Page 1 of 1

Atomic Bader charge and charge transfer using Quantum Espresso

Posted: Fri Dec 04, 2020 7:42 pm
by Hamza_ALLAL
Hello everybody,

I am trying to work with the bader-charge code (I AM A BEGINNER), I have modeled the adsorption of NH3 on the surface of the borophene (see attached Figure, FIG1) using Quantum espresso code.

I wanted:
1) first know the atomic charge for the NH3 molecule.
2) calculate the charge transfer
I have attached the three .ACF files
NH3_ACF = NH3
BS_ACF = Borophene
NH3-BS_ACF = NH3-Borophene

I can't figure out how to calculate the atomic charges on the atoms N, H1, H2, and H3 for the isolated molecule (before adsorption) and also after adsorption.

Knowing that, I generated the two .cube files NH3_valence.cube and NH3_allelec.cube, using plot_num = 0 and plot_num = 21 , respectively, with Quantum espresso code.
then, I used the BADER code command:
bader NH3_valence.cube -ref NH3_allelec.cube


Moreover, on the NH3-BS_ACF file, the last four lines 81, 82, 83 and 84 correspond respectively to atoms N, H1, H2, and H3.

I hope my question will be helpful for everyone.

Re: Atomic Bader charge and charge transfer using Quantum Espresso

Posted: Fri Dec 04, 2020 10:01 pm
by graeme
It looks to me like you don't have core charges included. Your N atoms has a charge over 8, which is including the charge on the 3 H atoms. You can still estimate the charge transfer from the Boron surface to the NH3 molecule as 8.1-8=0.1e, but you really need core charges to resolve the charge density on the H atoms.

Re: Atomic Bader charge and charge transfer using Quantum Espresso

Posted: Sat Dec 05, 2020 1:59 pm
by Hamza_ALLAL
Thank you very much dear graeme,

In the previous calculation I used: plot_num = 21. knowing that :

21 = all-electron charge density (valence + core).
For PAW calculations only; requires a very dense real-space grid.

So, I recalculated with plot_num = 17, I found the same thing.

17 = all-electron valence charge density
can be performed for PAW calculations only
requires a very dense real-space grid!


Do I have to use other option ??

Input File Description (pp.x)
https://www.quantum-espresso.org/Doc/IN ... html#idm24
=============================================
plot_num INTEGER

Selects what to save in filplot:

0 = electron (pseudo-) charge density

1 = total potential V_bare + V_H + V_xc

2 = local ionic potential V_bare

3 = local density of states at specific energy or grid of energies
(number of states per volume, in bohr ^ 3, per energy unit, in Ry)

4 = local density of electronic entropy

5 = STM images
Tersoff and Hamann, PRB 31, 805 (1985)

6 = spin polarization (rho (up) -rho (down))

7 = contribution of selected wavefunction (s) to the
(pseudo-) charge density. For norm-conserving PPs,
| psi | ^ 2 (psi = selected wavefunction). Noncollinear case:
contribution of the given state to the charge or
to the magnetization along the direction indicated
by spin_component (0 = charge, 1 = x, 2 = y, 3 = z)

8 = electron localization function (ELF)

9 = charge density minus superposition of atomic densities

10 = integrated local density of states (ILDOS)
from emin to emax (emin, emax in eV)
if emax is not specified, emax = E_closed

11 = the V_bare + V_H potential

12 = the sawtooth electric field potential (if present)

13 = the noncollinear magnetization.

17 = all-electron valence charge density
can be performed for PAW calculations only
requires a very dense real-space grid!

18 = The exchange and correlation magnetic field in the noncollinear case

19 = Reduced density gradient
(J. Chem. Theory Comput. 7, 625 (2011), doi: 10.1021 / ct100641a)
Set the isosurface between 0.3 and 0.6 to plot the
non-covalent interactions (see also plot_num = 20)

20 = Product of the electron density (charge) and the second
eigenvalue of the electron-density Hessian matrix;
used to colorize the RDG plot (plot_num = 19)

21 = all-electron charge density (valence + core).
For PAW calculations only; requires a very dense real-space grid.

22 = kinetic energy density (for meta-GGA and XDM only)

Re: Atomic Bader charge and charge transfer using Quantum Espresso

Posted: Mon Dec 07, 2020 2:02 am
by graeme
I think that you want the all-electron (valence+core) but based upon the total charge in your calculation (248 e) it seems that you only have the valence density in your analysis.

Re: Atomic Bader charge and charge transfer using Quantum Espresso

Posted: Mon Dec 07, 2020 5:14 pm
by Hamza_ALLAL
Thank you very much dear graeme, thank you for your time.
I will test again.