Radical localization in a series of symmetric Ni^II complexes with oxidized salen ligands

Square-planar nickel(II) complexes of salen ligands, N,N′-bis(3-tert-butyl-(5R)-salicylidene)-1,2-cyclohexanediamine), in which R=tert-butyl (1), OMe (2), and NMe2 (3), were prepared and the electronic structure of the one-electron-oxidized species [1-3]+. was investigated in solution. Cyclic voltammograms of [1-3] showed two quasi-reversible redox waves that were assigned to the oxidation of the phenolate moieties to phenoxyl radicals. From the difference between the first and second redox potentials, the trend of electronic delocalization 1+.>2+.>3+. was obtained. The cations [1-3]+. exhibited isotropic g tensors of 2.045, 2.023, and 2.005, respectively, reflecting a lower metal character of the singly occupied molecular orbital (SOMO) for systems that involve strongly electron-donating substituents. Pulsed-EPR spectroscopy showed a single population of equivalent imino nitrogen atoms for 1+., whereas two distinct populations were observed for 2+.. The resonance Raman spectra of 2+. and 3+. displayed the ν8a band of the phenoxyl radicals at 1612 cm−1, as well as the ν8a bands of the phenolates. In contrast, the Raman spectrum of 1+. exhibited the ν8a band at 1602 cm−1, without any evidence of the phenolate peak. Previous work showed an intense near-infrared (NIR) electronic transition for 1+. (Δν1/2=660 cm−1, ε=21 700 M−1 cm−1), indicating that the electron hole is fully delocalized over the ligand. The broader and moderately intense NIR transition of 2+. (Δν1/2=1250 cm−1, ε=12 800 M−1 cm−1) suggests a certain degree of ligand-radical localization, whereas the very broad NIR transition of 3+. (Δν1/2=8630 cm−1, ε=2550 M−1 cm−1) indicates significant localization of the ligand radical on a single ring. Therefore, 1+. is a Class III mixed-valence complex, 2+. is Class II/III borderline complex, and 3+. is a Class II complex according to the Robin-Day classification method. By employing the Coulomb-attenuated method (CAM-B3LYP) we were able to predict the electron-hole localization and NIR transitions in the series, and show that the energy match between the redox-active ligand and the metal d orbitals is crucial for delocalization of the radical SOMO