.; Matsuyama,
H.; Yumoto, I. Biochemistry 2007, 46, 11–22.
(25) Dunford, H. B. AdV. Inorg. Biochem. 1982, 4, 41–68.
Inorganic Chemistry, Vol. 48, No. 6, 2009 2617

Takahashi et al.
for [(TMP+•)FeIVO(L)](ClO4) and [(TMTMP+•)FeIVO(L)]-
(ClO4) showed normal Curie’s law behavior (Supporting
Information, Figure S7).
Direct evidence for the coordination of imidazole deriva-
tives to oxoiron(IV) porphyrin π-cation radical complexes
was obtained by the 2H NMR spectra of iron-bound
deuterium-labeled imidazole derivatives (Figure 6). The 2H
NMR shifts are summarized in Table 5, together with the
NMR shifts of iron-bound imidazole signals of oxoiron(IV)
and iron(III) porphyrin complexes (Supporting Information,
Figures S8 and S9). The paramagnetic shifts of the 2H NMR
signals of the deuterium-labeled imidazole derivatives clearly
show their coordination to the ferryl iron site as axial ligands.
The 2H NMR signals for the 2- and 4-positions of the iron-
bound imidazole derivative show upfield shifts, while those
for the 5-position show downfield shifts. With substitutions
of the 2-positions with methyl groups, the 2H NMR signals
of the methyl groups were observed downfield. The 2H NMR
paramagnetic shifts of iron-bound imidazole signals for
[(TMP+•)FeIVO(L)](ClO4) were larger than those of
[(TMTMP+•)FeIVO(L)](ClO4). Direct evidence for the phe-
nolate binding to oxoiron(IV) porphyrin π-cation radical
complexes was also obtained from their 19F NMR spectros-
copy (Supporting Information, Figure S10).
Figure 7 shows the EPR spectra of [(TMP+•)FeIVO(2-
MeIm)](ClO4) and [(TMTMP+•)FeIVO(2-MeIm)](ClO4) at 4
K. [(TMP+•)FeIVO(2-MeIm)](ClO4) showed the EPR signals
at g ) 4.47, 3.45, and 1.97, which were similar to those of
(TMP+•)FeIVO(ClO4).32 The EPR spectra of [(TMP+•)FeIVO(2-
MeIm)](ClO4) were typical of those of S ) 3/2 systems,
indicating strong ferromagnetic interactions between ferryl
iron (S ) 1) and the porphyrin π-cation radical (S ) 1/2).
TheEPRspectraof[(TMP+•)FeIVO(Im)](ClO4),[(TMP+•)FeIVO(5-
MeIm)](ClO4), and (TMP+•)FeIVO(3-F-4-NO2-PhO) were
very close to that of [(TMP+•)FeIVO(2-MeIm)](ClO4), sug-
gesting similar spin interactions (Table 3 and Supporting
Information, Figures S11 and S12).
Figure 4. 1H NMR spectra of iron porphyrin phenolate complexes in
dichloromethane-d2at-80°C:(a)(TMP)FeIII(3-F-4-NO2-PhO),(b)(TMP+•)FeIVO(3-
F-4-NO2-PhO), (c) (TMTMP)FeIII(3-F-4-NO2-PhO), (d) (TMTMP+•)FeIVO(3-
F-4-NO2-PhO).
large downfield shifts of the meso mesityl protons of
[(TMP+•)FeIVO(L)](ClO4) indicated the a2u porphyrin π-cat-
ion radical states, as the case of (TMP+•)FeIVO(3-F-4-NO2-
1
PhO). The H NMR shifts of [(TMP+•)FeIVO(2-MeIm)]-
(ClO4) were similar to those of [(TMP+•)FeIVO(Im)](ClO4)
and [(TMP+•)FeIVO(5-MeIm)](ClO4), but the 1H NMR
signals of [(TMP+•)FeIVO(2-MeIm)](ClO4) were broader and
were split at -90 °C (Figure 5c,d). This was because the
rotation of the axially bound 2-methylimidazole along the
Fe-N(2-MeIm) axis was hindered by a steric interaction
between the 2-methyl group of 2-methylimidazole and the
The EPR spectrum of [(TMTMP+•)FeIVO(2-MeIm)](ClO4)
differed from that of [(TMP+•)FeIVO(2-MeIm)](ClO4) but
was similar to that of [(TMTMP+•)FeIVO(MeOH)](ClO4)
reported previously (Figure 7b).32 The EPR spectra of
[(TMTMP+•)FeIVO(Im)](ClO4) and [(TMTMP+•)FeIVO(5-
MeIm)](ClO4) were similar to that of [(TMTMP+•)FeIVO(2-
MeIm)](ClO4), but the g⊥ values were changed by the axial
imidazole ligand (Table 4 and Supporting Information, Figure
S11). These EPR spectra were interpreted as a weak
ferromagnetic interaction between ferryl iron (S ) 1) and
the porphyrin π-cation radical (S ) 1/2). On the other hand,
the EPR spectrum of (TMTMP+•)FeIVO(3-F-4-NO2-PhO)
was different from that of [(TMTMP+•)FeIVO(2-MeIm)]+ and
exhibited an EPR signal at around g ∼ 2.0 (Supporting
Information, Figure S12), which was attributed to very weak
magnetic interaction between iron(IV) and porphyrin radical
spins. The EPR spectrum of (TMTMP+•)FeIVO(3-F-4-NO2-
PhO) was close to that of compound I of Pm-CAT.10 The
magnetic interactions between ferryl iron and the porphyrin
1
o-methyl group of the meso mesityl group. The H NMR
spectra of [(TMP+•)FeIVO(L)](NO3) were identical to those
of [(TMP+•)FeIVO(L)](ClO4) (Supporting Information, Figure
S4).
1
The H NMR spectra of [(TMTMP+•)FeIVO(L)](ClO4),
where L is Im, 2-MeIm and 5-MeIm, in dichloromethane-d2
at -80 °C are shown in Supporting Information, Figure S5.
1
The H NMR spectra of [(TMTMP+•)FeIVO(L)](ClO4) in-
dicated their a1u porphyrin π-cation radical states. Unlike
[(TMP+•)FeIVO(2-MeIm)](ClO4), the rotation of the 2-MeIm
in [(TMTMP+•)FeIVO(2-MeIm)](ClO4) is not hindered. This
is because the o-methyl group of the pyrrole ꢀ-mesityl group
in TMTMP is farther from the iron center than that of the
meso mesityl group in TMP. The 1H NMR spectra of
[(TMTMP+•)FeIVO(L)](NO3) are similar to those of
[(TMTMP+•)FeIVO(L)](ClO4) (Supporting Information, Fig-
(32) Fujii, H.; Yoshimura, T.; Kamada, H. Inorg. Chem. 1996, 35, 2373–
1
ure S6). The temperature dependence of H NMR signals
2377.
2618 Inorganic Chemistry, Vol. 48, No. 6, 2009

Oxoiron(IV) Porphyrin π-Cation Radical Complexes
1
Table 3. H NMR Chemical Shifts (ppm) and EPR Data of (TMP+•)FeIVO(L)
L
T, °C
pyrrole-H
ortho-CH3
meta-H
para-CH3
g
E/D
Ima
-60
-80
-60
-80
-60
-80
-80
-80
-80
-10.4
-11.3
-8.8
22.8, 23.8
26.4, 27.5
25.6
63.9, 65.4
73.1, 74.8
70.6, 72.2
broad
64.4, 65.8
73.2, 74.9
71.7
10.3
11.7
10.8
11.3
10.4
11.2
11.3
11.3
10.2
4.41, 3.55, 1.98
0.070
2-MeIma
4.47, 3.45, 1.97
4.42, 3.57, 1.98
0.085
0.070
-9.8
broad
5-MeIma
-9.8
22.8, 24.0
24.2, 25.8
25.4, 26.9
26.4, 23.9
25.3, 27.8
-11.1
-13.7
-27.4
-19.4
3-F-4-NO2-PhO
ClO4
NO3
4.59, 3.40, 1.95
4.42, 3.55, 1.98
4.45, 3.45, 1.96
0.100
0.075
0.090
-
66.4
72.5
-
a Counter anion is perchlorate ion (ClO4-). NMR; in dichloromethane-d2. EPR; in dichloromethane-toluene (5:1) at 4 K.
1
Table 4. H NMR Chemical Shifts (ppm) and EPR Data of (TMTMP+•)FeIVO(L)
L
pyrrole-CH3
meso-H
ortho-CH3
meta-H
para-CH3
g
J/D
Ima
113.2
123.7
111.8
101.9
135.5
127.7
132.1
0.0
-14.8
-4.0
-1.0
55.6
7.5, 8.7
8.0, 9.3
7.4, 8.6
6.6, 7.7
7.6, 9.6
7.4, 8.3
8.3, 9.6
12.6, 13.4
14.2, 15.0
12.6, 13.4
13.3, 13.7
14.2, 15.0
13.8, 14.6
13.7, 14.6
3.2
2.9
3.1
3.2
3.6
3.3
3.4
2.97, 2.00
3.67, 2.00
3.84, 2.00
2.0
3.23, 2.00
3.69, 2.00
0.3
0.7
0.9
0
0.4
0.7
2-MeIma
5-MeIma
3-F-4-NO2-PhO
ClO4
NO3
-
-
34.7
45.8
a Counter anion is perchlorate ion (ClO4-). NMR; in dichloromethane-d2. EPR; in dichloromethane-toluene (5:1) at 4 K.
Figure 5. 1H NMR spectra of [(TMP+•)FeIVO(L)](ClO4) in dichlo-
romethane-d2 at -60 °C. (a) L ) Im, (b) L ) 5-MeIm, (c) L ) 2-MeIm,
(d) L ) 2-MeIm at -90 °C.
π-radical for TMTMP complexes were weaker than those
for TMP complexes and became much weaker when the axial
ligand was changed from imidazole to phenolate.
The structural characterization of oxoiron(IV) porphyrin
π-cation radical complexes having imidazole and phenolate
Figure 6. 2H NMR spectra of (a) [(TMP+•)FeIVO(Im-d4)](ClO4), (b)
[(TMP+•)FeIVO(2-MeIm-d6)](ClO4), (c) [(TMP+•)FeIVO(5-MeIm-d6)](ClO4),
(d) [(TMTMP+•)FeIVO(Im-d4)](ClO4), (e) [(TMTMP+•)FeIVO(2-MeIm-
d6)](ClO4), and (f) [(TMTMP+•)FeIVO(5-MeIm-d6)](ClO4) in dichlo-
romethane. Temperature; (a-c) at -60 °C and (d-f) at -80 °C.
derivatives was examined using ESI-MS spectroscopy. ESI-
MS peaks corresponding to [(TMP+•)FeIVO(L)]+ and
[(TMTMP+•)FeIVO(L)]+ cations, where L ) Im, 2-MeIm, and
5-MeIm, were detected (Supporting Information, Figure S13
and Table S1). The isotope distribution patterns of the observed
ESI-MS peaks were almost identical to those calculated from
[(TMP+•)FeIVO(L)]+ and [(TMTMP+•)FeIVO(L)]+ structures.
All of the ESI-MS data supported the binding of imidazole
derivatives as an axial ligand. On the other hand, ESI-MS ion
peaks corresponding to (TMP+•)FeIVO(3-F-4-NO2-PhO) and
(TMTMP+•)FeIVO(3-F-4-NO2-PhO) could not be detected
because of their neutral character.
Reactivity of Compound I Models. To study the effect
of axial ligands on reactivity, we examined reactions of
oxoiron(IV) porphyrin π-cation radical complexes having
various axial ligands with cyclooctene. Figure 8a and 8b shows
the absorption spectral changes for the epoxidation reactions
of [(TMP+•)FeIVO(Im)](NO3) and [(TMTMP+•)FeIVO(Im)]-
(NO3) with cyclooctene at -80 °C, respectively. Upon the
addition of cyclooctene, the absorption spectra of [(TMP+•)-
Inorganic Chemistry, Vol. 48, No. 6, 2009 2619

Takahashi et al.
Table 5. NMR Chemical Shifts (ppm) of Iron-Bound Imidazole Signals
absorptions for the epoxidation reactions in the presence of large
excess of cyclooctene could be fitted well with a single
exponential function.33 The apparent rate constants linearly
depend on the concentration of cyclooctene (Supporting Infor-
mation, Figure S15), giving the second-order rate constants for
cyclooctene epoxidation reactions (Table 6). To confirm the
formation of the epoxidation product, GC-MS analysis was
applied to each reaction solution used in these kinetic experi-
ments, which showed the formation of cyclooctene oxide in
more than 80% yield (see Table 6).
We further examined the reactions of oxoiron(IV) por-
phyrin π-cation radical complexes with 3-fluoro-4-nitrophe-
nolate as an axial ligand. Panels c and d of Figure 8 show
absorptionspectralchangesforthereactionsof(TMP+•)FeIVO(3-
F-4-NO2-PhO) and (TMTMP+•)FeIVO(3-F-4-NO2-PhO), re-
spectively, with cyclooctene at 193 K. Upon the addition of
cyclooctene, the absorption spectra of (TMP+•)FeIVO(3-F-
4-NO2-PhO) and (TMTMP+•)FeIVO(3-F-4-NO2-PhO) changed
to those of (TMP)FeIII(3-F-4-NO2-PhO) and (TMTMP)FeIII(3-
F-4-NO2-PhO) with clear isosbestic points. The formation
of (TMP)FeIII(3-F-4-NO2-PhO) and (TMTMP)FeIII(3-F-4-
NO2-PhO) was further confirmed by 1H NMR measurements.
Kinetic analysis showed the second-order rate constants at
0.126 M-1 s-1 and 0.134 M-1 s-1 for the cyclooctene
epoxidation with (TMP+•)FeIVO(3-F-4-NO2-PhO) and
(TMTMP+•)FeIVO(3-F-4-NO2-PhO), respectively (Supporting
Information, Figure S15). Analysis of the product using GC-
MS showed that only cyclooctene oxide was detected, in 32
and 80% yield for (TMP+•)FeIVO(3-F-4-NO2-PhO) and
(TMTMP+•)FeIVO(3-F-4-NO2-PhO), respectively. The low
yield of cyclooctene oxide for (TMP+•)FeIVO(3-F-4-NO2-
PhO) may be due to its instability, as described in the section
above.
of Iron Porphyrin Complexes in Dichloromethane
complex
2-H(D) 4-H(D) 5-H(D) T/°C ref
(TMP+•)FeIVO(Im-d4)a
-30.2 -17.4
19.1 -80
12.4 -80
6.1d -80
5.9 -80
n.d.e -80
6.1d -80
(TMP+•)FeIVO(2-MeIm-d6)a
(TMP+•)FeIVO(5-MeIm-d6)a
(TMTMP+•)FeIVO(Im-d4)a
29.4d
-15.0
-34.3 -17.4
-14.5 -3.7
(TMTMP+•)FeIVO(2-MeIm-d6)a 23.9d
-2.7
(TMTMP+•)FeIVO(5-MeIm-d6)a -17.3 -2.5
(TMP)FeIVO(1-MeIm)b
-28.5 -21.6
n.d.e
-20.7
-29.1 -22.7
n.d.e
-22.5 -17.5
(TmTP)FeIVO(1,2-DiMeIm-d8)c n.d.e
-13.7
(TmTP)FeIVO(1,5-DiMeIm-d8)c -23.5 -15.8
-3.6 -80
(TMP)FeIVO(1,2-DiMeIm)b
(TMTMP)FeIVO(1-MeIm)b
-4.1 -80
-4.2 -80
-3.8 -80
-2.8 -50 38
-1.7 -50 38
-1.3d -50 38
(TMTMP)FeIVO(1,2-DiMeIm)b n.d.e
(TmTP)FeIVO(1-MeIm-d6)c
(TMP)FeIII(Im-d4)a, f
-28.6 33.6
18.4d
42.3
111.7 -60
89 -60
(TMP)FeIII(2-MeIm-d6)a, f
(TMP)FeIII(5-MeIm-d6)a, f
(TMTMP)FeIII(Im-d4)a, f
(TMTMP)FeIII(2-MeIm-d6)a, f
(TMTMP)FeIII(5-MeIm-d6)a, f
-36.3 33.8
24.6d -60
115.3 -80
91.3 -80
22.9d -80
-45.2 30.1
20.3d
45.9
-49.4 31.2
a Counter anion is perchlorate ion (ClO4-). b NMR shifts of heme protons
are summarized in Supporting Information, Table S2. c TmTP; Tetra-m-
tolylporphyrin, in toluene-d8. d Methyl signal. e Not determined. f Assignments
of the signals are based on the methyl substitutions (see Supporting
Information, Figure S6).
The reaction rates of oxoiron(IV) porphyrin π-cation radical
complexes with nitrate and chloride axial anions were also
determined at 193 K (Supporting Information, Figures S15 and
S16). The second-order rate constants and the yields of
cyclooctene oxide are summarized in Table 6. The reaction of
(TMTMP+•)FeIVO(Cl) could not be examined because the ozone
oxidation of (TMTMP)FeIIICl at -80 °C did not form
(TMTMP+•)FeIVO(Cl). The reaction rate constants and reaction
yields for imidazole and phenolate complexes were much larger
than those for nitrate and chloride. In the most extreme case,
the reaction rate was ∼400 times larger with the coordination
of 2-methylimidazole to (TMTMP+•)FeIVO(NO3).
To study the axial ligand effect on oxidation reactions other
than epoxidation, we examined the reactions of oxoiron(IV)
porphyrin π-cation radical complexes with N,N-dimethyl-p-
nitroaniline and 1,4-cyclohexadiene (Table 6 and Supporting
Information, Figure S17). Previous studies reported that ox-
oiron(IV) porphyrin π-cation radical complexes oxidize N,N-
dimethyl-p-nitroaniline to N-methyl-p-nitroaniline via an elec-
tron transfer mechanism and 1,4-cyclohexadiene to benzene via
Figure 7. EPR spectra of (a) [(P+•)FeIVO(2-MeIm)](ClO4) at 4K. (a) P )
TMP, in dichloromethane-toluene (5:1), (b) P ) TMTMP, in dichlo-
romethane-propionitrile (5:1). Signals denoted by asterisks are due to
[(TMP)FeIII(2-MeIm)](ClO4) and [(TMTMP)FeIII(2-MeIm)](ClO4).
FeIVO(Im)](NO3) and [(TMTMP+•)FeIVO(Im)](NO3) changed
to those of (TMP)FeIII(NO3) and (TMTMP)FeIII(NO3) contain-
ing 1 equiv of imidazole, respectively (Supporting Information,
1
Figure S1). H NMR measurements of the final reaction
solutions clearly showed that the final product was a 1:1 mixture
of nitrate and bis-imidazole iron(III) porphyrin complexes
(Supporting Information, Figure S14). The absorption spectral
changes for the reactions showed a slight shift in isosbestic
points because the reaction product changed from the monoimi-
dazole complex to a mixture of nitrate and bis-imidazole
complexes, as the reaction product increased (Scheme 2).
Similar absorption spectral changes were observed for cy-
clooctene epoxidation reactions when 2-methylimidazole and
4(5)-methylimidazole were used as axial ligands. In spite of a
slight shift in the isosbestic points, the time courses of the
(33) Time course of the absorptions for [(TMP+•)FeIVO(5-MeIm)](NO3)
could not be fit well with a single exponential function, but with a
bi-exponential function, suggesting some sequential reaction. The rate
constant for the first reaction was assigned as that for [(TMP+•)FeIVO(5-
MeIm)](NO3) with cyclooctene.
2620 Inorganic Chemistry, Vol. 48, No. 6, 2009

Oxoiron(IV) Porphyrin π-Cation Radical Complexes
Figure 8. UV-visible absorption spectral change for the reactions of (a) [(TMP+•)FeIVO(Im)](NO3), (b) [(TMTMP+•)FeIVO(Im)](NO3), (c) (TMP+•)FeIVO(3-
F-4-NO2-PhO), and (d) (TMTMP+•)FeIVO(3-F-4-NO2-PhO) with cyclooctene (2.5 × 10-2 M) in dichloromethane at -80 °C. Sample concentration ) 1.0
× 10-4 M in 1 cm path length cell. Insets; time dependence of the absorptions at (a) 667 nm, (b) 639 nm, (c) 667 nm, and (d) 643 nm.
Scheme 2
Discussion
Effect of Axial Ligand on Electronic Structure. All
oxoiron(IV) porphyrin π-cation radical complexes character-
ized in this study showed the Soret band with decreased
intensity around 400 nm and a characteristic absorption
a hydrogen abstraction mechanism.34,35 The second-order rate
around 650 nm. The peak positions for TMP complexes were
constants of [(TMTMP+•)FeIVO(Im)](NO3) and (TMTMP+•)-
FeIVO(3-F-4-NO2-PhO) for N,N-dimethyl-p-nitroaniline were
70 times higher than that of (TMTMP+•)FeIVO(NO3), and those
for 1,4-dihydrocyclohexadiene were 10 times higher. Coordina-
tion of imidazole and phenolate axial ligands enhanced the
reactivity of oxoiron(IV) porphyrin π-cation radical complex
not only for the epoxidation reaction but also for the aniline
N-demethylation and hydrogen abstraction reactions.
Regarding the catalase reactions, we examined the reactions
of (TMTMP+•)FeIVO(NO3), [(TMTMP+•)FeIVO(Im)](NO3), and
(TMTMP+•)FeIVO(3-F-4-NO2-PhO) with hydrogen peroxide in
dichloromethane at -80 °C. Whereas (TMTMP+•)FeIVO(NO3)
reacted in a few minutes with the addition of 10 equiv of
hydrogen peroxide in methanol (kapp ) 2.0 × 10-2 s-1 for 1
mM hydrogen peroxide), [(TMTMP+•)FeIVO(Im)](NO3), and
(TMTMP+•)FeIVO(3-F-4-NO2-PhO) reacted immediately to
form iron(III) porphyrin complexes. Because these reactions
occurred so rapidly, the reaction rate constants could not be
determined. While a drastic axial ligand effect was found for
the hydrogen peroxide oxidation, predominance of the phenolate
axial ligand over the imidazole axial ligand was not observed.
hardly changed by the axial ligand effect, however, those
for TMTMP complexes were sensitive to the axial ligand
effect (Table 1). In particular, the absorption peak at around
650 nm for TMTMP complexes showed ∼35 nm shift due
to the axial ligand. It is worth noting that the peak position
of (TMTMP+•)FeIVO(L) shifted to lower energy as the pKa
value of the axial ligand (L) increased (Figure 9).36 The axial
anion ligands showed a good correlation, while the axial
neutral ligands, L ) Im, 2-MeIm, and 5-MeIm, exhibited
slight shifts from the correlation for the axial anion ligands.
The absorption peak at around 650 nm may be a good marker
for estimation of the electron donor ability of the axial ligand.
Because of the structural similarity of protoporphyrin IX with
TMTMP, a similar correlation was also observed for peak
positions of compound I of peroxidases and catalases (Table
1 and 2). The peak positions of the compound I of
peroxidases and catalases also shifted from 650 to 688 nm
and were correlated with the pKa values of the proximal
ligands (Figure 9). Although other factors, such as the steric
and electrostatic environment around heme, must be con-
(36) (a) Albert, A. In Physical Methods in Heterocyclic Chemistry;
Katritzky, A. R., Ed.; Academic Press: New York, 1971; Vol. III, pp
1-108. (b) Pine, S. H.; Hendrickson, J. B.; Vram, D. J.; Hammond,
G. S. In Organic Chemistry, 4th ed; MacGraw-Hill: Auckland, 1981;
Chapter 6, pp 196-239. (c) Albert, A.; Serjeant, E. P. In The
Determination of Ionization Constants, 3rd ed.; Chapman and Hall:
London, 1962; Chapter 9, pp 136-175. (d) Chapman, E.; Bryan, M. C.;
Wong, C.-H. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 910–915.
(34) (a) Goto, Y.; Watamabe, Y.; Fukuzumi, S.; Jones, J. P.; Dinnocenzo,
J. P. J. Am. Chem. Soc. 1998, 120, 10762–10763. (b) Chiavarino, B.;
Cipollini, R.; Crestonei, M. E.; Fornarini, S.; Lanucara, F.; Lapi, A.
J. Am. Chem. Soc. 2008, 130, 3208–3217.
(35) Jeong, Y. J.; Kang, Y.; Han, A.-R.; Lee, Y.-M.; Kotani, H.; Fukuzumi,
S.; Nam, W. Angew. Chem., Int. Ed. 2008, 47, 7321–7324.
Inorganic Chemistry, Vol. 48, No. 6, 2009 2621

Takahashi et al.
Table 6. Second-Order Rate Constants for the Reaction of (P.+)FeIVO(L) with the Various Substrates in CH2Cl2 at -80 °C
k2, M-1 s-1
porphyrin
TMP
ligand
cyclooctenea
1,4-cyclohexadiene
N,N-dimethyl-p-nitroaniline
Im
0.216 ( 0.003 (87 ( 3)
0.030 ( 0.002 (72 ( 4)
0.240 ( 0.003 (71 ( 3)
0.126 ( 0.001 (32 ( 8)
0.011 ( 0.001 (62 ( 2)
8.21 ( 0.31 × 10-4 (64 ( 2)
0.095 ( 0.001 (82 ( 2)
0.306 ( 0.002 (89 ( 3)
0.126 ( 0.001 (67 ( 5)
0.134 ( 0.001 (80 ( 3)
7.13 ( 0.25 × 10-4 (48 ( 1)
b
2-MeIm
5-MeIm
3-F-4-NO2-PhOb
Cl-
-
NO3
Im
TMTMP
0.868 ( 0.005
3.470 ( 0.225
2-MeIm
5-MeIm
3-F-4-NO2-PhO
1.043 ( 0.016
0.082 ( 0.005
3.383 ( 0.050
0.048 ( 0.002
31
-
NO3
a Numbers in parentheses are yield (%) of cyclooctene oxide. Rate for the second slow reaction was 0.020 ( 0.001 M-1 s-1
.
Figure 10. Electron spin density distribution of the a2u and a1u porphyrin
π-cation radical orbitals. Red and blue colors represent signs of wave
functions.
form the a2u radical state, which shifts to the a1u state with
an increase in the electron-withdrawing effect of the meso-
substituent. Pyrrole ꢀ-substituted porphyrins, like TMTMP,
form the a1u radical state, which is not changed even with
an increase in the electron-withdrawing effect of the pyrrole
ꢀ-substituent. On the other hand, the present 1H NMR results
clearly show that the axial ligand effect does not change the
porphyrin π-cation radical state. Even if the electron-donating
axial ligand, such as imidazole and phenolate, were coor-
dinated to the ferryl iron, TMTMP and TMP complexes
remained in the a1u and a2u radical states, respectively. These
results indicate that compound I in all peroxidases and
catalases with pyrrole ꢀ-substituted porphyrin, like proto-
porphyrin IX, would be in the a1u radical state because one-
electron oxidation of protoporphyrin IX usually yields the
a1u radical state.
The paramagnetic shifts of the iron-bound axial ligands
allow the estimation of spin transfer from ferryl iron and
porphyrin π-cation radical spins to the iron-bound oxo and
axial ligand. The radical character of the oxo ligand has been
discussed in relation to the oxygenation reactivity of ox-
oiron(IV) porphyrin π-cation radical complex,37 but the spin
transfer from the porphyrin π-cation radical to the axial
ligand has not been studied in detail. As summarized in
Figure 6 and Table 5, the iron-bound imidazole signals
showed paramagnetic shifts, induced by ferryl iron and
porphyrin π-cation radical spins. We assume here that the
dipolar contribution of oxoiron(IV) porphyrin π-cation
Figure 9. Correlation of pKa value of the axial ligand (L) of
(TMTMP+•)FeIVO(L) with its absorption peak position. Red square; anion
ligand, L from left; ClO4-, NO3-, benzoate, 3-F-4-NO2-PhO, 4-NO2-PhO,
red circle; neutral ligand, L from left; Im, 5-MeIm, 2-MeIm. Black circle;
compound-I of heme enzymes listed in Table 1. The pKa values for
peroxidases other than CPO, catalases, and CPO are used those for histidine
(6.1), tyrosine (10.1), and cysteine (8.3) residues, respectively.36 Circle;
CPO compound-I for the pKa (10.6) of ethanethiol.36
sidered, the peak position observed for compound I can be
explained by the donor effect of the proximal ligand in these
heme enzymes. For peroxidases other than CPO, the observed
peak positions (650-658 nm) were close to the correlation
for the axial anion ligands and were reasonable for the
character of the imidazolate anion, resulting from hydrogen
bonding interactions of the proximal histidines with the
aspartate residues (see Figure 1a). The peak positions
(655-657 nm) for catalases can be explained by hydrogen
bonding between the proximal tyrosine ligand (phenolate)
and a nearby arginine residue (see Figure 1b), which shifted
the phenolate character of the proximal ligand to a phenol
and weakens the electron donor effect of the proximal
tyrosine ligand. The absorption peak (688 nm) for CPO
compound I indicates extremely strong electron donor effect
of the proximal cysteine ligand. The donor effect was much
stronger than that expected from the pKa value (8.3) of
cysteine but comparable to that expected from alkylthiol (pKa
10-11).36
(37) (a) Groves, J. T.; Nemo, T. E. J. Am. Chem. Soc. 1983, 105, 6243–
6248. (b) Champion, P. M. J. Am. Chem. Soc. 1989, 111, 3434–3436.
(c) Bernadou, J.; Fabiano, A.-S.; Robert, A.; Meunier, B. J. Am. Chem.
Soc. 1994, 116, 9375–9376.
(38) Balch, A. L.; La Mar, G. N.; Latos-Grazynski, L.; Renner, M. W.;
Thanabal, V. J. Am. Chem. Soc. 1985, 107, 3003–3007.
Previously, we showed that the porphyrin π-cation radical
(a1u or a2u) state depends on the porphyrin substitution pattern
(Figure 10).31a Meso-substituted porphyrins, such as TMP,
2622 Inorganic Chemistry, Vol. 48, No. 6, 2009

Oxoiron(IV) Porphyrin π-Cation Radical Complexes
radical complex is close to that of oxoiron(IV) porphyrin
complex.38 As previously analyzed for oxoiron(IV) porphyrin
complexes, two unpaired electrons in the ferryl iron dπ (dxz
and dyz) orbitals delocalize into the imidazole π-orbital (the
highest occupied π-molecular orbital), leading to upfield
shifts of all protons belonging to the iron-bound imidazoles.38
On the other hand, porphyrin π-cation radical spin would
be transferred to the axial imidazole σ-orbital via the
2
unoccupied ferryl iron dz orbital, resulting in downfield shifts
of the imidazole signals, especially the imidazole 1-H and
5-H signals, as observed for high spin iron(III) porphyrin
complexes (see Table 5 and Supporting Information, Figure
S9).39 Therefore, the upfield shifts of the imidazole 2-H and
4-H signals for oxoiron(IV) porphyrin π-cation radical
complexes were mainly induced by the spin delocalization
of the ferryl iron dπ spins. This result was further confirmed
by the large downfield shifts with the 2-methyl substitutions.
Compared with the NMR shifts of the iron-bound imidazole
signals of oxoiron(IV) porphyrin complexes, the upfield shifts
of the imidazole 2-H and 4-H signals for TMTMP complexes
become smaller with the formation of porphyrin π-cation
radicals, while those for TMP complexes hardly change.
Since the NMR shifts of the iron-bound imidazole signals
of oxoiron(IV) TMP complexes were similar to those of
TMTMP complexes (Table 5), the observed change suggests
that spin transfer of the ferryl iron spins to the axial ligands
is weakened with the formation of the a1u radical complex
(TMTMP) but is unchanged with the formation of the a2u
radical complex (TMP). On the other hand, with the
formation of the porphyrin π-cation radical, the imidazole
5-H(D) signals showed large downfield shifts for the TMP
complexes, but small shifts for the TMTMP complexes.
Large downfield shifts of the 5-H(D) signals for TMP
complexes indicate effective spin transfer from the a2u
porphyrin radical, but the small shifts for TMTMP complexes
indicate ineffective transfer from the a1u porphyrin radical.
Overall, the ferryl iron and porphyrin π-cation radical spins
were effectively transferred to the axial imidazole for the
a2u radical complex, but not for the a1u radical complex. This
is consistent with a previous EPR study of zinc(II) porphyrin
π-cation radical complexes.40 With molecular vibration of
the propyhrin plane, for example, doming, the a2u orbital is
Figure 11. Comparison of second-order rate constants for the cyclooctene
epoxidation of oxoiron(IV) TMP (brown bar) and TMTMP (blue bar)
π-cation radical complexes with various axial ligands. The numbers in
parentheses are yields of cyclooctene oxide.
between the TMP and TMTMP complexes might result from
differences in both the π-spin density on the pyrrole nitrogen
atom between the a2u and a1u radical states (see Figure 10)
and the flexibility of the porphyrin plane of TMTMP because
of less steric hindrance than in TMP. The EPR spectra of
compound I in heme enzymes more closely resemble those
of TMTMP complexes, confirming the a1u radical state. The
similarity of the EPR spectra of the compound I of heme
enzymes with those of model complexes with similar axial
ligands suggests that the axial ligand effect is an important
determinant of the intramolecular spin interaction of the
compound I, as well as the steric effect of amino acid
residues around the heme.
Effect of the Axial Ligand on Reactivity. Previously,
the axial ligand effect on styrene oxidation was studied using
(TMP+•)FeIVO(X), where X is either chloride, fluoride,
acetate, methanol, or trifluoromethylsulfonate.18 The second-
order rate constant for the styrene oxidation was increased
about 15 times when the axial ligand was changed from the
least reactive trifluoromethylsulfonate to the most reactive
fluoride. In this study, we also found similar increments when
the axial ligand was changed from nitrate to chloride (∼15-
fold). More importantly, this study showed a significant effect
of axial ligands of imidazole and phenolate derivatives on
the reactivity of oxoiron(IV) porphyrin π-cation radical
complex. As clearly shown in Figure 11, the effect of the
axial imidazole and phenolate was much more prominent
than that of the axial chloride. For the 3-fluoro-4-nitrophe-
nolate axial ligand, the second-order rate constants for the
TMP and TMTMP complexes were comparable. However,
for the imidazole and 4(5)-methylimidazole axial ligands,
the second-order rate constants for the TMP complexes were
about twice those of the TMTMP complexes. For the
2-methylimidazole axial ligand, the TMTMP complex was
much more reactive than the TMP complex. The low
reactivity of (TMP+•)FeIVO(2-MeIm) may be due to steric
interaction between the methyl groups of TMP and 2-me-
2
allowed to overlap with the ferryl iron dz orbital but the a1u
orbital is not. Moreover, the π-spin density of the pyrrole
nitrogen is large in the a2u orbital but not in the a1u orbital
(see Figure 10). These differences should result in differing
spin transfer between the a1u and a2u complexes. This
argument is applicable to the axial oxo ligand of oxoiron(IV)
porphyrin π-cation radical complex, suggesting different spin
density on the oxo ligand in the a1u and a2u radical complexes.
The effect of the axial ligand on spin interaction between
the ferryl iron and porphyrin π-cation radical spins of
oxoiron(IV) porphyrin π-cation radical complex is manifested
in its EPR spectrum. The difference in spin interaction
1
thylimidazole, which was detected in its H NMR spectra.
As reported for 1,2-dimethylimidazole complexes of ferric
TMP,41 the steric interaction might induce ruffling of the
(39) Cheng, R.-J.; Chen, P.-Y.; Lovell, T.; Liu, T.; Noodleman, L.; Case,
D. A. J. Am. Chem. Soc. 2003, 125, 6774–6783.
(40) Fujita, I.; Hanson, L. K.; Walker, F. A.; Fajer, J. J. Am. Chem. Soc.
1983, 105, 3296–3300.
(41) Munro, O. Q.; Marques, H. M.; Debrunner, P. G.; Mohanrao, K.;
Scheidt, W. R. J. Am. Chem. Soc. 1995, 117, 935–954.
Inorganic Chemistry, Vol. 48, No. 6, 2009 2623

Takahashi et al.
porphyrin plane of (TMP+•)FeIVO(2-MeIm), leading to a
drastic decrease in epoxidation reactivity.
drogen peroxide, we could not find a significant difference
between the imidazole and phenolate axial ligands. While
the axial ligand activates the compound I in heme enzyme,
the axial ligand would not be an essential factor to
discriminate the enzyme function.
A drastic axial ligand effect of imidazole and phenolate
derivatives was also found in the reactions with 1,4-
cyclohexadiene, N,N-dimethyl-p-nitroaniline, and hydrogen
peroxide. However, the axial ligand effect (∼10-fold) for
1,4-cyclohexadiene is not so significant as those (∼100-fold)
for N,N-dimethyl-p-nitroaniline and cyclooctene (Table 6).
Although we attempted to correlate these reactivities with
the present spectroscopic results of oxiron(IV) porphyrin
π-cation radical complexes, we could not find any clear
correlation. Therefore, the explanation for the dramatic axial
ligand effect on the reactivity must be effects other than the
electronic state of the oxiron(IV) porphyrin π-cation radical
complex. For example, differences in the reaction mecha-
nism, such as hydrogen abstraction and electron transfer,
would form different transition states, which might result in
different axial ligand effects. The axial ligand effect on
oxoiron(IV) porphyrin π-cation radical complex would be
different from that on iron(III) porphyrin complex as a
reaction product. Moreover, the thermodynamic parameters
for substrate oxidation must also be considered. For example,
the change in an enthalpy (∆H) for conversion of 1,4-
cyclohexadine to benzene and water is ∼ -260 kJ/mol, while
that of cyclohexene to cyclohexene oxide is ∼ -120 kJ/
mol.42 The large enthalpy contribution from the substrate
oxidation would weaken the enthalpy contribution from the
axial ligand effect.
Previously, the axial ligand effect of imidazole and
phenolate has been reported in catalytic oxygenation reac-
tions with iron and manganese porphyrin complexes.12-17
Previously, Watanabe et al. studied the axial ligand effect
on the former process, in which the coordination of the axial
imidazole ligand increased the rate of the O-O bond
cleavage of iron bound m-chloroperbenzoate and the rate
correlated with the donor effect of the axial imidazole
ligand.5 On the other hand, this study clearly showed that
the axial imidazole and phenolate ligands drastically in-
creased the reaction rates of oxoiron(IV) porphyrin π-cation
radical complexes with substrates. In the catalytic system,
the axial ligand works to accelerate not only the formation
of the high-valent oxo complex but also the oxo-transfer to
substrate. This would be the same in the heme enzymes. The
proximal histidine and tyrosine residues would drastically
increase the reactivity of compound I, resulting in much more
reactive compound I in peroxidases and catalases than in
the synthetic model complexes. The very reactive compound
I would lead to extremely high catalytic activity of these
heme enzymes. Although this study does not show the effect
of the thiolate axial ligand, the cysteine proximal ligands in
cytochrome P450 may also work to enhance the oxo-transfer
reactivity. Finally, in the oxidation reactions of cyclooctene,
1,4-cyclohexadiene, N,N-dimethyl-p-nitroaniline, and hy-
Experimental Section
Instrumentation. UV-visible absorption spectra were recorded
on an Agilent 8453 (Agilent Technologies) equipped with a USP-203
1
2
low-temperature chamber (UNISOKU). H, H, 13C, and 19F NMR
1
2
were measured on a Lambda-500 spectrometer (JEOL). H, H, and
13C NMR chemical shifts were reported versus tetramethylsilane (TMS)
and referenced to residual solvent peaks (dichloromethane: 5.32 ppm).
19F chemical shift values were referenced to external hexafluorobenzene
(C6F6) at 0 ppm. The concentrations of NMR samples were ∼5 mM.
EPR spectra were recorded at 4 K on Bruker E500 X-band spectrom-
eter with an Oxford Instruments EPR910 helium-flow cryostat
(Bruker). The concentrations of EPR samples were 1-3 mM. ESI-
MS spectra were obtained with a LCT time-of-flight mass spectrometer
equipped with an elecrospray ionization interface (Micromass). For
oxoiron(IV) porphyrin π-cation radical complexes, desolvation and
nebulizer gases were cooled with liquid nitrogen. The ESI-MS samples
were prepared by oxidation with ozone in acetonitrile or dichlo-
romethane at 233 K, followed by dilution with the same solvent, and
injected into the ESI-MS with a gastight syringe cooled with dry ice.
The sample solution was then transferred to the inlet of the mass
spectrometer through a precooled capillary tube. Gas chromatogra-
phy-mass spectroscopy (GC-MS) analysis was performed on a QP-
5000 GC-MS system (Shimadzu) equipped with a capillary gas
chromatograph (GC-17A, CBP5-M25-025 capillary column). Ozone
gas was generated by UV irradiation of oxygen gas (99.9%) with an
ozone generator PR-1300 (Clear Water) and used without further
purification.
Materials. Imidazole-d4 (98 atom % D) was purchased from
Aldrich. 2-Methylimidazole-d6 (98.1 atom % D) was purchased from
CDN ISOTOPES (Canada). 4-Methylimidazole-d6 was prepared from
4-methylimidazole by activated 10% Pd/C in D2O, according to the
literature.43 Deuterium incorporation into 4-methylimidazole, as ana-
lyzed by 1H and 13C NMR, was 2-H (95%), 5-H (95%), and 4-methyl
group (50%). Dichloromethane was purchased from a commercial
company as the anhydrous solvent and was stored in the presence of
4A molecular sieves. Other chemicals were purchased from com-
mercial companies and used without further purification. meso-
Tetramesitylporphyrin (TMPH2) and 2,7,12,17-teramethyl-3,8,13,18-
tetramesitylporphyrin (TMTMPH2) were prepared according to the
literature.44 (TMP)FeIIICl and (TMTMP)FeIIICl were prepared by the
insertion of iron into TMPH2 and TMTMPH2 with FeCl2 and sodium
acetate in acetic acid, respectively, and purified by silica gel column
using CH2Cl2/CH3OH as an eluent. (TMP)FeIIIOH and (TMTMP)-
FeIIIOH were obtained from (TMP)FeIIICl and (TMTMP)FeIIICl,
respectively, by passing through a basic alumina (10% water) column
with dichloromethane as an eluent.45 (TMP)FeIIIClO4 and (TMT-
MP)FeIIIClO4 were prepared by the reaction of (TMP)FeIIICl and
(TMTMP)FeIIICl, respectively, with silver(I) perchlorate in dichlo-
romethane solution, and purified by recrystallization from dichlo-
(43) Hardacre, C.; Holbrey, J. D.; McMath, S. E. J. Chem. Commun. 2001,
367–368.
(44) (a) Lindsey, J.; Wagner, R. J. Org. Chem. 1989, 54, 828–836. (b)
Ono, N.; Kawamura, H.; Bougauchi, M.; Maruyama, K. Tetrahedron
1990, 46, 7483–7496.
(45) Groves, J. T.; Gross, Z.; Stern, M. K. Inorg. Chem. 1994, 33, 5065–
5072.
(42) (a) Pedley, J. B.; Naylor, R. D.; Kirby, S. P. Thermochemical Data of
Organic Compounds; Chapman and Hall: London, 1986. (b) Stull,
D. R.; Westrum, E. F.; Sinke, G. C. The Chemical Thermodynamics
of Organic Compounds; John Wiley: New York, 1969. (c) Nagano,
H. In Kagaku-Binran, 5th ed.; Atake, T., Sorai, M., Matsuo, T., Eds.;
Maruzen: Tokyo; Vol. II, pp 301-313.
2624 Inorganic Chemistry, Vol. 48, No. 6, 2009

Oxoiron(IV) Porphyrin π-Cation Radical Complexes
romethane/n-hexane.46 (TMP)FeIIINO3 and (TMTMP)FeIIINO3 were
obtained by mixing (TMP)FeIIIOH and (TMTMP)FeIIIOH in toluene
with 1 M nitric acid, and purified by recrystallization from ether/n-
UV-vis. (CH2Cl2, 25 °C), 412, 511, 572 and 692 nm. [(TMP)FeIII(5-
1
MeIm)]ClO4; H NMR (CD2Cl2, 25 °C), 13.0 and 13.9 (meta-H),
4.4 (para-Me), 31.7 (pyrrole-H), -18.6 (5-MeIm 2-Me), 31.7 (5-
MeIm 4-H), 22.0 (5-MeIm 5-Me), 100.8 (5-MeIm N-H). UV-vis.
(CH2Cl2, 25 °C), 412, 510, 572, and 692 nm. [(TMTMP)FeIII-
hexane.47 (TMP)FeIII(NO3); H NMR (CD2Cl2, 25 °C), 4.0 and 6.2
1
(ortho-Me), 15.3 and 16.5 (meta-H), 4.3 (para-Me), 74.0 (pyrrole-H).
UV-vis. (CH2Cl2, 25 °C), 412, 514, 580, 647, and 691 nm.
(TMTMP)FeIII(NO3); 1H NMR (CD2Cl2, 25 °C), -49.6 (meso-H), 5.1
and 8.4 (ortho-Me), 12.9 and 13.3 (meta-H), 6.3 (para-Me), 58.7
(pyrrole-Me). UV-vis. (CH2Cl2, 25 °C), 388, 506, 532, and 637 nm.
Synthesis. Iron(III) Phenolate Complexes. Iron(III) phenolate
complexes were prepared by the published method.48 (TMP)FeIIIOH
and (TMTMP)FeIIIOH were dissolved in toluene and 1 equiv of
phenol derivative (4-nitrophenol, pentafluorophenol, 3,4,5-trifluo-
rophenol, or 3-fluoro-4-nitrophenol) was added with stirring. The
green or red solution changed to brown as the phenol dissolved.
The solution was evaporated to dryness, and the residue was purified
by crystallization from ether/hexane. Microcrystals of the iron(III)
phenol complex were collected by filtration. (TMP)FeIII(4-NO2-
PhO); 1H NMR (CD2Cl2, 25 °C), 3.0 and 5.2 (ortho-Me), 12.5 and
13.5 (meta-H), 3.7 (para-Me), 79.1 (pyrrole-H), -91.4 (phenol
ortho-H), 83.0 (phenol meta-H). UV-vis. (CH2Cl2, 25 °C), 420,
497, 559, 635, and 667 nm. (TMP)FeIII(F5-PhO); 1H NMR (CD2Cl2,
25 °C), 2.9 and 5.2 (ortho-Me), 12.7 and 13.6 (meta-H), 3.8 (para-
Me), 78.7 (pyrrole-H). UV-vis. (CH2Cl2, 298 K), 415, 500, 566,
642, and 673 nm. (TMP)FeIII(3,4,5-F3-PhO); 1H NMR (CD2Cl2, 25
°C), 3.0 and 5.1 (ortho-Me), 12.3 and 13.3 (meta-H), 3.7 (para-
Me), 79.8 (pyrrole-H), -101.0 (phenol ortho-H). UV-vis. (CH2Cl2,
25 °C), 420, 490, 557, 620, and 666 nm. (TMP)FeIII(3-F-4-NO2-
PhO); 1H NMR (CD2Cl2, 25 °C), 3.1 and 5.2 (ortho-Me), 12.7 and
13.7 (meta-H), 3.8 (para-Me), 78.9 (pyrrole-H), -85.1 and -81.5
(phenol ortho-H), 72.4 (phenol meta-H). 19F NMR (CD2Cl2, 25 °C),
58.4 (meta-F). UV-vis. (CH2Cl2, 25 °C), 419, 500, 560, 640, and
668 nm. (TMTMP)FeIII(F5-PhO); 1H NMR (CD2Cl2, 25 °C), -40.5
(meso-H), 4.8 and 7.4 (ortho-Me), 11.9 and 12.0 (meta-H), 6.0
(para-Me), 47.7 (pyrrole-Me), -83.1 and -79.9 (phenol ortho-
H), 67.9 (phenol meta-H). UV-vis. (CH2Cl2, 25 °C), 371, 398,
1
(Im)]ClO4; H NMR (CD2Cl2, 25 °C), -16.6 (meso-H), 4.8 and
8.0 (ortho-Me), 13.5 and 14.4 (meta-H), 5.5 (para-Me), 69.2
(pyrrole-Me), -16.1 (Im 2-H), 31.2 (Hm 4-H), 82.6 (Im 5-H), 98.5
(Im N-H). UV-vis. (CH2Cl2, 25 °C), 385, 503, and 626 nm.
1
[(TMTMP)FeIII(2-MeIm)]ClO4; H NMR (CD2Cl2, 25 °C), -31.9
(meso-H), 5.1 and 8.8 (ortho-Me), 13.8 and 14.5 (meta-H), 6.2
(para-Me), 70.0 (pyrrole-Me), 17.5 (2-MeIm 2-Me), 35.9 (2-MeIm
4-H), 66.6 (2-MeIm 5-H), 110.0 (2-MeIm N-H). UV-vis. (CH2Cl2,
298K), 385, 503, and 626 nm. [(TMTMP)FeIII(5-MeIm)]ClO4; 1H
NMR (CD2Cl2, 298 K), -20.1 (meso-H), 5.0 and 8.2 (ortho-Me),
13.7 and 14.5 (meta-H), 5.8 (para-Me), 70.3 (pyrrole-Me), -15.4
(5-MeIm 2-H), 32.6 (5-MeIm 4-H), 20.7 (5-MeIm 5-H), 99.5 (5-
MeIm N-H). UV-vis. (CH2Cl2, 25 °C), 385, 503, and 626 nm.
Oxoiron(IV) Porphyrin Complexes. Oxoiron(IV) porphyrin
complexes were prepared as previously described.45 (TMP)-
FeIII(ClO4)2 or (TMTMP)FeIII(ClO4)2, which was prepared from the
oxidation of (TMP)FeIIIOH or (TMTMP)FeIIIOH with solid ferric
perchlorate, in dichloromethane-d2 was transferred through a short
basic alumina (20% water) column (0.5 × 1 cm) at ambient
temperature, directly into a NMR tube in dry ice-acetone bath. One
equiv of 1-MeIm or 1,2-DiMeIm in dichloromethane-d2 was slowly
added to the cooled solution in the NMR tube.
Ozone Oxidation of Iron(III) Porphyrin. For NMR or EPR
measurement, iron(III) porphyrin complexes were dissolved in
dichloromethane and placed in NMR or EPR cell. The solution
was cooled by a dry ice acetone bath. Ozone gas was slowly
bubbled in the solution with a gastight syringe. With the formation
of oxoiron(IV) porphyrin π-cation radical complexes, the brown
1
solution changed to green. The oxidations were monitored by H
NMR or EPR spectra. For UV-visible absorption spectra, the
oxidation was carried out in a 1 cm quartz cuvette in a low-
temperature chamber set with a UV-visible absorption spectrometer
and monitored by absorption spectral change. Finally, excess ozone
gas was removed by bubbling argon gas with a gastight syringe.
Kinetics and Product Analysis. Oxoiron(IV) porphyrin π-cation
radical (100 µM) in dichloromethane was prepared in a 1 cm quartz
cuvette in a low-temperature chamber set with a UV-visible
absorption spectrometer, as described above. An excess of cy-
clooctene (20-1000 equiv) was then added to the solution with
vigorous stirring, and the reactions were monitored by the absorption
spectral change at constant time intervals. After confirming the
completion of the reactions, 10 equiv of tetrabutylammonium iodide
(n-Bu4N+I-) was added to the solution at the same temperature.
After warming to room temperature, quantitative product analyses
were subsequently performed with GC-MS using undecane as an
internal standard. The reaction rate constants were determined by
computer simulation of absorption versus time for the reactions.
1
498, 526, and 619 nm. (TMTMP)FeIII(3-F-4-NO2-PhO); H NMR
(CD2Cl2, 25 °C), -47.0 (meso-H), 4.5 and 7.3 (ortho-Me), 11.9
and 12.0 (meta-H), 5.0 (para-Me), 47.0 (pyrrole-Me), -83.1 and
-79.9 (phenol ortho-H), 67.9 (phenol meta-H). 19F NMR (CD2Cl2,
25 °C), 58.3 (meta-F). UV-vis. (CH2Cl2, 25 °C), 370, 401, 496,
524, and 619 nm.
Iron(III) Monoimidazole Complexes. Iron(III) porphyrin
monoimidazole complex was generated in situ by adding 1.0 equiv
of imidazole, 2-methylimidazole, or 4-methylimidazole to perchlo-
rate iron(III) porphyrin in dichloromethane.49 [(TMP)FeIII(Im)]ClO4;
1H NMR (CD2Cl2, 25 °C), 3.1 and 5.3 (ortho-Me), 12.6 and 13.4
(meta-H), 4.3 (para-Me), 19.9 (pyrrole-H), -15.5 (Im 2-H), 30.1
(Im 4-H), 85.1 (Im 5-H), 100.4 (Im N-H). UV-vis. (CH2Cl2, 25
1
°C), 412, 510, 575, and 690 nm. [(TMP)FeIII(2-MeIm)]ClO4; H
NMR (CD2Cl2, 298K), 2.7 and 5.7 (ortho-Me), 12.7 and 13.1 (meta-
H), 4.1 (para-Me), 34.9 (pyrrole-H), 16.6 (2-MeIm 2-Me), 31.8
(2-MeIm 4-H), 65.8 (2-MeIm 5-H), 109.0 (2-MeIm N-H).
Acknowledgment. This work was supported by grants
from the Japan Science and Technology Agency, CREST,
and from the Ministry of Education, Culture, Sports, Science
and Technology, Japan, the Global COE program.
(46) Reed, C. A.; Mashiko, T.; Bentley, S. P.; Kastner, M. E.; Scheidt,
W. R.; Spartalian, K.; Lang, G. J. Am. Chem. Soc. 1979, 101, 2948–
2958.
(47) Phillippi, M. A.; Baenziger, N.; Goff, H. M. Inorg. Chem. 1981, 20,
3904–3911.
(48) (a) Tang, S. C.; Koch, S.; Papaefthymiou, G. C.; Foner, S.; Frankel,
R. B.; Ibers, J. A.; Holm, R. H. J. Am. Chem. Soc. 1976, 98, 2414–
2434. (b) Ainscough, E. W.; Addison, A. W.; Dolphin, D.; James,
B. R. J. Am. Chem. Soc. 1978, 100, 7585–7591.
(49) (a) Quinn, R.; Nappa, M.; Valentine, J. S. J. Am. Chem. Soc. 1982,
104, 2588–2595. (b) Scheidt, W. R.; Geiger, D. K.; Lee, Y. J.; Reed,
C. A.; Lang, G. J. Am. Chem. Soc. 1985, 107, 5693–5699.
Supporting Information Available: Text (PDF) containing
additional figures (Figure S1-S17) and tables (Tables S1 and S2).
This material is available free of charge via the Internet at
http://pubs.acs.org.
IC802123M
Inorganic Chemistry, Vol. 48, No. 6, 2009 2625