The spin distribution in low-spin iron porphyrins

In many low-spin (S = 1/2) iron porphyrin derivatives, electron spin resonance (ESR) spectra indicate that one of the d(pi) orbitals of iron, either a d(xz) or d(yz), depending on the axial ligands of the porphyrin complex as well as their orientation, is essentially singly occupied; the unpaired electron is almost completely located at the metal. In contrast, nuclear magnetic resonance (NMR) and electron nuclear double resonance (ENDOR) spectroscopy convincingly show that a significant share of the unpaired electron is delocalized. This apparent contradiction is explained by the present density-functional-theory (DFT) calculations performed on a heme a model as well as on bis-imidazole-ligated iron porphyrin without substituents. The calculations show that the integrated spin density at the iron atom is nearly one, in agreement with the ESR measurements. However, significant areas with opposite (beta) spin are found along the Fe-N bond axes, thus evoking a need for additional alpha-spin density to be present in the porphyrin ring, ring substituents, and the axial ligands to keep the net amount of unpaired spin exactly one. The gross spin density, that is, the sum of unpaired alpha and beta spins, amounts to about 1.3 electrons. It seems that the degree to which alpha and beta spin dominate in different regions of the heme structure, as evidenced in these calculations, has not been previously observed.