wrong bader charge for N

[apci]

We did test of the bader analysis, and got the same results. We used bader analysis for several different systems, and got pretty reasonable results.

However, when we add Nitrogen in the system, the N charges seems to go crazy. The atomization energy calculated from VASP and Dmol3 agrees well, so it does not seem to be something wrong with the VASP runs. We wonder if anybody here studied N using VASP and bader analysis before.

in N2, the ACF.dat looks like
# X Y Z CHARGE MIN DIST
----------------------------------------------------------------
1 0.0000 0.0000 0.0484 2.6620 0.0484
2 0.0000 0.0000 1.1516 7.3380 1.0266
----------------------------------------------------------------
NUMBER OF ELECTRONS: 10.00000
one N atom gains 2.34 electrons from the other one, which is totally unreasonable.

in CH3NH2, the ACF.dat
# X Y Z CHARGE MIN DIST
----------------------------------------------------------------
1 0.6198 1.0856 0.8573 3.3050 0.0986
2 1.8990 1.7412 1.0286 7.8160 0.9282
3 9.9942 1.2612 1.7459 0.8985 0.2106
4 0.0869 1.5357 0.0059 0.9637 0.2549
5 0.6568 9.9894 0.6862 1.0167 0.2946
6 2.4912 1.6180 0.2064 0.0001 2.2578
7 2.4079 1.3573 1.8259 0.0000 2.3530
----------------------------------------------------------------
NUMBER OF ELECTRONS: 14.00000
again, the N gains 2.82 electrons, which is not reasonable.

[graeme]

Are you using core charges? If not, use the method described at:
http://theory.cm.utexas.edu/bader/vasp.php