The ABINIT Users Mailing List ( CLOSED )

[Xavier Gonze <gonze@pcpm.ucl.ac.be>]

Dear liuyudong,


liuyudong97@mails.tsinghua.edu.cn wrote:
> Dears all: I wish to know the charge density of an isolated atom, namely the 
> atoms are infinite far
> from each other. I guess that Abinit must have such a database of charge 
> density for each element,
> because the command "densty " CAN give "a rough description of the initial GS 
> density, for each
> type of atom" . Is it possible to get the charge density of an isolated atom 
> for each element
> without any SCF calculation? 1)If yes, how to read out the data? 2)If no, it 
> is necessary to setup
> a big supercell to get charge density of one atoms by simulating the lesson 1 of Abinit Tutorial? 
> Hopefully wish the first case is possible.
>
> ps. by the way, who knows the charge desnity file can be read by software 
> named AVS?

There is a database of all-electron LDA charge densities available
on the ABINIT Web site :
http://www.abinit.org/ABINIT/Psps/AE_DEN/aeden.html . It has been produced
thanks to the FHI98PP code by Matthieu Verstraete. For each element,
the first line gives
the number of radial point, than, in the first column, you have the
radii, and in the second column, the charge density (all in atomic units).
The utility cut3d uses such file to compute the Hirshfeld partition
of the charge of molecules or solids, see option 11 of cut3d  ,
routine hirsh.f , as well as the automatic test Test_v3#61 .

Inside the main ABINIT code, the database of charge densities is a very
rough guess, that should NOT be used for anything else than starting
SCF cycles.

If you want to have the charge density corresponding to a pseudopotential,
you might use FHI98PP , or perform an ABINIT run using a big supercell.

Xavier