Hello,
i am using abinit to calculate magnetization of cobalt i have done my job but i think my result are unphysical.
but i studied from paper that magnetization is strongly dependent to npc(no partial core),spc,mpc,bpc.but i can't underestand them in abintit.how can i choose them in abinit? could you please explain about them?
thank you.
small partial core
Re: small partial core
What pseudopotential or PAW data set are you using now?
Josef W. Zwanziger
Professor, Department of Chemistry
Canada Research Chair in NMR Studies of Materials
Dalhousie University
Halifax, NS B3H 4J3 Canada
jzwanzig@gmail.com
Professor, Department of Chemistry
Canada Research Chair in NMR Studies of Materials
Dalhousie University
Halifax, NS B3H 4J3 Canada
jzwanzig@gmail.com
Re: small partial core
Hello,
Thank you,as i said i am calculating magnetization of cobalt and i use HGH pesudopotential.27co.9.hgh.
Thank you for your help.
Thank you,as i said i am calculating magnetization of cobalt and i use HGH pesudopotential.27co.9.hgh.
Thank you for your help.
Re: small partial core
well, I have no experience with HGH so I may not be able to help. Have you tried using PAW? It should work well for this problem.
Josef W. Zwanziger
Professor, Department of Chemistry
Canada Research Chair in NMR Studies of Materials
Dalhousie University
Halifax, NS B3H 4J3 Canada
jzwanzig@gmail.com
Professor, Department of Chemistry
Canada Research Chair in NMR Studies of Materials
Dalhousie University
Halifax, NS B3H 4J3 Canada
jzwanzig@gmail.com
Re: small partial core
1) You need to explain what you mean by
npc(no partial core),spc,mpc,bpc
2) For many HGH potentials you can choose semicore states in the valence. If you need more you'll have to make the potential yourself. Non-linear core correction improves things too, and this is also easier to include in PAW, so I would follow Joe's advice
Matthieu
npc(no partial core),spc,mpc,bpc
2) For many HGH potentials you can choose semicore states in the valence. If you need more you'll have to make the potential yourself. Non-linear core correction improves things too, and this is also easier to include in PAW, so I would follow Joe's advice
Matthieu
Matthieu Verstraete
University of Liege, Belgium
University of Liege, Belgium