jp037139r_si_001.pdf (55.15 kB)

Development of a Theoretical Model of Nickel(II) Tris(oxime)amine Dichloride with Comparison to X-ray Crystallography and Vibrational Spectroscopy

Download (55.15 kB)
journal contribution
posted on 22.04.2004 by Rebecca M. Jones, Michael J. Baldwin
Using modern density-functional theory and vibrational spectroscopy, various theoretical models are evaluated for application to Ni(TRISOXH3)Cl2, the precursor to a novel oxygen-activating complex. Geometry optimizations and frequency calculations were performed using Gaussian 98 at the Hartree−Fock and B3LYP theory levels. Theoretical data are compared to the published X-ray crystal structure and to vibrational spectra obtained using Raman and infrared methods. The geometry closest to the crystal structure resulted from B3LYP/6-31G(d) & 6-311++G(d,p), where the higher basis set was applied to the metal center. This gas-phase model has a 1.0% average deviation from the experimental bond distances in the solid-phase structure. A lower level calculation (B3LYP/6-31G(d) & 6-311+G(d), again with the higher basis set applied to the Ni) yields vibrational frequencies that are as accurate as the higher level calculation. Three isotope-substituted TRISOXH3 ligands were synthesized (with 15N amine, 15N oximes, or deuterated oximes); the isotope shifts observed in the vibrational spectra of the Ni complexes were compared to the theoretical shifts. With this empirical information, 28 key normal modes of the vibrational spectra were assigned. A modified wavenumber−linear scaling procedure was applied to the assigned theoretical wavenumbers and yielded a 3-fold improvement in absolute deviation from experimental frequencies over the unscaled values.