RHOION file and RHOB file about vaspsol

[tbi-sci]

I have the same question.
What are the units of the numbers written to RHOION and RHOB?
It seems they are different from CHGCAR. Also it seems that the units in RHOION are different than in RHOB.
Many thanks for any ideas!

[rhennig]

The units of the data in the RHOIOIN and RHOB files are electrons*grid_volume, same as the units of the CHGCAR file. Based on this, you can convert the units to moles/liter, and we made the plots for the VASPsol paper that way. Note that instead of having absolute ion charge densities or bound charge densities, the RHOION and RHOB files have Net ion/bound charge densities.

Hence, in CHGCAR, the sum(CHGCAR_data/GRIDPTS) == total number of valence electrons.
However, in RHOION, the sum(RHOION_data/GRIDPTS) will be close to zero (order of 1E-10) as the positive and negative ions charge cancel to most parts.

And, the sum(RHOB_data/GRIDPTS) sum would be even smaller (order of 1E-15).

[tbi-sci]

Many thanks for the fast reply.
I just still don't understand why sum(RHOION_data/GRIDPTS) is so small.
If RHOION corresponds to the net ion charge density, then shouldn't sum(RHOION_data/GRIDPTS) be equal in magnitude to the net solute charge, so the Delta_NELECT?

As an example, in my calculation I have 308 electrons for the neutral system. Now I change that number to NELECT=308.1.
Then I get sum(CHGCAR_data/GRIDPTS)=308.10000 and sum(RHOION_data/GRIDPTS)=-0.00017924.
But I was expecting the latter to be equal to 0.1 to compensate in magnitude for the net electron number.
As a side note: My cell volume is 1287.7 A^3.

[tbi-sci]

UPDATE:
I think I found the missing scaling factors:
The data in RHOION must be multiplied with CELL_VOLUME / LN(10) to obtain the same definition as CHGCAR.
The data in RHOB must be multiplied with 4 * PI to obtain the same definition as CHGCAR.
Is this possible?

[bumpwy]

I wonder if there's any follow up this from the developer @rhennig ? I am trying to visualize the total solvent charge distribution at a charged Cu(100) surface. I also see that, when the metal is charged to +0.1 e, the RHOION integral sum(RHOION/GRIDPTS) gives some -1e-5 e, far from the supposed -0.1 e. I wonder what's missing.
Also, would the total solvent charge simply be RHOB + RHOION? Thanks for your help!

[veritas496]

Hi, I met a issue. When I opened the RHOION and LRHOB tags, RHOION and RHOB files weren't generated.
I was wondering if there is a solution for it?
I am using vasp 5.4.1 and vaspsol can work well, including lsol, ebk, lambda_d_k.
Thanks

[ericfonseca95]

Hello @veritas496,

I have not heard of this issue occurring before. Could you provide more detail if possible?
For now all I can recommend is conducting a fresh installation of VASPsol.

Let me know if I can be of further assistance,

Eric