pyiron_atomistics icon indicating copy to clipboard operation
pyiron_atomistics copied to clipboard

Vasp stresses are confusing and unit is a mystery

Open Leimeroth opened this issue 1 year ago • 2 comments

Lammps and vasp stores pressures/stresses somehow different and vasp stresses are confusing. Default lammps jobs store pressures as full matrix and stress =-pressure. For vasp the output contains a 3 vector pressure and a matrix stresses, but the stresses do not give any reasonable result when trying to caculate elastic constants with them and a short look in the code made me assume that units are not the same as for pressures.

Leimeroth avatar May 30 '23 14:05 Leimeroth

Yes, the current value of pressures makes no sense, I introduced stresses for backwards-compat, but it's been a while and imo we can throw the old pressures out. I'm not sure about the unit off the top my head, but I've used the stress output before successfully.

pmrv avatar May 31 '23 08:05 pmrv

To get stresses from the format written in hdf to GPa (which is the pyiron unit for pressure, so it makes sense for stresses too imo) I need to stress = stress/KBAR_TO_EVA*0.1

Leimeroth avatar May 31 '23 10:05 Leimeroth