about PAW

Message

xffan · #1 Post by **xffan** » Tue Mar 27, 2007 9:09 am

In VASP, The interaction between ions and electrons is described using US-PP or PAW. whether I should consider the GGA or LDA in calculation when I want to using PAW? If I should use the PAW-GGA for POTCAR file when I consider GGA using PAW?

When I download the PAW files, I find there are 3 files for PAW named potpaw_GGA, potpaw_PBE and potpaw. what is the meaning of the 3 files?

thanks!

#2 Post by **admin** » Tue Mar 27, 2007 11:55 am

please have a look at the pseudoppdatabase.pdf file which you can download from
http://cms.mpi.univie.ac.at/vasp-workshop

placeb0 · #3 Post by **placeb0** » Wed Jun 16, 2010 6:44 pm

2 Questions:

(1)
In the above link, under Session 11: Pseudopotential Data Base, slide 9 states:

"the PBE implementation follows strictly the PBE prescription, whereas the PW91
description is not that strict"

What exactly is meant by "strict"?

(2)

The fifth line in a given potpaw_PBE/X/POTCAR file reads:
LEXCH = PE
Does this mean that the "pseudopotentials" (i.e. the POTCARs) themselves were generated within the PBE formulation of the GGA?

If so, then I have a third question:

If I were to set the GGA flag in the INCAR file to say GGA = 91 and use a POTCAR from the potpaw_PBE data base, would the results reflect core states calculated within the PBE framework and valence (interstitial) states calculated withing the PW91 framework? This i more of a thought experiment type of question to help me understand the usage of the LEXCH and GGA flags. I don't actually plan on implementing it.

Thanks in advance...