Piezoelectric tensor calculations (LEPSILON)

[marcofronzi]

Hi all,

I recently calculated the piezoelectric tensor, the static dielectric matrix, ion-clamped piezoelectric tensor and the Born effective charges for several vdW hetero-structures, but for some of them I am having some problems.
After geometry relaxation is reached using use the conjugate gradient algorithm, I run an additional calculations using the following INCAR file

SYSTEM =Sys
ENCUT =700.0
NSW =10
ISIF =2
IBRION =6
POTIM =1.0
ISTART =1
ISPIN =2
EDIFF =1.0E-6
EDIFFG =-0.01
PREC =high
ISYM =0
ISMEAR =0
SIGMA =0.2
NPAR =8
LEPSILON =.TRUE.
LCALCPOL =.FALSE.
LBERRY =.FALSE.
EMIN = -20
EMAX = 20
NEDOS = 3000
LORBIT = 11
IVDW=12
LVDWSCS = .TRUE.
ADDGRID = .FALSE.
IDIPOL=3
LDIPOL= .TRUE.
LVTOT = .TRUE.

First issue: density functional perturbation theory (IBRION = does not work with the current INCAR setup.
Second issue: with the current setup (IBRION = 6), for a few structures, the "PIEZOELECTRIC TENSOR IONIC CONTR" do not appear in the OUTCAR file (I only get the PIEZOELECTRIC TENSOR).
I would appreciate if someone can help me to understand if I am doing the correct procedure.

Thanks and regards,

Marco

[martin.schlipf]

It is hard to see what you are exactly doing. A good strategy is typically to reduce the problem to a minimal one. Try starting with the tutorial example and see if that gets all the output you would expect. Then add flags from your INCAR file or change the other inputs (KPOINTS, vacuum) that might cause the issue.

If you find that there is indeed an issue, please post a bugreport including all the relevant input.