A detailed investigation into the electronic structures of macrocyclic iron(II)-nitrosyl compounds and their similarities to ferrous heme-nitrosyls

Abstract

This paper presents vibrational (IR and FT-Raman) and electronic (UV–Vis absorption and magnetic circular dichroism (MCD)) spectra of the five-coordinate (5C) low-spin ferrous iron-nitrosyl model complexes [Fe(L1)(NO)] (1), [Fe(L2)(NO)], (2) and [Fe(L3)(NO)] (3) that contain Schiff base type tetradentate macrocyclic ligands. These complexes show structural and vibrational data that are very similar to corresponding 5C ferrous heme-nitrosyls. For example, complexes 1–3 exhibit N–O and Fe–NO stretching frequencies of 1630–1680 and 520–595cm−1, respectively, which compares well with ν(N–O) and ν(Fe–NO) of [Fe(TPP)(NO)] at 1697 and 532cm−1. In complexes 1–3, a strong Fe–NO σ bond, mediated by the singly-occupied π∗ orbital of NO and dz2 of iron, and a medium strong Fe–NO π backbond are present. This is again in close agreement with the electronic structures of 5C ferrous heme-nitrosyls. Interestingly, a stepwise increase in the strength of the Fe–NO σ bond is observed in the order 1<2<3, as reflected by the direct correlation of the Fe–NO and N–O stretching frequencies in these complexes. Compared to the ferrous heme-nitrosyl model complex [Fe(TPP)(NO)], complexes 1–3 show a somewhat weaker Fe–NO σ bond and in particular, a stronger Fe–NO π backbond. The latter finding is reflected by the trend in ν(N–O) and ν(Fe–NO) vibrational frequencies between 3 and [Fe(TPP)(NO)]. In summary, complexes 1–3 serve as good spectroscopic models for 5C ferrous heme-nitrosyls. Correspondingly, these complexes can be used to locate d→π∗(NO) CT transitions, which would then provide evidence where such transitions could be expected for the heme-nitrosyls. In the heme complexes, the identification of CT transitions between Fe and NO is hampered by the very intense π→π∗ transitions of the heme macrocycle. In contrast, the ligands applied in 1–3 are not aromatic, and hence, show greatly reduced extinction coefficients for their π→π∗ transitions. Unfortunately, no d→π∗(NO) CT bands could be identified in the optical spectra of 1–3. The TD-DFT calculations predict that these features should be observed around 350nm, which would be to higher energy of the Soret band in 5C ferrous heme-nitrosyls.