Structural, magnetic, and dynamic characterization of the (dxz,dyz)4(dxy)1 ground-state low-spin iron(III) tetraphenylporphyrinate complex [(p-TTP)Fe(2,6-XylylNC)2]CF3SO3

Article abstract of DOI:10.1021/ja994190t

Contribution from the Laboratoire de Chimie Organometallique et Biologique,. The synthesis and characterization of the trifluoromethanesulfonate salt of bis(2,6-xylyl isocyanide)-tetrakis(p-tolyl)porphyrinatoiron(III), [(p-TTP)Fe(2,6-xylylNC)₂]CF₃SO₃ (1), is reported. The crystal structure shows that the porphyrinate ring is strongly ruffled. The equatorial Fe-N bond distances average to 1.961(7) A for 1, which is quite short for low-spin iron(III) porphyrinate derivatives. Two additional complexes, having TPP (2) and m-TTP (3) as the porphyrinates, were also synthesized and studied by NMR, EPR, and Moessbauer spectroscopy. All physical properties are consistent with a low-spin iron(III) porphyrinate with the less-common ground-state configuration (d_xz,d_yz)⁴(d_xy)¹. The ¹H NMR chemical shift of the pyrrole protons at 297 K is +10.7 ppm for 1. The EPR spectrum of 1 in solution is axial, with g_⊥ = 2.15 and g_∥ = 1.94, Σg² = 13.0, while in the solid state g_⊥ = 2.2 and g_∥ = 1.94, Σg² = 13.4. The Moessbauer spectrum of 1 at 190 K has an isomer shift of 0.14 mm/s and quadrupole splitting of 1.81 mm/s. Magnetic Moessbauer spectra analyzed in the intermediate spin-spin relaxation regime by the dynamic line-shape formalism of Blume and Clauser confirm this electron configuration and yield large negative quadrupole splittings, ΔE_Q = -1.8 to -2.0 mm/s for the three complexes. To our knowledge, this is the first case in which the Moessbauer spectra of low-spin ferriheme systems have been analyzed in terms of the effect of intermediate rates of spin fluctuations on the appearance of the spectra. Analysis of the temperature dependence of the quadrupole splitting, ΔE_Q, for 2 yielded a different estimate of the energy separation between the (d_xz,d_yz)⁴(d_xy)¹ ground state and an excited state than did the temperature dependence of the NMR isotropic shifts. It is postulated that the excited state is actually the planar transition state between the two ruffled conformations of the porphyrinate that are related by inversion . To explain the temperature dependence of both NMR isotropic shifts and Moessbauer quadupole splittings, the planar transition state must have the (d_xy)²(d_xz,d_yz)³ electron configuration. The energy barrier appears to be smaller in homogeneous solution than in the solid state and is considerably lower than that predicted for the (d_xz,d_yz)³(d_xy)² excited electronic state of the ruffled conformation on the basis of the EPR g values.

Full text of DOI:10.1021/ja994190t

4366
J. Am. Chem. Soc. 2000, 122, 4366-4377
Structural, Magnetic, and Dynamic Characterization of the
(dxz,dyz)4(dxy)1 Ground-State Low-Spin Iron(III)
Tetraphenylporphyrinate Complex [(p-TTP)Fe(2,6-XylylNC)2]CF3SO3
Ge´rard Simonneaux,*,† Volker Schu1nemann,‡ Christophe Morice,† Laurence Carel,†
Lo1ıc Toupet,§ Heiner Winkler,‡ Alfred X. Trautwein,*,‡ and F. Ann Walker*,
Contribution from the Laboratoire de Chimie Organome´tallique et Biologique, UMR CNRS 6509,
UniVersite´ de Rennes 1, Campus de Beaulieu, 35042 Rennes Cedex, France, Institut fu¨r Physik,
Medizinische UniVersita¨t zu Lu¨beck, 160 Ratzeburger Allee, D-23538 Lu¨beck, Germany, Groupe de
Physique Cristalline, UA CNRS 040804, UniVersite´ de Rennes 1, Campus de Beaulieu, 35042 Rennes
Cedex, France, and Department of Chemistry, UniVersity of Arizona, Tucson, Arizona 85721-0041
ReceiVed NoVember 30, 1999
Abstract: The synthesis and characterization of the trifluoromethanesulfonate salt of bis(2,6-xylyl isocyanide)-
tetrakis(p-tolyl)porphyrinatoiron(III), [(p-TTP)Fe(2,6-xylylNC)2]CF3SO3 (1), is reported. The crystal structure
shows that the porphyrinate ring is strongly ruffled. The equatorial Fe-N bond distances average to 1.961(7)
Å for 1, which is quite short for low-spin iron(III) porphyrinate derivatives. Two additional complexes, having
TPP (2) and m-TTP (3) as the porphyrinates, were also synthesized and studied by NMR, EPR, and Mo¨ssbauer
spectroscopy. All physical properties are consistent with a low-spin iron(III) porphyrinate with the less-common
1
ground-state configuration (dxz,dyz)4(dxy)1. The H NMR chemical shift of the pyrrole protons at 297 K is
+10.7 ppm for 1. The EPR spectrum of 1 in solution is axial, with g ) 2.15 and g ) 1.94, ∑g2 ) 13.0,
||
while in the solid state g ) 2.2 and g ) 1.94, ∑g2 ) 13.4. The Mo¨ssbauer spectrum of 1 at 190 K has an
||
isomer shift of 0.14 mm/s and quadrupole splitting of 1.81 mm/s. Magnetic Mo¨ssbauer spectra analyzed in the
intermediate spin-spin relaxation regime by the dynamic line-shape formalism of Blume and Clauser confirm
this electron configuration and yield large negative quadrupole splittings, ∆EQ ) -1.8 to -2.0 mm/s for the
three complexes. To our knowledge, this is the first case in which the Mo¨ssbauer spectra of low-spin ferriheme
systems have been analyzed in terms of the effect of intermediate rates of spin fluctuations on the appearance
of the spectra. Analysis of the temperature dependence of the quadrupole splitting, ∆EQ, for 2 yielded a different
estimate of the energy separation between the (dxz,dyz)4(dxy)1 ground state and an excited state than did the
temperature dependence of the NMR isotropic shifts. It is postulated that the excited state is actually the
planar transition state between the two ruffled conformations of the porphyrinate that are related by “inversion”.
To explain the temperature dependence of both NMR isotropic shifts and Mo¨ssbauer quadupole splittings, the
planar transition state must have the (dxy)2(dxz,dyz)3 electron configuration. The energy barrier appears to be
smaller in homogeneous solution than in the solid state and is considerably lower than that predicted for the
(dxz,dyz)3(dxy)2 excited electronic state of the ruffled conformation on the basis of the EPR g values.
Introduction
ing the two b cytochromes of mitochondrial “complex III” (also
known as ubiquinone-cytochrome c oxidoreductase) and chlo-
roplast cytochrome b6. EPR spectroscopy,6 and in some cases
Mo¨ssbauer spectroscopy,7 of well-defined low-spin heme model
compounds with high basicity pyridines, imidazoles, or cyanides
as the axial ligands has shown that their “large gmax” EPR signal
is indicative of the (dxy)2(dxz,dyz)3 electronic ground state and
near degeneracy of dxz and dyz. For complexes with planar axial
ligands this correlates with perpendicular alignment of these
ligands,7-10 while rhombic EPR signals are correlated with
parallel alignment of planar axial ligands.
Investigations of the structural and spectroscopic properties
of Fe(III) porphyrinates have been useful in understanding both
structural and electronic properties of the bis-histidine-
coordinated cytochromes b51 and c3,2,3 as well as the membrane-
bound cytochromes b with bis(histidine) coordination4,5 includ-
† Laboratoire de Chimie Organome´tallique et Biologique, Universite´ de
Rennes 1.
‡ Medizinische Universita¨t zu Lu¨beck.
§ Groupe de Physique Cristalline, UA CNRS 040804, Universite´ de
Rennes 1.
Tetraphenylporphinatoiron(III) bis(4-cyanopyridine)perchlo-
rate, [(TPP)Fe(4-CNPy)2]ClO4,11 for which the axial ligands
University of Arizona.
(1) Mathews, F. S.; Czerwinski, E. W.; Argos. P. In The Porphyrins;
Dolphin, D., Ed.; Academic Press: New York, 1979; Vol. VIII, p 108.
(2) Pierrot, M.; Haser, R.; Frey, M.; Payan, F.; Astier, J.-P. J. Biol. Chem.
1982, 257, 14341.
(5) Babcock, G. T.; Widger, W. R.; Cramer, W. A.; Oertling, W. A.;
Metz, J. Biochemistry 1985, 24, 3638.
(3) Higuchi, Y.; Kusunocki, M.; Matsuura, Y.; Yasuoka, N.; Kakudo
M. J. Mol. Biol. 1984, 172, 109.
(4) Widger, W. R.; Cramer, W. A.; Herrmann, R. G.; Trebst, A. Proc.
Natl. Acad. Sci. U.S.A. 1984, 81, 674.
(6) Innis, D.; Soltis, S. M.; Strouse, C. E. J. Am. Chem. Soc. 1988, 110,
5644.
(7) Walker, F. A.; Huynh, B. H.; Scheidt, W. R.; Osvath, S. R. J. Am.
Chem. Soc. 1986, 108, 5288.
10.1021/ja994190t CCC: $19.00 © 2000 American Chemical Society
Published on Web 04/25/2000

Ground-State Low-Spin [(p-TTP)Fe(2,6-xylylNC)2]CF3SO3
J. Am. Chem. Soc., Vol. 122, No. 18, 2000 4367
have strong π-acceptor properties, has been shown to have an
electronic stabilization of the (dxz,dyz)4(dxy)1 electronic ground
state. The corresponding tetramesitylporphyrinate, [(TMP)Fe-
(4-CNPy)2]ClO4,12 also reveals properties such as low-intensity
visible13 and near-IR14 MCD bands, as well as EPR g values
that resemble those of iron chlorins15 and proteins that contain
them.15,16 A Mo¨ssbauer, EPR, and NMR study of the tert-butyl
isocyanide complexes, [(TPP)Fe(t-BuNC)2]ClO4 and [(OEP)-
Fe(t-BuNC)2]ClO4,17 has shown that their g values (g ) 2.20-
NMR spectrum (0 to +10 ppm).25-29 It is also often possible
to determine the energy separation of the two electronic states
from the curvature of the Curie plot.30 Some time ago, La Mar
and co-workers31,32 recognized that the hyperfine shifts for both
the pyridine ligands and the tetraphenylporphyrin of a series of
[(TPP)Fe(R-Py)2]+ are very sensitive to the basicity of the axial
ligands. More recently, Safo et al.33 reported similar trends in
hyperfine shifts for tetramesitylporphinatoiron(III) bis(pyridine)
complexes as a function of ligand basicity, and Simonneaux
and co-workers reported that when two molecules of tert-butyl
2.28, g ) 1.94-1.83) are consistent with the purest (dxz,dyz)4-
||
1
(dxy)1 ground-state systems observed thus far, with the most
complete quenching of orbital momentum (∑g2 as small as 13.5).
Thus, it is now apparent that iron(III) tetraphenylporphyrinates,
when bound to strong π-acceptor ligands, are good models
of the “green hemes” that include (i) the iron chlorin that is
the prosthetic group of hydroperoxidase II18 and cytochrome
bd19 from Escherichia coli, (ii) the iron dioxoisobacterio-
chlorin of heme d1, the prosthetic group of the dissimila-
tory nitrite reductase of denitrifying bacteria,20 and (iii) the
siroheme prosthetic group of bacterial sulfite and nitrite reduc-
tases.21
isocyanide are bound to ferric tetraphenylporphyrin, the H
NMR spectrum is indicative of a low-spin complex,26,34 but with
isotropic shifts very different than those of any low-spin Fe-
(III) porphyrinate reported previous to that time. The hyperfine
shifts were found to reflect the very small magnetic anisotropy
of the iron, and an unusual pattern of unpaired spin density when
nitrogen donor axial ligands are exchanged for isocyanide
ligands, with a completely reversed pattern of delocalization.26
Subsequently, it was recognized that the unusual 1H NMR
behavior results from the existence of the (dxz,dyz)4(dxy)1 ground
state.11 Similar NMR results were later obtained for octaeth-
ylporphinatoiron(III) bis(tert-butyl isocyanide), where it could
readily be seen that there is very large π spin density at the
meso positions of the porphyrinate ring.17 This, combined with
the extremely small amount of spin density present at the pyrrole
positions of both the TPP26 and OEP17 analogues, indicates that
there is significant spin delocalization to the a2u(π) orbital, which
is possible only if the porphyrinate ring is highly ruffled, causing
twisting of the nitrogen pz orbitals from the normal to the
porphyrin ring.11,17,25
The relative energies of the dxy, dxz and dyz orbitals can be
calculated from the EPR g values on the basis of Griffith’s
theory22 by adopting the methods and “proper axis system” of
Taylor.23 For the complexes showing the “large gmax” EPR
signal, Mo¨ssbauer spectroscopy has been used to estimate the
components of the g-tensor which are not observable by EPR.7,9
Mo¨ssbauer spectroscopy in high magnetic fields and at low
temperatures also allows determination of the sign of the
quadrupole splitting, which has shown to be positive for the
(dxy)2(dxz,dyz)3 electronic ground state7,9 and negative for the
(dxz,dyz)4(dxy)1 electronic ground state.17
Although isocyanides are not natural substrates in biological
systems, their interactions with heme proteins have been studied
extensively. Thus, the reaction between protoheme and methyl
isocyanide was first reported by Warburg et al. in 1929.35 The
purpose was to study the effects of methyl isocyanide on the
photochemical dissociation of CO-hemoglobin. The formation
of the isocyanide-hemoglobin complex was later described
more thoroughly by Keilin,36 St. George and Pauling,37 and
Antonini and Brunori.38 Indeed, the idea that interactions with
protein residues control the binding of ligands to hemoglobin
was first recognized by St. George and Pauling37 using isocya-
nide binding with increasing steric bulk of the alkyl group. More
recent approaches have been developed by Olson, Gibson, and
NMR spectroscopy is useful for differentiating the two
electronic ground states at ambient temperatures, where the
(dxy)2(dxz,dyz)3 ground state exhibits pyrrole-H resonances at
negative chemical shifts (-12 to -30 ppm, depending on axial
ligands and temperature),24 while the (dxz,dyz)4(dxy)1 ground state
exhibits pyrrole-H resonances in the diamagnetic region of the
(8) Hatano, K.; Safo, M. K.; Walker, F. A.; Scheidt, W. R. Inorg. Chem.
1991, 30, 1643.
(9) Safo, M. K.; Gupta, G. P.; Walker, F. A.; Scheidt, W. R. J. Am.
Chem. Soc. 1991, 113, 5497.
(10) Munro, O. Q.; Serth-Guzzo, J. A.; Turowska-Tyrk, I.; Mohanrao,
K.; Walker, F. A.; Debrunner, P. G.; Scheidt, W. R. J. Am. Chem. Soc.
1999, 121, 11144.
(11) 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.
(25) Walker, F. A.; Simonis, U. In Biological Magnetic Resonance;
Berliner, L. E., Reuben, J., Ed.; Plenum: NY, 1993; Vol. 12, p 132 and
references therein.
(26) Simonneaux, G.; Hindre´, F.; Le Plouzennec, M. Inorg. Chem. 1989,
28, 823.
(12) Cheesman, M. R.; Walker, F. A. J. Am. Chem. Soc. 1996, 118, 7373.
(13) Bracete, A. M.; Kadkhodayan, S.; Sono, M.; Huff, A. M.; Zhuang,
C.; Cooper, D. K.; Smith, K. M.; Chang; C. K.; Dawson, J. H. Inorg. Chem.
1994, 33, 5042.
(14) Peng, Q.; Timkovich, R.; Loewen, P. C.; Peterson, J. FEBS Lett.
1992, 309, 157.
(15) Coulter, E. D.; Sono, M.; Chang, C. K.; Lopez, O.; Dawson, J. H.
Inorg. Chim. Acta 1995, 240, 603.
(16) Sutherland, J.; Greenwood, C.; Peterson, J.; Thomson, A. J. Biochem.
J. 1986, 233, 893.
(27) Nakamura, M.; Ikeue, T.; Fujii, H.; Yoshimura, T. J. Am. Chem.
Soc. 1997, 119, 6284.
(28) Nakamura, M.; Ikeue, T.; Fujii, H.; Yoshimura, T.; Tajima, K. Inorg.
Chem. 1998, 37, 7, 2405.
(29) Walker, F. A. In The Porphyrin Handbook; Kadish, K. M., Smith,
K. M., Guilard, R., Eds.; Academic Press: San Diego, CA, 2000; Vol. 5,
Chapter 36, pp 81-183.
(30) Shokhirev, N. V.; Walker, F. A. J. Phys. Chem. 1995, 99, 17795.
(31) La Mar, G. N.; Bold, T. J.; Satterlee, J. D. Biochim. Biophys. Acta
1977, 498, 189.
(17) Walker, F. A.; Nasri, H.; Turowska-Tyrk, I.; Mohanrao, K.; Watson,
C. T., Shokhirev, N., V.; Debrunner, P. G.; Scheidt, W. R. J. Am. Chem.
Soc. 1996, 118, 12109.
(18) Chiu, J. T.; Loewen, P. C.; Switala, J.; Gennis, R. B.; Timkovich,
R. J. Am. Chem. Soc. 1989, 111, 7046.
(19) Lorence, R. M.; Gennis, R. B. J. Biol. Chem. 1989, 264, 7135.
(20) Chang, C. K.; Wu, W. J. Biol. Chem. 1986, 261, 8593.
(21) Crane, B. R.; Siegel, L. M.; Getzoff, E. D. Science 1995, 270, 59.
(22) Griffith, J. S. Proc. R. Soc. London, A 1956, 235, 23.
(23) Taylor, C. P. S. Biochim. Biophys. Acta 1977, 491, 137.
(24) La Mar, G. N.; Walker, F. A. In The Porphyrins; Dolphin, D., Ed.;
Academic Press: New York, 1979; pp 61-157.
(32) La Mar, G. N.; del Gaudio, J.; Frye, J. S. Biochim. Biophys. Acta
1977, 498, 422.
(33) Safo, M. K.; Gupta, G. P.; Watson, C. T.; Simonis, U.; Walker, F.
A.; Scheidt, W. R. J. Am. Chem. Soc. 1992, 114, 7066.
(34) Geze, C.; Legrand, N.; Bondon, A.; Simonneaux, G. Inorg. Chim.
Acta 1992, 195, 73.
(35) Warburg, O.; Negelein, E.; Christian, W. Biochem. Z. 1929, 214,
26.
(36) Keilin, J. Biochem. J. 1949, 45, 440.
(37) St. George, R.; Pauling, L. Science 1951, 114, 629.
(38) Antonini, E.; Brunori, M. In Hemoglobin and Myoglobin in their
Reactions with Ligands; North-Holland Pub. Co.: Amsterdam, 1971; p 34.

4368 J. Am. Chem. Soc., Vol. 122, No. 18, 2000
Simonneaux et al.
others.39-52 Elegant studies of the structure-reactivity relation-
ships in myoglobin have been developed by assessing the effects
of substitution in the heme cavity using site-directed mutagen-
esis.45 In these studies, various alkyl isocyanides were tested,
along with more classical ligands such as dioxygen and carbon
monoxide. The large size of the alkyl isocyanide ligand was
used to amplify the effects of the mutation.
Extensive thermodynamic and kinetic studies of alkyl iso-
cyanide complexation to hemoglobin, myoglobin,38-45 and other
heme proteins53-55 have been reported, but many questions still
remain. First, despite the accuracy of these methods, conflicting
results persist with regard to the binding of aryl isocyanides to
myoglobin.56,57 Thus, the intrinsic electronic effect on ligand
binding associated with differences between alkyl and aryl
isocyanides is still under debate. Second, the reason complex-
ation of isocyanides with the ferric state is possible with
cytochromes P450, whereas no ligation is observed with
myoglobin and hemoglobin, is still not clear. Consequently,
studies of the reactions of isocyanides with simple synthetic
iron(III) porphyrinates,58-62 which have been more limited,
deserve special attention.
interactions. The magnetic hyperfine interaction couples the
nuclear spin I to the electron spin S, and transitions among the
electronic eigenstates ms ) +1/2 and ms ) -1/2 lead to
fluctuations in the nuclear Hamiltonian. In most cases, magnetic
Mo¨ssbauer spectra are analyzed either in the slow or the fast
relaxation limit. In the limit of slow spin relaxation (fluctuation),
typically at temperatures around 4.2 K, the Mo¨ssbauer spectra
may be calculated by assuming the electronic Hamiltonian Hˆ e
to be stationary and the two Zeeman levels to be populated
according to the Boltzmann factor. At higher temperatures the
fast fluctuation limit applies, and a thermal average of the spin
expectation value
S can be substituted into the nuclear
Hamiltonian. This is applicable for the case of spin lattice
relaxation typically at temperatures above 150 K, or for fast
spin-spin relaxation, the latter being often observed in powder
samples. For the case of intermediate relaxation, the resulting
changes in the energy spectrum of the Mo¨ssbauer transitions
are observable if the Larmor frequency of the iron nucleus ωL
is comparable to the spin transition rate w but large compared
to the nuclear decay rate, 1/τ ) 7 × 106 s -1 63
. In this study we
show that the complexes tetra(p-tolyl)porphyrinatoiron(III) bis-
(2,6-xylyl isocyanide) trifluoromethylsulfonate (1), tetraphe-
nylporphyrinatoiron(III) bis(2,6-xylyl isocyanide) trifluorome-
thylsulfonate (2), and tetra(m-tolyl)porphyrinatoiron(III) bis(2,6-
xylyl isocyanide) trifluoromethylsulfonate (3) possess the
(dxz,dyz)4(dxy)1 electronic ground state. We also demonstrate that
the Mo¨ssbauer spectra of these three complexes, even at
magnetic fields up to 5.35 T at 4.2 K, are influenced by
electronic spin-spin relaxation in the intermediate relaxation
regime. We have analyzed the Mo¨ssbauer spectra by using the
dynamic line shape formalism of Blume and Clauser.64 This
method has been applied previously to spin fluctuation rates of
high-spin ferrous rubredoxin63 and to catalytic intermediates of
horseradish peroxidase,65 cytochrome P450CAM, protocatechuate
dioxygenase, and horseradish peroxidase compounds I and II.66
Magnetic Mo¨ssbauer spectra of low-spin ferric hemes have been
analyzed in terms of either fast or slow relaxation limits in the
past.7,9,17 To our knowledge, this work is the first which studies
the influence of electron spin fluctuations on the Mo¨ssbauer
spectra of low-spin ferric heme compounds, either models or
proteins. The reasons these particular complexes should be in
the intermediate spin fluctuation regime and the likelihood of
observing this phenomenon in a heme protein are discussed.
We have found in this study that Mo¨ssbauer spectroscopy is
extremely informative concerning the effects of the electronic
structure of the low-spin Fe(III) center on the bonding interac-
tions with the ligands, as well as the nature of the ground and
excited states and an estimation of the energy separation between
them. This spectroscopic technique measures the nuclear energy
levels as they are perturbed by external fields and hyperfine
(39) For fast kinetics studies of myoglobin, see for example refs 40-
52.
(40) Gibson, Q. H.; Olson, J. S.; McKinnie, R. E.; Rohlfs, R. J. J. Biol.
Chem. 1986, 261, 10228.
(41) Olson, J. S.; Rohlfs, R. J.; Gibson, Q. H. J. Biol. Chem. 1987, 262,
12930.
(42) Belleli, A.; Blackmore, R. S.; Gibson, Q. H. J. Biol. Chem. 1990,
265, 13595.
(43) Jongeward, K. A.; Magde, D.; Taube, D. J.; Marsters, J. C.; Traylor,
T. G.; Sharma, V. S. J. Am. Chem. Soc. 1988, 110, 380.
(44) Wood, M. A.; Dickinson, K.; Willey, G. R.; Dodd, G. H. Biochem.
J. 1987, 247, 675.
(45) Olson, J. S.; Mathews, A. J.; Rohlfs, R. J.; Springer, B. A.; Egeberg,
K. D.; Sligar, S. G.; Tame, J.; Renaud, J. P.; Nagai, K. Nature 1988, 336,
265.
(46) Mathews, A. J.; Rohlfs, R. J.; Olson, J. S.; Tame, J.; Renaud, J. P.;
Nagai, K. J. Biol. Chem. 1989, 264, 16573.
(47) Rohlfs, R. J.; Mathews, A. J.; Carver, T. E.; Olson, J. S.; Springer,
B. A.; Egeberg, K. D.; Sligar, S. G. J. Biol. Chem. 1990, 265, 3168.
(48) Egeberg, K. D.; Springer, B. A.; Sligar, S. G.; Carver, T. E.; Rohlfs,
R. J.; Olson, J. S. J. Biol. Chem. 1990, 265, 11788.
(49) Carver, T. E.; Olson, J. S.; Smerdon, S. J.; Krzywda, S.; Wilkinson,
A. J.; Gibson, Q. H.; Blackmore, R. S.; Ropp, J. D.; Sliger, S. G.
Biochemistry 1991, 30, 4697.
(50) Smerdon, S. J.; Dodson, G. G.; Wilkinson, A. J.; Gibson, Q. H.;
Blackmore, R. S.; Carver, T. E.; Olson, J. S. Biochemistry 1991, 30, 6252.
(51) Carver, T. E.; Brantley, R. E.; Singleton, E. W.; Arduini, R. M.;
Quillin, M. L.; Phillips, G. N.; Olson, J. S. J. Biol. Chem. 1992, 267, 14443.
(52) Smerdon, S. J.; Krzywda, S.; Wilkinson, A. J.; Brantley, R. E.;
Carver, T. E.; Hargrove, M. S. Olson, J. S. Biochemistry 1993, 32, 5132.
(53) White, R. E.; Coon, M. J. J. Biol. Chem. 1982, 257, 3073.
(54) Sono, M.; Dawson, J. H.; Hall, K.; Hager, L. P. Biochemistry 1986,
25, 347.
Experimental Section
General Information. As a precaution against the formation of the
µ-oxo dimer [Fe(TPP)]2O,67 all reactions were carried out in dried
solvents in Schlenk tubes under an argon atmosphere. Solvents were
1
distilled from appropriate drying agents and stored under argon. H
NMR spectra were recorded on a Bruker AC 300P spectrometer in
CDCl3 at 300 MHz. Tetramethylsilane was used as internal reference.
Temperatures are given to within 1 K. Visible spectra were measured
on a Uvikon 941 spectrometer in CH2Cl2. Infrared spectra were recorded
on a Nicolet 205 FTIR instrument in CH2Cl2 solution. Elemental
analyses were performed by the Service Central of Analyses (CNRS)
at Vernaison, France. EPR spectra were recorded on a conventional
X-band spectrometer (Bruker 200D SRC) equipped with a He flow
cryostat (ESR 910, Oxford Instruments). Spectra were taken at 10 K
(55) Imai, Y.; Okamoto, N.; Nakahara, K.; Shoun, H. Biochim. Biophys.
Acta 1997, 1337, 66.
(56) Caughey, W. S.; Barlow, C. H.; O’Keeffe, D. H.; O’Toole, M. C.
Ann. N.Y. Acad. Sci. 1973, 206, 296-309.
(57) Patel, M. J.; Kassner, R. J. Biochem. J. 1989, 262, 959.
(58) For ferrous porphyrinate isocyanide derivatives, see for example
refs 59-61.
(63) Winkler, H.; Schulz, C.; Debrunner, P. G. Phys. Lett. 1979, 69A,
360.
(64) Clauser, M. J.; Blume, M. Phys. ReV. B 1971, 3, 583.
(65) Schulz, C. E.; Rutter, R.; Sage, J.P.; Debrunner, P. G.; Hager, L. P.
Biochemistry 1984, 23, 4743.
(66) Schulz, C. E.; Nyman, P.; Debrunner, P. G. J. Chem. Phys. 1987,
87. 5076.
(59) Caughey, W. S.; Barlow, C. H.; O’Keeffe, D. H.; O’Toole, M. C.
Ann. N.Y. Acad. Sci. 1973, 206, 296.
(60) Jameson, G. B.; Ibers, J. A. Inorg. Chem. 1979, 18, 1200.
(61) Le Plouzennec, M.; Bondon, A.; Sodano, P.; Simonneaux, G. Inorg.
Chem. 1986, 25, 1254.
(62) For ferric porphyrin isocyanide derivatives, see for example refs
17, 26, and 34.
(67) Fleischer, E. B.; Palmer, J. M.; Srivastava, T. S.; Chatterjee, A. J.
Am. Chem. Soc. 1971, 93, 3162.

Ground-State Low-Spin [(p-TTP)Fe(2,6-xylylNC)2]CF3SO3
J. Am. Chem. Soc., Vol. 122, No. 18, 2000 4369
Table 1. Crystallographic Data for
and a microwave power of 20 µW, a modulation frequency of 100
kHz, and a modulation amplitude of 5 G. Mo¨ssbauer spectra were
recorded using a conventional spectrometer in the constant acceleration
mode. Isomer shifts are given relative to R-Fe at room temperature.
The spectra obtained at 20 mT were measured in a He bath cryostat
(Oxford MD 306) equipped with a pair of permanent magnets. Spectra
obtained at 1 T were measured in an in-house-made bath cryostat
equipped with a superconducting magnet. For the high-field spectra
(5.35 T), a cryostat also equipped with a superconducting magnet
(Oxford Instruments) was used. Magnetically split spectra of paramag-
netic samples were simulated in the spin-Hamiltonian approximation
described below.
Reagents. The following iron porphyrins68 were prepared by
literature methods: [(TPP)FeCF3SO3],69 [(p-TTP)FeCF3SO3],70 and [(m-
TTP)FeCF3SO3].70 The 2,6-xylyl isocyanide (2,6-xylylNC) was obtained
from Fluka. Aromatic isocyanides are probably not very strong
coordinating ligands for the ferric state due to a decrease of basicity of
the ligand in comparison to that of alkyl isocyanides,71,72 thus making
iron(III) complexation more difficult. Using triflate, a weak axial
ligand,69,70 as an intermediate permits solving this problem.
Syntheses. [(p-TTP)Fe(2,6-xylylNC)2]CF3SO3 (1). To a solution
of 0.2 g (0.23 mmol) of [(p-TTP)FeCF3SO3] in 10 mL of dichlo-
romethane was added 8 equiv 2,6-xylylNC in 2 mL by a syringe while
the solution was stirred at room temperature. Then 40 mL of hexane
and 1 mL of toluene were added, and the solution was set aside for 2
days for crystallization at 25 °C. Purple crystals of 1 were collected
by filtration and washed with hexane. The yield was 0.22 g (85%).
UV-vis (toluene): λmax/nm 417 (ꢀ 118 mM-1 cm-1), 505 (ꢀ 12), 569
(ꢀ 8).
[(TPP)Fe(2,6-xylylNC)2]CF3SO3 (2). To a solution of 0.20 g (0.23
mmol) of [(TPP)Fe]CF3SO3 in 16 mL of dichloromethane was added
8 equiv of the ligand by syringe with stirring at room temperature.
Then 60 mL of pentane was added, and the solution was set aside for
2 days for crystallization at 0 °C. Purple crystals of 2 were collected
by filtration and washed with hexane. The yield was 0.20 g (72%).
Anal. Calcd for C63H46N6O3F3SFe: C, 70.06; H, 4.26; N, 7.78. Found:
C, 70.08; H, 4.56; N, 7.65. UV-vis (toluene): λmax/nm 418 (ꢀ 120
mM-1 cm-1), 505 (ꢀ 11), 569 (ꢀ 8).
For the preparation of the m-tolyl derivative, [(m-TTP)Fe(2,6-
xylylNC)2]CF3SO3 (3), the same procedure was used with the corre-
sponding triflate analogue in dichloromethane solvent and the ligand.
However, for the m-tolyl derivative, a partial decomposition was
observed; the solutions were reduced in volume to ensure precipitation,
due to the high solubility of the m-tolyl derivative in dichloromethane.
Data for [(m-TTP)Fe(2,6-xylylNC)2]CF3SO3 follow. Yield: 78%. UV-
vis (toluene): λmax/nm 417 (ꢀ 120 mM-1 cm-1), 505 (ꢀ 12), 569 (ꢀ 7).
X-ray Structure Determination. The X-ray study was carried out
on an Enraf-Nonius CAD-4 diffractometer using graphite-monochro-
matized Mo KR radiation. The cell parameters were obtained by fitting
a set of 25 high-θ reflections. Crystallographic data are collected in
Table 1. Crystals of the compound were obtained as reported above.
Atomic scattering factors were taken from the International Tables for
X-ray Crystallography. All calculations were performed on a Silicon
Graphics Indy computer with the Molen package73 for refinement and
ORTEP calculations.
[(p-TTP)Fe(2,6-xylylNC)2]CF3SO3 (1)
compound
empirical formula
1‚1/2(CH2Cl2)
C67H54FeF3N6O3S, 1/2(CH2Cl2)
FW
1259.19
monoclinic
P21/c
11.330(2)
30.343(7)
18.096(9)
102.34(3)
6077(5)
4
1.376
4.295
294
0.080
crystal system
space group
a, Å
b, Å
c, Å
â, deg
V, Å3
Z
P
calc, g‚cm-3
µ, cm-1
T, K)
Rw
final R
0.075
polarization corrections the structure was solved by direct methods,74
which revealed the non-hydrogen atoms of the cation and the anion.
After scale factor calculations and difference Fourier, several peaks
were found and assigned to the nonstoichiometric dichloromethane and
toluene. After anisotropic refinement, all the hydrogen atoms of the
cation moiety were found with a Fourier difference map. The whole
structure was refined by the full-matrix least-squares techniques (use
of F magnitude; x, y, z, âi,j for Fe, N, and C atoms; x, y, z, and B for
triflate anion; and x, y, z for hydrogen atoms; 780 variables and 3613
observations; w ) 1/σ(Fo)2 ) [σ2(I) + (0.04Fo )2]-1/2) with the resulting
2
R ) 0.075, Rw ) 0.080, and Sw ) 2.95 (residual ∆F < 0.45 e Å-3).
Mo1ssbauer Theory and Spectral Analysis. (a) Spin-Hamiltonian
Formalism. The Zeeman interaction of a spin BS with an applied field
BB, with gc representing the electronic gc-tensor, is described by the
Hamiltonian
Hˆ eff ) µBBSgcBB
(1)
where µB denotes the Bohr magneton.
Magnetic Mo¨ssbauer spectra were simulated using eq 1 to derive
the spin-expectation S together with the nuclear Hamiltonian,75,76
eQVzz
Hˆ N
)
[3ˆI2z - I(I + 1) + η(ˆIx2 - ˆIy2)] - gNµNBIBB + BS ACBI
4I(2I - 1)
(2)
Here, I denotes the nuclear spin-quantum number of the ground and
excited nuclear states, Q the nuclear quadrupole moment of the excited
nuclear state, Vzz the main component of the electric field gradient (efg)
tensor, η ) (Vxx - Vyy)/Vzz the asymmetry parameter of the efg, and gN
the nuclear g-factor of the ground and excited statez. AC is the hyperfine
coupling tensor.
(b) Analysis of Mo1ssbauer Spectra in the Presence of Electronic
Relaxation. If spin-lattice relaxation is the dominant process, Mo¨ss-
bauer spectra are analyzed in general for two limiting cases. In the
limit of slow spin fluctuation, typically at temperatures around 4.2 K,
the Mo¨ssbauer spectra may be calculated by assuming Hˆ e to be
stationary and the two Zeeman levels to be populated according to the
Boltzmann factor. At higher temperatures the fast fluctuation limit
applies and a thermal average of the spin expectation value BS can be
substituted in eq 2. However, if Mo¨ssbauer spectra are influenced by
intermediate spin fluctuation rates caused by temperature-independent
spin-spin relaxation, neither fast nor slow relaxation limits may apply.
Such samples then exhibit Mo¨ssbauer spectra in the intermediate
relaxation regime where the nuclear Larmor frequency is comparable
to the spin transition rate over the whole experimentally accessible
Single-Crystal Structure Determination on [(p-TTP)Fe(2,6-
xylylNC)2]CF3SO3 (1). The data collection (2θmax ) 54°, scan ω/2θ
) 1, tmax ) 60 s, h,k,l range h ) 0.14, k ) 0.38, l ) -23.23, intensity
controls without appreciable decay (0.15%)) gave 14 105 reflections
from which 3613 reflections satisfied I > 3σ(I). After Lorentz and
(68) Abbreviations used: H2TPP, tetraphenylporphyrin; H2(m-TTP),
tetrakis(m-tolyl)porphyrin; H2(p-TTP), tetrakis(p-tolyl)porphyrin; (2,6-xy-
lylNC), 2,6-xylyl isocyanide.
(69) Reed, C. A.; Mashiko, T.; Bentley, S. P.; Kasner, M. E.; Scheidt,
W. R.; Spartalian, K.; Lang, G. J. Am. Chem. Soc. 1979, 101, 2948.
(70) Goff, H.; Shimomura, E. J. Am. Chem. Soc. 1980, 102, 31.
(71) Malatesta, L.; Bonati, F. Isocyanide Complexes of Metals; Wiley
(Interscience): New York, 1969.
(74) Sheldrick, G. M. In Crystallographic Computing 3: Data Collection,
Structure Determination, Proteins and Databases; Sheldrick, G. M., Kru¨ger,
C., Goddard, R., Eds; Clarendon Press: Oxford, 1985.
(75) Wickman, H. H.; Klein, M. P.; Shirley, D. A. Phys. ReV. 1966,
152, 345.
(72) Singleton, E.; Oosthuizen, E. AdV. Organomet. Chem. 1983, 22,
209.
(73) ENRAF-NONIUS Molecular Determination Package, Delft Uni-
versity, The Netherlands, 1990.
(76) Trautwein, A. X.; Bill, E.; Bominaar, E. L.; Winkler, H. Struct.
Bonding 1991, 78, 1.

4370 J. Am. Chem. Soc., Vol. 122, No. 18, 2000
Simonneaux et al.
Table 2. Selected Bond Distances (Å) and Bond Angles (deg) for
[(p-TTP)Fe(2,6-xylylNC)2]CF3SO3 (1)a
Distances
Fe-N(1)
Fe-N(2)
Fe-N(3)
Fe-N(4)
1.937(7)
1.965(7)
1.967(7)
1.974(7)
Fe-C(49)
Fe-C(58)
C(49)-N(5)
C58-N(6)
1.95(1)
1.92(2)
1.13(1)
1.16(1)
Angles
89.9(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)
C(49)-Fe-N(1)
C(49)-Fe-N(2)
C(49)-Fe-C(58)
C(49)-Fe-N(3)
C(49)-Fe-N(4)
C(58)-Fe-N(1)
C(58)-Fe-N(2)
C(58)-Fe-N(3)
C(58)-Fe-N(4)
Fe-C(49)-N(5)
Fe-C(58)-N(6)
86.5(4)
92.2(4)
93.5(4)
91.9(4)
86.0(4)
87.8(4)
172.8(9)
172.8(9)
179.4(4)
89.6(4)
90.4(4)
179.4(4)
90.1(4)
94.0(4)
88.1(4)
172.5(4)
a Standard deviations in parentheses.
Figure 2. Formal diagram of the porphyrinato core in [(p-TTP)Fe-
(2,6-xylylNC)2]CF3SO3 (1) showing deviations of each unique atom
from the mean plane of the core (units 0.01 pm).
dihedral angles between the porphyrinate plane and the planes
of the four phenyl rings are 119.4(3), 67.7(3), 108.9(3), and
66.1(3)° for 1; these values are well removed from 90° but are
not unusual.
Selected individual bond distances and angles for [(p-TTP)-
Fe(2,6-xylylNC)2]CF3SO3 (1) are given in Table 2, and the
complete lists of distances and angles are given in Tables S1
and S2 (Supporting Information), respectively. Positional pa-
rameters and general displacement parameters, B, are given in
Tables S3 and S4 (Supporting Information), respectively. The
metal-carbon bond distances of the isocyanide ligands average
to 1.935(10) Å for 1 and are slightly longer than those observed
in the analogous iron(III) complex containing tert-butyl iso-
cyanide ligands (1.915 (20) Å).17 This is consistent with the
greater σ donor ability of alkyl isocyanide ligands compared
with that of aryl isocyanides and with a concomitant increase
in σ bonding from the axial ligand to iron.
The equatorial Fe-N bond distances average to 1.961(6) Å
for 1, which is quite short for low-spin iron(III) porphyrinate
derivatives.77,78 These short distances are consistent with the
significant ruffling of the porphyrinate core, which is very
similar to that recently reported in perchlorato derivatives of
bis(4-cyanopyridine)tetraphenylporphyrinatoiron(III)11,33 and of
bis(tert-butyl isocyanide) octaethyl- and tetraphenylporphyri-
natoiron(III).17 Deviation of each unique atom from the mean
plane of the core of 1 is shown in Figure 2. The ruffling of the
core is quite apparent in the complex and similar to the ruffling
observed in bis(tert-butyl isocyanide) octaethyl- and tetraphe-
Figure 1. ORTEP diagram of [(p-TTP)Fe(2,6-xylylNC)2]CF3SO3 (1)
(a) looking at the complex along the mean plane of the porphyrin and
(b) looking down on the porphyrin ring. Labels assigned to the
crystallographically unique atoms are displayed. Atoms are contoured
at the 50% probability level.
temperature range. The ferric low-spin samples discussed in this work
exhibit such spectra. The case of intermediate spin fluctuation is treated
by using the dynamic line shape formalism of Blume and Clauser,64
which has been elaborated by Winkler et al.63
Results and Discussion
Structure Determination. The molecular structure of the
cation of [(p-TTP)Fe(2,6-xylylNC)2]CF3SO3 (1) is shown in the
ORTEP diagrams of Figure 1a and 1b (and of the complete
molecular unit in Figure S1 in the Supporting Information),
along with the numbering scheme for the crystallographically
unique atoms. As shown in Figure 1b, the xylyl rings of the
axial ligands are not in parallel planes. The dihedral angles
between the planes of the individual xylyl groups and the
N1,N3 axis are 39.6(2) and 15.5(2)°, yielding a dihedral angle
between the planes of the two xylyl groups of 55.1°. The
(77) Scheidt, W. R.; Reed, C. A. Chem. ReV. 1981, 81, 43.
(78) Scheidt, W. R.; Lee, Y. J. Struct. Bonding (Berlin) 1987, 64, 1.

Ground-State Low-Spin [(p-TTP)Fe(2,6-xylylNC)2]CF3SO3
J. Am. Chem. Soc., Vol. 122, No. 18, 2000 4371
Figure 3. 1H NMR spectrum of [(p-TTP)Fe(2,6-xylylNC)2]CF3SO3 (1) recorded at 297 K in CDCl3. Assignments of the various resonances are
indicated; S marks the residual solvent peaks (chloroform and dichloromethane), and x marks minor impurities.
Table 3. Observed Shifts and Separation of the Isotropic Shift into
nylporphyrinatoiron(III)17 and [(4-CN-Py)2Fe(TPP)]ClO4.11 The
Contact and Dipolar Contributions in
[(p-TTP)Fe(2,6-xylylNC)2]CF3SO3 (1) in CDCl3 at 297 K
55.1° dihedral angle between xylyl planes mentioned above is
apparently not related to the ruffling of the porphyrin ring, and
hence the observed angle may arise from crystal packing
considerations. A packing diagram is shown in Figure S2
(Supporting Information).
a
b
c
proton type
δobs
δiso
δdip
δcon
o-H
m-H
-3.42
16.51
(0.86)d
12.36
(1.1)e
10.69
-11.44
8.9
(-1.65)
9.72
(-6.53)
2.49
0.76
0.35
(0.22)
0.23
(0.31)
1.48
-12.20
8.55
(-1.87)
9.49
(-6.84)
1.01
IR Spectroscopy. Determination of the isocyanide stretching
frequencies in heme isocyanides provides an opportunity to
demonstrate clearly the sensitivity of ligand binding to the
electronic properties of the second axial ligand and the metal-
loporphyrin. For example, the ν(CN) stretching frequency of
CNR is increased upon coordination of the aromatic isocyanide
to iron(III) porphyrin from 2135 cm-1 for the free ligand to
2195 cm-1 in [(p-TTP)Fe(2,6-xylylNC)2]CF3SO3 (1). This
increase of the stretching frequency, which indicates a higher
bond order in the complex than in the free ligand, is consistent
with the formal positive charge on the iron.71 However, the data
suggest that the observed isocyanide stretching frequencies upon
complexation are also influenced by the electronic nature of
the phenyl group on the isocyanide ligand. As expected, the
ν(CN) stretching frequency of ligated 2,6-xylylisocyanide is
lower than that of ligated t-BuNC upon coordination of the
isocyanide to [(p-TTP)FeCF3SO3]; ν(NC) increases from 2195
cm-1 for 1 to 2220 cm-1 in [(TPP)Fe(t-BuNC)2]CF3SO3 (∆ν
) 25 cm-1).17 This is consistent either with a greater π-acceptor
ability of the aryl isocyanide ligand as compared with that of
the alkyl isocyanide ligand and with a concomitant increase in
π back-bonding from iron to the CN bond, or with a weaker
σ-donating ability of the aromatic isocyanide ligand relative to
that of the alkyl isocyanide ligand.
p-CH3
Pyrr-H
a Chemical shifts in ppm. b Isotropic shifts with the diamagnetic
complex (TPP)Fe(t-BuNC)2 as reference.61 c Calculated from the o-H
dipolar shift of the [TPPFe(NMeIm)2]+ complex,29 scaled by the ratio
of the axial g-anisotropies of the isocyanide complexes of this study
divided by that of the NMeIm complex.29 Other dipolar shifts calculated
from that of o-H, based on the relative geometric factors (3 cos2 θ -
1)/r3.24,29 d m-CH3 shift of [(m-TTP)Fe(2,6-xylylNC)2]+ in parentheses.
e p-H shift of [(m-TTP)Fe(2,6-xylylNC)2]+ and [(TPP)Fe(2,6-xy-
lylNC)2]+ in parentheses.
at a downfield position, 10.7 ppm, as compared to 9.7 ppm for
[(TPP)Fe(t-BuNC)2]CF3SO3.26 This contrasts with the pyrrole
proton signals of most other low-spin Fe(III) porphyrinates at
ambient temperature, including [(TPP)Fe(N-MeIm)2]ClO4 (δ )
-17 ppm)25 and [(TPP)Fe(P(Me)2Ph)2]ClO4 (δ ) -19.4 ppm),79
and provides confirmation of the (dxz,dyz)4(dxy)1 electronic
structure in these aryl isocyanide derivatives. Magnetic mea-
surements using the method of Evans80,81 were made for 0.03
M CD2Cl2 solutions of 1 at 297 K, employing Me4Si as the
reference. The solution magnetic moment (µ ) 1.9 µB) is
1
compatible with the low-spin state, S ) /2.
The temperature dependences of the isotropic shifts of the
protons of 1 in CDCl3 are shown in Figure 4. The isotropic
shifts appear to vary almost linearly with 1/T over the range of
measurement, but the linearly extrapolated lines do not pass
through the origin at 1/T ) 0. The deviations of the intercepts
are +8.5 (o-H), -11 (p-CH3), -4.5 (m-H), and -3.5 ppm (pyrr-
H). Hence, the temperature dependence of 1 was analyzed in
terms of the possibility of a thermally accessible excited state.30
1
NMR Spectroscopy. The H NMR spectrum of [(p-TTP)-
Fe(2,6-xylylNC)2]CF3SO3 (1) at 297 K is shown in Figure 3.
The isotropic shifts are listed in Table 3. The peaks for the
phenyl protons of the porphyrin ring are assigned completely
by methyl substitution in combination with 1D proton NMR
decoupling experiments (not shown). For the isocyanide ligands,
the relative intensities determine the assignment. The shifts of
the isocyanide ligand are independent of the presence of excess
ligand, and hence axial ligand dissociation does not appear to
be significant at ambient temperature. The spectrum of [(p-TTP)-
Fe(2,6-xylylNC)2]CF3SO3 (1) shows the pyrrole proton signal
(79) Sodano, P.; Simonneaux, G. Inorg. Chem. 1988, 27, 3956.
(80) Evans, D. F. J. Chem. Soc. 1959, 2003.
(81) For superconducting magnets, the revised equation must be used:
Sur, S. K. J. Magn. Reson. 1989, 82, 169.

4372 J. Am. Chem. Soc., Vol. 122, No. 18, 2000
Simonneaux et al.
Figure 4. Curie plot of the isotropic shifts vs 1000 × reciprocal
temperature (K) for [(p-TTP)Fe(2,6-xylylNC)2]CF3SO3 (1) in CDCl3,
showing extended Curie law fits30 to ∆E ) 70 (- - -) and 740 cm-1
(s). Symbols: 9, p-CH3; 2, m-H; b, pyrrole-H; +, o-H.
Figure 5. EPR spectrum of (a) [(p-TTP)Fe(2,6-xylylNC)2]CF3SO3 (1)
and (b) [(m-TTP)Fe(2,6-xylylNC)2]CF3SO3 (3) in the solid state,
recorded at 10 K. Experimental conditions: microwave frequency, 9.649
GHz; microwave power, 20 µW; modulation frequency, 100 kHz;
modulation amplitude, 1 G.
This analysis showed that acceptable fits could be obtained
assuming ∆E values between ground and excited states of 70
to greater than 700 cm-1. Example fits for 70 and 740 cm-1
values of ∆E are included in Figure 4. Although the smaller
∆E value appears to fit somewhat better, the high-temperature
data points may be less reliable due to the possibility of some
ligand dissociation (although none was detected). As is evident
from the fits shown in Figure 4, the relatively gentle temperature
dependence of the experimental data down to -50 °C does not
allow the determination of a unique fit. As discussed below,
the temperature dependence of the quadrupole splitting, ∆EQ,
obtained from the Mo¨ssbauer data, are consistent with ∆E )
740 cm-1, but there may be reasons why the value of ∆E is
larger in the solid state, and the process represented by this ∆E
value may not be simply an excited electronic state (see below).
To characterize the iron bis-isocyanide electronic structure,
the isotropic shifts were calculated for 1 by using [(TPP)Fe(t-
BuNC)2] and related p- and m-methyl substituted diamagnetic
complexes as references.61 Analysis of the isotropic shift was
made according to the method of La Mar,31 and as found for
the related tert-butyl isocyanide complex,26 there is a relatively
large contact contribution to the meso-phenyl-H resonances of
1-3. The mechanism of spin transfer is thus, as expected,
significantly different from that observed for low-spin ferric bis-
imidazole24,25,29 and bis-phosphine79,82 complexes of synthetic
porphyrins, where the meso-phenyl shifts were found to be
essentially completely dipolar in origin.24,79 Using the o-H
dipolar shift of [(TPP)Fe(N-MeIm)2]+ at 298 K,24 scaled by the
ratio of the axial g-anisotropies of the isocyanide complexes
1-3 divided by that of the imidazole complex,29 the dipolar
shift of the o-H of complexes 1-3 could be calculated. Then,
using the relative geometric factors,29 the dipolar contribution
to the pyrrole-H and all other protons can be obtained. The
resulting dipolar and contact contributions are included in Table
3. The chemical shift of the pyrrole-H resonance of complexes
1-3 at ambient temperatures is larger (more positive) than that
for [(TPP)Fe(t-BuNC)2]+,17 which leads, in Table 3, to a positive
value for the contact shift of the pyrrole-H of about 1 ppm.
This is not consistent with spin delocalization via porphyrin π
orbitals29 and suggests that either the dipolar shift has been
underestimated or there is a small amount of σ spin delocal-
ization to the pyrrole-H position. The latter possibility seems
unlikely, because the dxy orbital has nodes at the porphyrin
nitrogens with respect to σ-bonding.
The chemical shifts observed for the axial ligands (δo-CH
)
3
-0.85 ppm; δm-H ) 5.46 ppm; δp-H ) 6.70 ppm) are close to
those expected for groups located in the shielding region of the
porphyrin ring. There should be only very small dipolar shifts
for those protons because of their distance from the iron. Thus,
there appears to be essentially no spin density on the axial
ligands, as expected because of the orthogonality of the
isocyanide pπ orbitals and the metal dxy orbital that contains
the unpaired electron.
EPR Spectroscopy. [(p-TTP)Fe(2,6-xylylNC)2]CF3SO3 (1)
and [(TPP)Fe(2,6-xylylNC)2]CF3SO3 (2) exhibit identical axial
EPR spectra, with g values of g ) 2.2 and g ) 1.94 (Figure
||
5a). These values are identical with those reported for [(TPP)-
Fe(t-BuNC)2]ClO4,17 which also exhibits the (dxz,dyz)4(dxy)1
ground state. Assuming this ground state, the mixing coefficients
a, b, and c of the Fe dyz, dxz, and dxy atomic orbitals, respectively,
can be calculated from the g values22 according to
gxx ) 2[a2 - (b + c)2]
gyy ) 2[(a + c)2 - b2]
(3)
gzz ) 2[(a + b)2 - c2]
a2 + b2 + c2 ) 1
(82) Simonneaux, G. Coord. Chem. ReV. 1997, 165, 447.

Ground-State Low-Spin [(p-TTP)Fe(2,6-xylylNC)2]CF3SO3
J. Am. Chem. Soc., Vol. 122, No. 18, 2000 4373
Table 4. Rhombicity and Tetragonality Parameters as Obtained
from the EPR Data According to Eq 4
2 and 1
3
V/λ
0
0
-
100%
∆/λ
-8.90
98%
% dxy
We follow the convention that the heme normal is the z-axis.23
The g values can be used to calculate the relative energies of
the Fe formerly t2g orbitals and their tetragonal (∆/λ) and
rhombic (V/λ) splitting parameters,23 in terms of the spin-orbit
coupling constant λ, according to
gx
gy
V/λ )
+
gz + gy gz - gx
(4)
gx
gz
1
2
∆/λ )
+
-
V/λ
gz + gy gy - gx
The parameters calculated in this way are listed in Table 4. The
EPR spectrum only provides the absolute values of the diagonal
elements of the g-tensor. The g-tensor consistent with a (dxz,dyz)4-
(dxy)1 ground state is gc ) (-2.2, 2.2, -1.94) according to eqs
3.
The [(m-TTP)Fe(2,6-xylylNC)2]CF3SO3 complex (3) exhibits
an isotropic EPR signal at g ) 2.14 in the solid state. The EPR
spectra of CH2Cl2 frozen solutions of complexes 1, 2, and 3 all
show axial EPR signals with g ) 2.15-2.18 and g| ) 1.94-
1.95 (data not shown). The difference between solution- and
solid-state EPR spectra of the porphyrin complex 3 may be
caused by a slightly different structure in the solid state than in
solution, or, more likely, it may be a result of exchange
narrowing due to spin-spin interactions in the solid state. On
the basis of the NMR data discussed above, we have found that
the ground states of all three complexes are of the dxy type.
This is supported by the fact that the EPR spectra of their CH2-
Cl2 frozen solutions are almost identical. Therefore, based on
the strong evidence from the NMR and the frozen solution EPR
spectra of complex 3, we assume the g-tensor to be gc ) (-2.14,
2.14, -2.14), which is consistent with a dxy HOMO.
Mo1ssbauer Spectroscopy. The Mo¨ssbauer spectrum of
[(TPP)Fe(2,6-xylylNC)2]CF3SO3 (2) obtained at 190 K in an
applied field of 20 mT is shown in Figure 6a. At 4.2 K, also at
20 mT, basically the same magnetically broadened asymmetric
doublet is observed, with an isomer shift of δ ) 0.08 mm/s
and a quadrupole splitting |∆EQ| ) 1.99 mm/s (Figure 6b). This
lack of change with temperature is explained by the fact that
temperature-independent spin-spin relaxation is reflected in the
Mo¨ssbauer spectra. To slow the spin-spin relaxation rate,
spectra were collected at 4.2 K in fields of up to 5.34 T parallel
and perpendicular to the γ-beam; representative spectra are
displayed in Figure 6c-e. It should be noted that no consistent
hyperfine parameter set could be found to simulate the spectra
shown in Figure 6 according to eq 2, assuming either slow or
fast relaxation limits. Even extensive variation of the Euler
angles that describe the rotation from the principal axis of gc
and AC to the axis of the efg tensor, a method used successfully
for dissolved [(TPP)Fe(t-BuNC)2]ClO4,17 did not reproduce the
Mo¨ssbauer pattern.
Figure 6. Mo¨ssbauer spectra of [(TPP)Fe(2,6-xylylNC)2]CF3SO3 (2)
obtained at (a) 190 K and 4.2 K in fields of (b) 20 mT, (c) 1 T, and (d)
5.34 T perpendicular to the γ-beam, as well as in a field of 5.34 T
parallel to the γ-beam (e). The solid lines are simulations with the spin-
Hamiltonian parameters given in Table 5 and the relaxation rates (a,
b) 2.2 × 108 s-1, (c) 1.9 × 108 s-1, and (d, e) 6.0 × 107 s-1
.
Table 5. Mo¨ssbauera and Spin-Hamiltonian Parameters Obtained
from the Simulations Shown in Figures 6, 8, and 9
δ
∆EQ
Γ
gc
(mm/s) (mm/s) (mm/s) η AC/gNµN (T)
1 (-2.2, 2.2, -1.94)
2 (-2.2, 2.2, -1.94)
0.14 -1.81
0.08 -1.99
0.26 0 (-5.5, 5.5, 27)
0.26 0 (-5.5, 5.5, 40)
3 (-2.14, 2.14, -2.14) 0.12 (1.94b 0.26 1 (-2.5, 2.5, 32)
a The spectrum of [(TPP)Fe(2,6-xylylNC)2]CF3SO3 (2) obtained at
190 K (Figure 6a) has been analyzed with δ ) 0.05 mm/s and ∆EQ
)
-1.90 mm/swith all other parameters as quoted in the table. That of
[(p-TTP)Fe(2,6-xylylNC)2]CF3SO3 (1) (Figure 8a) has been analyzed
with δ ) 0.10 mm/s and ∆EQ ) -1.76 mm/s and that of [(m-
TTP)Fe(2,6-xylylNC)2]CF3SO3 (3) (Figure 9a) with δ ) 0.10 mm/s
and ∆EQ ) (1.75 mm/s.b b The sign of the quadrupole splitting cannot
be determined since η ) 1.
spin relaxation rates given in the caption of Figure 6. The
negative sign of the quadrupole splitting, ∆EQ ) -1.99 mm/s,
is indicative of a singly occupied dxy HOMO, which confirms
the analysis of the EPR spectra. As already discussed above,
the Griffith model yields for this situation the g-tensor gc )
(-2.2, 2.2, -1.94). The g-tensor thus obtained was used to
simulate the Mo¨ssbauer spectra in Figure 6, and an experimental
A-tensor of AC/gNµN ) (-5.5, 5.5, 40) T was found.
As reported by Lang, it is possible to calculate the A-tensor
from the expressions83-85
Since relaxation effects influence the Mo¨ssbauer spectra under
all conditions applied, they have been analyzed by using the
dynamic line shape formalism of Blume and Clauser.63,64 The
solid lines shown in Figure 6 are spin-Hamiltonian simulations
for an S ) 1/2 system with the parameters given in Table 5 and
(83) Oosterhuis, W. T.; Lang, G. J. Chem. Phys. 1969, 50, 4381.
(84) Oosterhuis, W. T.; Lang, G. Phys. ReV. 1969, 178, 439.
(85) Rhynard, D.; Lang, G.; Spartalian, K.; Yonetani, T. J. Chem. Phys.
1979, 71, 3715.

4374 J. Am. Chem. Soc., Vol. 122, No. 18, 2000
Simonneaux et al.
Ax ) -P -4bc - (1 + k)(a2 - b2 - c2) +
[
3
7
6
7
(a2 - 3b2 - 3c2) + a(b + c)
]
]
Ay ) +P -4ac - (1 + k)(b2 - a2 - c2) +
[
(5)
3
7
6
(b2 - 3a2 - 3c2) + b(a + c)
7
Az ) +P -4ab - (1 + k)(c2 - a2 - b2) +
[
3
6
(c2 - 3a2 - 3b2) + c(a + b)
]
Figure 7. Relaxation rates obtained from the Mo¨ssbauer analysis of
[(TPP)Fe(2,6-xylylNC)2]CF3SO3 (2) (Figure 6) as a function of the
Boltzmann population of the upper Zeeman level P(ms ) +1/2)
calculated from eq 6 for the different indicated applied fields.
7
7
The proportionality constant P is taken as -62 T and the Fermi
contact constant κ ) 0.35 as for most heme proteins86 and heme
models.9
If one calculates the A-tensor according to this model one
obtains AC/gNµN ) (-7.4, 7.4, 50.9) T. The experimental values
quoted above are reduced with respect to the calculated values
(eqs 5) by approximately 25%, which may be caused by
covalency effects in the iron-ligand bonds and/or spin delo-
calization by porphyrin f Fe π donation.24
The relaxation process can be identified as follows: The
population P(ms ) +1/2) of the upper ms ) +1/2 Zeeman level
is given by the Boltzmann factor,
exp(-∆E/kBT)
P(ms ) +1/2) )
(6)
1 + exp(-∆E/kBT)
with the Zeeman energy
∆E ) gµBB
(7)
Figure 7 shows a linear relationship of the relaxation rates
obtained from the Mo¨ssbauer analysis as a function of P(ms )
+1/2) for the different applied fields. This linear relationship is
characteristic for spin-spin relaxation, or to be more specific,
cross relaxation,87 because the probability for the latter rises
linearly with the occupation of the upper level at one of the
neighboring ions. Thus, this situation can only arise in a
concentrated sample such as a polycrystalline solid sample of
a low-spin iron(III) porphyrinate and is not expected to occur
for either frozen solutions of 57Fe-enriched samples of ferriheme
models and proteins or polycrystalline heme proteins, where
the hemes are well separated from each other.
The Mo¨ssbauer spectra of [(p-TTP)Fe(2,6-xylylNC)2]CF3SO3
(1), obtained at 190 and 4.2 K in fields of 20 mT and 5.34 T,
are shown in Figure 8. The solid lines are spin-Hamiltonian
simulations with the parameters given in Table 5. The spectra
have been analyzed with the relaxation rates corresponding to
those used for [(TPP)Fe(2,6-xylylNC)2]CF3SO3 (2). Again, the
negative sign of the quadrupole splitting, ∆EQ ) -1.81 mm/s,
is indicative of a singly occupied dxy HOMO. The experimental
A-tensor is AC/gNµN ) (-5.5, 5.5, 27) T. Az is reduced from the
theoretical value of 50.9 T by approximately 50%, indicating
that the covalency effects in [(p-TTP)Fe(2,6-xylylNC)2]CF3SO3
(1) are more important than those in [(TPP)Fe(2,6-xylylNC)2]-
CF3SO3 (2).
Figure 8. Mo¨ssbauer spectra of [(p-TTP)Fe(2,6-xylylNC)2]CF3SO3 (1)
obtained at (a) 190 K and 4.2 K in fields of (b) 20 mT and (c) 5.34 T
parallel to the γ-beam. The solid lines are simulations with the spin-
Hamiltonian parameters given in Table 5 and the relaxation rates 2.2
× 108 s-1 for (a) and (b) and 6.0 × 107 s-1 for (c).
shown in Figure 9. The solid lines are spin-Hamiltonian
simulations with the parameters given in Table 5. The spectra
have been analyzed with relaxation rates corresponding to those
used for [(TPP)Fe(2,6-xylylNC)2]CF3SO3 (2) and [(p-TTP)Fe-
(2,6-xylylNC)2]CF3SO3 (1). The magnetic Mo¨ssbauer pattern
of Figure 9b can best be simulated by taking the asymmetry
parameter η ) 1. Therefore, we cannot determine the sign of
the quadrupole splitting for complex 3. A pure dxy hole would
exhibit η ) 0, as in the cases of complexes 1 and 2. However,
the complex [(OEP)Fe(t-BuNC)2]ClO4 exhibits η ) 0.41 in the
solid state, but an almost zero asymmetry parameter of η )
0.14 in solution.17 The reason for the increase in the asymmetry
parameter in the solid state is not clear. The ruffling of the
porphyrin plane present in [(OEP)Fe(t-BuNC)2]ClO4 and prob-
ably also in complex 3 may lead to strong delocalization of the
dxy iron hole to the pz orbitals of the porphyrin nitrogens via
the a2u(π) filled porphyrin orbital, but whether this admixture
can lead to such strong changes of η may be questionable. This
The Mo¨ssbauer spectra of [(m-TTP)Fe(2,6-xylylNC)2]CF3-
SO3 (3) obtained at 4.2 K in fields of 20 mT and 5.34 T are
(86) Lang, G. Q. ReV. Biophys. 1970, 3, 1.
(87) Boyle, A. J. F.; Gabriel, J. R. Phys. Lett. 1965, 19, 451.

Ground-State Low-Spin [(p-TTP)Fe(2,6-xylylNC)2]CF3SO3
J. Am. Chem. Soc., Vol. 122, No. 18, 2000 4375
Figure 10. Temperature dependence of the quadrupole splitting of
[(TPP)Fe(2,6-xylylNC)2]CF3SO3 (2). The solid line has been calculated
according to eq 4 with ∆E ) 740 cm-1, the dash-dotted line with ∆E
) 70 cm-1, and the dotted line with ∆E ) 1500 cm-1
.
Table 6. Magnitude Per d Electron of the Diagonal Electric Field
Gradient (efg)val Tensor Elementsa
orbital
(Vxx)val/e r-3
(Vyy)val/e r-3
(Vzz)val/e r-3
dxy
dxz
dyz
-2/7
-2/7
+4/7
-2/7
+2/7
-2/7
+4/7
-2/7
-2/7
+2/7
+4/7
-2/7
-2/7
+4/7
-4/7
2
2
Figure 9. Mo¨ssbauer spectra of [(m-TTP)Fe(2,6-xylylNC)2]CF3SO3
(3) obtained at (a) 190 K and at 4.2 K in fields of (b) 20 mT and (c)
5.34 T perpendicular to the γ-beam. The solid lines for subspectra I
are simulations with the spin-Hamiltonian parameters given in Table
5 and the relaxation rates 2.2 × 108 s-1 for (a) and (b) and 6.0 × 107
s-1 for (c). Subspectrum II exhibits δ ) 0.24 mm/s at 4.2 K in (b) and
δ ) 0.19 mm/s at 190 K in (a), as well as ∆EQ ) 0 mm/s and Γ )
0.26 mm/s and was simulated in (c) assuming S ) 0; it accounts for
4% of the total area and is attributed to ferrous low-spin [(m-TTP)Fe-
(2,6-xylylNC)2] caused by autoreduction of the ferric form.
dx -y
2
dz
a Note that for a hole the signs of Vii /e<r-3> reverse.
∆EQ is expected, while for a single unpaired electron in either
dxz or dyz, a positive sign for ∆EQ is expected. Thus, population
of an excited state with an unpaired electron in dxz or dyz should
cause a decrease in the magnitude of the observed quadrupole
splitting.
Figure 10 shows the temperature dependence of the quadru-
pole splitting of [(TPP)Fe(2,6-xylyl)2]CF3SO3 (2) in the range
from 4.2 to 300 K. The simulations of the spectra obtained at
4.2 K in high magnetic fields yield a negative sign of the
quadrupole splitting, which is consistent with a dxy electronic
ground state. Since the rhombicity, V, of 1, 2, and 3 has been
determined to be zero from the EPR analysis, the first excited
state represents a hole either in the dyz or in the dxz orbital, which
means that the first excited state is doubly degenerate. Thermal
population of the doubly degenerate excited state leads to a
decrease in the overall quadrupole splitting, if one assumes fast
relaxation (or almost fast, as in our case), because the excited
states with the hole in a dyz or dxz orbital exhibit a positive
quadrupole splitting.
may be addressed further by molecular orbital calculations once
the structure of 3 has been solved. Nevertheless, the Mo¨ssbauer
spectra have been measured in the solid state, and therefore we
have used for the analysis reported here the g-factors measured
for the solid complex 3, gc ) (-2.14, 2.14, -2.14). From these,
the theoretical A-tensor according to eq 5 follows as AC ) (-4.2,
4.2, 61.1) T. The experimental A-tensor is AC/gNµN ) (-2.5,
2.5, 32) T. As for complexes 1 and 2, this deviation of about
50% is also attributed to covalency effects.
The NMR analysis discussed above and shown in Figure 4
indicates that the complexes under investigation exhibit a
relatively low-lying excited state, presumably the (dxz,dyz)3(dxy)2
state, at some energy above the (dxz,dyz)4(dxy)1 ground state. Low-
lying states may influence the relaxation behavior of paramag-
netic centers via Orbach-like spin-lattice relaxation processes.
Such processes exhibit a relaxation behavior that is field
independent but strongly temperature dependent. The Mo¨ssbauer
analysis of the solid iron porphyrinates presented in this
study clearly shows that a field-dependent and temperature-
independent relaxation process is effective at 4.2 K. The
mechanism is identified above as spin-spin or cross relaxation.
This type of relaxation process, however, is not influenced by
low-lying excited crystal field states. However, at higher
temperatures, the quadrupole splitting, ∆EQ, may vary if an
excited state is populated, if the quadrupole splitting of that
excited state is opposite in sign to that of the ground state. In
fact, according to the Townes-Dailey approximation, for a
single unpaired electron in the dxy orbital, a negative sign for
To calculate the temperature dependence of ∆EQ, we assume
that the electric field gradient (efg) tensor has only orbital
contributions. If we denote the efg tensor originating from a
hole in a dxy, dxz, and dyz orbital as 6Vxy, 6Vxz, and 6Vyz and consider
that the dxz and dyz excited states are degenerate, twice the
temperature dependence of the efg tensor 6V(T) is given by
1
Z
6V(T) ) (6Vxy + 6Vxz e-∆E/k T + 6Vyz e-∆E/k T
)
(8)
B
B
with Z being the statistical sum Z ) 1 + 2e-∆E/k T for a three-
level system with a degenerate excited state separated from the
ground state by ∆E. Insertion of the expectation values Vii /
e r-3 (see Table 6) into eq 8 yields
B

4376 J. Am. Chem. Soc., Vol. 122, No. 18, 2000
Simonneaux et al.
the value of 300 cm-1 for low-spin ferrihemes,93 while Levin
and Brill obtained a value of 350 cm-1 for metmyoglobin
complexes.94 Walker and co-workers have previously used 400
cm-1 as the maximum possible value of λ for low-spin
ferrihemes having the (dxy)2(dxz,dyz)3 electronic ground state,7
Horrocks and Greenberg95 have used 340 cm-1, and Turner and
co-workers96 have used 270 cm-1 for the same types of systems.
The MO calculations92 show that covalency effects indeed lead
to a reduction of λ from the free ion value. Thus, covalency
certainly plays a role for the complexes studied here. However,
we believe that its importance, as suggested by the low value
of λ ) 83 cm-1, is greatly overestimated; rather, a value 2-4
times this is more reasonable, based upon previous work.90-96
Thus, we expect the crystal field value ∆ to be of the order of
1500-3000 cm-1, perhaps nearer the smaller to emphasize the
importance of covalency. Therefore, the temperature dependence
of the quadrupole splitting must be attributed to effects other
than Boltzmann population of excited orbital states, i.e., to
dynamic effects.
Such dynamic effects as rotation of axial ligands at higher
temperatures have previously been observed in low-spin (S )
0) Fe(II) porphyrins.97,98 However, the axial ligands of this study
have cylindrical symmetry at the binding site to the metal, and
it is very unlikely that rotation of the xylyl groups would in
any way affect the energies of the orbitals of the metal center.
Furthermore, the samples of this study are polycrystalline solids,
in which large-scale motions of parts of the molecule are
prevented by the packing in the crystalline lattice. However,
the strongly ruffled porphyrin ring, which is stabilized in the
ruffled conformation by porphyrin-to-metal π donation from
the 3a2u(π) orbital to the hole in the metal dxy orbital,11 should
have an associated low-energy “inversion” of the ruffled ring.
Such an “inversion” is clearly observed when bulky planar axial
ligands are present on tetramesitylporphyrinatoiron(III), for it
is required for their rotation.99-101 And even in the absence of
planar axial ligands, such “inversion” must occur and should
be a relatively low-energy process. This “inversion” would move
meso positions that are displaced above the mean plane of the
porphyrin ring shown in Figure 2 down, while those displaced
below the mean plane are moved up, with concomitant
lengthening of the Fe-N(por) bonds at the planar transition
state. Although such large-scale movements from one ruffled
conformation to the other are unlikely in the solid state, a partial
“inversion,” in which the porphyrin ring approaches planarity,
with Fe-N(por) bonds lengthened, may be possible. From what
has been learned about ruffled vs planar porphyrin rings in this
and previous work, it is reasonable to expect that the planar
porphyrin ring would stabilize the unpaired electron in the dxz,dyz
orbitals, yielding a (dxy)2(dxz,dyz)3 electron configuration for this
transition state, the same electron configuration as expected for
the excited electronic state of the ruffled conformation, discussed
6V(T)/e r-3
)
2
4
7
2
7
-
+
7
2
2
7
4
7
‚e-∆E/k T
+
‚e-∆E/k T
B
B
-
+
7
4
2
7
2
7
(
) (
(
)
(
)
-
7
1 + 2e-E/k T
B
2
7
2
-
7
2
7
2
‚e-∆E/k T
B
-
+
7
4
4
) (
)
-
7
7
)
(9)
1 + 2e-E/k T
B
The quadrupole splitting ∆EQ(T) is proportional to the main
component Vzz(T) of the efg tensor:
E (T) ∼ V (T) 1 + η(T)2/3
(10)
x
Q
zz
Insertion of the efg elements obtained in eq 8 gives in the present
case η(T) ) 0 for all temperatures. Therefore, ∆EQ is propor-
tional to Vzz, and combining eqs 9 and 10 yields, for the
temperature dependence of the electronic contribution to ∆EQ:
exp(-∆E/kBT) - 1
∆EQ(T) )
∆EQ(4.2 K) (11)
2 exp(-∆E/kBT) + 1
The solid and dashed lines in Figure 10 have been calculated
from eq 11 using ∆E ) 740 cm-1 and ∆E ) 70 cm-1, the
limiting values derived from the temperature dependence of the
NMR spectra. For comparison, we also show the dotted line
which corresponds to ∆E ) 1500 cm-1 as an upper limit.
Clearly, the value for ∆E ) 740 cm-1 fits the temperature
dependence of the quadrupole splitting well.
The energy difference ∆E should correspond to the crystal
field splitting ∆, which has been calculated from the electronic
g-tensors (see Table 4). If we set ∆E ) ∆ ) 740 cm-1 and
insert the value of ∆/λ ) 8.90 obtained for complex 2 (Table
4), we obtain λ ) 83 cm-1. This value is much lower than the
free ion value (460 cm-1).88 Nevertheless, for K3FeCN6 a value
of λ ) 80 ( 5 cm-1 was determined experimentally by a
combined EPR, NMR, and Mo¨ssbauer study.89 However,
Bleaney et al. used 278 cm-1 in the interpretation of their EPR
(93) Maltempo, M. M. J. Chem. Phys. 1974, 61, 2540.
(94) Levin, P. D.; Brill, A. S. J. Phys. Chem. 1988, 92, 5103.
(95) Horrocks, W. DeW.; Greenberg, E. S. Biochim. Biophys. Acta 1973,
322, 38; Mol. Phys. 1974, 27, 993.
(96) Turner, D. L. Eur. J. Biochem. 1993, 211, 563.
(97) Schappacher, M.; Ricard, L.; Fischer, J.; Weiss, R.; Bill, E.; Montiel-
Mantoya, R.; Winkler, H.; Trautwein, A. X. Eur. J. Biochem. 1987, 168,
419.
data for K3FeCN6,90 and Figgis et al. reported 275-375 cm-1 91
,
the range of the latter values being consistent with MO
calculations performed by Reschke et al.92 Maltempo has used
(88) Figgis, B. N.; Lewis, J. In Techniques of Inorganic Chemistry;
Jonassen, H. B., Weiss- berger, A., Eds.; Weiley Interscience: New York,
1965; Vol. IV, p 159.
(98) Grodzicki, M.; Flint, H.; Winkler, H.; Walker, F. A. Trautwein, A.
X. J. Phys. Chem. A 1997, 101, 4202.
(89) Merrithew, P. B.; Modestino A. J. J. Am. Chem. Soc. 1972, 94,
3361.
(90) Bleany, B.; O’Brien, Proc. Phys. Soc. 1956, B69, 1216.
(91) Figgis, B. N.; Gerlach, M.; Mason, B. Proc. R. Soc. 1969, A309,
91.
(99) Walker, F. A.; Simonis, U. J. Am. Chem. Soc. 1991, 113, 8652;
1992, 114, 1929.
(100) Shokhirev, N. V.; Shokhireva, T. Kh.; Polam, J. R.; Watson, C.
T.; Raffii, K.; Simonis, U.; Walker, F. A. J. Phys. Chem. A 1997, 101,
2778.
(92) Reschke, R.; Trautwein A.; Harris, F. E.; Date, S. K. J. Magn. Magn.
Mater. 1979, 12, 176.
(101) Polam, J. R.; Shokhireva, T. Kh.; Raffii, K.; Simonis, U.; Walker,
F. A. Inorg. Chim. Acta 1997, 263/1-2, 109.

Ground-State Low-Spin [(p-TTP)Fe(2,6-xylylNC)2]CF3SO3
J. Am. Chem. Soc., Vol. 122, No. 18, 2000 4377
in the previous paragraph. Thus, such a partial “inversion” in
the solid state, and a complete “inversion” in solution, provide
an explanation for the temperature dependence of the quadrupole
splitting in the solid state and the isotropic shifts of the
porphyrinate protons in homogeneous solution. It also allows
for a different (lower) barrier to inversion of the ruffled
porphyrinate ring in solution than in the solid state, consistent
with the fact that the lower value of ∆E ≈ 70 cm-1 appears to
fit the NMR data better (Figure 4), while a much higher value
of ∆E ≈ -740 cm-1 fits the Mossbauer data better (Figure 10).
It is interesting to note that a too-small value of ∆E ≈ 296
cm-1 was found and commented on from the temperature
dependence of the isotropic shifts of [(OEP)Fe(t-BuNC)2]+, for
which the tetragonal splitting ∆/λ was found to be -8.33,
similarly suggesting a value of ∆ in the range of 1500-3000
“inversion” is undoubtedly responsible for the small value of
∆E, and that this energy term represents an activation energy
for reaching the planar transition state of the ruffled complex.
Acknowledgment. The financial support of the MESR
(C.M.), the Deutsche Forschungsgemeinschaft (A.X.T.), and the
National Institutes of Health (USA), DK31038 (F.A.W.), is
gratefully acknowledged.
Supporting Information Available: Figures showing an
ORTEP drawing of the porphyrinate cation and anion of 1 and
the packing diagram and unit cell of 1, and four tables showing
the bond distances, bond angles, positional parameters, and
displacement parameters for 1 (PDF). This material is available
free of charge via the Internet at http://pubs.acs.org.
cm-1 17
that in that case, as in the present complexes 1-3, porphyrin
.
It is now possible to conclude with some confidence
JA994190T