Bis(imino)pyridine Iron Dinitrogen Compounds Revisited: Differences in Electronic Structure Between Four- and Five-Coordinate Derivatives.

The electronic structures of the four- and five-coordinate
aryl-substituted
bis­(imino)­pyridine iron dinitrogen complexes, (iPrPDI)­FeN2 and (iPrPDI)­Fe­(N2)2 (iPrPDI = 2,6-(2,6-iPr2–C6H3–N=CMe)2C5H3N), have been investigated by a combination of spectroscopic techniques
(NMR, Mössbauer, X-ray Absorption, and X-ray Emission) and
DFT calculations. Homologation of the imine methyl backbone to ethyl
or isopropyl groups resulted in the preparation of the new bis­(imino)­pyridine
iron dinitrogen complexes, (iPrRPDI)­FeN2 (iPrRPDI = 2,6-(2,6-iPr2–C6H3–N=CR)2C5H3N; R = Et, iPr), that are exclusively four coordinate
both in the solid state and in solution. The spectroscopic and computational
data establish that the (iPrRPDI)­FeN2 compounds
are intermediate spin ferrous derivatives (SFe = 1) antiferromagnetically coupled to bis­(imino)­pyridine
triplet diradical dianions (SPDI = 1).
While this ground state description is identical to that previously
reported for (iPrPDI)­Fe­(DMAP) (DMAP = 4-N,N-dimethylaminopyridine) and other four-coordinate
iron compounds with principally σ-donating ligands, the d-orbital
energetics determine the degree of coupling of the metal-chelate magnetic
orbitals resulting in different NMR spectroscopic behavior. For (iPrRPDI)­Fe­(DMAP) and related compounds, this coupling is strong
and results in temperature independent paramagnetism where a triplet
excited state mixes with the singlet ground state via spin orbit coupling.
In the (iPrRPDI)­FeN2 family, one of the iron
singly occupied molecular orbitals (SOMOs) is essentially dz2 in character resulting in poor overlap
with the magnetic orbitals of the chelate, leading to thermal population
of the triplet state and hence temperature dependent NMR behavior.
The electronic structures of (iPrRPDI)­FeN2 and
(iPrPDI)­Fe­(DMAP) differ from (iPrPDI)­Fe­(N2)2, a highly covalent molecule with a redox noninnocent
chelate that is best described as a resonance hybrid between iron(0)
and iron­(II) canonical forms as originally proposed in 2004.