[Pw_forum] Generating the Cs ultrasoft PP

Emine Kucukbenli kucukben at sissa.it
Wed May 18 02:43:00 CEST 2011

Dear Tram,
first, try adding more channels:
(as the warnings in the output of ld1.x have been asking you to)
6S 1 0 1.00 0.00 4.20 4.50
> add another s
6P 2 1 0.00 0.00 6.10 6.10
> add another p if you want to have different rcut and rcutus.
5D 3 2 0.00 0.00 3.00 3.00

try other local channels,
try tm=.true. to use Troullier-Martins pseudization,
try the ionized version like in the pseudo database,
try adding 5s and 5p as two other channels,
change rcuts, energies and trytrytry.

it might be of help if you find the generation file of the NC pseudo  
in the database, try asking the author.
have fun,
emine kucukbenli, phd student, sissa, italy

ps:you probably need non linear core correction with this pp.
pps:are cesium orbitals really that huge to use 6au for rcut?

   SISSA Webmail https://webmail.sissa.it/
   Powered by Horde http://www.horde.org/

More information about the users mailing list