FNO : another challenging molecule for correlation methods

Although FNO has long been considered a computationally challenging molecule, no thorough study of this species has been performed with methods of electron correlation beyond second-order Mo/ller–Plesset perturbation theory (MP2) and the singles and doubles configuration interaction (CISD) method. W...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:The Journal of chemical physics 1993-07, Vol.99 (1), p.397-402
Hauptverfasser: DIBBLE, T. S, FRANCISCO, J. S
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Although FNO has long been considered a computationally challenging molecule, no thorough study of this species has been performed with methods of electron correlation beyond second-order Mo/ller–Plesset perturbation theory (MP2) and the singles and doubles configuration interaction (CISD) method. We have investigated the equilibrium structure, vibrational spectra, and relative energetics of FNO using high-levels of ab initio theory. Methods include third and fourth-order Mo/ller–Plesset perturbation (MP3 and MP4), singles and doubles coupled-cluster (CCSD) theory, and the CCSD(T) method, which incorporates a perturbational estimate of the effects of connected triple excitations. The quadratic configuration interaction method including single, double, and triple excitations [QCISD(T)] method is also used. Computations using these methods are performed with medium and large basis sets. Geometries calculated at the Hartree–Fock level are grossly in error, while going to the CISD, MP3, MP4, and CCSD methods produces increasingly better results. The CCSD(T) and QCISD(T) methods yield quite accurate structures; surprisingly, the MP2 results are almost as good. The quality of the calculated relative energetics closely follows that of the structure predictions, except that the MP2 level of theory now performs slightly worse than MP3, MP4, and CCSD methods.
ISSN:0021-9606
1089-7690
DOI:10.1063/1.465763