benjamin.hourahine at strath.ac.uk
Wed Jun 1 22:49:49 CEST 2011
On 01/06/11 17:46, Vladimir Yushkevich wrote:
> Dear DFTB+ users,
> By using waveplot I obtain total charge density (according to manual).
> Indeed, I think it is electronic charge density (integration over all
> volume of supercell produces the total number of electrons in my
> system). Then in my calculations ionic charge density is added as a
> set of point charges. Now I can get "total charge density" (sum of
> electronic and ionic parts) and analyze the average distribution over
> one of Cartesian coordinate. My problem is that point charges cause
> cusps near atomic positions. I suppose that point charges in DFTB+
> code approximated by (for example) Gaussian function.
> I need to know what function is used for approximation of spatial
> distribution of the nuclear charge and core electrons?
> Best regards, Vladimir Yushkevich
The number of electrons is only those in valence states, and the
waveplot density is from
only these electrons, the core states and nuclear charge are not
provided. For the standard
non-relativistic parameters, the nuclear charge is at a point and the
core states are atomic
Kohn-Sham orbitals, but these are discarded during the parameterization
parameters use atomic Slater orbitals as a basis so the cusp is
implicitly present). Can
you explain what property you are calculating that requires the total
Dr. B. Hourahine, SUPA, Department of Physics,
University of Strathclyde, John Anderson Building,
107 Rottenrow, Glasgow G4 0NG, UK.
+44 141 548 2325, benjamin.hourahine at strath.ac.uk
The University of Strathclyde is a charitable body,
registered in Scotland, with registration number SC015263
More information about the DFTB-Plus-User