Page 1 of 1
cell optimization with vdW-DF functional
Posted: Thu Jul 12, 2012 12:19 am
by tibeten
Hi,
In the manual, it is pointed out that
"The optimisation of the cell (ISIF=3 and higher) is currently not possible. "
Has it become possible now? Or cell optimization has to be done elsewhere?
Thanks,
B
cell optimization with vdW-DF functional
Posted: Thu Jul 12, 2012 11:14 am
by kelum
Hi,
the calculation of the stress tensor is available in the latest release (VASP 5.2.12 from November), it's not in the 5.2.12 released in July last year. The code works for spin non-polarised systems. For spin-polarised it works as well but the code needs to be changed since there is a bug. The line 138 with
stress=0.d0
needs to be deleted, then the cell optimisation works for spin-polarised systems as well.
This will be fixed in the next release.
Again, for no spin-polarisation nothing needs to be changed in the 5.2.12 (November) version.
You can test it in the same way as cell optimisation with PBE or LDA, do a direct cell optimisation and compare to a Murnaghan (or other) fit to energies at different volumes.
Remember that the cell optimisation requires a higher plane wave cut-off than a usual calculation, this is true for PBE and also for the vdW-DF cell optimisations. (This should be somewhere in the manual.)
Best,
j