Comparison of electronic structures and light-induced excited spin state trapping between [Fe(2-picolylamine)(3)](2+) and its iron(III) analogue.

We investigated mer- and fac-Fe(II)(2-pic)(3) (pic = picolylamine) and Fe(iii) analogue, mer-Fe(III)(2-pic)(3), by the DFT method to clarify the mechanism of light-induced excited spin state trapping (LIESST). In mer-Fe(II)(2-pic)(3), the potential energy surface (PES) of the triplet state is the least stable but it is close to the PESs of the singlet and quintet states at the equilibrium geometry of the triplet state within 5 kcal mol(-1). This indicates that intersystem crossing occurs from the triplet state to either the singlet state or the quintet state around the equilibrium geometry of the triplet state. The quintet state is as stable as the singlet state in their equilibrium geometries. All Fe-N bonds of the quintet state are longer than those of the singlet state by about 0.19 A. These are consistent with the general understanding that the Fe-ligand distances are considerably different but the relative stability is little different between the low spin and high spin states in LIESST complexes. Actually, a large activation barrier is calculated for the conversion between the singlet and quintet states, which is enough to suppress thermal spin transition and tunneling between them. The d-d transition energies are calculated with the TD-DFT method to be 2.05, 2.07, and 2.09 eV in the singlet state and 1.46 and 1.64 eV in the quintet state. Because of the significantly large difference in excitation energy between the singlet and quintet states, irradiation of visible light with different wavelengths selectively induces the excitation to the singlet excited state or the quintet one. All these results are consistent with the fact that both LIESST and reverse-LIESST are observed in mer-Fe(II)(2-pic)(3). The fac-isomer is also useful for the LIESST/reverse-LIESST, though the mer-isomer is better. In the Fe(iii) analogue, mer-Fe(III)(2-pic)(3), the DFT-computational results indicate small activation barriers and a large overlap of absorption spectra between the doublet and sextet states. Also, the Fe(III)-N bond distances are less different between the low spin and high spin states than the Fe(II)-N ones, leading to the narrow potential wall between the doublet and sextet states. As a result, the LIESST and reverse-LIESST cannot be observed in this Fe(iii) complex.