> > i) which method is correct whether doing doing vc-relax alone or collect > all volume and total energy from vc-relax output and fitting this with B-M > equation ? > In my opinion, the biggest difference would come from systems with hard pseudopotentials when stress tensor convergence is achieved at a much higher plane wave cut off than the one required to converge the total energy. If this difference between required cut offs are too high one might as well obtain energy(volume) and use a valid equation of state, rather than vc-rlx.
emine kucukbenli, postdoc at theos, epfl, switzerland
