Page 1 of 1

Units of magnetic moment on CHG file

Posted: Tue Dec 26, 2006 6:46 pm
by salva
Dear users/developers.
I am working with a magnetic system, having Mn atoms on it. I wrote a script to get the magnetic moment within a sphere centered on the Mn atoms. Specifics follow:

The calculation is spin-polarized. I take the CHG file and read the density, dividing each entry by the volume of the cell.

The radii around each Mn atom is about 1.2 Ang. The calculation is done with PAW pseudos.

After adding up the entries from the second dataset on the CHG file (related to the magnetization) around the radius mentioned about the selected ions, I get for the magnetization some big values, say, ~14000 for the magnetization about this ion.

I am confident about the script as also the electronic charges around the ions are overestimated by a factor of thousands. So even though the values are high, they seem to be consistent within each other (consistently bigger I should say).

The value should be close to three, as this is the value assigned through the MAGMOM flag. So, even though I can just "normalize" the value I get so that it becomes the value I am expecting to get, in this case 3, I'd rather learn about the units on the file so that I can make the necessary change of units that (hopefully) will give me a consistent answer.

Thank you for your time.

Salvador

Units of magnetic moment on CHG file

Posted: Wed Dec 27, 2006 1:42 pm
by admin
why do you not make use of the LORBIT input tag? It will give the local charges and magnetisations (spin-up - spin-down charge) in OUTCAR.
please mind that
if you set LORBIT >= 10, the sphere radius will be chosen to be RWIGS as read from POTCAR
to set RWIGS explicitely, please define it in INCAR (for each atom) and set LORBIT=1