Abstract
Homolytic bond dissociation energies (BDE) have been calculated for a series of organometallic XH 3-Y compounds, where X is one of the Group 14 atoms (carbon, silicon, germanium, tin and lead) and Y is one of 68 ligands ranging from the hydrogen atom to the methylacetate and OClO 3 groups. Several ab initio and hybrid density functional theory (DFT) levels and methods have been tested and these are compared with each other and with experiment, where available. For the ab initio studies all atoms beyond hydrogen used a Hartree-Fock (HF) or relativistic analogue compact effective potential (CEP) to replace the core electrons. Extended basis sets were used for the valence electrons including polarization and diffuse functions. For the DFT calculations both all electron (AE) and CEP methods were applied. The results show that, except for the fragment radicals (Y) that have excessively large calculated 〈 S 2〉 values, MP2 gives the best average agreement with experiment for the CH 3-Y and SiH 3-Y sets, compared to MP3, MP4 and CCSD(T). For the fragment radicals with large 〈 S 2〉 values projected-MP2 and CCSD(T) give generally comparable results. The calculated DFT/AE BDE are similar but usually slightly lower than MP4/CEP while the DFT/CEP binding energies are consistently lower and farther from experiment. CCSD(T) usually gives lower binding energies than MP n. The appropriateness of using HF-type effective potentials and valence electron basis sets with DFT needs to be examined closely.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have