Grand-canonical energy calculations of charged systems

[alex-aziz]

I have been been attempting to repeat the calculations performed on the Pt(111) slab from the JCP 2019 (doi.org/10.1063/1.5132354) paper.

Due to what is likely a trivial misconception, I am having trouble simulating your graph of the grand-canoncial energy of a charged slab shown in Fig. 2.

I have firstly corrected for the DFT energy of the charged slab based on Q V, "where Q is the net charge of the simulation cell and V the shift in reference potential, e.g., V = FERMI_SHIFT." (If I have increased NELECT by 0.05 I take the charge as -0.05).

This then gives me a linear dependance of the energy of the system with respect to the charge (q).

I then note the reference to "The grand canonical electronic energy, F, is the Legendre transformation of the free energy, A, of the system, F(U) = A(n)-n_electrode x U" where n_electrode is the net charge of the electrode slab and U is the applied potential.

But I am unsure if this needs to be applied and if so I am still not getting the quadratic behaviour. I am using the Pt(111) surface from Materials Project (9 atoms) with a vacuum of 30 Angstroms.

Much appreciated Alex Aziz

[alex-aziz]

I have managed to do this now by applying the correction as you have stated in the other post.

Many thanks Alex

[zhichao-1699]

Hi, Alex, I am also currently repeating this grand canonical potential calculation , and I just got a similar linear relationship (even with the additional -n_electrode*U term). Could you please give some suggestions how to solve the problem?

Best, Zhichao