Abstract

Since the late 1960s molecular orbital (MO) theory has dominated the development of ab initio quantum chemistry. Specifically, the Hartree­ Fock-Roothaan (HFR) (1-3) formalism of the self-consistent field (SCF) method, which was greatly facilitated by the development of gaussian basis sets as suggested by Boys (4, 5), and the post-Hartree-Fock methods of configuration interaction (CI) (6-8), multiconfiguration SCF (MCSCF) (9, 10), many-body perturbation theory (MBPT) and coupled-cluster methods (CCM) (11), and Moller-Plesset perturbation theory (MPPT) (12). With the advent of high speed vector supercomputers the field has virtually exploded; larger molecules or very high accuracy for small molecules are now possible. In addition to such capability has come more accurate knowledge of the size of I-particle and N-particle basis sets required for high accuracy (13-16). Not only does the use of large basis sets require large memory and file-storage capability, but HFR and post-HFR algorithms contain significant bottlenecks to vectorization and parallelization. Although many new and efficient matrix methods have been developed

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call