[ase-users] Bader charge analysis problem with ase SIESTA

Boubacar Traore traorehade at yahoo.fr
Fri Oct 24 16:47:32 CEST 2014


Dear ase users,
I would like to perform the bader charge analysis with siesta. I have tried to follow the exemple on ase website (https://wiki.fysik.dtu.dk/ase/ase/dft/bader.html) and generate the cube file from my siesta *.RHO file. However, the charges that I am obtaining are not correct at all. In my exemple, I am looking at HfO2 with Hf (+4) and O (-2). However the ACF.dat file give zero charge for all my Hf atoms and more than +7 for my O atoms.
Since siesta generates the *.RHO file in e-/Bohr**3, I did not need to convert it to Bohr.Here is the script that I used:
----------------------------from ase import *from ase.calculators.siesta import Siestaimport osfrom ase.io import write
os.environ['SIESTA_PP_PATH'] = "%s/pseudo"%(os.getcwd())os.environ['SIESTA_SCRIPT']  = "%s/run_siesta.py"%(os.getcwd())

bulk = Atoms('Hf4O8')
bulk.set_cell([( 5.1457998828458198,    0.0000000000000000,    0.0000000000000000),               (0.0000000000000000,    5.1951698640744004,    0.0000000000000000),               (-0.8964782985261750,    0.0000000000000000,    5.2553447611848201)])
calc = Siesta(label='HfO2', write_fdf=False) # It uses my own fdf file             bulk.set_calculator(calc)e = bulk.get_potential_energy()density = calc.get_pseudo_density() # siesta density is in e-/Bohr**3write('density.cube', bulk, data=density)
 ------------------------
Siesta-3 can generate SystemLabel.BADER file and recommends grid2cube to convert it to cube format. But grid2cube from siesta works for orthogonal cells only. Since my HfO2 is not orthogonal, I am not able to convert the SystemLabel.BADER file to cube format. I was hoping that the bader analysis would work with ASE script that converts the *.RHO file to cube format. 
So, has anyone managed to do the bader analysis with ase SIESTA? and did anyone encounter  similar problem? and if yes, how was it solved?


Thank you for your help!
Boubacar

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.fysik.dtu.dk/pipermail/ase-users/attachments/20141024/0c365cea/attachment.html>


More information about the ase-users mailing list