• Print
• Login

Introduction:

FLEUR is a new FLAPW-Code the development of which is partly funded by the EUROPEAN RESEARCH NETWORK (Psik). The FLAPW-Method (Full Potential Linearized Augmented Plane Wave Method) is an all-electron method which within density functional theory is universally applicable to all atoms of the periodic table and to systems with compact as well as open structures. It is widely considered to be the most precise electronic structure method in solid state physics. Due to the all-electron nature of the method, partially occupied f-states in Lanthanides can be treated, magnetism is included rigorously and nuclear quantities e.g. isomer shift, hyperfine field, electric field gradient (EFG), and core level shift are calculated routinely. Also open systems such as surfaces, deposited and free wires and chains, tubular systems, clusters or inorganic molecules represent no basic problem. The capability of calculating the forces exerted on the atoms within the LAPW method opens the gate to structure optimization and molecular dynamics and puts this method up on the same category as the widespread pseudopotential method, but able of treating systems pain-full or unattainable by the pseudopotential method. The FLEUR-code is partly developed in collaboration with other groups, please refer to the group page. In particular, it comprises the following features.

Features in detail:

1. Bulk, film and wire geometry. FLEUR can use a truly two-dimensional basis-set or one dimensional basis-set and can thus calculate ultra-thin films or wires, without the need of supercell setups.
2. More than one energy panel
3. Local orbitals to supplement the LAPW-basis.
4. Forces exerted on the atoms
5. Electric fields applied to surfaces.
6. Spin-Orbit interaction in combination with magnetism.
7. Non-collinear magnetic structures with and without external constraints, spiral spin density waves.
8. Full-relativistic treatment of core electrons (Dirac + Magnetism).
9. LDA XC: X-α, Wigner, MJW, VWN, Perdew-Zunger, BH; GGA XC: PW91, PBE.
10. LDA+U
11. Quasi-Newton-Methods to accelerate charge self-consistency.

Technical features:

1. Second variation techniques to speed up eigenvalue problem
2. Downfolding into subspace to speed up eigenvalue problem
3. Runs on a wide range of architectures due to the use of standard libraries, including Linux PCs and DEC Unix Workstations as well as Cray T3E and IBM supercomputers.
4. Massive parallelization: Both the k-point loop and the eigenvalue-problem are parallelized and scale very well on MPI machines.