Partial charge densities in VASP

[madhu_menon2]

Hi,
I am trying to calculate partial charge densities. I went to the topic: wiki/index.php/LORBIT
I see that due to the issues with version >=5.4.4, they suggest the following steps:
1. Self-consistent calculation with symmetry switched on (ISYM=2)
2. Recalculation of the partial charge density with symmetry switched off (ISYM=0)
What do I calculate in step 1? I.e., what output am I looking for? Do I calculate DOS in step 1 or just the optimization? Then what do I take from step 1 as input into step 2? Obviously, I am assuming that I calculate DOS in step 2. Do I also copy CONTCAR to POSCAR in step2. Your help would be greatly appreciated. Thanks.

[henrique_miranda]

Hi,

All the input files for step 1 + the resulting WAVECAR need to be transfered to step 2.
If you are doing a relaxation calculation then you should do it before doing step 1.
You should start the step 1 already with the relaxed ionic positions.
Hope this answers your question.

[madhu_menon2]

Thanks for the suggestion. So, I should do a static calculation with ISYM=2 in step 1 and use the resulting WAVECAR as the input along with ISYM=0 to calculate the partial charges in step 2, right?

[henrique_miranda]

Yes, that should do the trick.
I also recommend that you carefully check the partial densities that you obtain in both step 1 and 2.
The two steps are only needed because of a bug in the symmetrization of the partial densities so at least you should check that the two steps indeed yield different results.
The issue should have been fixed in VASP6

[madhu_menon2]

Great. Thank you!