Friedeloszillationen und Bildungsenergien von Punktdefekten in Metallen

Deutz, J.; Zeller, R.; Dederichs, P. H.

Book/Report

FZJ-2018-01938

Friedeloszillationen und Bildungsenergien von Punktdefekten in Metallen

Deutz, J. (Corresponding author)FZJ* ; Zeller, R.FZJ* ; Dederichs, P. H.FZJ*

1982
Kernforschungsanlage Jülich, Verlag Jülich

Jülich : Kernforschungsanlage Jülich, Verlag, Berichte der Kernforschungsanlage Jülich 1805, 98 p. (1982)

Please use a persistent id in citations: http://hdl.handle.net/2128/17702

Report No.: Juel-1805

Abstract: Charge density oscillations ("Friedeloscillations") around substitutional 3d-impurities in Cu are calculated as well as formation energies of point defects in Cu. The calculations are based on density functional theory in the local spin density approximation and on the KKR-Green function method. The defect is described by a single perturbed muff in tin potential in an otherwise unperturbed muffin-tin lattice. As an important improvement in electronic strucutre calculations we introduce the method of complex energy integration, which represents an efficient and reliable way of determining charge density perturbations at neighbouring sites. Charge density- and magnetisation oscillations around 3d defects depend strongly on the position of the virtual bound state relative to the Fermi energy. In general our calculations confirm the preasymptotic behaviour as predicted by simple jellium models, although additional oscillations show up due to the orthogonalisation on thecore wave functions. Using the charge densities of the perturbed crystal in single site approximation, we calculate the energy of formation of the vacancy and 3d impurities as well as Al in Cu by taking advantage of the extemal properties of the total energy. The single particle energies are calculated using Lloyd's formula, while the Coulomb contributions are summed up explicitely over four neighbouring shells. In comparison with experimental formation energies our results are generally too high (e.g. 2.4 eV for the vacancy in Cu instead of 1.3 eV experimentally). The derivations from the experiments are likely to be caused by systematic errors, the removal of which is difficult but not impossible.

Contributing Institute(s):