Re: Kohn Sham energy levels and Hartree potential ( No.1 ) 
 Date: 2014/01/28 09:43
 Name: T. Ozaki
 Hi,
I think that comparison between the KohnSham energy and the Hartree potential in the absolute value may lead to misreading, while there must be some correlation. Also, it is noted that OpenMX calculates the Hartree potential arising from difference charge density.
Regards,
TO

Re: Kohn Sham energy levels and Hartree potential ( No.2 ) 
 Date: 2014/01/28 12:37
 Name: Le The Anh
 Dear Ozaki sensei,
Thank you very much for your reply. I understand. Another thing, the zero energy in the KohnSham energy is the zero energy in the Hartree potential. Am I right? In OpenMX, I see four output files: *.v0.cube, *.vhart.cube, *.vna.cube, and *.vxc0.cube. I assume these files corresponding to KohnSham potential (excluded nonlocal part), difference Hartree potential (arising from difference charge density), neutral atom potential, and exchange correlation potential. Does the KohnSham potential equal the sum of three other potentials? Is the absolute zero energy in those potentials calculated for the vacuum level? If I have a supercell and I put my structure at the center, the empty space between my structure and the supercell is the vacuum. Dose the vacuum level correspond to this empty space?

Re: Kohn Sham energy levels and Hartree potential ( No.3 ) 
 Date: 2014/01/28 13:13
 Name: T. Ozaki
 Hi,
> Another thing, the zero energy in the KohnSham energy is the zero energy in the Hartree > potential. Am I right?
Though I cannot understand well what you meant, they can be related to each other. Could you explain more precisely about "the zero energy in the KohnSham energy is the zero energy in the Hartree potential"?
> In OpenMX, I see four output files: *.v0.cube, *.vhart.cube, *.vna.cube, and > *.vxc0.cube. I assume these files corresponding to KohnSham potential (excluded > nonlocal part), difference Hartree potential (arising from difference charge density), > neutral atom potential, and exchange correlation potential. > Does the KohnSham potential equal the sum of three other potentials?
The relations are given by v0 = vhart + vna + vxc0 veff = v0 + vnl where vnl is the nonlocal pseudopotential.
> Is the absolute zero energy in those potentials calculated for the vacuum level?
The vacuum level is generally obtained from the value of difference Hartree potential around vacuum region being far from the surface of a slab model, and the work function is defined to be difference between the value and the Fermi energy. In this sense, your statement may be true.
> If I have a supercell and I put my structure at the center, the empty space between > my structure and the supercell is the vacuum. Dose the vacuum level > correspond to this empty space?
Yes it does, as I wrote above.
Regards,
TO

Re: Kohn Sham energy levels and Hartree potential ( No.4 ) 
 Date: 2014/01/28 14:11
 Name: Le The Anh
 Dear Ozaki sensei,
>Could you explain more precisely about "the zero energy in the KohnSham energy is the zero energy in the Hartree potential"?
The KohnSham energy level is the KohnSham eigenvalue in the *.out file. I meant, for example, the LUMO is at 0.1eV with respect to the chemical potential 3eV. That means the absolute value of the LUMO level is 2.9eV with respect to the vacuum level. On the other hand, the vacuum level is set to be 0eV of the difference Hartree potential. That means the LUMO level stays at the 2.9eV of the difference Hartree potential if I plot the the KohnSham eigenvalue and the difference Hartree potential in the same figure. Am I right?
I thank you very much.
Sincerely,
The Anh

Re: Kohn Sham energy levels and Hartree potential ( No.5 ) 
 Date: 2014/01/28 14:37
 Name: T. Ozaki
 Hi,
I think that your statements are correct. (Please make sure that the KohnSham eigenvalues stored in the *.out file are not relative, but absolute values.)
Regards,
TO

Re: Kohn Sham energy levels and Hartree potential ( No.6 ) 
 Date: 2014/01/29 11:57
 Name: Le The Anh
 Dear Ozaki sensei,
I thank you very much for your reply.
As for the vnl nonlocal pseudopotential, is there any way to obtained this potential in the *.cube format?

Re: Kohn Sham energy levels and Hartree potential ( No.7 ) 
 Date: 2014/01/29 16:16
 Name: Le The Anh
 From the output file *.out, I see many kinds of energies:
Uene: band energy Ukin: kinetic energy UH0: electric part of screened Coulomb energy UH1: difference electronelectron Coulomb energy Una: neutral atom potential energy Unl: nonlocal potential energy Uxc0: exchangecorrelation energy for alpha spin Uxc1: exchangecorrelation energy for beta spin Ucore: corecore Coulomb energy Uhub: LDA+U energy Ucs: constraint energy for spin orientation Uzs: Zeeman term for spin magnetic moment Uzo: Zeeman term for orbital magnetic moment Uef: electric energy by electric field UvdW: semiempirical vdW energy
Is the UH1 the difference Hartree potential? If so, how can I get the UH0 in the cube file format?

Re: Kohn Sham energy levels and Hartree potential ( No.8 ) 
 Date: 2014/01/29 16:28
 Name: T. Ozaki
 Hi,
As for the nonlocal potential, please refer a discussion at http://www.openmxsquare.org/forum/patio.cgi?mode=view&no=1525
> Is the UH1 the difference Hartree potential?
Yes it is.
> If so, how can I get the UH0 in the cube file format?
UH0 is the energy which is partly related to the neutral atom potential. Except for the nonlocal potential, all the information about potentials can be accessed through the available cube files.
Regards,
TO

Re: Kohn Sham energy levels and Hartree potential ( No.9 ) 
 Date: 2014/01/29 18:34
 Name: Le The Anh
 Thank you very much, Ozaki sensei.
The Anh
