Answers are grounded in this paper and cite their sources.
One of J. A. Pople's leading contributions in ab initio quantum chemistry was his 1979 paper in which he and his colleagues presented calculations for the analytic second derivatives of the Self Consistent Field Potential Energy surface. This led to the automatic characterization of stationary points as minima or transition states. Recently there has been an upsurge of interest in computational density functional theory (DFT). Here we describe our implementation of analytic gradients and second derivatives for the Kohn‐Sham potential energy surface. This parallels similar developments by Pople and his coworkers. We examine Pople's idea of differentiating the weights in the quadrature scheme to ensure that the energy gradient is exactly zero at the energy minimum. We present some calculations on small molecules which are as near ‘exact’ as possible for the LDA and BLYP functional which we use. In other words we use large basis sets and a large number of quadrature points to evaluate the extra (nonanalytical) integrals. We do not use any fitting procedures and no other approximations are introduced.
No takes yet. Share an insight, caveat, or question.
Handy et al. (1993) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: