This is the authors' abstract. We don't add key points for this paper.
We present a full-potential band-structure scheme based on the linear combination of overlapping nonorthogonal orbitals. The crystal potential and density are represented as a lattice sum of local overlapping nonspherical contributions. The decomposition of the exchange and correlation potential into local parts is done using a technique of partitioning of unity resulting in local shape functions, which add exactly to unity in the whole crystal and which are very easily treated numerically. The method is all-electron, which means that core relaxation is properly taken into account. Nevertheless, the eigenvalue problem is reduced to the dimension of a minimum valence orbital basis only. Calculations on sp and transition metals give results comparable to other full-potential methods. The calculations on the diamond lattice demonstrate the applicability of our approach to open structures. The consequent local description of all real-space functions allows the treatment of substitutional disordered materials.
No takes yet. Share an insight, caveat, or question.
Koepernik et al. (1999) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: