VSC School Seminar: Simulating solids with WIEN2k


    Nowadays advanced materials become more and more complex, because specific applications require corresponding materials with sophisticated properties. An understanding of the relation between those interesting properties and certain materials characteristics (e.g. structure or composition) is crucial in order to predict new properties or design advanced materials. Simulations of materials properties are one way to gain such insights. Many different methods based on different models and different approximations exist. The program package WIEN2k [1] utilizes Density Functional Theory (DFT) to describe materials and allows to calculate their properties using quantum mechanical simulations of the electronic structure of the material. During these simulations the Schrödinger equation of a given system is solved numerically. The computationally most demanding part of these calculations is the solution of a hermitian (or symmetric) generalized eigenvalue problem – where the sizes of the corresponding matrices depend on the number of atoms in (and the complexity of) the unit cell of the material. Unfortunately the simulation of complex materials depends cubically on the number of atoms in the unit cell and is computationally expensive for cells with more than 100 atoms.

    In the first part of the talk an introduction to DFT is given. Then the performance of WIEN2k in a high performance computing (HPC) environment is discussed and modifications (e.g. concerning algorithms or libraries) that have been implemented within the scope of the VSC School Project “Simulation of solids using WIEN2k” to improve this performance are presented. In the second part of the talk two applications that benefit from an efficient usage of HPC resources will be presented: For a study on the adsorption behaviour of water on mixed alkaline earth oxide surfaces (with both Mg- and Ca-atoms at the surface) [2] large supercells were simulated leading to an understanding of the improved solvation of MgO/CaO mixtures. For this project efficient MPI-parallel eigensolvers for the solution of the corresponding large eigenvalue problems were necessary. Another project concerns the design of a big test set to compare the accuracy of different existing DFT-codes and validate new ones as specified within the Delta Project [3]. This test set should contain experimentally known solids as well as hypothetical ones with specific properties in order to capture the vast chemical and structural diversity of real materials. To find suitable experimental structures a large number of known solids have to be screened and although a single calculation of that type can be done in a few minutes on a single core, the sheer number of necessary calculations requires efficient high-throughput execution.

    This talk is based on work supported by the VSC Research Center funded by the Austrian Federal Ministry of Education, Science and Research (BMBWF) and reports about the VSC School Project "Simulation of solids using WIEN2k". Additional funding has been obtained from the Research Foundation Flanders (FWO) (project No. G0E0116N).

    [1]   Blaha, P., et.al., www.wien2k.at .
    [2]   Müller, D., et al., Adv. Sustainable Syst. 2018, 2, 1700096.
    [3]   Lejaeghere, K., et al., Science 351, aad3000 (2016).


    Thomas Ruh
    Institute of Materials Chemistry, TU Wien, and Center for Molecular Modelling, Ghent University



    Date, Time, and Location:

    16.04.2018, 15:00 - 17:00, FH Hörsaal 2 (TU Wien, Wiedner Hauptstraße 8-10, 2nd floor, yellow area)


    If you would like to join us for this event, please email to: vsc-seminar@list.tuwien.ac.at
    (so we will know how many persons will come)


    The course material will be available for registered attendees only.


    Upcoming and past events of the VSC School Seminar:

    Opens external link in new windowvsc.ac.at/training