Dear Heather and QE Users Many thanks for your attention and answer. The case of study is a Ni atom which is connected to 4 carbon atoms in a graphite-like structure. The projected DOS calculation by projwfc.x (lowdin charge analysis) after relaxation procedure shows that the Ni electronic configuration is : 4s[0.4023] + 3d[8.8971]. As I know the initial guess of this pseudo-potential is 3d[9] 4s[1] for Ni and I wanna to know that if the initial guess was 3d[8] 4s[2], would I get 4s[0.4023] + 3d[8.8971] again after the relaxation procedure or not ? I appreciate any suggestion in advance.
Best Wishes * ----------------- Masoud Nahali PhD Student of Physical Chemistry Sharif University of Technology * Heather Kulik wrote : > > Dear Masoud, > > You are correct in that the pseudopotential was generated in the state 3d9 > 4s1. > > One minor point is that according to NIST ( > http://physics.nist.gov/PhysRefData/ASD/levels_form.html), the low-lying > experimental atomic energy levels of Ni are: > > ^3F(3d84s2) ~0 and ^3D(3d94s1) ~ <0.1 eV, ^1D(3d94s1) ~ <0.5 eV. > > However, the good news is that, the state in which the pseudopotential was > generated should not prevent you from accurately describing states that > have > 3d84s2 character since the core electrons from a 3d94s1 and 3d84s2 Ni > should > be described fairly similarly. > > You can also verify the character of your electronic states using any of > the > postprocessing codes (e.g. projected density of states from projwfc.x) > > Note: A more serious problem is that if your extended system (i.e. more > than > one atom) has numerous solutions close in energy comparable to the > character > of the atomic system, most x-c functionals will likely fail to predict the > correct ordering. > > You should try comparing the relative energies you get for a few > representative systems to experimental data where possible to identify if > the GGA (or LDA) suffers from any particular bias in the relative ordering > of these states. > > Hope that helps! > -- > Heather K > Postdoc at Stanford University > > > On Wed, Dec 15, 2010 at 12:31 AM, Masoud Nahali <masoudnahali at > live.com<http://www.democritos.it/mailman/listinfo/pw_forum> > >wrote: > >* Dear Quantum Espresso Users *>* *>* I have used Ni.pbe-nd-rrkjus.UPF pseudo-potential for nickel in my project. *>* In the info part of this pseudopotential has been written : *>* *>* Info: Ni PBE 3d9 4s1 RRKJ3 US *>* *>* But I know that the well established electron configuration for nickel is *>* "3d8 4s2" and now I am too worry that it may cause an error in my work. May *>* it ? *>* I appreciate your help to understand the point in advance. Many Thanks *>* *>* *>* *>* Best Wishes *>* * *>* *>* ----------------- *>* *>* Masoud Nahali *>* *>* PhD Student of Physical Chemistry *>* Sharif University of Technology *>* * *>* * ** ** -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.democritos.it/pipermail/pw_forum/attachments/20101216/6adaa95d/attachment.htm
