GW calculation

Moderators: gmatteo, bruneval, rigna

Post Reply
Newlife1982
Posts: 37
Joined: Thu Jun 24, 2010 2:13 pm

GW calculation

Post by Newlife1982 » Mon Jun 28, 2010 1:42 pm

Dear All,

I have a problem with input variable 'kssform' and the error is as followings:

The present version of abinit does not accept pseudopotentials with more than one projector per angular momentum channel when nbandkss/=0 and kssform ==1,2.
Use kssform = 3 in your input file


For the pseudopotentials, I used the one with semi-core. If I change the kssform to 3, there is no .KSS file.

Please give me some suggestions.

Thank you in advance for your replying.

Wang

bruneval
Posts: 40
Joined: Mon Aug 17, 2009 11:38 am

Re: GW calculation

Post by bruneval » Tue Jun 29, 2010 1:46 pm

Which element? which semicore? which pseudo?

Newlife1982
Posts: 37
Joined: Thu Jun 24, 2010 2:13 pm

Re: GW calculation

Post by Newlife1982 » Tue Jun 29, 2010 5:06 pm

Thank you for your replying.
I calculate the band gap of the crystal TiO2 using LDA.
For Ti, I use Hartwigsen-Goedecker-Hutter pseudopotentials with zion = 12. While I use Teter "extended norm-conserving" pseudopotentials for Oxygen with zion = 6.

Wang

bruneval
Posts: 40
Joined: Mon Aug 17, 2009 11:38 am

Re: GW calculation

Post by bruneval » Tue Jun 29, 2010 5:46 pm

In this type of pseudo, you have several projectors per angular momentum. This is not allowed so far in the GW implementation of abinit.
Try to use instead traditional troullier-martins pseudos.

Fabien

Newlife1982
Posts: 37
Joined: Thu Jun 24, 2010 2:13 pm

Re: GW calculation

Post by Newlife1982 » Wed Jun 30, 2010 9:40 am

Fabien

Thank you very much!

Wang

Post Reply