bad spin dependent partial charge density for hybrid functional
Posted: Tue Jun 28, 2011 9:15 pm
There seems to be a problem with the PARCHG files generated for hybrid functionals in spin polarized configurations. Specifically, both densities (total and spin) are identical in the PARCHG file when exact exchange is active. Total charge density (CHGCAR) is OK, and PARCHG for conventional functionals are OK too.
From poking around the code, which frankly I admit to not fully understanding, the problem seems to be related to the ISYM, which is set differently by default for conventional PAW (2) and hybrid functionals (3).
This is with the latest 5.2.11 I could find, dated Jan 18 2011 on the ftp site. The web portal version claims to be older (Nov 2010).
Is this a known problem? Any ideas for fixes or workaround?
thanks,
Noam
From poking around the code, which frankly I admit to not fully understanding, the problem seems to be related to the ISYM, which is set differently by default for conventional PAW (2) and hybrid functionals (3).
This is with the latest 5.2.11 I could find, dated Jan 18 2011 on the ftp site. The web portal version claims to be older (Nov 2010).
Is this a known problem? Any ideas for fixes or workaround?
thanks,
Noam