Answers are grounded in this paper and cite their sources.
In the original formulation of the MNDO method, the approximation of two-center two-electron integrals over p orbitals is not rotationally invariant, and the coulombtype integrals do not match their one-center counterparts. In recent computer codes the rotational invariance has been restored by dropping the original approximation for two-center exchange integrals over π orbitals, which at the same time disconnected these too from the one-center values. It is shown in this paper that the rotational invariance problem can be overcome in the framework of the original approximation. It is also demonstrated that the mismatch between the one- and two-center two-electron integrals may be the reason for the too short Si-Si bond lengths predicted by MNDO.
No takes yet. Share an insight, caveat, or question.
De�k et al. (1987) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: