WIEN2k

WIEN2k
Name	WIEN2k
Developer	R. L. and Other Developers
Released	1990s
Latest release	ongoing
Programming language	Fortran, C, shell
Operating system	Unix, Linux
License	Proprietary Academic, Commercial

WIEN2k is a software package for electronic structure calculations using the full-potential (linearized) augmented plane-wave and local orbitals method. It is widely used in condensed matter physics, materials science, and chemistry for first-principles simulations of solids, surfaces, and interfaces. The package is associated with numerous studies involving crystal structure analysis, magnetism, and optical properties across academic and industrial research groups.

Overview

WIEN2k was developed by researchers associated with institutions such as the Vienna University of Technology, and its development and dissemination intersect with groups at the Max Planck Institute, Harvard University, Massachusetts Institute of Technology, Stanford University, University of Cambridge, California Institute of Technology, University of Tokyo, ETH Zurich, University of California Berkeley, and Los Alamos National Laboratory. The method implemented links to foundational work by Walter Kohn, Pierre Hohenberg, and Lev Landau through theoretical lineage including density functional approximations by John P. Perdew, Kieron Burke, and Matthias Ernzerhof. Users often compare WIEN2k outputs to results from codes like VASP, Quantum ESPRESSO, ABINIT, SIESTA, CASTEP, ELK, FPLO, FLEUR, GPAW, CRYSTAL, WIEN97, WIEN96, WIEN, DMFT implementations, and WienX. Collaborations and citations connect to journals such as Physical Review Letters, Physical Review B, Journal of Chemical Physics, Nature Materials, Science, and Journal of Physics: Condensed Matter.

Features and Capabilities

The package implements full-potential linearized augmented plane-wave (FP-LAPW) and augmented plane-wave plus local orbitals (APW+lo) techniques, enabling precise total-energy, band-structure, density-of-states, and Fermi-surface computations. It supports exchange-correlation functionals like LDA, GGA, meta-GGA, and hybrid approaches influenced by Perdew–Burke–Ernzerhof and Heyd–Scuseria–Ernzerhof formulations. WIEN2k handles spin-polarized, spin–orbit-coupled, non-collinear magnetism, and relativistic treatments relevant to studies involving heavy elements such as uranium, plutonium, gold, platinum, mercury, and lead. Advanced modules facilitate optical spectra, electron energy-loss spectra, hyperfine interactions, electric field gradients, and NMR shifts, comparable to analyses performed with methods in Dynamical Mean Field Theory (DMFT), GW, Bethe–Salpeter Equation, and time-dependent density functional theory as seen in collaborations with codes like YAMBO, BerkeleyGW, and TRIQS.

Methodology and Algorithms

The core algorithm partitions space into muffin-tin spheres and interstitial regions, expanding wavefunctions in spherical harmonics inside spheres and plane waves outside, a lineage traceable to methods refined by Slater and Andersen. Linearization and variational principles draw on the work of Bloch, Kohn, and Sham, while numerical strategies use iterative diagonalization, Davidson or Lanczos methods, and Broyden density mixing. Brillouin-zone sampling leverages Monkhorst–Pack grids, tetrahedron methods, and Wannier functions for interpolation, echoing techniques employed in studies at Oxford, Cambridge, Columbia, Princeton, and MIT. Convergence control uses k-point meshes, RKmax parameters, and augmented basis optimizations, with parallelization strategies employing MPI, OpenMP, and BLAS/LAPACK libraries optimized on clusters from Cray, IBM, Dell, Hewlett-Packard, and Intel-based supercomputers.

Installation and System Requirements

WIEN2k is distributed for Unix-like systems, primarily Linux distributions such as Ubuntu, CentOS, Red Hat Enterprise Linux, SUSE, and scientific clusters running SLURM or PBS schedulers. Compilation requires Fortran compilers such as GNU Fortran, Intel Fortran, or PGI, and libraries including FFTW, BLAS, LAPACK, ScaLAPACK, and sometimes MKL for performance on Intel architectures. Typical deployments occur on workstations from Lenovo, HP, Dell Precision, and on HPC resources at national labs like Oak Ridge, Argonne, Lawrence Berkeley, and at university clusters at Imperial College, University of Toronto, EPFL, and Seoul National University. System administrators often integrate modules for environment management via Lmod or Environment Modules.

Usage and Workflow

Users begin with crystal structure input from databases and tools like ICSD, Materials Project, AFLOW, COD, and Crystallography Open Database, and visualization via VESTA, XCrySDen, Jmol, or ParaView. Workflows follow geometry setup, initialization, self-consistent field cycles, band-structure and density-of-states calculations, and post-processing for properties such as optical spectra or transport coefficients, often interfacing with BoltzTraP, Wannier90, Phonopy, and ASE. Scripts automate tasks for high-throughput studies, reproducibility, and data management in environments linked to GitHub, GitLab, and institutional repositories at CERN, SLAC, and DESY. Training and user support occur at summer schools and workshops organized by institutions including the European XFEL, Max Born Institute, and the American Physical Society.

Applications and Examples

Applications span prediction of electronic band gaps in semiconductors like silicon, gallium arsenide, and indium phosphide; magnetic anisotropy in iron, cobalt, and nickel compounds; topological phases in bismuth, antimony, and topological insulators; superconductivity studies in cuprates, iron pnictides, and MgB2; and defect energetics in oxides such as TiO2, ZnO, and SrTiO3. Case studies connect to experiments at facilities including ESRF, APS, Diamond Light Source, and PETRA III, and to technologies developed by Siemens, Samsung, IBM Research, Toyota, and General Electric. Cross-disciplinary work links to catalysis research at MIT and Caltech, battery materials at Argonne and Toyota, and thermoelectrics investigated at the University of Houston and Oak Ridge National Laboratory.

Development and Licensing

Development historically involves academic groups, consortia, and contributors from institutions like TU Wien, MPI, and various university research groups in Europe, North America, and Asia. Licensing is proprietary with academic and commercial options; institutional agreements and site licenses govern distribution, with training, documentation, and user forums maintained by developer teams and community mailing lists. Integration with open-source tools and collaborative projects continues through joint efforts with groups contributing to open data initiatives such as Materials Project, NOMAD, and the Open Quantum Materials Database.

Category:Science software