Answers are grounded in this paper and cite their sources.
An improved method for calculating analytic approximations to the self-consistent-field (SCF) functions is developed. The numerically given function is divided by a polynomial having the same nodes as the function, and the monotonically decreasing quotient is then expressed as a sum of exponentials. An exact method for expanding an arbitrary function into exponentials is discussed, but the results turn out to be extremely sensitive even to rounding-off errors. For practical purposes a method of successive approximations is used instead; it works inwards from the tail towards the origin and, in each step, two exponentials are simultaneously determined. Applications are carried out to the argon-like ions (Cl^-, A, K⁺, Ca⁺²) and to the copper ion. By expressing a SCF function fₙₗ(r) in terms of ($n+1$) exponentials, the error may be brought below the order of magnitude 0.001-0.002. By using the data available for Ca⁺ and Cu⁺, analytic wave functions have finally been interpolated for the first row of the transition metals (Sc, Ti, V, Cr, Mn, Fe, Co, Ni); the degree of accuracy is discussed.
No takes yet. Share an insight, caveat, or question.
Löwdin et al. (1956) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: