H atom using SCAN

[shaochunlee]

Hi,

I wanted to simulated water using SCAN functional, but I couldn't find POTCAR in "potpaw_PBE.54.tar.gz" for H atom containing the following lines as mentioned in vaspwiki:
kinetic energy-density
mkinetic energy-density pseudized
kinetic energy density (partial)

I am sure that the Potential database is the latest one.
Could someone help me with this?

Thanks,
Shao-Chun Lee

[fabien_tran1]

Hi,

If I search for "kinetic" in the POTCAR files for the H atom (from potpaw_PBE.54), I get:

"grep -i kinetic H/POTCAR":
Error from kinetic energy argument (eV)
kinetic energy-density

"grep -i kinetic H_*/POTCAR":
H_GW/POTCAR: Error from kinetic energy argument (eV)
H_GW/POTCAR: Kinetic energy density valence
H_GW/POTCAR: kinetic energy-density
H_h_GW/POTCAR: Error from kinetic energy argument (eV)
H_h_GW/POTCAR: Kinetic energy density valence
H_h_GW/POTCAR: kinetic energy-density
H_h/POTCAR: Error from kinetic energy argument (eV)
H_h/POTCAR: kinetic energy-density
H_s/POTCAR: Error from kinetic energy argument (eV)
H_s/POTCAR: kinetic energy-density

Don't you get the same?

[shuang_luo]

Hi there, I am also using SCAN to simulate a system that includes hydrogen atoms. However, I have been unable to locate a potcar file for H atoms that contains the three lines of kinetic energy information required by SCAN. Have you managed to find a solution to this issue?

[yong_wang1997]

Hi Guys,

I recently tried the same procedure for water. Here is an answer that I found on another forum talking about using METAGGA for H.

"I think the differnece between kinetic energy-density and mkinetic energy-density pseudized lies in the fact that the former indicates the representation of the kinetic energy density for the all-electron part of the pseudopotential, while the pseudized kinetic energy density represents the kinetic energy density associated with the pseudopotential and is used to describe the behavior of the valence electrons.

Since hydrogen has only one electron in its 1s orbital, which is considered a valence electron, there is no need for a separate pseudopotential approximation. Therefore, the kinetic energy-density in the POTCAR file for hydrogen represents the kinetic energy of the valence electron without the need for a pseudized representation. Thus, I think you don't need to worry about 'mkinetic energy-density pseudized' in this case."

Hope this answer could address your concerns. Have a good day!

Yong.