The orientation of the principal axes of the g tensor
with respect to the relationship of axial ligand
planes to the porphyrin nitrogens has been studied in the framework of
the one-electron crystal field model for
tetragonal and rhombic low-spin d5 complexes such as
ferriheme centers. All five d atomic orbitals were
taken into account for two different ground-state electronic
configurations, the “normal”
(d
xy
)2(d
xz
,d
yz
)3
and
the “novel”
(d
xz
,d
yz
)4(d
xy
)1
configurations. The expressions for the g tensor,
g values, and magnetic axes were
derived on the basis of first-order perturbation theory. The
conditions for co- and counterrotation of magnetic
axes with rotation of planar axial ligands away from the porphyrin
nitrogens toward the meso positions and
beyond, as well as the order of g values, have been
analyzed. It is found that counterrotation is the
only
possibility for the
(d
xz
,d
yz
)4(d
xy
)1
configuration and that it is also by far more common for the
(d
xy
)2(d
xz
,d
yz
)3
electron configuration. The possibilities of nonlinear
co-/counterrotation are also explored. The
predictions
of this treatment are then compared to experimental results obtained
from single-crystal EPR, glassy sample
ESEEM, and solution NMR spectroscopic studies. It is clear that
the majority of experimental systems reported
thus far follow the major predictions of this treatment: Most systems
exhibit angle-for-angle (linear)
counterrotation of the g or χ tensor with rotation of
planar axial ligands away for the N−Fe−N axes.
Hence,
knowing the structure of a model heme or heme protein, and in
particular, the orientation of (fixed) axial
ligand planes, one should be able to predict the approximate
orientation of the in-plane magnetic axes. This
knowledge provides a check on the values obtained in new solution NMR,
single-crystal EPR or frozen solution
ESEEM experiments.