ABINIT Discussion Forums

Dear abinit users,
Hi, I am doing some calculations with Gw for In2O3. I am using the 49-In-4d-LDA-fhi.txt and o8-o-LDA-fhi.txt pseudo potential s for In and O respectively. I want to know, is it suitable these pseudo potential s for Gw calculations?
Best wishes,
Asadi

GW pseudopotentials should include semicore states. For Indium, I'd advise to have one including 4s, 4p and 4d as valence.
I guess the one you propose is not suitable.
Concerning oxygen, no problem.

Fabien

Dear fabien,
Hi,thank you for your reply.
In the download part of abinit , i find the following semi-core pseudopotentials:LDA_HGH/49-In with pspcod=3 that couldn't run the program,LDA-FHI/49In and GGA_FHI/49In .
please guide me GGA_FHI/49In is Proper?
With good regards,
Asadi.

For sure, the HGH pseudopotential is not compatible with the current implementation of the KSS generator.

You should take a look at the header of the pseudopotential for Indium and discover how many electrons they consider as valence. 3 ? 13 ? or 21 ?
For proper GW calculation, I'd bet that you need a 21-electron pseudo.

Fabien

Dear Fabien ,

Thanks for your reply.

Best regards,
Asadi.

Dear Fabien,

In the download part of abinit , pseudopotentials including semicore states that is compatible with the current GW calculation is limited. Where can we find the more suitable pseudopotentials? It would be very nice if you could give me some suggestions. Thank you very much.

Best regards,

Wang