Total number of electrons different in Bader and in DFT

[diego.lopez]

Dear Bader users,

I am using Bader code to obtain the electron transfer in a complex formed by a carbon nanotube and a transition metal oxide encapsulated. When I run the DFT calculations in CP2K I obtain a number of electrons equal to 4316 but when I run Bader on the electron density cube file i have a total number of electrons equal to 4318.8554. How should I interpretate this result? My DFT calculations are done under periodic boundary conditions and using NC PP.

Thank you in advanced

[graeme]

If this is the total charge density, there can be errors with respect to integration around cusps at the nuclei. With plane wave calculations, the valence charge will add up properly. I can tell you that the Bader code simply adds the numbers in the charge density file and multiplies by the volume to get the total number of electrons - nothing fancy going on there.

[diego.lopez]

Thanks for your suggestion!

Then I suspect that my Bader results are not reliable since the integration of the electronic density of my system is not accurate. How could I improve this result?

[graeme]

I'm assuming the charge density that you have is the total charge. See if you can get the valence charge as well. Then you can do the analysis on the total charge but integrate the valence. The valence charge should be smoother and will likely sum properly.