Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
O: Fachverband Oberflächenphysik
O 64: Focussed session: Theory and computation of electronic structure: new frontiers V (jointly with HL, DS)
O 64.2: Vortrag
Donnerstag, 17. März 2011, 11:45–12:00, TRE Phy
What is the G0W0 band gap of ZnO? — Martin Stankovski1, •Gabriel Antonius2,1, David Waroquiers1, Anna Miglio1, Hemant Dixit3, Patrick Rinke4, Hong Jiang4, Matteo Giantomassi1, Xavier Gonze1, Michel Côté2, and Gian-Marco Rignanese1 — 1IMCN-NAPS, Université catholique de Louvain,B-1348 Louvain-la-Neuve, Belgium — 2Département de physique, Université de Montréal , Montréal, Canada — 3CMT-EMAT, Departement Fysica, Universiteit Antwerpen, Groenenborgerlaan 171, B-2020, Antwerpen, Belgium — 4Fritz-Haber-Institut, Berlin-Dahlem, Germany
Zinc oxide is known to be a challenging system for G0W0 calculations. Its theoretical description has been widely discussed recently, and authors do not agree on the value of the band gap one should obtain from the G0W0 method. In an attempt to clarify the situation, we study the accuracy and the convergence properties of many schemes or approximations used at each level of the calculation, and show how different procedures may lead to very different conclusions. We first invest the sensitivity of the final band gap on the initial exchange-correlation potential used to generate the Kohn-Sham structure. We then study the behaviour of various plasmon pole models used to reproduce the dynamical properties of the dielectric matrix and discuss their validity for this particular system. Finally, the pseudopotential approach is compared to the PAW formalism, equivalent to an all-electrons calculation.