[Fe(OEP?)(X)]+ π-Cation Radicals: Characterization and Spin-Spin Interactions

Article abstract of DOI:10.1021/ic961053d

The preparation and characterization of the π-cation radical derivatives of [Fe(OEP)(X)] (X = Cl , Br^-) is reported. Three different derivatives have been prepared: [Fe(OEP?)(Cl)]ClO₄, [Fe(OEP?)(Cl)][SbCl₆], and [Fe(OEP?)(Br)][SbCl₆]. All derivatives have been characterized by UV - vis, IR, and Mo?ssbauer spectroscopy. In addition. [Fe(OEP?)(Cl)]ClO₄ has been characterized by a single-crystal structure determination, and [Fe(OEP?)(Cl)][SbCl₆] and [Fe(OEP)(Br)][SbCl₆] have been studied by temperature-dependent magnetic susceptibility measurements and Mo?ssbauer measurements in an applied magnetic field. The X-ray structure of [Fe(OEP?)(Cl)]ClO₄ reveals a five-coordinate porphyrinate species that forms tight cofacial π-π dimers in which the two porphyrin rings are almost exactly overlapped with an inter-ring separation of 3.24 A?, a lateral shift of 0.2 A?, and a twist angle between the two rings of 31.2°; the two iron atoms are 4.112 A? apart. Crystal data: C₃₇H₄₆FeCl₄O₄N₄,a = 27.454(7) A?, b = 15.322(3) A?, c = 19.802(3) A?, β= 116.14 (2)°, monoclinic, C2/c, Z = 8. Iron(III) is found to be in the high-spin state in all derivatives. The magnetic data (susceptibility and Mo?ssbauer) have been interpreted in terms of two spin coupling models. Both models give a picture of strong coupling between the various spins in the dimeric species. In the model judged to best fit all data with a physically meaningful zero-field splitting, there are three terms in the total Hamiltonian: an axial zero-field splitting parameter D for the high-spin iron, an intramolecular antiferromagnetic coupling -2J_Fe-r(S · s) between the iron spin S = 5/2 and the π-cation radical s = 1/2 spin, and an intermolecular antiferromagnetic coupling -2J_R-R(S + s)· (S′ +-s)-(5′ +s′) between the total spins on each half of the dimer.

Full text of DOI:10.1021/ic961053d

+
+
406
Inorg. Chem. 1997, 36, 406-412
[Fe(OEP•)(X)]+ π-Cation Radicals: Characterization and Spin-Spin Interactions
Charles E. Schulz,*,† Hungsun Song,‡ Anil Mislankar,‡ Robert D. Orosz,§
Christopher A. Reed,*,§ Peter G. Debrunner,*,| and W. Robert Scheidt*,‡
Departments of Physics, Knox College, Galesburg, Illinois 61401, and University of Illinois,
Urbana, Illinois 61801, Department of Chemistry and Biochemistry, University of Notre Dame,
Notre Dame, Indiana 46556, and Department of Chemistry, University of Southern California,
Los Angeles, California 90089-0744
ReceiVed August 28, 1996X
The preparation and characterization of the π-cation radical derivatives of [Fe(OEP)(X)] (X ) Cl-, Br-) is reported.
Three different derivatives have been prepared: [Fe(OEP•)(Cl)]ClO4, [Fe(OEP•)(Cl)][SbCl6], and [Fe(OEP•)(Br)]-
[SbCl6]. All derivatives have been characterized by UV-vis, IR, and Mo¨ssbauer spectroscopy. In addition,
[Fe(OEP•)(Cl)]ClO4 has been characterized by a single-crystal structure determination, and [Fe(OEP•)(Cl)][SbCl6]
and [Fe(OEP)(Br)][SbCl6] have been studied by temperature-dependent magnetic susceptibility measurements
and Mo¨ssbauer measurements in an applied magnetic field. The X-ray structure of [Fe(OEP•)(Cl)]ClO4 reveals
a five-coordinate porphyrinate species that forms tight cofacial π-π dimers in which the two porphyrin rings are
almost exactly overlapped with an inter-ring separation of 3.24 Å, a lateral shift of 0.2 Å, and a twist angle
between the two rings of 31.2°; the two iron atoms are 4.112 Å apart. Crystal data: C37H46FeCl4O4N4, a )
27.454(7) Å, b ) 15.322(3) Å, c ) 19.802(3) Å, â ) 116.14 (2)°, monoclinic, C2/c, Z ) 8. Iron(III) is found
to be in the high-spin state in all derivatives. The magnetic data (susceptibility and Mo¨ssbauer) have been interpreted
in terms of two spin coupling models. Both models give a picture of strong coupling between the various spins
in the dimeric species. In the model judged to best fit all data with a physically meaningful zero-field splitting,
there are three terms in the total Hamiltonian: an axial zero-field splitting parameter D for the high-spin iron, an
5
intramolecular antiferromagnetic coupling -2JFe-r(bS‚bs) between the iron spin S ) /2 and the π-cation radical s
) 1/2 spin, and an intermolecular antiferromagnetic coupling -2JR-R(bS + bs)‚(bS′ + bs′) between the total spins on
each half of the dimer.
Introduction
Hund’s rule-like maximum spin multiplicity. Case ii arises in
lower symmetry systems where the metal and ligand magnetic
In earlier work on metalloporphyrin π-cation radicals,1-6 we
have investigated the variety of spin couplings between the metal
ion and porphyrin ring in a number of derivatives that differ in
metal ion, axial ligation, and/or porphyrin ligand. Both intra-
and intermolecular spin coupling can be significant.7 Two
distinct intramolecular coupling mechanisms are found: (i)
ferromagnetic coupling between the metal and ligand magnetic
centers, leading to maximal spin multiplicity and (ii) antifer-
romagnetic coupling, effectively leading to a lower multiplicity
state. It is found that case i is observed when the porphyrin
ring is essentially planar and approximates D4h symmetry. In
this symmetry, the partially filled metal orbitals and the
porphyrin π-radical orbital are orthogonal, which then leads to
orbitals are not strictly forbidden by symmetry from net overlap.
The copper(II) π-cation radicals of TPP2,8 and TMP,4 which
are S ) 0 and S ) 1 species, respectively, nicely illustrate these
important core conformation effects on magnetic coupling. Fajer
and co-workers9 have recently confirmed the effects of core
ruffling with the use of a sterically congested copper(II) radical
π-cation complex; the compound is diamagnetic, as expected
for this severely ruffled derivative. In well-characterized
systems that display intermolecular coupling, it is the unpaired
electrons on the porphyrin rings that are coupled. The
magnitude of the coupling ranges from 2J ) -14.7 cm-1 in
one crystalline form of weakly coupled [Zn(TPP•)(OClO3)]
dimers to greater than -200 cm-1 in the strongly coupled
(diamagnetic) dimer [Zn(OEP•)(OH2)]2(ClO4)2.4
Cases where both intra- and intermolecular spin coupling can
be significant are less well understood. The [Cu(OEP•)]+ and
[Cu(OEC•)]+ radical systems exhibit Cu-Cu triplet EPR
spectra10 with no evidence for unpaired radical electrons in either
the EPR10 or the bulk magnetic susceptibility.11 The results
are consistent with pairwise interacting complexes. It is
presumed that these copper species have strongly overlapped,
cofacial ring pairs like those found in the zinc and nickel
† Know College.
‡ University of Notre Dame.
§ University of Southern California.
| University of Illinois.
X Abstract published in AdVance ACS Abstracts, January 15, 1997.
(1) Gans, P.; Buisson, G.; Due´e, E.; Marchon, J.-C.; Erler, B. S.; Scholz,
W. F.; Reed, C. A. J. Am. Chem. Soc. 1986, 108, 1223.
(2) Erler, B. S.; Scholz, W. F.; Lee, Y. J.; Scheidt, W. R.; Reed, C. A. J.
Am. Chem. Soc. 1987, 109, 2644.
(3) Song, H.; Rath, N. P.; Reed, C. A.; Scheidt, W. R. Inorg. Chem. 1989,
28, 1839.
(4) Song, H.; Reed, C. A.; Scheidt, W. R. J. Am. Chem. Soc. 1989, 111,
6865.
(5) (a) Song, H.; Reed, C. A.; Scheidt, W. R. J. Am. Chem. Soc. 1989,
111, 6867. (b) Song, H.; Orosz, R. D.; Reed, C. A.; Scheidt, W. R.
Inorg. Chem. 1990, 29, 4274.
(6) Schulz, C. E.; Song, H.; Lee, J. Y.; Mondal, J. U.; Mohanrao, K.;
Reed, C. A.; Walker, F. A.; Scheidt, W. R. J. Am. Chem. Soc. 1994,
116, 7196.
(7) Reed, C. A.; Orosz, R. D. In Research Frontiers in Magnetochemistry;
O’Connor, C. J., Ed.; World Scientific: Singapore, 1993; pp 351-
393.
(8) Abbreviations used: TMP, dianion of tetramesitylporphyrin; TPP,
dianion of tetraphenylporphyrin; OEP, dianion of octaethylporphyrin;
OEC, dianion of octaethylchlorin; a raised dot in the formula of the
porphyrin indicates a π-cation radical, e.g., OEP• for the OEP π-cation
radical.
(9) Renner, M. W.; Barkigia, K. M.; Zhang, Y.; Medforth, C. J.; Smith,
K. M.; Fajer, J. J. Am. Chem. Soc. 1994, 116, 8582.
(10) Mengerson, C.; Subramanian, J.; Fuhrhop, J.-H. Mol. Phys. 1976, 32,
1299.
S0020-1669(96)01053-1 CCC: $14 00 © 1997 American Chemical Society

+
+
[Fe(OEP•)(X)]+ π-Cation Radicals
Inorganic Chemistry, Vol. 36, No. 3, 1997 407
complexes5 and in related examples of other [M(OEP•)] π-cation
radicals.12 In the six-coordinate complex [VO(OH2)(OEP•)]-
SbCl6,6 where the two axial ligands inhibit the face to face
approach of the two porphyrin rings, there is only a “saw-
toothed” edge to edge contact between the two radical rings.
Nonetheless, the most satisfactory model that explains the
physical data is one with intramolecular ferromagnetic coupling
(between the metal and radical) and fairly strong antiferromag-
netic intermolecular coupling (2Jr-r) -139 cm-1) between two
radical spins, rather than just simple mononuclear, intramo-
lecular coupling.
In an earlier communication,14 we reported the molecular
structure, magnetic susceptibility, and Mo¨ssbauer spectra of the
π-cation radical derivative of chloro(octaethylporphinato)iron-
(III), [Fe(OEP•)(Cl)]+. Two distinct models with different
intermolecular coupling constants were found to provide
adequate fits of the Mo¨ssbauer and magnetic susceptibility data;
both are fully consistent with high-spin iron(III) species. To
confirm the applicability of these models, we have now
characterized a related bromo derivative, [Fe(OEP•)(Br)]+, as
its hexachloroantimonate salt. We also fully report on our
characterization of the chloro derivative [Fe(OEP•)(Cl)]+ and
the further development of models for the magnetic interactions.
Table 1. Crystallographic Data
empirical formula: C37H46FeCl4O4N4
a ) 27.454(7) Å
b ) 15.322(3) Å
c ) 19.802(3) Å
â ) 116.14(2)°
space group ) C2/c
T ) -150(2) °C
λ ) 0.710 73 Å
F
calcd ) 1.44 g cm-3
µ ) 7.3 cm-1
V ) 7477.4 Å3
Z ) 8
R1(Fo)a ) 0.069
R2(Fo)b ) 0.063
2
a R1 ) ∑||Fo| - |Fc||/∑|Fo|. b R2 ) |∑w(|Fo| - |Fc|)2/∑wFo ]1/2
.
contained hexanes as the nonsolvent and the other one had pentane.
Both bottles were placed in a refrigerator (4 °C) for crystallization.
Two days later, both bottles were opened and the crystals were
harvested. A suitable crystal for X-ray diffraction study was obtained
from the crystallization with pentane as the nonsolvent. These
crystallization experiments always resulted in the formation of both
[Fe(OEP•)(Cl)]ClO4 and [Fe(OEP•)(OClO3)2]. The two products could
be differentiated by the morphology of the crystals. In our experiments,
crystals of [Fe(OEP•)(Cl)]ClO4 have hexagonal shapes while those of
[Fe(OEP•)(OClO3)2] have parallelepiped shapes. UV-vis and IR
spectra were measured on samples comprised of selected crystals.
UV-vis (CH2Cl2 solution): λmax 356 (Soret), 518, 581 nm. IR (KBr):
ν(OEP•) 1533 cm-1 (broad); ν(ClO4) 1144 (strong, broad), 623 cm-1
ν(Fe-Cl) 353 cm-1 (broad).
;
Bulk samples were prepared by the following procedure: [Fe(OEP)-
(Cl)] (97.5 mg, 0.156 mmol) and thianthrenium perchlorate (49.2 mg,
0.156 mmol) were dissolved in dichloromethane in a 100 mL Schlenk
flask. The solution was placed in a sonicator for 1 min and then filtered
into pentane. The resulting precipitate was filtered and washed with
pentane until the filtrate was colorless. The product was dried under
vacuum (yield 75%).
Experimental Section
General Information. UV-vis spectra were recorded on Perkin-
Elmer Lambda 6 and 19 spectrophotometers and IR spectra on a Perkin-
Elmer 883 spectrometer. EPR spectra were measured on a Varian
E-Line spectrometer operating at X-band frequency. Mo¨ssbauer spectra
were measured on ground crystals as Apiezon grease suspensions at
4.2 and 200 K in a zero field and with a 4.5 T applied magnetic field
at a number of temperatures. All solid-state samples for the spectros-
copy measurements were prepared in a Vacuum Atmospheres drybox.
All reactions were performed under an argon atmosphere with Schlen-
kware and cannula techniques. Dichloromethane was dried by distil-
lation from CaH2, and pentane and hexanes were dried by distillation
from sodium-benzophenone. Tris(p-bromophenyl)aminium hexachlo-
roantimonate was purchased from Aldrich. Iron was inserted into
H2OEP by standard techniques.15 Thianthrenium perchlorate was
prepared by literature procedures.16 Caution: These perchlorate-con-
taining materials can detonate spontaneously and should be handled
only in small quantities; other safety precautions are also warranted.
We have experienced an explosion of thianthrenium perchlorate under
mild heating but have had no difficulties with the porphyrin radical
salts.17
Preparation of [Fe(OEP•)(Cl)][SbCl6]. [Fe(OEP)(Cl)] (89.7 mg,
0.144 mmol) and tris(p-bromophenyl)aminium hexachloroantimonate
(115.9 mg, 0.142 mmol) were dissolved in dichloromethane (∼15 mL)
in a 100 mL Schlenk flask. The solution was stirred for an hour, and
the solvent was removed under vacuum until ∼5 mL of solvent
remained. The solution was filtered, and the product was washed with
a small amount of dichloromethane (yield 92%). UV-vis (CH2Cl2
solution): λmax 356 (Soret), 518, 576 nm. IR (KBr): ν(OEP•) 1532
cm-1 (strong); ν(SbCl6) 344 cm-1 (strong).
Preparation of [Fe(OEP•)(Br)]SbCl6. [Fe(OEP)(Br)] (200 mg,
0.299 mmol) and tri-(p-bromophenyl)aminium hexachloroantimonate
(248 mg, 0.304 mmol) were placed in a Schlenk flask and degassed
for 1 h. About 20 mL of dichloromethane was then introduced into
the flask through a cannula under argon. Immediately a brown-colored
precipitate began to appear. The mixture was stirred for 1.5 h. The
solution was filtered, and the product was washed with dichloromethane
to remove unreacted [Fe(OEP)(Br)] and NR3 SbCl6. The product was
then dried under vacuum (yield 90%). UV-vis (CH2Cl2 solution): λmax
358 (Soret), 519, 586, 623 nm. IR (KBr): ν(OEP•) 1535 cm-1 (strong);
Preparation of [Fe(OEP•)(Cl)]ClO4. [Fe(OEP)(Cl)] (51.3 mg,
0.0822 mmol) and thianthrenium perchlorate (25.4 mg, 0.0804 mmol)
were placed in a 100 mL Schlenk flask, and dichloromethane (∼15
mL) was added. The solution was stirred for 10 min, filtered, and
transferred into 10 mL beakers, which were placed in crystallizing
bottles under argon for crystallization. One crystallizing bottle
ν(SbCl6) 344 cm-1
.
Magnetic Susceptibility Measurements. Measurements were
performed on lightly compressed samples (∼30 mg) in an aluminum
bucket on a SHE Model 905 SQUID susceptometer at 2 and 10 kG.
When samples were ground vigorously, as is usual for magnetic
susceptibility measurements, inconsistent data were obtained. An
apparent lattice disruption occurred, and these data were not used.
X-ray Diffraction Studies. A suitable dark-brown single crystal
of [Fe(OEP•)(Cl)]ClO4‚CH2Cl2, with approximate dimensions 0.07 ×
0.27 × 0.31 mm, was subjected to preliminary examination at 123 (
2 K. An Enraf-Nonius CAD4 diffractometer equipped with a locally
modified Syntex LT-1 cooling system and with graphite-monochro-
mated Mo KR radiation (λ ) 0.710 73 Å) was used. Final cell con-
stants and space group are reported in Table 1. The choice of the
centrosymmetric space group was supported by the E-statistics and the
subsequent successful solution and refinement of the structure. The
intensity data were collected using the θ/2θ scan technique. A total
of 5787 reflections were considered observed and were corrected for
the effects of absorption. Full crystallographic details are given in
Table S1.
(11) The 6-300 K magnetic susceptibility data for [Cu(OEP•)][SbCl6] are
cleanly fit by the Bleaney-Bowers model as a pair of weakly
interacting Cu(II) ions with no residual magnetism attributable to the
radical spins: Mondal, J. U.; Scheidt, W. R. Unpublished results.
(12) Brancato-Buentello, K.; Cheng, B.; Reddy, K. V.; Scheidt, W. R.
Manuscript in preparation. We have also carried out spectroscopic
studies measuring dimerization constants for a number of [M(OEP•)]+
derivatives; Kdim for the copper complex is ∼100. There is no anion
dependence for this complex with the perchlorate and hexachloroan-
timonate anions.13
(13) Brancato-Buentello, K. E.; Kang, S.-J.; Scheidt, W. R. Submitted for
publication.
(14) Scheidt, W. R.; Song, H.; Haller, K. J.; Safo, M. K.; Orosz, R. D.;
Reed, C. A.; Debrunner, P. G.; Schulz, C. E. Inorg. Chem. 1992, 31,
939.
(15) Adler, A. D.; Longo, F. R.; Kampas, F.; Kim, J. J. Inorg. Nucl. Chem.
1970, 32, 2443.
(16) Murata, Y.; Shine, H. J. J. Org. Chem. 1969, 34, 3368.
(17) Wolsey, W. C. J. Chem. Educ. 1973, 50, A335. Chem. Eng. News
1983, 61 (Dec. 5), 4; 1963, 41 (July 8), 47.
The structure was solved using the Patterson interpretation routine
of SHELXS-8618 followed by tangent formula recycling, which revealed

+
+
408 Inorganic Chemistry, Vol. 36, No. 3, 1997
Schulz et al.
the positions of all 50 non-hydrogen atoms of the cation, anion, and
solvate. All hydrogen atoms were located from difference electron
density Fourier maps calculated after a preliminary least-squares
refinement of the non-hydrogen atoms. The final cycles of refinement
assigned anisotropic temperature factors to the non-hydrogen atoms
and included the hydrogen atoms as fixed, idealized contributors (dC-H
) 0.95 Å, BH ) 1.1 × BC). The final data:variable ratio was 12.8:1.
Refinement converged with unweighted and weighted agreement factors
of R1 ) 0.069 and R2 ) 0.063; the estimated standard deviation of an
observation of unit weight was 1.45. The final difference electron
density Fourier map was judged free of significant features with the
highest peak ) 0.64 e/Å3. Final atomic coordinates are listed in Table
S2. Anisotropic thermal parameters for all non-hydrogen atoms and
the fixed hydrogen atom coordinates are also available as Supporting
Information (Tables S3 and S4).
Results
Small quantities of pure [Fe(OEP•)(Cl)]ClO4 can be pre-
pared by the hand selection of single crystals, and indeed,
these crystals are of X-ray quality. However, bulk samples of
[Fe(OEP•)(Cl)]ClO4 (quantities large enough for Mo¨ssbauer and
magnetic susceptibility studies) were always contaminated
with [Fe(OEP•)(OClO3)2], presumably because of labile anions
and differential solubilities. Although the low solubility of
the [Fe(OEP•)(Cl)][SbCl6] and [Fe(OEP•)(Br)][SbCl6] deriva-
tives prevented us from obtaining single-crystal specimens
adequate for X-ray analysis, bulk samples of both hexachloro-
antimonate salts are readily prepared and are pure by Mo¨ssbauer
criteria.
Mo¨ssbauer spectra of all species in a zero applied magnetic
field display a single quadrupole doublet, except for that of
[Fe(OEP•)(Cl)]ClO4 which also shows a smaller, second doublet
consistent with the presence of a small amount of [Fe(OEP•)-
(OClO3)2]. The quadrupole doublets and isomer shifts at 4.2
K are ∆Eq )0.59 mm/s and δ ) 0.41 mm/s for [Fe(OEP•)(Cl)]-
ClO4, ∆Eq ) 0.71 mm/s and δ ) 0.42 mm/s for [Fe(OEP•)-
(Cl)][SbCl6], and ∆Eq ) 0.77 mm/s and δ ) 0.41 mm/s for
[Fe(OEP•)(Br)][SbCl6]. Mo¨ssbauer spectra of these radicals
have also been obtained in a 4.5 T applied magnetic field. These
spectra show a temperature-dependent magnetic broadening of
the quadrupole doublet (Figure 1).
Figure 1. Mo¨ssbauer spectra of [Fe(OEP•)(Br)][SbCl6] taken in an
applied magnetic field of 45.6 kG parallel to the γ-ray direction at
temperatures of 35, 75, and 150 K. The solid lines are simulations using
the first model described in the text.
Temperature-dependent magnetic susceptibilities have been
obtained from 2-300 K for [Fe(OEP•)(Cl)][SbCl6] and 6-300
K for [Fe(OEP•)(Br)][SbCl6]. Measurements were made at 2
and 10 kG. The lack of any field dependence confirmed that
the samples were free of ferromagnetic impurities. Replicate
measurements were made, and a Mo¨ssbauer spectrum was
obtained on a portion of each sample. All samples showed only
a single quadrupole doublet. Both species show similar
temperature-dependent values for the magnetic moment. Data
for [Fe(OEP•)(Br)][SbCl6] are given in graphical form in Figure
2, and data for [Fe(OEP•)(Cl)][SbCl6] are given in Figure 3.
Data from Nakashima et al.19 are also displayed in the top panel
of Figure 3. Possible spin coupling models for the complexes
were derived from consideration of both the magnetic suscep-
tibility data and the Mo¨ssbauer data in an applied field.
Complementary physical data are necessary in limiting the
possible models.
The molecular structure of the five-coordinate [Fe(OEP•)-
(Cl)]+ cation as the perchlorate salt is shown in Figure 4. The
labeling scheme for the atoms employed in all tables is also
shown in the diagram. Individual values of the bond distances
and angles are given in Tables 2 and 3. The average value of
the four Fe-Np bonds is 2.058(5) Å, and the axial Fe-Cl bond
distance is 2.186(1) Å. The displacement of the iron(III) atom
out of the 24-atom mean plane is 0.43 and 0.46 Å from the N4
plane. Figure 5 shows the displacements, in units of 0.01 Å,
of the atoms of the 24-atom porphinato core from the best least-
squares plane. The porphyrin core is slightly domed toward
the iron atom, i.e., the core exhibits a rather unusual reverse
doming. Nonetheless, all deviations from exact planarity are
less than 0.08 Å. [Fe(OEP•)(Cl)]+ cations exist as cofacial
dimers in the solid state. Two orthogonal views of the cofacial
π-π dimer are given in Figures 6 and 7. The two iron(III)
atoms in the dimer are separated by 4.112(1) Å, the Ct‚‚‚Ct
(18) Sheldrick, G. Acta Crystallogr., Sect. A 1990, A46, 467. Other
programs used in this study included local modifications of Jacobson’s
ALLS, Zalkin’s FORDAP, Busing and Levy’s ORFFE and ORFLS,
the SDP package of Enraf-Nonius, and Johnson’s ORTEP2. Atomic
form factors were from: Cromer, D. T.; Mann, J. B. Acta Crystallogr.,
Sect. A 1968, A24, 321. Real and imaginary corrections for anomalous
dispersion in the form factor of the iron and chlorine atoms were
from: Cromer, D. T.; Liberman, D. J. J. Chem. Phys. 1970, 53, 1891.
Scattering factors for hydrogen were from: Stewart, R. F.; Davidson,
E. R.; Simpson, W. T. J. Chem. Phys. 1965, 42, 3175. All calculations
were performed on VAXstation 3200 and 4000-90 computers.
(19) Nakashima, S.; Ohya-Nishiguchi, H.; Hirota, N.; Fujii, H.; Morishima,
I. Inorg. Chem. 1990, 29, 5207.

+
+
[Fe(OEP•)(X)]+ π-Cation Radicals
Inorganic Chemistry, Vol. 36, No. 3, 1997 409
Figure 2. Magnetic susceptibility data for [Fe(OEP•)(Br)][SbCl6] in
an applied field of 10 kG. The experimental data are shown as filled
circles, while the solid lines are calculated using multiparameter spin
coupling models. Panel A shows the fit with the spin coupling model
in which there is intermolecular coupling of the total spins and using
parameters given in the text. Panel B shows the fit with the second
model described in the text with three spin coupling (J) parameters.
Figure 3. Magnetic susceptibility data for [Fe(OEP•)(Cl)][SbCl6] in
an applied field of 10 kG. The experimental data are shown as filled
circles, while the solid lines are calculated using multiparameter spin
coupling models. Panel A shows the fit with the spin coupling model
in which there is intermolecular coupling of the total spins and using
parameters given in the text. Panel B shows the fit with the second
model described in the text with three spin coupling (J) parameters.
Panel A also shows the data reported by Nakashima et al.19 (filled
triangles). The dashed line is the fit to the data using the parameters
presented by Nakashima et al.19
separation is 3.25 Å, the separation between the two mean planes
is 3.24 Å, and the lateral shift is 0.2 Å.
Discussion
The coordination geometry of the five-coordinate [Fe(OEP•)-
(Cl)]+ cation, as the perchlorate salt, is that expected for a high-
spin iron(III) porphyrinate derivative: a square pyramidal FeN4X
group with values for the Fe-Np and Fe-X bonds and the iron
atom displacement within the range of previously observed
values.20 Thus, in agreement with earlier observations, the
formation of a porphyrin π-cation radical seems to have a
minimal effect on the metal ion coordination geometry. Fur-
thermore, physical properties that probe the spin state, notably
Mo¨ssbauer spectroscopy, also unambiguously lead to an as-
signment of a high-spin state to the iron. Thus, all attempts to
describe the electronic structure and magnetic properties of these
Figure 4. ORTEP diagram of the [Fe(OEP•)(Cl)]+ π-cation radical
showing the atom labeling scheme. Thermal ellipsoids of all atoms
are contoured at the 50% probability level. Hydrogen atoms have been
omitted for clarity.
5
species should utilize an S ) /2 state for iron(III).
As we have observed for other four- and five-coordinate
[M(OEP•)]+ radicals, the cation of [Fe(OEP•)(Cl)]ClO4 forms
a solid-state cofacial dimer in which the two porphyrin rings
interact quite strongly. There are two previously structurally
characterized species of this type, the [Zn(OEP•)(OH2)]+ and
[Ni(OEP•)]+ cations.5 Both also form cofacial dimers, and we
believe that such solid-state dimer formation is a very general
phenomenon for OEP π-cation radicals. Figure 6 shows that
two porphyrin planes in the [Fe(OEP•)(Cl)]22+ dimer approach
each other closely; the interplanar separation between the two
least-squares planes is found to be a strikingly small 3.24 Å.
Corresponding values in [Zn(OEP•)(OH2)]+ and [Ni(OEP•)]+
are 3.31 and 3.19 Å, respectively. In all cases, the interplanar
separation is smaller than that of the ∼3.35 Å spacing of
graphite; moreover, the C‚‚‚C separations in the radicals are
much smaller than those in graphite. Although not required
by the crystallographic symmetry, the two porphyrin planes are
parallel to each other within 0.2°. (The two planes of Figures
6 and 7 are related by a crystallographic 2-fold axis.)
The perpendicular view of the cofacial dimer (Figure 7) shows
that the two rings are directly above each other (i.e., the lateral
shift is effectively zero). Such completely overlapped rings are
(20) Scheidt, W. R.; Reed, C. A. Chem. ReV. 1981, 81, 543.

+
+
410 Inorganic Chemistry, Vol. 36, No. 3, 1997
Schulz et al.
a
Table 2. Bond Distances (Å) in [Fe(OEP•)(Cl)]ClO4
Table 3. Bond Angles (deg) in [Fe(OEP•)(Cl)]ClO4
Fe-Cl(1)
2.186(1)
2.055(3)
2.063(3)
2.052(3)
2.062(3)
1.364(4)
1.384(5)
1.357(5)
1.386(4)
1.356(4)
1.398(5)
1.350(5)
1.394(4)
1.461(5)
1.402(5)
1.459(5)
1.371(5)
1.450(5)
1.403(5)
1.457(5)
1.372(5)
1.456(5)
1.408(5)
1.452(5)
1.371(5)
1.439(5)
1.410(5)
1.450(5)
C(a8)-C(m4)
C(b1)-C(b2)
C(b3)-C(b4)
C(b5)-C(b6)
C(b7)-C(b8)
C(b1)-C(11)
C(b2)-C(21)
C(b3)-C(31)
C(b4)-C(41)
C(b5)-C(51)
C(b6)-C(61)
C(b7)-C(71)
C(b8)-C(81)
C(11)-C(12)
C(21)-C(22)
C(31)-C(32)
C(41)-C(42)
C(51)-C(52)
C(61)-C(62)
C(71)-C(72)
C(81)-C(82)
C1(2)-O(1)
C1(2)-O(2)
C1(2)-O(3)
C1(2)-O(4)
C(9)-C1(3)
C(9)-C1(4)
Fe-Fe′ b
1.373(5)
1.355(5)
1.346(5)
1.356(5)
1.355(5)
1.487(5)
1.501(5)
1.502(5)
1.504(5)
1.500(5)
1.490(5)
1.505(5)
1.499(5)
1.519(5)
1.534(6)
1.530(6)
1.528(6)
1.525(5)
1.550(6)
1.525(6)
1.524(5)
1.449(3)
1.439(3)
1.447(3)
1.439(3)
1.758(4)
1.767(4)
4.112(1)
Cl(1)-Fe-N(1)
Cl(1)-Fe-N(2)
Cl(1)-Fe-N(3)
Cl(1)-Fe-N(4)
N(1)-Fe-N(2)
N(1)-Fe-N(3)
N(1)-Fe-N(4)
N(2)-Fe-N(3)
N(2)-Fe-N(4)
N(3)-Fe-N(4)
Fe-N(1)-C(a1)
Fe-N(1)-C(a2)
C(a1)-N(1)-C(a2)
Fe-N(2)-C(a3)
Fe-N(2)-C(a4)
C(a3)-N(2)-C(a4)
Fe-N(3)-C(a5)
Fe-N(3)-C(a6)
C(a5)-N(3)-C(a6)
Fe-N(4)-C(a7)
Fe-N(4)-C(a8)
C(a7)-N(4)-C(a8)
N(1)-C(a1)-C(b1)
103.10(9) C(a1)-C(b1)-C(11) 125.6(3)
104.47(9) C(b2)-C(b1)-C(11) 128.1(3)
102.29(9) C(a2)-C(b2)-C(b1) 106.9(3)
101.20(9) C(a2)-C(b2)-C(21) 123.7(3)
87.5(1) C(b1)-C(b2)-C(21) 129.4(3)
154.6(1) C(a3)-C(b3)-C(b4) 106.7(3)
87.0(1) C(a3)-C(b3)-C(31) 124.9(3)
86.6(1) C(b4)-C(b3)-C(31) 128.4(3)
154.3(1) C(a4)-C(b4)-C(b3) 106.6(3)
87.6(1) C(a4)-C(b4)-C(41) 125.3(3)
126.1(2) C(b3)-C(b4)-C(41) 128.1(3)
125.8(2) C(a5)-C(b5)-C(b6) 107.0(3)
105.5(3) C(a5)-C(b5)-C(51) 125.8(3)
126.7(2) C(b6)-C(b5)-C(51) 127.2(3)
126.2(2) C(a6)-C(b6)-C(b5) 106.2(3)
105.2(3) C(a6)-C(b6)-C(61) 123.9(3)
127.0(2) C(b5)-C(b6)-C(61) 129.8(3)
125.2(2) C(a7)-C(b7)-C(b8) 106.3(3)
105.2(3) C(a7)-C(b7)-C(71) 125.0(3)
126.3(2) C(b8)-C(b7)-C(71) 128.7(3)
125.7(2) C(a8)-C(b8)-C(b7) 106.8(3)
105.0(3) C(a8)-C(b8)-C(81) 126.0(3)
111.1(3) C(b7)-C(b8)-C(81) 127.2(3)
Fe-N(1)
Fe-N(2)
Fe-N(3)
Fe-N(4)
N(1)-C(a1)
N(1)-C(a2)
N(2)-C(a3)
N(2)-C(a4)
N(3)-C(a5)
N(3)-C(a6)
N(4)-C(a7)
N(4)-C(a8)
C(a1)-C(b1)
C(a1)-C(m4)
C(a2)-C(b2)
C(a2)-C(m1)
C(a3)-C(b3)
C(a3)-C(m1)
C(a4)-C(b4)
C(a4)-C(m2)
C(a5)-C(b5)
C(a5)-C(m2)
C(a6)-C(b6)
C(a6)-C(m3)
C(a7)-C(b7)
C(a7)-C(m3)
C(a8)-C(b8)
N(1)-C(a1)-C(m4) 124.5(3) C(a2)-C(m1)-C(a3) 125.7(4)
C(b1)-C(a1)-C(m4) 124.3(3) C(a4)-C(m2)-C(a5) 125.8(3)
N(1)-C(a2)-C(b2)
110.2(3) C(a6)-C(m3)-C(a7) 126.4(3)
N(1)-C(a2)-C(m1) 125.3(3) C(a1)-C(m4)-C(a8) 126.1(3)
C(b2)-C(a2)-C(m1) 124.5(4) C(b1)-C(11)-C(12) 113.3(3)
N(2)-C(a3)-C(b3)
111.3(3) C(b2)-C(21)-C(22) 113.3(3)
a Estimated standard deviations in the least significant digits are
given in parentheses. b Coordinates of primed atoms are related to those
of the corresponding unprimed atoms by the transformation 1 - x, y,
0.5 - z.
N(2)-C(a3)-C(m1) 124.9(3) C(b3)-C(31)-C(32) 114.0(3)
C(b3)-C(a3)-C(m1) 123.7(3) C(b4)-C(41)-C(42) 113.3(3)
N(2)-C(a4)-C(b4)
110.1(3) C(b5)-C(51)-C(52) 112.8(3)
N(2)-C(a4)-C(m2) 124.3(3) C(b6)-C(61)-C(62) 112.0(3)
C(b4)-C(a4)-C(m2) 125.6(3) C(b7)-C(71)-C(72) 113.3(3)
N(3)-C(a5)-C(b5)
111.1(3) C(b8)-C(81)-C(82) 112.6(3)
found in both of the previously characterized OEP cation
radicals, but are otherwise unknown in porphyrin structures. It
is interesting to note that the unusual reverse doming of the
core in [Fe(OEP•)(Cl)]ClO4 leads to slightly closer inter-ring
contacts than would be the case for a more planar species,
suggesting the probable stabilizing effects of the inter-ring
contacts. Moreover, bond distances in the inner 16-membered
ring show a quite unusual alternating pattern that must result
from a strong inter-ring interaction (vide infra). The great inner
16-membered ring consists of eight N-Ca and eight Ca-Cm
bonds which, in metalloporphyrins, are invariably equivalent
and are consistent with a completely delocalized system. In
the present case, however, the N-Ca and Ca-Cm bonds divide
into alternating “long” and “short” sets. The pattern and
individual values of the bond distances are shown in Figure 5.
Average values are 1.357(6) or 1.391(7) Å for N-Ca bonds,
and average Ca-Cm bond distances are 1.372(1) or 1.406(4)
Å. The number in parentheses following each averaged value
is the estimated standard deviation calculated on the assumption
that the averaged values are drawn from the same population;
the differences in the sets are clearly statistically significant.
Further evidence for the experimental significance of the
differences in these sets is given by the close agreement in all
values for the Ca-Cb and Cb-Cb bond distances: average values
are 1.453(7) and 1.353(5) Å, respectively.
A similar alternating bond distance pattern has been observed
for the zinc radical dimer but not for the nickel radical dimer
and naturally leads to the question of the basis for the effect. A
comparison of the inter-ring geometry and concomitant inter-
ring interactions in [Fe(OEP•)(Cl)]ClO4 with the previously
characterized radical dimers is especially revealing. The relative
orientation of the two rings of the dimer (twist angle), as
measured by the average N-Fe-Fe′-N′ dihedral angle, is
31.2°. The value in the Zn derivative is virtually identical at
31.3°, while the Ni complex has a twist angle of 40.4°. The
differences in twist angles leads to differing, close C‚‚‚C inter-
N(3)-C(a5)-C(m2) 124.6(3) O(1)-Cl(2)-O(2)
C(b5)-C(a5)-C(m2) 124.3(3) O(1)-Cl(2)-O(3)
109.5(2)
109.3(2)
109.5(2)
109.4(2)
109.7(2)
109.5(2)
111.7(2)
176.53(4)
77.14(9)
78.99(9)
77.49(9)
75.33(9)
N(3)-C(a6)-C(b6)
110.4(3) O(1)-Cl(2)-O(4)
N(3)-C(a6)-C(m3) 124.6(3) O(2)-Cl(2)-O(3)
C(b6)-C(a6)-C(m3) 125.0(3) O(2)-Cl(2)-O(4)
N(4)-C(a7)-C(b7)
112.1(3) O(3)-Cl(2)-O(4)
N(4)-C(a7)-C(m3) 124.3(3) Cl(3)-C(9)-Cl(4)
C(b7)-C(a7)-C(m3) 123.6(3) Fe′-Fe-Cl(1)
N(4)-C(a8)-C(b8)
109.8(3) Fe′-Fe-N(1)
N(4)-C(a8)-C(m4) 124.4(3) Fe′-Fe-N(2)
C(b8)-C(a8)-C(m4) 125.8(3) Fe′-Fe-N(3)
C(a1)-C(b1)-C(b2) 106.2(3) Fe′-Fe-N(4)
ring contacts. Magnetic data for all of the OEP dimers are
consistent with strong inter-ring coupling, but this does not lead
to the formation of a carbon-carbon bond as has been recently
observed21 for a (octaethyloxophlorin radical) nickel(II) deriva-
tive. In the nickel complex, the inter-ring interaction pattern is
a simple 8-fold set of Ca‚‚‚Ca interactions. If the inter-ring
interaction is between π-cation ring atoms with the highest
unpaired electron density, this Ca‚‚‚Ca is precisely the pattern
expected for the interaction of a pair of a1u radical cations.
However, the iron and zinc species have a different inter-ring
orientation, one in which the close contacts are between the Ca
and Cm atoms in the pattern Ca‚‚‚Ca, Ca‚‚‚Cm, Cm‚‚‚Ca, etc.,
around the inner 16-membered ring (cf. Figure 7).22
The small Mo¨ssbauer quadrupole splitting constants for the
two [Fe(OEP•)(Cl)]+ derivatives and for [Fe(OEP•)(Br)][SbCl6]
clearly indicate that the iron centers are high spin. The isomer
shift values require the assignment of a +3 oxidation state,
(21) Balch, A. L.; Noll, B. C.; Reid, S. M.; Zovinka, E. P. J. Am. Chem.
Soc. 1993, 115, 2431.
(22) Czernuszewicz et al. (Czernuszewicz, R.; Macor, K. A.; Li, X.-Y.;
Kincaid, J. R.; Spiro, T. G. J. Am. Chem. Soc. 1995, 111, 3860) have
suggested that owing to the near-degeneracy of the A1u and A2u ground
states, a pseudo-Jahn-Teller effect should arise and could lead to bond
alternation. We are continuing our studies on this interesting bond
alternation effect.

+
+
[Fe(OEP•)(X)]+ π-Cation Radicals
Inorganic Chemistry, Vol. 36, No. 3, 1997 411
Figure 5. Formal diagram of the porphinato core displaying perpen-
dicular displacements, in units of 0.01 Å, of the core atoms from the
24-atom mean plane. Also entered on the diagram are the values of
the individual bond distances in the core and coordination group. Note
the alternating short-long pattern of the N-Ca and Ca-Cm bond
distances in the great inner 16-membered ring.
Figure 7. ORTEP diagram of the cofacial dimer showing the overlap
between the two porphyrin rings. The least-squares porphyrin plane is
parallel to the plane of the paper. The axial chloride ligand has not
been drawn.
similar temperature dependence. We have been able to produce
acceptable fits to the data for the bromide and chloride
complexes with two models that treat the coupling between the
spins somewhat differently. The first model is that which we
described previously14 for [Fe(OEP•)(Cl)]2[SbCl6]2 and which
had been developed by Lang et al.24 for [Fe(TPP•)(Cl)][SbCl6].
In this model, there are three terms in the total Hamiltonian:
an axial zero-field splitting parameter D for the high-spin
iron, an intramolecular magnetic coupling -2JFe-r(bS‚bs) be-
5
1
tween the iron spin S ) /2 and the π-cation radical s ) /2
spin, and an intermolecular coupling -2JR-R(bS + bs)‚(bS′ + bs′)
between the total spins on each half of the dimer. The second
model uses a Hamiltonian with D and -2JFe-r(bS‚bs), as before,
along with additional coupling of the spins of the two radicals,
-2Jr-r(bs‚bs′), and the interaction between the two iron atoms,
-2JFe-Fe(bS‚bS′).
Figure 6. An edge-on ORTEP diagram of the cofacial dimer formed
by two [Fe(OEP•)(Cl)]+ π-cation radicals. The horizontal axis is parallel
to and the vertical axis is perpendicular to the 24-atom mean plane.
Atoms are contoured at the 50% probability level.
The magnetic susceptibility data do not allow us to distinguish
between the two coupling models. With the first coupling model
there is a range of parameters that gives acceptable fits to the
susceptibility data. Two such sets for [Fe(OEP•)(Br)]2[SbCl6]2
are D ) 35 cm-1, 2JFe-r ) -83 cm-1, and 2JR-R ) -3cm-1
thereby further confirming the π-cation radical assignment. Most
importantly, the fact that the Mo¨ssbauer spectra are not split
by the presence of a weak magnetic field strongly suggests that
iron(III) must be coupled to other spins so that the systems no
longer have half-integer spins. This latter point places definite
restrictions on spin-spin coupling models that can be used to
fit the magnetic susceptibility data. However, analysis of the
data is complicated by the effects of intermediate spin fluctuation
rates at temperatures below 35 K, even in a large applied field.
We were able to get good fits to the data above 35 K using the
ω-tensor model of Kent et al.23 The ω-tensor components are
temperature dependent and should be proportional to the
thermal-averaged spin expectation values for the iron spin.
Knowledge of the ω-tensor components and the spin expectation
values allows one to calculate the components of the magnetic
hyperfine tensor A of the iron.
or D ) 10 cm-1, 2JFe-r ) -83 cm-1, and 2JR-R ) -4 cm-1
.
Since the calculated iron spin expectation values are quite
different for these two parameter sets, they will produce different
components of the hyperfine tensor, as calculated using the
Mo¨ssbauer-determined ω-tensor. For the set with D ) 35 cm-1
,
)
the tensor components are Axx ) Ayy ) -21Tgnân, and Azz
-40Tgnân, while for the set with D ) 10 cm-1, the components
are Axx ) Ayy ) Azz ) -18Tgnân. Since the high-spin ferric A
tensor is expected to be nearly isotropic, we conclude that the
parameter set with the smaller zero-field splitting constant is
the more reasonable one. It is also to be noted that this value
of D is comparable to values of D measured for other high-
spin iron(III) porphyrinates.25 Applying the same Mo¨ssbauer
and susceptibility fit criteria to the data for [Fe(OEP•)(Cl)]2-
[SbCl6]2, we obtain D ) 10 cm-1, 2JFe-r ) -63 cm-1, 2JR-R
The magnetic susceptibility data for [Fe(OEP•)(Br)]2[SbCl6]2
are shown in Figure 2; [Fe(OEP•)(Cl)]2[SbCl6]2 shows a quite
(24) Lang, G.; Boso, B.; Erler, B. S.; Reed, C. A. J. Chem. Phys. 1986,
84, 2998.
(23) Kent, T. A.; Spartalian, K.; Lang, G.; Yonetani, T. Biochem. Biophys.
Acta 1977, 490, 331.
(25) Debrunner, P. G. In Iron Porphyrins, Part III; Lever, A. B. P., Gray,
H. B., Eds.; VCH: New York, 1989; pp 139-227.

+
+
412 Inorganic Chemistry, Vol. 36, No. 3, 1997
Schulz et al.
) -4 cm-1, Axx ) Ayy ) -24Tgnân, and Azz ) -25Tgnân. The
value of D is now smaller than in our previous fit14 and is now
high-spin iron(III) in the [Fe(OEP•)(X)]+ systems. Given the
effective planarity of the porphinato core, it might have been
thought that the iron(III) and radical spins should display
ferromagnetic coupling.7 However, unlike [Cu(TMP•)]+ 4 or
[Fe(TPP•)(OClO3)2],1 the metal atom is substantially out of the
porphyrin plane. The effective local symmetry of five-
coordinate [Fe(OEP•)(X)]+ can be no higher than C4V, rather
than D4h. In this symmetry (and any possible lower symmetry),
more consistent with the value expected for iron(III).
A
comparison of the chloro and bromo complex values shows that
the antiferromagnetic iron-radical coupling is slightly weaker
in the chloro complex, while the intermolecular magnetic
coupling is about the same. Most significantly, the chloro and
bromo complexes have quite similar values that fit both the
measured magnetic susceptibilities and Mo¨ssbauer parameters.
The second model also leads to acceptable calculated fits for
the magnetic susceptibility and the magnetic Mo¨ssbauer data
for both the chloride and bromide complexes. The values
2
the porphyrin a1u orbital and the iron dz orbital are not strictly
orthogonal. There will be net overlap, and thus, antiferromag-
netic coupling is allowed. It is important to note that for this
system, the planarity or nonplanarity of the porphyrin macro-
cycle is unimportant. Lastly, we note that the magnitude of
the coupling between the nominal a1u radical26 and Fe(III) in
[Fe(OEP•)(X)]+ (|2J| ∼ 100 cm-1) is larger than might be
expected.7,25,27 More data on a1u vs a2u comparisons are
necessary, however, before any conclusions can be reached.
obtained for [Fe(OEP•)(Br)]2[SbCl6]2 are 2Jr-r ) -348 cm-1
,
2JFe-r ) -118 cm-1, 2JFe-Fe ) 3 cm-1 and D ) 1 cm-1, Axx
) Ayy ) -24Tgnân, and Azz ) -16Tgnân. The values
determined for [Fe(OEP•)(Cl)]2[SbCl6]2 are 2Jr-r ) -278 cm-1
,
2JFe-r ) -90 cm-1, 2JFe-Fe ) 1cm-1 and D ) 3 cm-1, Axx
)
Ayy ) -24Tgnân, and Azz ) -22Tgnân. This model explicitly
shows the strong coupling between the two porphyrin radicals
in the dimer, in agreement with other derivatives. However,
the calculated value of D appears to be unrealistically low for
an iron(III) complex. When the value of D is increased to
reasonable values in this model, there is a significant disparity
in the fit of the very low temperature susceptibility data (T <
20 K) for all J values.
The first model thus gives better fits to the magnetic data
when the additional requirement of a physically meaningful
value of D is included. This model also yields a qualitatively
correct result for the temperature variation of the magnetic
hyperfine splitting in the Mo¨ssbauer spectra. The solid lines
in Figure 1 are simulations of the [Fe(OEP•)(Br)][SbCl6] spectra
using the electronic parameters that fit the susceptibility data
for the first model, assuming that the fast spin fluctuation
approximation is valid. In these simulations, we varied only
the parameters Axx ()Ayy) and Azz to match the experimental
data. The resulting hyperfine tensor, Axx ) Ayy ) 18.4Tgnân
and Azz ) 18.8Tgnân, is a very reasonable result for high-spin
Fe(III). We note that the simulation best matches the data at
the higher temperatures and conclude that a dynamic model
which allows intermediate spin fluctuation rates would be
necessary to accurately reproduce the details of the Mo¨ssbauer
spectra.
Summary
We have reported the molecular structure of the π-cation
radical complex [Fe(OEP•)(Cl)]ClO4 and complete magnetic
susceptibility and Mo¨ssbauer studies on two related species
[Fe(OEP•)(Cl)][SbCl6] and [Fe(OEP•)(Br)][SbCl6]. The struc-
tural analysis shows that [Fe(OEP•)(Cl)]ClO4 exists as a dimer
in the crystalline state, and this was also assumed in the analysis
of the magnetic properties of the related hexachloroantimonate
salts. All data are consistent with the presence of high-spin
iron(III) centers strongly coupled to the porphyrin radical with
additional interaction between the pair of rings in the solid-
state dimers.
Acknowledgment. We acknowledge, with thanks, the sup-
port of the National Institutes of Health under Grants GM-38401
(W.R.S.), GM-23851 (C.A.R.), GM-16406 (P.G.D.), and GM-
48513 (C.E.S.).
Supporting Information Available: Tables of crystal data and
intensity collection parameters, fractional coordinates, atomic displace-
ment parameters, and fixed hydrogen coordinates for [Fe(OEP)Cl]-
ClO4‚CH2Cl2 (5 pages) and an X-ray crystallographic file, in CIF format,
is available. Access and/or ordering information is given on any current
masthead page.
At first glance, it might seem surprising that the second
model, with its three coupling constants, should not fit the
susceptibility data better than the first model, with its two
coupling constants. However, the total spin formulation of the
first model implicitly includes contributions from all possible
couplings between the four different spin centers of the dimer.
It is to be noted that both coupling models lead to a picture
of an antiferromagnetically coupled porphyrin radical and a
IC961053D
(26) An a1u radical assignment is made by comparison to other OEP radical
cations; the ground state is primarily the result of the peripheral
substiutents.
(27) (a) Fujii, H.; Yoshimura, T.; Kamada, H. Inorg. Chem. 1996, 35, 2373.
(b) Jayaraj, K.; Terner, J.; Gold, A.; Roberts, D. A.; Austin, R. N.;
Mandon, D.; Weiss, R.; Bill, E.; Mu¨ther, M.; Trautwein, A. X. Inorg.
Chem. 1996, 35, 1632 and references therein.