Hi All, I am trying to calculate piezoelectric tensor of tetragonal BaTiO3 with the help of VASP package program, I've written LEPSILON=T in INCAR file, after end of the job, I can get PIEZOELECTRIC TENSOR in the unit of C/m^2. It doesn't match with the literature works.
I obtained the elastic constants, and took the inverse of it via MATLAB for get to compliance tensor, then multiply the result with piezoelectric tensor that is coming from the VASP. What would be the possible reasons behind my miscalculation?
VASP isn't my expertise, but the most common piezoelectric tensors ( d & g) don't have units of C/m2. Check VASPs definition of its piezoelectric tensor to ensure that your literature parameter matches. Second thought--are your VASP and literature values in full tensor form or in condensed matrix form? There are factors of 2 or 4 when converting between the two for some coefficients. See Nye or Newnham for the condensed matrix to tensor conversions
Dear Pilgrim, thank you for your helpful answer, VASP gives e units of C/m2, literature mostly gives C/N. I've check it out.
Dear Dibs, thank you for your interest.
I've noticed that LEPSILON=T tag calculates just ion-clamped piezoelectric tensor.
http://cms.mpi.univie.ac.at/vasp/vasp/LEPSILON_static_dielectric_matrix_ion_clamped_piezoelectric_tensor_Born_effective_charges.html
Piezoelectric constants can be decomposed into purely electronic "clamped-ion" and nuclear "internal-strain" contributions. See:
http://journals.aps.org/prb/pdf/10.1103/PhysRevB.88.035102
In light of these papers, VASP should be calculate to contribution of nuclear on piezoelectric constants as like ABINIT, but WHICH tag, or HOW?
ABINIT output :
http://www.abinit.org/documentation/helpfiles/for-v7.4/tutorial/lesson_ffield.html#1
Thank you!
- Badges
- Science topic