An efficient method for calculating spatially extended electronic states of large systems with a divide-and-conquer approach

Abstract

We present an efficient post-processing method for calculating the electronic structure of nanosystems based on the divide-and-conquer approach to density functional theory (DC-DFT), in which a system is divided into subsystems whose electronic structure is solved separately. In this post process, the Kohn-Sham Hamiltonian of the total system is easily derived from the orbitals and orbital energies of subsystems obtained by DC-DFT without time-consuming and redundant computation. The resultant orbitals spatially extended over the total system are described as linear combinations of the orbitals of the subsystems. The size of the Hamiltonian matrix can be much reduced from that for conventional calculation, so that our method is fast and applicable to general huge systems for investigating the nature of electronic states.

9 pages, 9 figures

References in corpus (6)

• Nearsightedness of Electronic Matter
• O(N) methods in electronic structure calculations
• Linear scaling Krylov subspace method for large scale {\it ab initio} electronic structure calculations of metals
• Comparative study of hybrid functionals applied to structural and electronic properties of semiconductors and insulators
• Optimized multi-site local orbitals in the large-scale DFT program CONQUEST
• Efficient Calculations with Multisite Local Orbitals in the large-scale DFT Code CONQUEST