Answers are grounded in this paper and cite their sources.
In many situations, the most long-ranged parts of the exchange and correlation holes surrounding an electron cancel one another. Apparently for this reason, local spin density and generalized gradient approximations are more accurate for exchange and correlation together than for either alone. A study is made of the ability of these density functionals, and also the unmodified second-order gradient expansion, to describe various short-range effects in atoms: the correlation contribution to the interacting kinetic energy, the antiparallel-spin correlation energy, and the correction to the random phase approximation. Generalized gradient approximations, constructed with no adjustable parameter from the electron gas of slowly varying density, are found to give results of useful accuracy for real atoms. Prospects are discussed for use of the new functionals to improve the accuracy of electronic-structure calculations.
No takes yet. Share an insight, caveat, or question.
John P. Perdew (1993) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: