Hello,

I have a question about the calculation of Mulliken Gross Populations
in the Adf-BAND version 1.1.0.

I perform a calculation with a FCC structure of N : my unit cell is N4.
One atom is on a vertex, and the three others are in the middle of faces.
Keywords in the input file are therefore

  LATTICE

   4.8 0.0 0.0
   0.0 4.8 0.0
   0.0 0.0 4.8

   **

   ATOMS N

   0.0 0.0 0.0
   2.4 2.4 0.0
   2.4 0.0 2.4
   0.0 2.4 2.4

The four atoms N are equivalents in this structure, so I expect the same values
in the Mulliken Gross Populations for these atoms.

I actually obtained the same values with the VERSION 0.5 of BAND, but not with the
new adf band program version 1.1.0.

Atom q-atom

N1 4.989
N2 5.016
N3 4.981
N4 5.015

I was surprised by this result and I wanted to have opinions about that.

I can add that in the charge analysis by numerical integration, the four
atoms are equivalents in both versions of adf band.

Thanks for answering me.

---
Stanislas REMY
LSDSMS University of Montpellier
France