Structures and properties of octaethylporphinato(phenolate)iron(III) complexes with NH?O hydrogen bonds: Modulation of Fe-O bond character by the hydrogen bond

Article abstract of DOI:10.1016/j.ica.2004.09.014

Iron(III) porphinate complexes of phenolate that have NH?O hydrogen bonds on the coordinating oxygen were synthesized as models of heme catalase. The NH?O hydrogen bonds in the complexes lower an electron donation from the coordinating oxygen to iron and induce a long Fe-O bond and a positive redox potential. Iron(III) porphinate complexes of phenolate that have NH?O hydrogen bonds on the coordinating oxygen, [Fe^III(OEP){O-2,6-(RCONH) ₂C₆H₃}] (R = CF₃ (1), CH₃ (3)) and [Fe^III(OEP)(O-2-RCONHC₆H₄)] (R = CF₃ (2), CH₃ (4)) (OEP = 2,3,7,8,12,13,17,18-octaethyl- 21H, 23H-porphinato), were synthesized and characterized as models of heme catalase. The presence of NH?O hydrogen bonds was established by their crystal structures and IR shifts of the amide NH band. The crystal structure of 1 shows an extremely elongated Fe-O bond, 1.926(3) ?, compared to 1.887(2) ? in 2 or 1.848(4) ? in [Fe^III(OEP)(OPh)]. The NH?O hydrogen bond decreases an electron donation from oxygen to iron, resulting in a long Fe-O bond and a positive redox potential.

Full text of DOI:10.1016/j.ica.2004.09.014

Inorganica Chimica Acta 358 (2005) 331–338
www.elsevier.com/locate/ica
Structures and properties of octaethylporphinato(phenolate)iron(III)
complexes with NHꢀ ꢀ ꢀO hydrogen bonds:
modulation of Fe–O bond character by the hydrogen bond
Daisuke Kanamori, Yusuke Yamada, Akira Onoda, Taka-aki Okamura, Seiji Adachi,
*
Hitoshi Yamamoto, Norikazu Ueyama
Department of Macromolecular Science, Graduate School of Science, Osaka University, 1-1 Machikaneyama-cho, Toyonaka, Osaka 560-0043, Japan
Received 27 July 2004; accepted 6 September 2004
Available online 30 November 2004
Abstract
Iron(III) porphinate complexes of phenolate that have NHꢀ ꢀ ꢀO hydrogen bonds on the coordinating oxygen, [FeIII(OEP){O-2,6-
(RCONH)2C6H3}] (R = CF3 (1), CH3 (3)) and [FeIII(OEP)(O-2-RCONHC6H4)] (R = CF3 (2), CH3 (4)) (OEP = 2,3,7,8,12,13,17,18-
octaethyl-21H, 23H-porphinato), were synthesized and characterized as models of heme catalase. The presence of NHꢀ ꢀ ꢀO hydrogen
bonds was established by their crystal structures and IR shifts of the amide NH band. The crystal structure of 1 shows an extremely
elongated Fe–O bond, 1.926(3) A, compared to 1.887(2) A in 2 or 1.848(4) A in [FeIII(OEP)(OPh)]. The NHꢀ ꢀ ꢀO hydrogen bond
decreases an electron donation from oxygen to iron, resulting in a long Fe–O bond and a positive redox potential.
ꢀ 2004 Elsevier B.V. All rights reserved.
˚
˚
˚
Keywords: Hydrogen bond; Phenolate complexes; Porphyrin complexes; Iron complexes; Catalase; X-ray analysis
1. Introduction
considered to be predominant in catalytic activity [13].
However, the NHꢀ ꢀ ꢀO hydrogen bonds on the axial phe-
nolate must modulate the Fe–O bond character and the
redox potential of iron because of chemical similarity.
Computational studies examining the axial phenolate
and its hydrogen bonds have recently reported that the
NHꢀ ꢀ ꢀO hydrogen bond positively shifts the oxidation
potential of the axial ligand and lowers the Fe–O bond or-
der in high-valent states, compound I and II [14,15]. In
our previous report, model complexes having NHꢀ ꢀ ꢀO
hydrogen bonds on the coordinating phenolate, [FeIII
(TPP)(O-2-CF3CONHC6H4)] and [FeIII(TPP){O-2,6-
(CF3CONH)2C6H3}] (TPP = tetraphenylporphinato),
were synthesized [11]. In that study, the positive shift of
the redox potential of FeIII/FeII by the NHꢀ ꢀ ꢀO hydrogen
bonds to the coordinating oxygen has been established
but modulation of the Fe–O bond character has not.
The present study provides an indepth discussion of
the effects of NHꢀ ꢀ ꢀO hydrogen bonds, especially regard-
Heme catalases have a high catalytic ability to dispro-
portion harmful hydrogen peroxide with surprisingly
high acceleration [1,2]. They have a tyrosinate, which
functions as an axial ligand, with double NHꢀ ꢀ ꢀO hydro-
gen bonds from the arginine guanidinium group [3–9].
The ‘‘push–pull’’ concept is generally used to explain
the activation of substrates in heme enzymes. The major
role of axial ligands is considered to be ‘‘push’’. However,
we have previously established additional roles of an axial
thiolate in P450 model complexes [10–12]. In these com-
plexes, NHꢀ ꢀ ꢀS hydrogen bonds on the axial thiolate
modulate the Fe–S bond character and the redox poten-
tial of iron. In the case of catalase, the roles of axial phe-
nolate have been poorly investigated because ‘‘pull’’ is
*
Corresponding author. Tel.: +81 6 6850 5449; fax: +81 6 6850
5474.
E-mail address: ueyama@chem.sci.osaka-u.ac.jp (N. Ueyama).
0020-1693/$ - see front matter ꢀ 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2004.09.014

332
D. Kanamori et al. / Inorganica Chimica Acta 358 (2005) 331–338
2.2.2. 2,6-Bis(acetylamino)phenol
To an aqueous solution (10 ml) of 2,6-diaminophenol
dihydrochloride (1.8 g, 9.1 mmol), acetic anhydride (3
ml, 32 mmol) and sodium acetate trihydrate (3.5 g, 26
mmol) were added and heated. After cooling down to
room temperature, there appeared white needles. Yield:
1.2 g (70%); m.p. 178–179 ꢁC. Anal. Calc. for
C10H12N2O3: C, 57.68; H, 5.81; N, 13.45. Found: C,
57.64; H, 5.77; N, 13.41%. 1H NMR (DMSO-d6); d
ppm 9.99 (OH), 9.70 (NH), 7.40 (m-H), 6.82 (p-H),
2.18 (CH3).
Scheme 1. Phenolate ligands with hydrogen bonds and steric
hindrance.
ing Fe–O bond character, by examining what role the
number of hydrogen bonds plays and by making a com-
parison with a steric effect. Various simple 2-mono(acyl-
amino)- and 2,6-bis(acylamino)phenolate ligands were
used as models of the axial of catalase. The 2,6-di-isopro-
pylphenolate ligand was also used as non-hydrogen
bonding model with a similar bulkiness in order to
approximate the steric effect of the acylamino group
(Scheme 1). A series of X-ray analyses of novel model
complexes with different numbers of hydrogen bonds,
[FeIII(OEP){O-2,6-(CF3CONH)2C6H3}], [FeIII(OEP)-
2.2.3. 2,6-Bis(trifluoroacetylamino)phenol
To a CH2Cl2 solution (20 ml) of 2,6-diaminophenol
dihydrochloride (0.5 g, 2.5 mmol), including triethylam-
ine (1.6 ml, 11 mmol), trifluoroacetic anhydride (0.90 ml,
5.6 mmol) was slowly added at 0 ꢁC. After the solution
was stirred for 60 min, 10 ml of water was added and
then CH2Cl2 was removed under reduced pressure.
The residue was dissolved in ethyl acetate (100 ml) and
the organic layer was washed with 100 ml of water twice.
After drying over anhydrous Na2SO4, the solution was
concentrated. The obtained residue was recrystallized
from diethyl ether/n-hexane. Yield: 340 mg (42%); m.p.
136–138 ꢁC. Anal. Calc. for C10H6N2O3F6: C, 37.99;
H, 1.91; N, 8.86. Found: C, 37.84; H, 1.83; N, 9.47%.
1H NMR (DMSO-d6); d ppm 10.70 (NH), 9.81 (OH),
7.37 (m-H), 6.96 (p-H).
(O-2-CF3CONHC6H4)],
[FeIII(OEP){O-2,6-(i-Pr)2C6
H3}] and [FeIII(OEP)(OPh)] (OEP = 2,3,7,8,12,13,17,
18-octaethyl-21H,23H-porphinato), have been per-
formed. Changes in the Fe–O bond character that appear
to be connected to the number of hydrogen bonds were
discussed based on a comparison of the structures. Other
properties associated with these complexes are also pre-
sented in this paper.
2.2.4. 2-Trifluoroacetylaminophenol
2-Trifluoroacetylaminophenol was synthesized from
2-aminophenol in a similar method to 2,6-bis(trifluoro-
acetylamino)phenol. 1H NMR (DMSO-d6); d ppm
10.51 (NH), 9.93 (OH), 7.38 (3-H), 7.20 (5-H), 6.99 (6-
H), 6.89 (4-H).
2. Experimental
2.1. Materials
All procedures were performed under Ar atmosphere
by the Schlenk technique except for ligand syntheses. All
organic solvents were dried over CaH2 and distilled un-
der Ar atmosphere before use. [FeIII(OEP)]2O and
[FeIII(OEP)(OPh)] (5) were synthesized by the reported
method [16]. Other reagents including 2-acetylaminophe-
nol were purchased from Nacalai Tesque or Tokyo
Chemical Industry and used without further purification.
2.3. Synthesis of [FeIII(OEP){O-2,6-(CF3CONH)2
C6H3}] (1), [FeIII(OEP)(O-2-CF3CONHC6H4)] (2),
[FeIII(OEP){O-2,6-(CH3CONH)2C6H3}] (3), [FeIII
(OEP)(O-2-CH3CONHC6H4)] (4), and [FeIII(OEP)
{O-2,6-(i-Pr)2C6H3}] (6)
[FeIII(OEP)]2O was mixed with corresponding phenol
(1 equiv. for 1–4 and 5 equiv. for 6) in CH2Cl2 at room
temperature. After being stirred for 1 h, the solution was
concentrated to dryness. 1 was recrystallized from tolu-
2.2. Synthesis of axial ligands
2.2.1. 2,6-Diaminophenol dihydrochloride
ene/n-hexane,
2 was from dichloromethane/diethyl
2,6-Dinitrophenol (1.5 g, 8.2 mmol) was suspended in
concentrated hydrochloric acid (36 ml). Then, tin pow-
der (3.0 g) was added slowly to the mixture with stirring
at room temperature. After this solution turned yellow,
it was slowly cooled to 4 ꢁC. The obtained white needles
were collected and recrystallized from hot water/concen-
trated hydrochloric acid. Yield: 1.5 g (91%) H NMR
(DMSO-d6); d ppm 9.02 (OH, NH3), 7.11 (m-H), 6.89
(p-H).
ether, 3 and 4 were from toluene, and 6 was from tolu-
ene/acetonitrile. 1–4 were dark purple crystals and 6 was
black.
2.3.1. [FeIII(OEP){O-2,6-(CF3CONH)2C6H3}] (1)
Anal. Calc. for C46H49N6O3FeF6: C, 61.13; H, 5.46;
N, 9.30. Found: C, 60.18; H, 5.30; N, 9.14%. UV/Vis
(CH2Cl2, r.t.); kmax (e · 10ꢁ4 cmꢁ1 Mꢁ1) 625 (6.8), 527
1

D. Kanamori et al. / Inorganica Chimica Acta 358 (2005) 331–338
333
1
(11.5), 497 (12.0), 384 (110). H NMR (CDCl3); d ppm
lows. The same volume of Triton X-100 was added
to DMF solution (100 mM) of each sample. This solu-
tion was diluted by four times the volume of water.
The pH of the solution (10 mM) was determined using
a Metrohm 716 DMS titrino, which is combined with
Metrohm 728 stirrer and a saturated calomel LL micro
pH glass electrode. The saturated calomel glass elec-
trode purchased from Metrohm was calibrated by
0.05 M C6H4(COOH)(COOK) buffer (pH 4.01) and
0.025 M KH2PO4 and 0.025 M Na2HPO4 buffer (pH
6.86).
71.9 (3-H), ꢁ81.6 (4-H), ꢁ33.8 (meso), 38.7, 42.8
(–CH2–), 6.1 (–CH3), 3.6 (NH).
2.3.2. [FeIII(OEP)(O-2-CF3CONHC6H4)] (2)
Anal. Calc. for C44H49N5O2FeF3: C, 66.66; H, 6.23;
N, 8.83. Found: C, 66.10; H, 6.15; N, 8.77%. UV/Vis
(CH2Cl2, r.t.); kmax (e · 10ꢁ4 cmꢁ1 Mꢁ1) 614 (5.7), 523
(8.52), 496 (8.67), 391 (79.0). 1H NMR (CDCl3); d
ppm ꢁ89 (6-H), 90.4 (3-H), 65.4 (5-H), ꢁ90.9 (4-H),
ꢁ35.5 (meso), 39.3, 36.4 (–CH2–), 5.9 (–CH3).
2.3.3. [FeIII(OEP){O-2,6-(CH3CONH)2C6H3}] (3)
Anal. Calc. for C46H55N6O3Fe Æ C7H8: C, 71.69; H,
7.15; N, 9.46. Found: C, 71.44; H, 7.16; N, 9.52%. UV/
Vis (CH2Cl2, r.t.); kmax (e · 10ꢁ4 cmꢁ1 Mꢁ1) 617 (7.7),
2.6. Ligand exchange reaction
The ligand exchange reactions were performed in
CDCl3 at 30 ꢁC. To a solution of [Fe(OEP)(OPh)] in
CDCl3 (0.6 ml, 5 mM), each of 1–4 was added in the
1
524 (11.7), 497 (13.2), 392 (117). H NMR (CDCl3); d
1
ppm 83.9 (3-H), ꢁ98.5 (4-H), ꢁ34.0 (meso), 38.5, 36.6
solid state using small funnel. H NMR spectra were
(–CH2–), 5.7 (–CH3), 3.3 (NH), 11.6 (acetyl-H).
measured after storing for 30 min at 30 ꢁC. The ex-
change ratio was determined by the integrated intensity
of the peaks in paramagnetic region. When phenol
was added to a solution of 1–4, 500 mM solution of
phenol in CDCl3 was added successively via microsy-
ringe. Other conditions were the same to the above
experiment.
2.3.4. [FeIII(OEP)(O-2-CH3CONHC6H4)] (4)
Anal. Calc. for C44H52N5O2Fe Æ (C7H8)0.25: C, 72.13;
H, 7.14; N, 9.19. Found: C, 72.30; H, 7.18; N, 9.56%.
UV/Vis (CH2Cl2, r.t.); kmax (e · 10ꢁ4 cmꢁ1 Mꢁ1) 617
(7.7), 524 (11.7), 497 (13.2), 392 (117). 1H NMR
(CDCl3); d ppm ꢁ108 (6-H), 96.2 (3-H), 78.5 (5-H),
ꢁ103.0 (4-H), ꢁ34.8 (meso), 35.7 (–CH2–), 5.4 (–CH3),
3.3 (NH), 9.8 (acetyl-H).
2.7. X-ray structure determination
Each single crystal of [FeIII(OEP){O-2,6-(CF3
CONH)2C6H3}] (1), [FeIII(OEP)(O-2-CF3CONHC6H4)]
(2), [FeIII(OEP)(OPh)] (5), and [FeIII(OEP){O-2,6-(i-
Pr)2C6H3}] (6) was sealed in a glass capillary under ar-
gon atmosphere for the X-ray measurement, which
was performed at 23 ꢁC on a Rigaku AFC5R diffrac-
tometer equipped with a rotating anode X-ray genera-
tor. The radiation used was Mo Ka monochromatized
2.3.5. [FeIII(OEP){O-2,6-(i-Pr)2C6H3}] (6)
UV/Vis (CH2Cl2, r.t.); kmax (e · 10ꢁ4 cmꢁ1 Mꢁ1) 593
1
(16.3), 482 (17.6), 396 (165). H NMR (CDCl3); d ppm
123.4 (3-H), ꢁ142.6 (4-H), ꢁ42.0 (meso), 30.8 (–CH2–),
5.1 (–CH3), 1.30, 0.91 (i-Pr, CH3).
2.4. Physical measurements
˚
with graphite (0.71069 A). An empirical absorption cor-
Visible spectra were recorded in dichloromethane
solution on a SHIMADZU UV-3100PC spectrophotom-
rection was applied. The basic crystallographic parame-
ters for 1, 2, 5, and 6 are listed in Table 1. Unit cell
dimensions were refined with 25 reflections. These stand-
ard reflections were chosen and monitored with every
150 reflections and did not show any significant change.
The structure was solved by heavy atom Patterson meth-
ods (PATTY [17]) for 1, 2, and 5 or a direct method
1
eter. H NMR spectra were taken on a JEOL EX-270
spectrometer in chloroform-d solution at 30 ꢁC. Tetra-
methylsilane was used as a standard (0 ppm). IR spectra
were recorded on a Jasco FT/IR 8300 spectrometer.
Samples were prepared as KBr pellets. Electrochemical
measurements were carried out using a BAS 100B/W
instrument in dichloromethane solution (1 mM) that
contained 0.1 M tetra-n-butylammonium perchlorate as
a supporting electrolyte. E1/2 value was referenced to
an SCE electrode at room temperature and a value
uncorrected for junction potential was obtained.
(
SHELXS-86 [18]) for 6 and expanded using Fourier
techniques. Non-hydrogen atoms were refined aniso-
tropically. Hydrogen atoms were placed at calcu-
lated positions. All calculations were performed using
1
the teXsan crystallographic software and SHELXL-97
[19].
2.5. pH titration
pKa measurements were performed in 10 mM aque-
ous or 10% Triton X-100 aqueous micellar solution at
25 ꢁC. The preparation of micellar solutions is as fol-
1
Crystal Structure Analysis Package, Molecular Structure Corpo-
ration, 1985 and 1992.

334
D. Kanamori et al. / Inorganica Chimica Acta 358 (2005) 331–338
Table 1
Crystallographic data for [FeIII(OEP){O-2,6-(CF3CONH)2C6H3}] (1), [FeIII(OEP)(O-2-CF3CONHC6H4)] (2), [FeIII(OEP)(OPh)] (5), and
[Fe(OEP){O-2,6-(i-Pr)2C6H3}] (6)
1
2
5
6
Chemical formula
Formula weight
Color
C46H49O3N6FeF6
903.77
black
C44H49O2N5FeF3
792.75
black
C42H49ON4Fe
681.72
black
C48H61ON4Fe
765.88
black
Crystal system
Lattice parameter
monoclinic
triclinic
triclinic
orthorhombic
˚
a (A)
14.298(4)
18.765(4)
17.462(4)
90
13.926(3)
14.868(3)
10.053(3)
99.74(2)
12.119(4)
15.122(4)
10.834(3)
97.65(2)
101.35(3)
111.05(2)
1771(1)
18.988(3)
25.312(3)
18.142(3)
90
˚
b (A)
˚
c (A)
a (ꢁ)
b (ꢁ)
c (ꢁ)
108.80(2)
90
103.16(2)
94.14(2)
1984.4(8)
90
90
3
˚
V (A )
4435(1)
P21/n (#14)
4
8719(2)
Pbca (#61)
8
ꢀ
ꢀ
Space group
Z
Dcalc (g cmꢁ3
l (Mo Ka) (cmꢁ1
Temperature (K)
Scan type
2hmax (ꢁ)
P1 ð#2Þ
P1 ð#2Þ
2
2
)
1.353
4.12
296
1.327
4.37
296
1.278
4.64
296
1.167
3.84
296
)
x–2h
52.0
x–2h
55.0
0.043
0.132
9482
9108
5319
1.03
x–2h
50.0
0.057
0.183
6573
6249
2730
0.96
x–2h
47.0
R1
0.051
0.143
9365
0.051
0.161
7079
wR2
Number of reflections (total)
Measured (unique)
8997
2824
7079
2359
Number of observations [I > 3r(I)]
Goodness-of-fit
0.95
0.92
3. Results and discussion
gen bonding decreases an electron donation from the
coordinating oxygen to iron. Fe–N4 plane distances
were shortened by the NHꢀ ꢀ ꢀO hydrogen bond. The
short Fe–N4 plane distance shows that the iron has a
low electron density and behaves like a high-valent state
in the hydrogen-bonded complexes. If the elongation of
the Fe–O bond was caused by the steric hindrance, the
Fe–N4 plane distances must be elongated as well as
Fe–O bond lengths [23]. A shortening of the Fe–N4
plane distances was also observed in P450 model com-
plexes [10,12]. In contrast to the large Fe–O–C bond an-
gle (170.6(4)ꢁ, nearly straight) of 6, the Fe–O–C bond
angles of 1, 2, and 5 are 122.8(3)ꢁ, 125.5(2)ꢁ, and
142.2(3)ꢁ, respectively. In terms of orbital overlapping,
a small Fe–O–C bond angle is not suitable. The small
Fe–O–C bond angles in 1 and 2 are probably geometri-
cally favorable to form the NHꢀ ꢀ ꢀO hydrogen bonds in
place of orbital overlapping. The elongated Fe–O bond
lengths and the small Fe–O–C bond angles show that
the NHꢀ ꢀ ꢀO hydrogen bonds reduce the s-character of
the coordinating oxygen.
3.1. Molecular structures
The molecular structures of 1, 2, 5, and 6 are shown
in Fig. 1. The crystal parameters of these complexes are
listed in Table 1, and the selected bond lengths and an-
gles are listed in Table 2. No significant intermolecular
interaction was detected in any of these cases. The amide
NH protons in 1 and 2 direct toward the coordinating
oxygen. Short distances between the coordinating oxy-
˚
gen and the amide NH protons (2.31 and 2.33 A in 1
˚
and 2.25 A in 2) suggest the existence of NHꢀ ꢀ ꢀO hydro-
gen bonds. The presence of hydrogen bonds was con-
firmed using IR spectra as discussed below.
The Fe–O bond length of the NHꢀ ꢀ ꢀO hydrogen
˚
bonded complexes (1.926(3) A for 1 and 1.887(2) A
˚
˚
for 2) is elongated by 0.039 A in proportion to the num-
ber of hydrogen bonds compared to that of the unsubsti-
˚
tuted phenolate complex (5), 1.848(4) A. The Fe–O
bond length in 1 is extremely long compared with other
˚
reported complexes (1.847–1.868 A) [11,20–22]. On the
other hand, the Fe–O bond length of non-hydrogen-
bonded complex with similar bulkiness (6) is short
3.2. Detection of NHꢀ ꢀ ꢀO hydrogen bonds using IR
spectroscopy
˚
(1.816(4) A), although 6 has methyne CH protons in a
similar situation to the NH protons in 1 and 2. The elon-
gation of the Fe–O bond in 1 and 2 is attributed to the
NHꢀ ꢀ ꢀO hydrogen bonds rather than to steric hin-
drance. These results indicate that the NHꢀ ꢀ ꢀO hydro-
Hydrogen bonds in doubly (1) and singly (2) hydro-
gen-bonded complexes were established by a shift of
the NH band in IR spectra. The m(NH) bands of 1
and the corresponding phenol were observed at 3379

D. Kanamori et al. / Inorganica Chimica Acta 358 (2005) 331–338
335
Fig. 1. Molecular structures of: (a) [Fe(OEP){O-2,6-(CF3CONH)2C6H3}] (1); (b) [Fe(OEP)(O-2-CF3CONHC6H4)] (2); (c) [Fe(OEP)(OPh)] (5); (d)
[Fe(OEP){O-2,6-(i-Pr)2C6H3}] (6). Hydrogen atoms except for amide NH and methyne CH are omitted for clarity.
Table 2
Selected bond lengths and angles of [FeIII(OEP){O-2,6-(CF3CONH)2C6H3}] (1), [FeIII(OEP)(O-2-CF3CONHC6H4)] (2), [FeIII(OEP)(OPh)] (5), and
[Fe(OEP){O-2,6-(i-Pr)2C6H3}] (6)
1
2
5
6
˚
Fe–O (A)
1.926(3)
122.8(3)
92
1.887(2)
125.5(2)
113
1.848(4)
142.2(3)
168
1.816(4)
170.6(4)
98
Fe–O–C (ꢁ)
Fe–O–C–C (ꢁ)
Fe–Nip (A) (mean)
˚
2.051
2.31, 2.33
0.427
2.060
2.25
0.443
2.061
2.071
2.32, 2.34
0.484
˚
N–Hꢀ ꢀ ꢀO for 1, 2 or CHꢀ ꢀ ꢀO for 6 (A)
˚
Fe to Nip plane (A)
0.467
and 3390 cmꢁ1, respectively. The low-wavenumber shift
of m(NH) (11 cmꢁ1) supports the suggestion that 1 has
NHꢀ ꢀ ꢀO hydrogen bonds, as indicated by X-ray analy-
sis. Similarly, the m(NH) band of 2 at 3359 cmꢁ1 is lower
than that of the corresponding phenol, 3388 cmꢁ1. The
larger shift of m(NH) in 2 (29 cmꢁ1) as compared with
the shift in 1 indicates the formation of a stronger
NHꢀ ꢀ ꢀO hydrogen bond in 2. This finding is consistent
with the results obtained from studying [(TPP)FeIII
{O-2,6-(CF3CONH)2C6H3}] and [(TPP)FeIII(O-2-CF3
CONHC6H4)], where a single NHꢀ ꢀ ꢀO hydrogen bond
is stronger than double hydrogen bonds [11].
[Fe(OEP)(S-2-RCONHC6H4)] (R = CF3, CH3), show a
larger shift of m(NH), 54–123 cmꢁ1 [12]. The results
show that the NHꢀ ꢀ ꢀS hydrogen bond is stronger than
the NHꢀ ꢀ ꢀO hydrogen bond. Even though NHꢀ ꢀ ꢀO
hydrogen bonds are weaker, the elongation of the Fe–
˚
O bond (0.078 A) is larger than that of the Fe–S bond
˚
(0.057 A) by NHꢀ ꢀ ꢀS hydrogen bonds. The coordinating
sulfur has a low s-character even in a non-hydrogen
bonded complex, [Fe(OEP)(SPh)], as suggested by a
small Fe–S–C bond angle (102.5(3)ꢁ). Elongation of
the Fe–O or Fe–S bond length depends on not only a de-
cline in the electron donation from the ligand to iron but
also a decline in the s-character of the coordinating
atom. In P450 model complexes, the Fe–S bond length
The P450 model complexes that have NHꢀ ꢀ ꢀS hydro-
gen bonds, [Fe(OEP){S-2,6-(RCONH)2C6H3}] and

336
D. Kanamori et al. / Inorganica Chimica Acta 358 (2005) 331–338
mostly depends on electron donation because the Fe–S–
C bond angles indicate that the s-character of the coor-
dinating sulfur exhibits no significant change as a result
of the introduction of the NHꢀ ꢀ ꢀS hydrogen bonds. This
is because the elongation of the Fe–O bond is larger
than that of the Fe–S bond.
complexes, namely absorption maxima [26], equilibrium
constant of ligand exchange reaction [26,27], and redox
potentials [28]. In an aqueous Triton X-100 micellar
solution, the pKa values of 2,6-(t-BuCONH)2C6H3OH,
2-t-BuCONHC6H4OH and 4-t-BuCONHC6H4OH were
determined (9.1, 9.6, and 10.1, respectively). Under
hydrophobic conditions, the pKa value of a phenol
derivative decreases in the presence of an NHꢀ ꢀ ꢀO
hydrogen bond. The hydrophobic conditions used here
realize the inside of catalase where the heme group ex-
ists. pKa values in the hydrophobic environments are
important because they are affected by the permittivity
of the surroundings. The results show that the neighbor-
ing amide groups to phenolic OH increase the acidity of
the phenolic OH as a result of NHꢀ ꢀ ꢀO hydrogen bond
formation during and after deprotonation. This effect of
neighboring amide groups has been reported in our pre-
vious papers on thiophenol and benzoic acid derivatives
[29,30]. Even in an aqueous solution, neighboring amide
groups decrease the pKa value of phenolic OH. For
example, we found the pKa values of 2,6-(CH3CONH)2-
C6H3OH and unsubstituted phenol in an aqueous 0.1 M
NaClO4 solution at 25 ꢁC to be 8.4 and 9.8, respectively.
A low pKa value indicates that a stabilization of the phe-
nolate anion by delocalization of the negative charge on
phenolate oxygen has occurred. This finding is consist-
ent with previous discussions, where the NHꢀ ꢀ ꢀO hydro-
gen bond decreases the electron-donating ability of the
coordinating oxygen.
1
3.3. H NMR spectra
1
Table 3 lists the H NMR chemical shifts of com-
plexes 1–6 in chloroform-d at 30 ꢁC. The general pattern
of the chemical shifts of phenolate ligand, which does
not seem to be affected by a spatial proximity to iron,
is described as follows: The ortho- and para-protons
show upfield shifts, whereas meta-protons show down-
field shifts [24,25]. The alternating shift pattern, which
is the opposite sign of the chemical shifts for meta- ver-
sus ortho- and para-protons, is indicative of p spin delo-
calization on the phenolate ligand. If the general pattern
was followed, the amide NH protons of 1–4 should ap-
pear downfield. However, the amide NH protons did
not show significant downfield shifts but instead ap-
peared at 3.3–3.6 ppm, slightly upfield compared to
the corresponding phenol. This upfield shift has been
observed in P450 model complexes that have NHꢀ ꢀ ꢀS
hydrogen bonds and has been explained by a direct
through NHꢀ ꢀ ꢀS hydrogen bond contact [12]. The re-
sults show the NHꢀ ꢀ ꢀO hydrogen bonding in 1–4.
The contact shifts of aryl protons in 1–4 are signifi-
cantly smaller than in the non-hydrogen-bonded com-
plex (6). The small contact shifts in 1–4 are because of
the weakened Fe–O bond by the NHꢀ ꢀ ꢀO hydrogen
bonds, which was found as the elongated Fe–O bond
in the crystal structure.
1
3.5. Ligand exchange reaction detected by H NMR
The ligand exchange reaction, defined as Eq. (1), be-
tween [Fe(OEP)(OAr)] and 10 equivalent PhOH was
monitored in CDCl3 solution at 30 ꢁC. The equilibrium
constants (K) for 1–4 are summarized in Table 4. The
reaction was monitored based on the integral intensity
of meta- or 3,5-aromatic protonsꢀ signals, and the order
of K was determined to be 4 ꢂ 2 > 3 > 1. The results
indicate that strongly NHꢀ ꢀ ꢀO hydrogen-bonded com-
3.4. pKa values of the axial ligand in aqueous solutions
and aqueous micellar solutions
It has been reported that pKa values of axial ligands
influence various characteristics of Fe(III) porphinato
Table 3
1H NMR chemical shifts of [FeIII(OEP)(OAr)] in CDCl3 at 30 ꢁC
b
3-Ha 5-H
4-H
ꢁ81.6
6-H
meso
–CH2–
–CH3
6.1
NH
3.6
COCH3
CH(CH3)2
CH(CH3)2
1
2
3
4
6
71.9
ꢁ33.8
38.7
42.8
39.3
36.4
38.5
36.6
35.7
c
90.4
65.4
83.9
ꢁ90.9
ꢁ98.5
ꢁ89
ꢁ35.3
ꢁ34.0
ꢁ34.8
ꢁ42.0
5.9
5.7
5.4
5.1
3.3
3.3
11.6
9.8
96.2
78.5
ꢁ103.0
ꢁ142.6
ꢁ108
c
123.4
30.8
1.30, 0.91
a
b
c
The peaks for 3-H and 5-H were distinguished by the difference of their longitudinal relaxation time (T1).
The peak for 6-H was also observed utilizing the difference of T1 from that of the 4-H.
Not detected.

D. Kanamori et al. / Inorganica Chimica Acta 358 (2005) 331–338
337
Table 4
Equilibrium constants between [Fe(OEP)(OAr)] and [Fe(OEP)(OPh)]
Table 6
Redox potentials of [FeIII/II(OEP)(OAr)] vs. SCE in CH2Cl2
in CDCl3 at 30 ꢁC ([Fe] = 5 mM, [PhOH]0 = 50 mM)
Epc
Epa
E1/2
[Fe(OEP)(OAr)]
K
1
2
3
4
5
6
ꢁ0.71
ꢁ0.74
ꢁ0.81
ꢁ0.82
ca. ꢁ0.15a
ꢁ0.49
ca. ꢁ0.43a
ꢁ0.61
[FeIII(OEP){O-2,6-(CF3CONH)2C6H3}] (1)
[FeIII(OEP)(O-2-CF3CONHC6H4)] (2)
[FeIII(OEP){O-2,6-(CH3CONH)2C6H3}] (3)
[FeIII(OEP)(O-2-CH3CONHC6H4)] (4)
a
2.56 · 10ꢁ4
7.09 · 10ꢁ3
1.66 · 10ꢁ3
>103a
ꢁ0.38
ꢁ0.59
ꢁ0.58
ꢁ0.70
ꢁ0.75b
ꢁ0.88
ꢁ0.91
Tentatively assigned by poor shape.
ꢁ0.84
Completely converted to [Fe(OEP)(OPh)] under this condition.
a
b
Ref. [28].
plexes have a smaller K value than weakly hydrogen-
bonded ones and that doubly NHꢀ ꢀ ꢀO hydrogen-bonded
complexes have a smaller K value than singly NHꢀ ꢀ ꢀO
hydrogen-bonded ones. The results show that the
NHꢀ ꢀ ꢀO hydrogen bonds support complexation and
are consistent with our previous findings concerning fer-
redoxin model complexes [31].
has a more positive FeIII/FeII redox potential. The redox
potentials of 1–6 in CH2Cl2 at room temperature are
summarized in Table 6. As previously determined with
NHꢀ ꢀ ꢀS hydrogen bonds [10–12,31], the NHꢀ ꢀ ꢀO hydro-
gen bonds also shift the FeIII/FeII redox potentials to the
positive side. Although the redox couples of FeIII/FeII in
1 are irreversible, a process that occurs most likely be-
cause of bond dissociation of the weakened Fe–O bond
during reduction, the positively shifted reduction poten-
tial of the ferric species supports the positive shift of the
redox potential. The results show that the NHꢀ ꢀ ꢀO
hydrogen bonds decrease the electron density on iron
as a result of decreasing the electron donation from
the coordinating oxygen to iron.
K
½FeIIIðOEPÞðOArÞꢃþPhOHꢀ½FeIIIðOEPÞðOPhÞꢃþArOH
½FeIIIðOEPÞðOPhÞꢃ½ArOHꢃ
K ¼
½FeIIIðOEPÞðOArÞꢃ½PhOHꢃ
ð1Þ
3.6. Electronic absorption spectra and electrochemical
properties
The maxima in the electronic absorption spectra of
the complexes, 1–6, are listed in Table 5. All spectra
show four absorption maxima in the range from 380
to 650 nm and one shoulder around 360 nm. It has been
reported that the wavelength of absorption maxima of
(OEP)Fe(III) phenolate complexes is proportional to
the acidity of the axial phenolate ligand [26]. In particu-
lar, the maximum around 600 nm is sensitive to the acid-
ity. The maxima of NHꢀ ꢀ ꢀO hydrogen-bonded
complexes (1–4) were observed in longer wavelengths
than that of non-hydrogen-bonded complexes (5 and
6) and shifted to longer wavelengths in proportion to
the strength or number of NHꢀ ꢀ ꢀO hydrogen bonds.
The correlation between the redox potentials of
[FeIII(OEP)(phenolato)] and the acidity of 4-substituted
phenol has also been reported [28]. The phenolate com-
plex, which has a more acidic phenol as an axial ligand,
4. Conclusions
The NHꢀ ꢀ ꢀO hydrogen bonding effects were deter-
mined to elongate the Fe–O bond and decrease the s-
character of the coordinating oxygen. The effects are
caused by a delocalization of the negative charge on
the coordinating oxygen by the hydrogen bonds and
leads to a positive shift of redox potential. The presence
of the NHꢀ ꢀ ꢀO hydrogen bonds induces significant
changes in various properties, although the direction
of the NH group toward the coordinating oxygen in
model complexes is not very suitable for hydrogen
bonding. In catalase, the NHꢀ ꢀ ꢀO hydrogen bond
should be more effective than in model complexes be-
cause the NH protons of the guanidinium group in cat-
alase direct to a lone pair of the coordinating oxygen in
a more suitable orientation.
Table 5
Absorption maxima of [FeIII(OEP)(OAr)] in CH2Cl2 at room
temperature
Appendix A. Supplementary material
kmax (e · 10ꢁ4/cmꢁ1 Mꢁ1
)
Crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallographic
Data Centre, CCDC Nos. 245729–245732 for com-
pounds 1, 2, 5, and 6. Copies of this information may
be obtained free of charge from The Director, CCDC,
12 Union Road, Cambridge, CB2 1EZ, UK (fax: +44
1223 336 033; e-mail: deposit@ccdc.cam.ac.uk or
1
625 (6.8)
614 (5.7)
617 (7.7)
611 (7.3)
603 (8.23)
593 (16.3)
527 (11.5)
523 (8.52)
524 (11.7)
519 (11.2)
520 (10.77)
497 (12.0)
496 (8.67)
497 (13.2)
494 (11.1)
490 (12.21)
482 (17.6)
384 (110)
2
391 (79.0)
392 (117)
393 (110)
3
4
5a
6
396 (165)
a
Ref. [26].

338
D. Kanamori et al. / Inorganica Chimica Acta 358 (2005) 331–338
[16] J.W. Buchler, H.H. Schneehage, Z. Naturforsch. B 28 (1973)
433.
www.ccdc.cam.ac.uk). Supplementary data associated
with this article can be found, in the online version, at
doi:10.1016/j.ica.2004.09.014.
[17] P.T. Beurskens, G. Admiral, G. Beurskens, W.P. Bosman, S.
Garcia-Granda, R.O. Gould, J.M.M. Smits, C. Smykalla, The
DIRDIF Program System, Technical Report of the Crystallog-
raphy Laboratory, University of Nijmegen, The Netherlands,
1992.
References
[18] G.M. Sheldrick, in: G.M. Sheldrick, C. Kruger, R. Goddard
(Eds.), Crystallographic Computing 3, Oxford University Press,
Oxford, 1985, p. 175.
[1] H. Sies, Angew. Chem. Int. Ed. 13 (1974) 706.
[2] G.R. Schonbaum, B. Chance, Enzymes, third ed., vol. 13, 1976, p.
363.
[19] G.M. Sheldrick, SHELXL-97: Program for the Refinement of
Crystal Ed., University of Gottingen, Germany, 1997.
¨
[3] I. Fita, A. Silva, M.R.N. Murthy, M.G. Rossmann, Acta
Crystallogr. Sect. B: Struct. Sci. B42 (1986) 497.
[20] H.M. Goff, E.T. Shimomura, Y.J. Lee, W.R. Scheidt, Inorg.
Chem. 23 (1984) 315.
[4] G.N. Murshudov, W.R. Melik-Adamyan, A.I. Grebenko, V.V.
Barynin, A.A. Vagin, B.K. Vainshtein, Z. Dauter, K.S. Wilson,
FEBS Lett. 312 (1992) 127.
[21] M.P. Byrn, C.J. Curtis, Y. Hsiou, S.I. Khan, P.A. Sawin, S.K.
Tendick, A. Terzis, C.E. Strouse, J. Am. Chem. Soc. 115 (1993)
9480.
[5] J. Bravo, N. Verdaguer, J. Tormo, C. Betzel, J. Switala, P.C.
Loewen, I. Fita, Structure (London) 3 (1995) 491.
[22] A.M. Helms, W.D. Jones, G.L. McLendon, J. Coord. Chem. 23
(1991) 351.
[6] P. Gouet, H.-M. Jouve, O. Dideberg, J. Mol. Biol. 249 (1995) 933.
[7] G.N. Murshudov, A.I. Grebenko, V. Barynin, Z. Dauter, K.S.
Wilson, B.K. Vainshtein, W. Melik-Adamyan, J. Bravo, J.M.
[23] W.R. Scheidt, D.K. Geiger, Y.J. Lee, C.A. Reed, G. Lang, J.
Am. Chem. Soc. 107 (1985) 5693.
[24] R.D. Arasasingham, A.L. Balch, R.L. Har, L. Latos-Grazyn´ski,
J. Am. Chem. Soc. 112 (1990) 7566.
´
Ferran, J.C. Ferrer, J. Switala, P.C. Loewen, I. Fita, J. Biol.
Chem. 271 (1996) 8863.
[25] R.D. Arasasingham, A.L. Balch, C.R. Cornman, J.S. De Ropp,
K. Eguchi, G.N. La Mar, Inorg. Chem. 29 (1990) 1847.
[26] T. Uno, K. Hatano, Y. Nishimura, Y. Arata, Inorg. Chem. 29
(1990) 2803.
´
[8] M.J. Mate, M. Zamocky, L.M. Nykyri, C. Herzog, P.M. Alzari,
C. Betzel, F. Koller, I. Fita, J. Mol. Biol. 286 (1999) 135.
[9] C.D. Putnam, A.S. Arvai, Y. Bourne, J.A. Tainer, J. Mol. Biol.
296 (2000) 295.
[27] M.W. Nee, J.R. Lindsay Smith, J. Chem. Soc., Dalton Trans.
(1999) 3373.
[10] N. Ueyama, N. Nishikawa, Y. Yamada, T. Okamura, A.
Nakamura, J. Am. Chem. Soc. 118 (1996) 12826.
[28] H. Sugimoto, N. Ueda, M. Mori, Bull. Chem. Soc. Jpn. 55 (1982)
3468.
[11] N. Ueyama, N. Nishikawa, Y. Yamada, T. Okamura, A.
Nakamura, Inorg. Chim. Acta 283 (1998) 91.
[29] N. Ueyama, M. Inohara, A. Onoda, T. Ueno, T. Okamura, A.
Nakamura, Inorg. Chem. 38 (1999) 4028.
[12] N. Ueyama, N. Nishikawa, Y. Yamada, T. Okamura, S. Oka, H.
Sakurai, A. Nakamura, Inorg. Chem. 37 (1998) 2415.
[13] W. Melik-Adamyan, J. Bravo, X. Carpena, J. Switala, M.J. Ma´te,
I. Fita, P.C. Loewen, Proteins 44 (2001) 270.
[30] A. Onoda, Y. Yamada, J. Takeda, Y. Nakayama, T. Okamura,
M. Doi, H. Yamamoto, N. Ueyama, Bull. Chem. Soc. Jpn. 77
(2004) 321.
[14] C. Rovira, I. Fita, J. Phys. Chem. B 107 (2003) 5300.
[15] M.T. Green, J. Am. Chem. Soc. 123 (2001) 9218.
[31] N. Ueyama, Y. Yamada, T. Okamura, S. Kimura, A. Nakamura,
Inorg. Chem. 35 (1996) 6473.