One more thing: you should use the options VX_TAU and VX_TAU-TAUW only
with PBE clmsum/vrespsum (i.e., post-PBE plotting). Otherwise the plots
will be wrong.

On Fri, 12 Feb 2016, wrote:


Concerning units, it's a mess in WIEN2k. In the file
$WIENROOT/SRC_lapw0/vxclm2.f (version 14)
I can see that if the keyword VX_TAU is used then the units of
tau^{KS} are Hartree/bohr^3.
However, with VX_TAU-TAUW, the units of tau^{KS}-tau^{W} are Ryd/bohr^3.

I guess that ATU and ANG should provide values in /bohr^3 and /Ang^3,

For spin-unpolarized calculations, case.r2v contains tau_{up} such that
the total tau is tau=2*tau_{up}.
For spin-polarized calculations, case.r2v and case.r2vdn contain
tau_{up} and tau_{dn} such that tau=tau_{up}+tau_{dn}.

F. Tran

On Fri, 12 Feb 2016, Nikola Novakovic wrote:

Hello everybody,

just to be sure, which units are used for TAU (and TAUW) in r2v file
from lapw0?
According to UG, the units for energy density should be Ryd/bohr^3?
does ATU/ANG flag in in5 file in any way change units if r2v file with
TAU(-TAUW) data is passed to lapw5 as input in r2v file (since lapw5
not know the type of data and their units)?

To my knowledge, r2v file contain only spin-up data, even in non spin
polarized case? So every output should be multiplied by two?

Thanks in advance,

Novakovic Nikola
Laboratory for Nuclear and
Plasma Physics
VINCA Institute of Nuclear Sciences
POB 522, 11001 Belgrade
Phone: +381 11 3408 610

Wien mailing list

Reply via email to