paw-pbe at expanded pressures. As volume inceases, pressure increases as well!

[akberta]

Dear vasp users:

I have encountered a rather unusual behavior when using the vasp code. When running fixed volume runs with the paw-pbe pseudopotential, I have found that in the expanded pressure region as the volume increases, the pressure increases as well! (In the compressed pressure region, the pressure-volume relationship was correct -- as volume increased, pressure decreased. This non-physical behavior shows up only in the expanded pressure region) Clearly, this isn't physical, and I am wondering if there is a bug in the paw-pbe pseudopotentials at expanded pressures?

I have searched the forum, but have not seen any other reports of this behavior. We are using vasp.4.6.28 on Opteron 64-bit dual processor machines, compiled with the Intel Fortran 9.0 code.

I am using these pseudopotentials:
PAW_PBE O_s 07Sep2000, PAW_PBE Fe_pv 06Sep2000 and PAW_PBE Si 05Jan2001

Any advice would be greatly appreciated. Thanks!
Amy

[admin]

first of all, please carefully check if your basis set size is converged:
all data calculated on the basis of the stress tensors require very
high precision (set ENCUT to at least 1.3 * max(ENMAX) in INCAR)
as a hint for insufficient basis set size: the total drift in the forces should be smaller than the forces acting on the ions themselves (OUTCAR).