The vibrational spectrum of FeO{sub 2}{sup +} isomers—Theoretical benchmark and experiment

• Published in: The Journal of chemical physics 2014-05, Vol.140 (20)
• Main Authors: Maier, Toni M., Boese, A. Daniel, Sauer, Joachim, Wende, Torsten, Fagiani, Matias, Asmis, Knut R., Wilhelm-Ostwald-Institut für Physikalische und Theoretische Chemie, Universität Leipzig, Linnéstrasse 2, D-04103 Leipzig
• Format: Article
• Language: English
• Subjects: CORRECTIONS; DISSOCIATION; ENERGY DENSITY; INORGANIC, ORGANIC, PHYSICAL AND ANALYTICAL CHEMISTRY; IRON IONS; IRON OXIDES; ISOMERS; MATHEMATICAL METHODS AND COMPUTING; MOLECULES; MULTIPLICITY; PERTURBATION THEORY; PHOTOLYSIS; RELATIVISTIC RANGE; SPECTRA; SPIN; VALENCE; YIELDS
• ISSN: 0021-9606; 1089-7690
• DOI: 10.1063/1.4878667

Infrared photodissociation is used to record the vibrational spectrum of FeO{sub 2}{sup +}(He){sub 2–4} which shows three bands at 1035, 980, and 506 cm{sup −1}. Quantum chemical multi-reference configuration interaction calculations (MRCISD) of structures and harmonic frequencies show that these bands are due to two different isomers, an inserted dioxo complex with Fe in the +V oxidation state and a side-on superoxo complex with Fe in the +II oxidation state. These two are separated by a substantial barrier, 53 kJ/mol, whereas the third isomer, an end-on complex between Fe{sup +} and an O{sub 2} molecule, is easily converted into the side-on complex. For all three isomers, states of different spin multiplicity have been considered. Our best energies are computed at the MRCISD+Q level, including corrections for complete active space and basis set extension, core-valence correlation, relativistic effects, and zero-point vibrational energy. The average coupled pair functional (ACPF) yields very similar energies. Density functional theory (DFT) differs significantly from our best estimates for this system, with the TPSS functional yielding the best results. The other functionals tested are BP86, PBE, B3LYP, TPSSh, and B2PLYP. Complete active space second order perturbation theory (CASPT2) performs better than DFT, but less good than ACPF.