[Wien] Exchange-correlation energy

[tran at theochem.tuwien.ac.at]

Hi,

For 1 there is IGRAD.EQ.33 in vxclm2.f which is for exchange-only PBE
(then correlation c is c=xc-x). For such thing the correct procedure is to
plug the PBE orbitals (obtained from an usual PBE calculation) into the
exchange-only PBE:
1) replace 13 by 33 in case.in0
2) x lapw0

I' m not really sure to understand for 2.

For 3 there is the contributions from the spheres and interstitial
in case.output0. This is the 3rd column when you grep for "TOTAL="
(the values are in Hartree). This is not really possible to get easily
the contrbution from an arbitrary region of space. A solution
would be to use lapw5 to write on a file the values of exc on a
regaular 3d grid an then integrate yourself. But then the integration
close to the nuclei will be very inaccurate.

One thing: It is dangerous and not recommended to compare two values
of Exc when they were obtained from two different electron densities.

F. Tran


On Tue, 26 Nov 2013, Martin Gmitra wrote:

> Dear Wien2k users,
>
> Some time ago there was discussed topic about xc energy
> http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg01719.html.
> Since then Wien2k has KVC flag allowing to print :EXC contribution to
> the total energy. Could you please navigate me how:
>
> 1. one gets separate exchange and correlation contributions?
>
> 2. to get the separate contributions orbital resolved for each atom in the cell?
>
> 3. to get the xc contributions in a particular region (say cylinder =
> including part of interstitial and muffin-tins) of the cell?
>
> I would like to do it for PBE (option 13).
> Thanks in advance for help.
> Martin Gmitra
> Uni Regensburg
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:  http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
>