Crystal structure of bis(4-methylimidazole)tetraphenylporphyrinatoiron(III) chloride and related compounds. Correlation of ground state with Fe-N bond lengths

Article abstract of DOI:10.1021/ic990848s

The crystal structure of the title compound is presented and shown to be one of a class of low-spin iron porphyrin complexes having a ground-state electronic configuration of (d(xy))²(d(xz))²(dyz)¹. If their Fe-N bond lengths (average N-porphyrin plotted against average N-axial) are considered, this class of low-spin iron(III) porphyrins of general formula [Fe(III)Por(L)₂]⁺X^- and of ²B ground state is shown to be distinctly different crystallographically from a similar class of compounds with the same general formula but with a ²E or a (d(xy))²(d(xz),d(yz))³ ground state. A third group of compounds with the same general formula have a (d(xz),d(yz))⁴(d(xy))¹ ground state and again are in a different region of the plot. Compounds showing intermediate properties can be forecast from the simple relationship presented in this work. The electron paramagenetic resonance data are shown to be dependent on the ground state, and those of configuration (d(xy))²(d(xz),d(yz))³ and the ²B ground state obey a correlation previously suggested in the literature.

Full text of DOI:10.1021/ic990848s

2874
Inorg. Chem. 2000, 39, 2874-2881
Crystal Structure of Bis(4-methylimidazole)tetraphenylporphyrinatoiron(III) Chloride and
Related Compounds. Correlation of Ground State with Fe-N Bond Lengths
Jack Silver,*,† Paul J. Marsh,† Martyn C. R. Symons,† Dimitri A. Svistunenko,‡
Christopher S. Frampton,*,§ and George R. Fern†
School of Chemical and Life Sciences, Woolwich Campus, University of Greenwich, Wellington Street,
Woolwich, London, SE18 6PF, U.K., Department of Biological Sciences, Central Campus, University of
Essex, Wivenhoe Park, Colchester, Essex, C04 3SQ, U.K., and Roche, Discovery Welwyn,
Broadwater Road, Welwyn Garden City, Hertfordshire, AL7 2AY, U.K.
ReceiVed July 16, 1999
The crystal structure of the title compound is presented and shown to be one of a class of low-spin iron porphyrin
complexes having a ground-state electronic configuration of (dxy)2(dxz)2(dyz)1. If their Fe-N bond lengths (average
N-porphyrin plotted against average N-axial) are considered, this class of low-spin iron(III) porphyrins of general
2
formula [FeIIIPor(L)2]+X- and of B ground state is shown to be distinctly different crystallographically from a
similar class of compounds with the same general formula but with a 2E or a (dxy)2(dxz,dyz)3 ground state. A third
group of compounds with the same general formula have a (dxz,dyz)4(dxy)1 ground state and again are in a different
region of the plot. Compounds showing intermediate properties can be forecast from the simple relationship
presented in this work. The electron paramagenetic resonance data are shown to be dependent on the ground
2
state, and those of configuration (dxy)2(dxz,dyz)3 and the B ground state obey a correlation previously suggested
in the literature.
Introduction
model compounds have been extremely useful as aids to the
understanding of the bonding and properties of the haems in
such proteins.
There have been a number of studies of bis-ligated porphy-
rinato iron(III) complexes, [FeIIIPor(L)2]+, where the axial
ligands (L) are aliphatic amines,1,2 histidine,3 imidazole, or
substituted imidazoles,4-20 as models for cytochromes b. These
It is well-known, from studies of cytochromes b from various
mitochondrial and chloroplast sources, that the haem (iron
protoporphyrin IX) in these proteins is coordinated to two
histidine residues.21-28 The principal mechanisms of fine control
of haem iron reactivity in haemoproteins arise from the
electronic and steric influences of these ubiquitous ligands.29
Studies of the physical properties of the cytochrome b proteins
have focused on the orientation of the imidazole planes of the
axial histidine ligands.8,15,18,24,30 Cytochromes b from complex
III of mitochondria give electron paramagnetic resonance (EPR)
† University of Greenwich.
‡ University of Essex.
§ Roche, Discovery Welwyn.
(1) Marsh, P. J.; Silver, J.; Symons, M. C. R.; Taiwo, F. A. J. Chem.
Soc., Dalton Trans. 1996, 2361-2369.
(2) Marques, H. M.; Munro, O. Q.; Crawcour, M. L. Inorg. Chim. Acta
1992, 96, 221-229.
(3) Medhi, O. K.; Silver, J. J. Chem. Soc., Dalton Trans. 1990, 555-
559.
(4) Medhi, O. K.; Silver, J. J. Chem. Soc., Dalton. Trans. 1990, 263-
270.
(17) Epstein, L. M.; Straub, D. K.; Marricondi, C. Inorg. Chem. 1967, 6,
1720-1724.
(5) Scheidt, W. R.; Osvath, S. R.; Lee, Y. J. J. Am. Chem. Soc. 1987,
109, 1958-1963.
(18) Scheidt, W. R.; Chipman, D. M. J. Am. Chem. Soc. 1986, 108, 1163-
1167.
(6) Collins, D. M.; Countryman, R.; Hoard, J. L. J. Am. Chem. Soc. 1972,
94, 2066-2072.
(19) Soltis, S. M.; Strouse, C. E. J. Am. Chem. Soc. 1988, 110, 2824-
2829.
(20) Scheidt, W. R.; Kirner, J. L.; Hoard, J. L.; Reed, C. A. J. Am. Chem.
Soc. 1987, 109, 1963-1968.
(21) Matthews, F. S.; Czerwinski, E. W.; Argos, P. In The Porphyrins;
Dolphin, D., Ed.; Academic Press: New York, 1979; Vol. 7, pp 108-
147.
(22) Iyanagi, T. Biochemistry 1977, 16, 2725-2730.
(23) Widger, W. R.; Cramer, W. A.; Herrman, R. G.; Trebst, A. Proc.
Natl. Acad. Sci. U.S.A. 1984, 81, 674-678.
(7) Hatano, K.; Safo, M. K.; Walker, F. A.; Scheidt, W. R. Inorg. Chem.
1991, 30, 1643-1650.
(8) Walker, F. A.; Huynh, B. H.; Scheidt, W. R.; Osvath, S. R. J. Am.
Chem. Soc. 1986, 108, 5288-5297.
(9) Safo, M. K.; Gupta, G. P.; Walker, F. A.; Scheidt, W. R. J. Am. Chem.
Soc. 1991, 113, 5497-5510.
(10) Yoshimura, T.; Ozaki, T. Arch. Biochem. Biophys. 1984, 230, 466-
482.
(11) Higgins, T. B.; Safo, M. K.; Scheidt, W. R. Inorg. Chim. Acta 1990,
178, 261-267.
(24) Babcock, G. T.; Widger, W. R.; Cramer, W. A.; Oertlings, W. A.;
Mertz, J. Biochemistry 1985, 24, 3638-3645.
(25) Keller, R.; Groudinsky, O.; Wu¨thrich, K. Biochim. Biophys. Acta 1973,
328, 233-238.
(12) Nakamura, M.; Tajima, K.; Tada, K.; Ishizu, K.; Nakamura, N. Inorg.
Chim. Acta 1994, 224, 113-124.
(13) Quinn, R.; Valentine, J. S.; Byrn, M. P.; Strouse, C. E. J. Am. Chem.
Soc. 1987, 109, 3301-3308.
(26) Blumberg, W. E.; Peisach, J. In Cytochrome Oxidase; King, T. E.,
Orii, Y., Chance, B., Okuniki, K., Eds.; Elsevier: Amsterdam, 1979;
pp 153-159.
(14) Quinn, R.; Strouse, C. E.; Valentine, J. S. Inorg. Chem. 1983, 22,
3934-3940.
(27) Babcock, G. T.; Callahan, P. M.; Ondrias, M. R.; Salmeen, I.
Biochemistry 1981, 20, 959-966.
(15) Walker, F. A.; Reis, D.; Balke, V. L. J. Am. Chem. Soc. 1984, 106,
6888-6898.
(28) Moore, G. R.; Williams, R. J. P. FEBS Lett. 1977, 79, 229-232.
(29) Perutz, M. F.; Ten Eyck, L. F. Cold Spring Harbor Symp. Quant.
Biol. 1971, 36, 295.
(16) Little, R. G.; Dymock, K. R.; Ibers, J. A. J. Am. Chem. Soc. 1975,
97, 4532-4539.
10.1021/ic990848s CCC: $19.00 © 2000 American Chemical Society
Published on Web 06/02/2000

Crystal Structure of Fe(III) Compounds
Inorganic Chemistry, Vol. 39, No. 13, 2000 2875
Table 1. X-band EPR Data for Compounds of the Type [FeIIIPor(L)2]+X- Where L ) Imidazole or Subsituted Imidazole Presented in Order of
Increasing gz Values
V
no.a
complexb
temp (K)
gz
gy
gx
φ (deg)
/
ref
∆
7
[K(K222)][FeIIITPP(4-MeIm-)2]
[FeIIITPP(HIm)2]Clc
77
77
110
100
77
7
86
77
77
77
86
77
2.6
2.24
2.33
2.3
1.82
1.84
1.59
1.59
1.54
1.57
1.55
1.57
1.47
1.49
1.48
1.48
1/17
0.66
0.76
0.65
0.64
0.64
0.64
0.64
0.63
0.58
0.61
0.57
0.55
14
this work
this work
15
11
7
19
this work
13
13
19
13
2.67
2.86
2.85
2.87
2.9
2.92
2.95
2.96
2.97
2.99
3.00
6
[FeIIITPP(4-MeIm)2]Cl (crystalline)
[FeIIITPP(4-MeIm)2]Cl (CHCl3)
[FeIIITPP(1-MeIm)2]ClO4
[FeIIIT2,6-Cl2PP(1-VinIm)2]ClO4
[FeIIITPP(HIm)2]Cl
4d
2.29
2.28
2.27
2.31
2.33
2.27
2.30
2.27
2.27
9
8
1
22/32
14/20
5
[FeIIITPP(HIm)2]Cl
3
2
5
4
[FeIIITPP(t-MU)2]SbF6
22
15
41
29
[FeIIITPP(c-MU)2]SbF6
[FeIIITPP(HIm)2]Cl
[FeIIITPP(c-MU)2]SbF6
a Figure 1. b K222 ≡ Kryptofix 222. PPIX ≡ protoporphyrinato IX dianion. T2,6-Cl2PP ≡ 5,10,15,20-tetrakis(2,6-dichlorphenyl)porphyrinato
dianion. c Site found in [FeIIITPP(Him)2]Cl that is ascribed to the structure reported by Hoard et al.6 d Average value of molecules 6a and 6b.
spectra31-34 known as “large gmax”. A “large gmax” spectrum
consists of an intense gz value at g g 3.3; the gy and gx values
are often not seen at all in the spectrum. Complexes in which
the imidazole planes on either side of the iron porphyrin plane
are perpendicular to each other give rise to this type of
spectrum.9,15 The Mo¨ssbauer quadrupole splitting (∆EQ) values
are around 1.8 mm s-1.4 Such perpendicular orientation of the
imidazole ligands relative to each other is found for iron
porphyrin complexes where the binding of the imidazoles is
sterically hindered. Cytochrome b5 proteins have rhombic EPR
spectra with gz ≈ 3.0, gy ≈ 2.2, and gx ≈ 1.4.35-38 These g
values are in the range observed for model compounds4-15 in
which the imidazole planes are in parallel orientation. Com-
plexes with ∆EQ values of ∼2.25-2.4 mm s-1 have been
assigned to this orientation.4,17 The bis(histidine)protoporphy-
rinato(IX)iron(III) complex gave a ∆EQ value of 2.14 mm s-1
and large line widths, which suggested that the histidines bind
as sterically hindered imidazoles and that there was a large angle
between the two imidazole planes.3
Hoard et al.6 first used the symbol “φ” to describe the angle
from the intersection of the plane of the axial ligand and the
porphyrin plane to the nearest Fe-Npor vector. Strouse et al.13,19
have reported a correlation of the crystal field parameter V/∆
with φ for five [FeIIITPP(L)2]+ complexes (1-5) (TPP )
5,10,15,20-tetraphenylporphyrinato dianion, L ) imidazole or
substituted imidazole) with different φ values (Table 1, Figure
1). In this paper, we present an extended study of the effect of
φ on crystal field parameters. This includes complexes 1-5 used
in the studies by Strouse et al.13,19 and, using both literature
Figure 1. Ligands in [FeIIITPP(L)2]X complexes 1-9.
Experimental Section
data7,11,13,14,19 and new results, a further four [FeIIIPor(L)2]+
complexes (6-9) (Table 1, Figure 1). We also report the crystal
and molecular structure of [FeIIITPP(4-MeIm)2]Cl (6), and by
comparing this structure with those of other known low-spin
complexes, we show a simple relationship between their
structures and the ground-state electronic configuration.
Materials. Imidazole and 4-methylimidazole were purchased from
Aldrich Chemical Co., and chloroform and hexane were from Fisher
Scientific, U.K. All chemicals were used without further purification.
a. Preparation of Bis(imidazole)tetraphenylporphyrinatoiron(III).
The complexes were prepared using the method of Scheidt et al.5 as
follows. [FeIIITPP(Cl)] (0.07 g; 0.1 mmol) and imidazole (0.04 g; 0.6
mmol) were dissolved in chloroform (4 mL). A 1:2 chloroform/hexane
mixture (9 mL) was allowed to diffuse into the solution. Very small
single crystals were obtained after 1-2 days. Anal. Calcd for [FeIII-
TPP(Him)2]Cl‚CHCl3‚H2O: C, 62.81; H, 3.99; N, 11.49. Found: C,
62.44; H, 3.93; N, 11.66. Crystal data: a ) 11.064(2) Å, b ) 13.147-
(2) Å, c ) 17.648(4) Å, R ) 70.01(1)°, â ) 72.52(2)°, γ ) 86.29(1)°,
V ) 2298.8(8) Å3. The cell is identical with that published by Scheidt
et al.5 Our cell is slightly smaller because it was collected at 123 K.
This shows that the sample contained material of the same structure as
that reported by Scheidt et al.5 However, as will be seen in the following
text, this material was prepared several times. The analysis was always
similar to that given above, but in one sample a second form must
have been present (see Results and Discussion).
(30) Carter, K. R.; Tsai, A.; Palmer, G. FEBS Lett. 1981, 132, 243-246.
(31) Tsai, A.-L.; Palmer, G. Biochim. Biophys. Acta 1983, 722, 349-363.
(32) Salerno, J. C.; McGill, J. W.; Gerstile, G. C. FEBS Lett. 1983, 162,
257-261.
(33) Leigh, J. S.; Erecinska, M. Biochim. Biophys. Acta 1975, 387, 95-
106.
(34) Erecinska, M.; Oshino, R.; Oshino, N.; Chance, B. Arch. Biochem.
Biophys. 1973, 157, 431-445.
(35) Rivera, M.; Barillas-Mury, C.; Christensen, K. A.; Little, J. W.; Wells,
M. A.; Walker, F. A. Biochemistry 1992, 31, 12233-12240.
(36) Von Bodman, S. B.; Schuler, M. A.; Jollie, D. R.; Sligar, S. G. Proc.
Natl. Acad. Sci. U.S.A. 1986, 83, 9443-9447.
(37) Ikeda, M.; Izuka, T.; Takao, H.; Hagihara, B. Biochim. Biophys. Acta
1974, 336, 15-24.
(38) Passon, P. G.; Reed, D. W.; Hultquist, D. E. Biochim. Biophys. Acta
1972, 275, 51-61.

2876 Inorganic Chemistry, Vol. 39, No. 13, 2000
Silver et al.
Table 2. Mo¨ssbauer Data for Compounds of the Type [FeIIIPor(L)2]+X- Where L ) Imidazole Type Ligand (Values Recorded at ∼77 K
Unless Otherwise Stated)
a
b
no.
compound
solvent
δ (mms-1
)
∆EQ (mm s-1
)
Γ (mm s-1
)
ref
[FeIIITPP(HIm)2]Cl
crystalline
solid
solid
solid
solid
crystalline
crystalline
0.28(2)
0.50
0.40
0.25(1)
0.17(1)
0.34(3)
0.28
2.22(2)
2.23
2.11
2.09(1)
2.03(1)
2.26(3)
2.28
0.33(2)/0.46(3)
not given
not given
0.29(1)/0.45(1)
0.20(1)/0.26(1)
0.46(4)/0.96(6)
0.71/0.94
this work
17
17
this work
this work
this work
9
[FeIIITPP(HIm)2]Cl
[FeIIITPP(HIm)2]Cl (room temp)
[FeIIITPP(HIm)2]Clc
[FeIIITPP(HIm)2]Clc (room temp)
[FeIIITPP(4-MeIm)2]Cl
[FeIIITMPd
6
11/12
(1-MeIm)2]ClO4
[FeIIIPPIX(HIm)2]+
dmso
0.22(2)
0.24(1)
0.24
0.23(1)
0.26(1)
0.16(2)
0.26(6)
2.38(2)
2.35(1)
2.30
2.24(1)
2.34(1)
1.87(2)
1.99(6)
0.21(3)/0.26(4)
0.31(1)/0.32(1)
not given
0.37(1)/0.49(2)
0.16(1)/0.18(1)
10.29(1)/0.59(3)
0.19(8)/0.30(9)
4
4
17
4
4
4
3
water/ethanold
solid
[FeIIIPPIX(1-MeIm)2]+
dmso
water/ethanold
water/ethanold
water
[FeIIIPPIX(2-MeIm)2]+
FeIIIPPIX(histidine)2]+
(pH 10.1)
water/ethanol
(pH 8.4)
water
(pH 11.0)
water
[FeIIIPPIX
0.21(3)
0.28(5)
0.26(2)
2.09(3)
2.28(5)
2.22(2)
0.40(3)/0.61(6)
0.32(4)/0.42(8)
0.36(2)/0.44(4)
3
3
3
(NR-histidine)2]+
[FeIIIPPIX(histamine)2]+
{FeIIIPPIX(pilocarpate)]+
(pH 10.1)
a Relative to metallic iron at 298 K. b Half-width at half-height. c Additional site in EPR spectrum. d TMP ) 5,10,15,20-tetramesitylporphyrinato
dianion.
b. Preparation of Bis(4-methylimidazole)tetraphenylpor-
phyrinatoiron(III) (6). The preparation is as for the preparation of
the bis(imidazole) complex except that 4-methylimidazole (0.05 g; 0.6
mmol) was used.
Mo1ssbauer Spectroscopic Measurements. Mo¨ssbauer spectra were
recorded at room temperature or at 77 K. The apparatus and methodol-
ogy have been described previously.39
(where L ) imidazole or substituted imidazole) are presented
in Table 1. The bis(imidazole complex) was prepared several
times. In all three, different EPR signals were observed, i.e.,
from three different structural forms (two of which are in the
triclinic structure reported by Scheidt et al.5). Most samples give
an EPR spectrum with only one site, which was the overlap of
the two sites observed by Strouse et al.13,19 In one case an
additional site was observed. The parameters of this second site
were reasonably similar to the g values observed for [K(K222)]-
[FeIIITPP(4-MeIm-)2],14 where both the axial ligands were
deprotonated. It is extremely doubtful that there was formation
of a bis(imidazolato) complex under the conditions used, and
in addition a cation would be required to balance the charge.
There was no analytical evidence for this. There are two other
possible explanations. Either only one axial ligand was depro-
tonated and that the species that gave rise to the second site is
[FeIIITPP(HIm)(Im-)] (this we believe is still unlikely because
the counterion would have to be H+) or a second [FeIIITPP-
(HIm)2]Cl solid was synthesized that had a structure similar to
that reported by Hoard et al.6 We believe this has occurred
because the analysis indicated this formulation. We note that
others could not prepare this crystalline form. We were only
able to prepare it together with the other form and not by itself.
The parameters we obtained of solid [FeIIITPP(4-MeIm)2]Cl
agree with the values presented by Walker, Reis, and Balke15
for a frozen solution study.
EPR Spectroscopic Measurements. The apparatus and methodology
are as described in our previous paper.1
Crystal Structure Determination of 6. A black plate crystal was
selected and mounted on the end of a glass fiber. The crystal was
transferred to a Rigaku AFC7R four-circle diffractometer equipped with
an Oxford Cryosystems cryostream cooler,40 operating at 123(1) K.
Unit cell constants were determined prior to data collection by least-
squares refinement of 25 reflections well-positioned throughout recipro-
cal space, 51.35° e θ e 54.50°. The data were corrected for Lorentz
polarization effects. An analytical absorption correction was applied;
maximum and minimum correction factors were 0.734 and 0.234,
respectively. The intensity of three standard reflections monitored every
150 reflections showed an overall decrease in the intensity over the
period of the data collection of 3.84%, and the data were adjusted
accordingly. A total of 10 038 reflections were measured of which 9354
were unique, Rint ) 0.0254. A total of 7719 reflections were observed
with I > 2σ(I).
The structure of 6 was solved by direct methods (SHELXS-86),41
with full-matrix least-squares refinement on F2 (SHELXL-94).42 Non-
hydrogen atoms were refined with anisotropic displacement parameters,
and hydrogen atoms were included using a riding model. A correction
for secondary extinction, x ) 0.00179(14), according to the method in
SHELXL-97 was included. S ) 1.015, w ) 1/[σ2(Fo ) + (0.0722P)2 +
2
Mo1ssbauer Data. The Mo¨ssbauer parameters for the bis-
(imidazole), bis(4-methylimidazole) derivatives, and similar
compounds3,4,9,17 are presented in Table 2. The parameters for
the bis(imidazole) compound for all but one sample were in
good agreement with those of Epstein et al.17 In one case the
∆EQ value was lower than expected. This was the material that
showed the presence of two EPR signals.
Crystal and Molecular Structure of [FeIIITPP(4-meth-
ylimidazole)]Cl. The crystal structure of 6 reveals that there
are two independent molecules in the unit cell. The single-crystal
crystallographic data are reported in Table 3, and a comparison
of selected bond lengths and angles in the two molecules is
given in Table 4. The coordination around the Fe atom in
2
3.1518P] where P ) (Fo + 2Fc2)/3. Maximum and mean ∆/σ were
0.001 and 0.000, respectively. Maximum and minimum ∆/F were 0.78
and -0.74 e Å-3
.
Results and Discussion
EPR Data. The EPR g-values for [FeIIITPP(HIm)2]Cl, [FeIII-
TPP(4-MeIm)2]Cl, and related [FeIIIPor(L)2]+ complexes7,11,13-15,19
(39) Hamed, M. Y.; Hider, R. C.; Silver, J. Inorg. Chim. Acta 1982, 66,
13-18.
(40) Cosier, J.; Glazer, A. M. J. Appl. Crystallogr. 1986, 19, 105-107.
(41) Sheldrick, G. M. Acta Crystallogr. 1990, A46, 467-473.
(42) Sheldrick, G. M. FHELXTL, version 5; Brooker AXF: Madison, WI,
1994.

Crystal Structure of Fe(III) Compounds
Inorganic Chemistry, Vol. 39, No. 13, 2000 2877
Table 3. Crystallographic Data for Compound 6
chemical formula
a, Å
b, Å
c, Å
R, deg
â, deg
γ, deg
V, Å3
C54 H42 C17 Fe N8
12.0001(14)
21.204(4)
10.505(2)
91.51(2)
106.601(12)
88.84(2)
2560.3(7)
1106.96
P1h
123(1)
1.541 78
1.436
0.234-0.734
0.0468
fw
space group
temp, K
λ, Å
F (calcd), mg/cm3
transm coeff
R
Rw
0.1245
Table 4. Selected Bond Lengths and Angles in Molecules 6a and
6b
iron coordination
6a
6b
Bond Length (Å)
2.008(2)
Fe-Npor
Fe-Npor
Fe-Naxial
2.002(2)
1.988(2)
1.987(2)
1.993(2)
1.975(2)
Bond Angle (deg)
89.77(10)
Figure 2. Crystal structure of [FeIIITPP(4-methylimidazole)2]Cl.
Npor-Fe-Npor
Naxial-Fe-Npor
89.76(10)
90.24(10)
92.06(9)
87.94(9)
89.28(10)
90.73(10)
90.23(10)
91.91(10)
88.09(10)
89.71(10)
90.29(10)
molecule 6a (Figure 2) shows it to be very similar to that of
compound 1, the bond lengths being the same within experi-
mental error and the φ angle also similar. Molecule 6b has a
similar φ angle but has a slightly longer Fe-Naxial distance.
However the distances in 6b are in the range of distances in
compounds 1-3 and the φ angle is close to 1. Hence, molecules
6a and 6b are very similar, and so we might not expect to see
two distinct EPR signals, and indeed we do not.
Other than the differences in the length of the Fe-Npor bonds
in molecules 6a and 6b all other distances within the porphyrin
ring are very similar. Both porphyrin rings are flat, with the
largest deviations from the plane for 6a being -0.061(2) Å (N1)
and for molecule 6b being +0.077(2) Å for N5. The packing
diagram of 6 is shown in Figure 3. There are four CHCl3
molecules (two equivalent pairs) in the unit cell and two
equivalent Cl- ions. The latter are each hydrogen-bonded
through one hydrogen of a protonated nitrogen of an imidazole
ring to both 6a and 6b and to the hydrogen atoms of two
different CHCl3 molecules. Thus, each imidazole ring of both
molecules 6a and 6b bonds to a Cl- ion via a hydrogen bond.
The shortest H bonds are from molecule 6a to Cl- (2.125(1)
Å), whereas those from 6b to Cl- are 2.264(1) Å long. The
CHCl3 to Cl- H bonds are longer at 2.308(1) and 2.549(1) Å,
respectively. The presence of the H bonds from the imidazole
rings is consistent with the findings in the structures of
compounds 1-5 and further adds evidence to the possibility of
such bonding being a control mechanism of haem iron reactivity
in cytochrome b43,44 and other haemproteins.29
Figure 3. Packing diagram of [FeIIITPP(4-methylimidazole)2]Cl.
latter attracts less negative charge from the imidazole ring. This
allows stronger π-bonding (electron donation) from the methyl
imidazole ligands to the iron atom in 6a and hence shorter
Fe-Nax bonds. In molecule 6b, where the H bonds are longer,
the imidazole character from H-bonding will be larger (hence,
negative charge is increased). This might have been expected
to shorten the Fe-N bond on an electrostatic argument. Clearly,
this is not observed and indicates that π-bonding rather than
σ-bonding is dominant.
It is worth noting that the shorter H bonds found in molecule
6a may explain why the Fe-Naxial bonds are also the shortest
in this molecule. As the Cl- pulls the proton toward itself, the
(43) Brautigan, D. L.; Feinberg, B. A.; Hoffman, B. M.; Margoliash, E.;
Peisach, J.; Blumberg, W. E. J. Biol. Chem. 1977, 252, 574-582.
(44) De Ropp, J. S.; Thanabal, V.; La Mar, G. N. J. Am. Chem. Soc. 1985,
107, 8268-8270.

2878 Inorganic Chemistry, Vol. 39, No. 13, 2000
Silver et al.
that the trend is a significant finding. The fact that the correlation
breaks down for ∆φ angles greater than zero suggests that
metal-ligand bonding involves t2g orbitals, which make an
important contribution to the EPR parameters.
It is obvious that for parallel planes bonding of both ligands
to the singly occupied d orbital is important. This will be smaller
if the ∆φ angle is larger than zero. An examination of the bond
lengths in Table 5 for compounds 1-6 is instructive. As φ
increases, the difference between the long and short Fe-Npor
bonds diminishes. Compounds 1, 2, and 6 (which have smaller
φ values) have larger differences between the long and short
Fe-Npor bonds than compounds 3-5. In all the compounds 1-6
the Fe-Nax bond lengths are shorter than the Fe-Npor. The bond
lengths suggest that the strongest bonding is in the axial direction
in 1-6. This most probably results from strong π-bonding from
the axial ligands, whereas the porphyrin π-bonding to the iron
V
Figure 4. Plot of φ vs /∆.
t
2g orbitals is manifestly weaker. This would agree with previous
Effects of Orientation of the Axial Ligands. It has been
shown that when the electron configuration of low-spin Fe(III)
is (dxy)2(dxz,dyz)3 even for metalloporphyrins that are equatorially
symmetrical and have perpendicular bound identical axial
ligands, then the iron is in a rhombically distorted system.9,45
The electronic structure of these complexes may be described
in terms of the crystal field parameters V and ∆ where V is the
rhombic splitting parameter and ∆ is the tetragonal splitting
parameter (Figure 4). The wave functions are linear combina-
tions of the three states with coefficients a (for dyz), b (for dxz),
Mo¨ssbauer spectroscopic measurements on the sign of the field
gradient that suggests that the highest electron population lies
in the axial direction.47
For compounds 8 and 9, which do not fit the curve in Figure
4, the situation is very different. The Fe-Npor bonds are overall
noticeably shorter than in compounds 1-6, showing stronger
bonding between the porphyrin and the iron atom; in addition
the axial bonding distances are more equal in length to the
Fe-Npor distances. The extent that this is caused by or in fact
causes the ∆φ angle cannot be deduced. Compound 7, which
also does not fit the relationship in Figure 4, has a much less
accurately known structure that limits the validity of any
deductions. The Fe-Npor distances are long, but the errors could
bring them in line with compounds 8 and 9. However, the axial
bonds are very short, even allowing for the errors. This would
be expected for strong ligand π-bonding to the iron. In this case
the ligands are very electron-rich, having electron donors on
the second nitrogen atom rather than a hydrogen atom. We will
discuss this point further in the next section of the paper.
There are no EPR measurements in the literature for
compounds 10-12. Compounds 10 and 11 have similar
structural features to compounds 8 and 9, whereas compound
12 is very similar to compounds 4 and 5. Compound 13 is very
likely to be responsible for the second EPR site we recorded
when we prepared the material of this formula. This compound
2
2
2
and c (for dxy). The dx -y and dz are considered to be of too
high energy to contribute. By use of the axis system of Taylor,45
it has been shown9 that
gx ) 2[a2 - (b + c)2], gy ) 2[(a + c)2 - b2],
gz ) 2[(a + b)2 - c2]
and the crystal field parameters V/λ and ∆/λ may be described as
V/λ ) Eyz - Exz and ∆/λ ) Eyz - Exy - (1/2)(V/λ)
Therefore, V/λ ) gx/(gz + gy) + gy/(gz - gx) and ∆/λ ) gx/(gz +
gy) + gz(gy - gx) - (1/2)(V/λ). Strouse and co-workers13,19
proposed that V/∆, readily derived from EPR parameters,45 could
V
be related to φ. Scheidt et al.11 calculated
/ ) 0.635 for
∆
V
complex 9. This agreed poorly with the correlation of Strouse
and co-workers,13,19 who predicted a value of V/∆ ) 0.58-0.54
for φ ) 22-32°. Scheidt et al.11 suggested that the poor
agreement might be due to two possible reasons. First, the
correlation was derived from complexes 1-5 that have parallel
imidazole planes, whereas there is an angle between the ligand
would then have a gz of 2.673 and a /∆ of 0.76. Clearly, this
does not fit the correlation in Figure 4, but if the axial bond
lengths are averaged, then it is similar to compounds 8-10.
Is the correlation useful? Only if it is assumed that a
compound or haem protein of unknown structure has a ∆φ value
of zero can an EPR gz value be translated into a φ angle.
However, it could be argued that this is of limited value because
the only way (to date) to measure ∆φ is a structural analysis,
which of course would also give φ. For cases where EPR spectra
change with pH as in cytochrome b547 or where they differ for
the low- and high-potential forms as in chloroplast cytochrome
V
planes of complexes 7-9. These compounds give /∆ values
that do not agree with the correlation (Figure 4). Second, the
correlation was derived for imidazole derivatives unsubstituted
at the 1-position and may not be expected to work for other
derivatives, especially as Walker, Lo, and Ree46 have described
pKa versus ligand binding equilibrium constant relationships for
imidazole derivatives substituted at the 1-position and unsub-
stituted derivatives. The values for the two types are significantly
different. This latter reason is certainly wrong because complex
7 has ligands that are not substituted in the 1-position and is
related to complex 6.
b559,
24 using the correlation and assuming ∆φ is 0 may aid the
understanding of the role of the haem. Of course, the EPR
spectra must be of the rhombic type before this correlation is
applied.
It should be noted that the compounds that do not fit the
correlation (8 and 9) are substituted in the 1-position (Figure
1) and that compound 7 carries a negative charge that will be
predominantly focused at the 1-position. The 1-position is not
substituted in histidine residues bonding in haemproteins, and
Our complex (6) does follow the original trend, and the angle
between the ligand planes (∆φ) is near zero. It therefore appears
(45) Taylor, C. P. S. Biochim. Biophys. Acta. 1977, 491, 137-149.
(46) Walker, F. A.; Lo, M. W.; Ree, M. T. J. Am. Chem. Soc. 1976, 98,
5553-5560.
(47) Peisach, J.; Mims, W. B. Biochemistry 1977, 16, 2795-2799.

Crystal Structure of Fe(III) Compounds
Inorganic Chemistry, Vol. 39, No. 13, 2000 2879
Table 5. Comparison of Major Bond Lengths and Angles around the Iron Atoms and gz Values Where Known
complex
gz
2.6
Fe-Npor (Å)
2.031(12)
Fe-Nax (Å)
φ (deg)
∆φ (deg)
ref
7
[K(K222)]
1.974(11)
1.982(11)
1.993(2)
1.988(2)
1.969(3)
1.977(3)
1.971(4)
1.973(4)
1.985(3)
1.988(4)
1.983(7)
1.990(3)
1.996(7)
1.973(6)
1.985(5)
1.974(2)
1.958(12)
1.928(12)
1.975(2)
1.987(2)
1.970(3)
1.978(3)
1.976(4)
1.968(4)
1.977(3)
1.983(4)
1.979(7)
1.964(3)
1.967(7)
1.988(5)
1.966(5)
1.975(3)
1
17
3.1(1)
4.6(2)
22
32
14
20
5
22
15
41
29
3
16
23
18
14
[FeIIITPP(4-MeIm)2]
[FeIIITPP(4-MeIm)2]Cl
[FeIIITPP(4-MeIm)2]Cl
[FeIIITPP
2.006(12)
2.008(2)
2.002(2)
1.988(3)
1.993(3)
1.988(4)
1.981(4)
2.002(3)
1.995(4)
2.007(6)
1.995(3)
1.998(7)
2.012(5)
1.992(5)
2.002(3)
6a
6b
9
2.859
2.859
2.866
0
10
6
this work
11
7
(1-MeIm)2]ClO4
[FeIIIT2,6-Cl2PP
(1-VinIm)2]ClO4
[FeIIITPP(HIm)2]Cl.
[FeIIITPP(t-MU)2]SbF6
[FeIIITPP(c-MU)2]SbF6
[FeIIITPP(HIm)2]Cl.
[FeIIITPP(c-MU)2]SbF6
[FeIIIPPIX
8
2.9
1
3
2
5
4
2.916
2.964
2.965
2.988
2.999
0
0
0
0
0
5, 19
13
13
5, 19
13
10
13
16
(1-MeIm)2]ClO4
11
12
13
14
15
16
17
18
19
20
21
22
23
[FeIIITMP
a
a
0
0
9
(1-MeIm)2]ClO4
[FeIIITMP
2.005(3)
1.999(3)
1.965(3)
36
9
(1-MeIm)2]ClO4
[FeIIITPP(HIm)2]Cl
1.987(4)
1.999(4)
1.988(5)
1.982(4)
1.976(4)
1.968(4)
1.968(7)
1.971(7)
1.957(5)
1.971(5)
1.966(3)
1.968(3)
1.969(4)
1.966(4)
1.950(4)
1.957(4)
2.023(4)
1.990(4)
1.980(4)
1.986(5)
1.972(5)
1.966(4)
1.972(4)
1.969(7)
1.964(7)
1.955(6)
1.959(6)
1.958(4)
1.962(4)
1.950(4)
1.973(4)
1.944(4)
1.958(4)
2.004(4)
1.991(5)
1.957(4)
2.005(5)
2.001(5)
2.015(4)
2.010(4)
2.018(7)
2.006(7)
2.021(6)
2.001(5)
2.002(4)
1.989(4)
1.989(4)
1.978(4)
2.008(4)
1.997(4)
2.057(5)
18
39
34
38
32
32
29
48
43
44
44
44
37
42
36
35
23
57
86
89
77
90
90
79
89
0
6
[FeIIITPP
3.70
3.56b
3.07
2.53
2.89
3.48
2.54
3.24
2.818
58
20
59
47
59
9
(Py)2]ClO4
[FeIIITPP
(2-MeIm)2]ClO4
[FeIIITMP
(3-ClPy)2]ClO4
[FeIIITMP
(4-CNPy)2]ClO4
[FeIIITMP
(3-EtPy)2]ClO4
[FeIIITMP
(4-Me2NPy)2]ClO4
[FeIIITPP
60
57
9
(4-CNPy)2]ClO4
[FeIIIOEP((C5H5)
(C4NH4)Fe)2]O3SCF3
[FeIIIOEP
1.987(2)
1.999(2)
1.986(2)
1.990(2)
1.995(2)
2.031(2)
41
41
0
(4-NMe2Py)2]ClO4
[FeIIIOEP
0
61
(3-Clpy)2]ClO4
a Solid-state data were too complex to fit but are said to be the rhombic type (ref 9). b Reference 8.
thus, the fact that compounds 7-9 do not fit the correlation
does not negate its use because it could be argued that they are
not relevant. If this is the case, then all the relevant compounds
(1-6) fit the correlation and in all cases ∆φ equals 0. Therefore,
we would expect that in haemproteins (containing two histidines)
∆φ would equal zero (if the EPR spectra were rhombic) and
the correlation can be applied.
It is useful to see to what extent this is true. A number of
crystal structures of cytochromes b5 and cytochromes bc1
complexes have been solved to varying degrees of resolution.
Of the three known X-ray structures of cytochromes b5, all show
that the histidine ligands to the b haems are very close to parallel,
with ∆φ angles close to or equal to zero (allowing for the
resolution of the structures).48-50 Moreover, the φ values are
close to 40°. This value is similar to that expected from the ∆/λ
value for neutral cytochrome b5 of 0.52.35,36,47,51 In addition, a
cytochrome b5 structure derived from solution NMR studies52
also shows a ∆φ angle close to parallel and a φ value of ∼40°.
There have been a number of reports of the crystal structures
of cytochromes bc1.53-56 The cytochrome b haems in these
structures have ∆φ close to or equal to 90° and φ angles around
40-45°. The EPR spectra of these b haems are of the “large
gmax” type,13 typical of iron porphyrin complexes where the
imidazole rings of the axial ligands are perpendicular to one
another, and so of course, we would not expect them to obey
the correlation in Figure 4.
It is clear that compounds 14-20, which are “large gmax
”
class, all have much smaller Fe-Npor distances arising from
strong iron t2g π-bonding to the Npor orbitals and long bonds to
the axial ligands. This suggests in all these cases (compounds
14-20), with the possible exception of compound 19, that the
iron to Nax π-bonding is very weak or does not exist at all,
which is in agreement with our previous findings.1
(52) Arnesano, F.; Banci, L.; Bertini, I.; Felli, I. C. Biochemistry 1998,
37, 173-184.
(53) Zhang, Z.; Huang, L.; Shulmeister, V. M.; Chi, Y. I.; Kim, K. K.;
Hung, L.; Crofts, A. R.; Berry, E. A.; Kim, S. H. Nature 1998, 392,
677-684.
(54) Xia, D.; Yu, C. A.; Kim, H.; Xia, J. Z.; Kachurin, A. M.; Zhang, L.;
Yu, L.; Deisenhofer, J. Science 1997, 277, 60-66.
(55) Iwata, S.; Lee, J. W.; Okada, K.; Lee, J. K.; Iwata, M.; Rasmussen,
B.; Link, T. A.; Ramasawamy, S.; Jap, B. K. Science 1998, 281, 64-
71.
(56) Crofts, A. R.; Hong, S. J.; Ugulava, N.; Baraquera, B.; Gennis, R.;
Geurgova-Kuras, M.; Berry, E. A. Proc. Natl. Acad. Sci. U.S.A. 1999,
96, 10021-10026.
(48) Matthews, F. S.; Argos, P.; Levine, M. Cold Spring Harbor Symp.
Quant. Biol. 1971, 36, 387.
(49) Rodriguez-Maranon, M. J.; Qiu, F.; Stark, R. E.; White, S. P.; Zhang,
X.; Foundling, S. I.; Rodriguez, V.; Schilling, C. L.; Bunce, R. A.;
Rivera, M. Biochemistry 1996, 35, 16378-16390.
(50) Rivera, M.; Seetharaman, R.; Girdhar, D.; Wirtz, M.; Zhang, X.; Wang,
X.; White, S. Biochemistry 1998, 37, 1485-1494.
(51) Guzov, V. M.; Houston, H. L.; Muratalier, M. B.; Walker, F. A.;
Feyeriesen, R. J. Biol. Chem. 1996, 271, 26637-26645.

2880 Inorganic Chemistry, Vol. 39, No. 13, 2000
Silver et al.
line broadening of features in this region, and no other feature.
This clear difference between the structures fits well with the
two kinds of geometry. The first arises from the 2B ground state.
2
The second, when nonsterically hindered, arises from the E
ground state and when sterically hindered from the (dxz,dyz)4-
(dxy)1 ground state.
Compound 21, which we recently reported,57 is unusual. It
has the longest axial Fe-N bond lengths (sterically hindered)
and has a rhombic type EPR spectrum but with its gz value
near those found for “large gmax” type EPR spectra. Thus, it
has the (dxy)2(dxz,dyz)3 ground state. As can be seen from Figure
5, it is no closer to the 45° line (ideal octahedral structure) than
the other compounds. The axial azaferrocene ligands in this
compound are poor π-acceptors but good σ-donors. These axial
ligands are sterically hindered from approaching the Fe atom,
though like the imidazoles, they are five-membered rings and
so might otherwise have been expected to get closer to the iron
atom. The large “gmax” compound closest to 21 is compound
14. Compound 14 is a pyridine complex of tetraphenylporphy-
rinatoiron(III). It is worth noting that the [FePPIX(py)2]+
complex has in fact a (dxy)2(dxz,dyz)3 ground state.9,17 The latter
porphyrin, unlike TPP, has no bulky side groups to interact
sterically with the pyridine, and so the pyridine N atom can
bind more strongly. Compounds 22 and 2361 are the closest
compounds to compound 21. They are also FeIIIOEP complexes.
Here, the iron-nitrogen porphyrin bond lengths are shorter than
compound 21 even though the axial ligands are weak, supporting
the fact that the ligands in compound 21 are sterically hindered.
Compounds 22 and 23 are known to have a (dxy)2(dxz, dyz)3
ground state.62-64
Figure 5. Plot of average Fe-Npor distance vs average Fe-Nax
distance.
Simple Explanation of the Different Ground States. The
plot of average Fe-Npor distance (average of all four Fe-N
porphyrin distances) against Fe-Nax distance (Figure 5) is
instructive. The compounds on the plot can be divided into four
groups except for compound 2157 (which is discussed below).
Compounds 1-13 and 225-7,9,11,13,14,16,19 have all been shown
to have a (dxy)2(dxz,dyz)3 ground state, which is an orbital singlet
(2B), and the unpaired electron is localized in the dyz orbital.1,57
Of these compounds, 1-13 are within the same area (presented
with an arbitrary oval line) of the plot. Two other areas are
also shown on the plot. In the area between the smaller and
larger areas are compounds 14, 15, and 19.9,20,58 These are the
“large gmax” class and have an orbital doublet (2E) ground state.
Thus, from this plot the different ground states can be simply
explained in terms of crystal field effects of the six ligand atoms
(the nitrogen atoms) nearest the iron. When the axial ligands
are sterically hindered, then they are more weakly bonded than
the porphyrin nitrogen ligands and the (dxz,dyz)4(dxy)1 ground
state is preferred. Such compounds are 16-18 and 209,59,60
shown within the smaller arc. For compounds in the latter two
cases the axial ligands are perpendicular and the Fe-Nax bond
lengths differ (their average is plotted) in Figure 5.
It thus appears that Figure 5 may be useful in considering
which axial ligands would control the ground state of the iron
for a given porphyrin. For instance, it may be possible to choose
an axial ligand that would change its ground state with an iron
porphyrin as a function of temperature. [FeIIIOEP(4-NMe2Py)2]-
ClO4, compound 22, has been shown to exist in two forms, one
of which is said to exist in a high- and low-spin equilibrium.9,62
Moreover, a range of FeIIIOEP-substituted bispyridine com-
pounds have been shown to exhibit such behavior.63,64 It is
significant that it is the FeIIIOEP compounds that lie between
The perpendicular arrangement arises for the (dxy)2(dxz,dyz)3
ground state either from an electronic interaction of the Fe (dxy)2
orbital and the axial ligand π-orbitals, though donation from
the latter to the former would be very weak, or more likely
through dπ-pπ weak bonding interactions from the iron dxz or
dyz orbitals donating each to different axial ligands. For the
(dxz,dyz)4(dxy)1 case the perpendicular arrangement is probably
due to interactions with the filled iron dxz and dyz orbitals and
the axial ligands.
The EPR results fit clearly into two distinct types, both
relating to low-spin Fe(III) complexes. One set fits nicely onto
the correlating graph shown in ref 1. There are always three
well-defined features, gy falling close to 2.2 while gx and gz
vary together, with gz decreasing as gx increases. These range
from ca. 1.95 to 1.5 for gx and ca. 2.5 to 3.0 for gz. All these
compounds are on the left of the plot in Figure 5.
2
2
the B and E ground states. Compounds 21 and 22 may lie
2
2
near the true boundary between B and E states, where an
intermediate spin complex has been reported.59
This explanation of the different ground states also gives
further insight into the compounds that obey the relationship
between V and ∆. For the imidazole ligands in Figure 5 it is
apparent that compounds 1-6 (which have the relationship
shown in Figure 4) all have a large Fe-Npor average distance
(greater than 1.99 Å). Of the other compounds with imidazoles
as axial ligands (that have known EPR spectra), only compound
7 has such a value. This compound is very electron-rich and is
a good π-donor. For one of the two axial ligands, φ ) 1° and
the ligand is thus very suitable for π-bonding to the half-
occupied iron orbital and does this very well, and the other is
not too far away, φ ) 17°, and cannot donate more density
because the iron is probably saturated. Hence, compound 7 does
The other set (compounds 14-20 on the right-hand side of
the plot in Figure 5) is quite different, the key difference being
the very large high-field shifts for gz (g|), together with major
(57) Cesario, M.; Giannotti, C.; Guilhem, J.; Silver, J.; Zakrzewski, J. J.
Chem. Soc., Dalton Trans. 1997, 47-53.
(61) Scheidt, W. R.; Geiger, D. K.; Haller, K. J. J. Am. Chem. Soc. 1982,
104, 495-499.
(58) Inniss, D.; Soltis, S. M.; Strouse, C. E. J. Am. Chem. Soc. 1988, 110,
5644-5650.
(62) Scheidt, W. R.; Geiger, D. K.; Hayes, R. G.; Lang, G. J. Am. Chem.
Soc. 1983, 105, 2625-2632.
(59) Safo, M. K.; Gupta, G. P.; Watson, C. T.; Simonis, U.; Walker, F.
A.; Scheidt, W. R. J. Am. Chem. Soc. 1992, 114, 7066-7075.
(60) Safo, M. K.; Walker, F. A.; Raitsimring, A. M.; Walters, W. P.; Dolata,
D. P.; Debrunner, P. G.; Scheidt, W. R. J. Am. Chem. Soc. 1994, 116,
7760-7770.
(63) Hill, H. A. O.; Skyte, P. D.; Buchler, J. W.; Lueken, H.; Tonn, M.;
Gregson, A. K.; Pellizer, G. J. Chem. Soc., Chem. Commun. 1979,
151.
(64) Gregson, A. K. Inorg. Chem. 1981, 20, 81-87.

Crystal Structure of Fe(III) Compounds
Inorganic Chemistry, Vol. 39, No. 13, 2000 2881
in fact fit the relationship in Figure 5 rather well if the angle of
the second axial ligand is neglected.
1. There is a simple plot of average Fe-Npor distance against
average Fe-Nax distance that rationalizes the underlying dif-
ference in the low-spin FeIII ground states. Further, this plot
may be used to forecast which axial ligands control which
ground state and what type of properties need to be designed
into a ligand to approach these boundary conditions.
Because the relationship in Figure 4 therefore appears
significant, it may be possible to read off a φ angle for a
V
V
measured value of /∆. For instance, cytochrome b5 has /∆ )
0.52 in neutral solution and a value of 0.66 in alkaline media.47
From Figure 4 this may mean that the φ angle of the “parallel”
ligand planes changes from 35° to about 0° as the solution is
made alkaline. Similarly, cytochrome b55924 may have an angle
around 25° in the low-potential form and a very different angle
2. A previously reported correlation is significant and may
be used to forecast the angles between parallel imidazole planes
and the Fe-Npor vector (the φ angles) in proteins if the EPR
spectra are known. It has been shown that complex 6,
[FeIIITPP(4-MeIm)2]Cl, which has parallel imidazole planes, fits
the correlation shown Figure 4.
V
where /∆ ) 0.41 in the high-potential form.
At first sight it may be thought that this ignores effects such
as imidazolate character (which may be important in the
experiment that is in alkaline media), tilt distortion, changes in
degree of tetragonality or in ruffling. However, all these physical
properties will have affected the compounds in the correlation,
and yet the correlation holds. So it is reasonable to assume that
they are minor effects and that the dominant effect is the φ angle
The angle between the planes of the ligands is clearly
important for complexes 7-9 that do not fit the correlation well.
The ligands of complexes 7 and 9 can be thought of as being
electron-rich and so would be good electron donors through
the nitrogen atom lone pair. 1-Vinylimidazole, the ligand of
complex 8, might be expected to be electron-poor if the vinyl
group is electron-withdrawing. Thus, the reasons for these
complexes having nonparallel ligand planes are obviously
complicated.
V
and /∆ value (though the correlation to some extent may be
based on a small contribution to these factors).
Conclusions
We have been able to demonstrate two findings for
[FeIIIPor(L)2]+ complexes.
IC990848S