Thomas, In the past, I filled a small set of cubes within the cores in the valence charge density CUBE file made by PP, with the contribution of Slater effective core states:
http://dx.doi.org/10.1007/s00214-011-0955-3 But I am sure that now there are much more elegant solutions. At that time all the PAW stuff was not working properly through PP. I am quite confident that a uniform spreading of the core charge within the same cubes would have the same effect within the numerical errors of the later charge analysis performed in direct space. Maybe, even a single-point addition in the cube to which the nucleus belongs would give the same result. I can send you the barbarian code, if you want. Regards, Giovanni La Penna National research council of Italy (CNR) Institute for chemistry of organometallic compounds (ICCOM) Sesto Fiorentino (Firenze), Italy tel.: +39 0555225264 / skype: giovannilapenna On Tue, 6 May 2014, Thomas Gruber wrote: > Dear all, > > I still have a problems with pp.x in version 5.0.3 and also in version 5.0.99 > for Si bulk. Since I used PAW pseudopotentials I wanted to run the bader > analysis from "http://theory.cm.utexas.edu/bader/"; and like to have all > electrons and not just the 4 valence electrons. But when I create the cube > file I do not get all electron with the bader analysis. Did anyone tried the > all electron reconstruction? Even with an FFT grid of 200x200x200 instead of > the standard 36^3 I just get a total number of electrons of 8 for 2 Si-Atoms. > Can someone tell me, if it works for any version of QE with the PAW > pseudopotentials from the website to get all electron charge density cube > file? > > Thanks in advise. > > -- > TU Bergakademie Freiberg > > Dipl. Chem. Thomas Gruber > Institut f?r Theoretische Physik > Leipziger Str. 23 / OG. 14 > D-09599 Freiberg > > Tel: +49 3731 39-2006 > Email: thomas.gruber at physik.tu-freiberg.de > Webseite: tu-freiberg.de > >
