Dear Dr. Blaha, thank you for the reply. > No, the Berry phase approach is not implemented.
Would that be a useful extension? Is it possible to expect developers' support while working on implementation of this functionality? In would be nice to have that option in Wien2k since major PW packages offer that Berry phase polarization, but I am somewhat skeptical about the performance of pseudopotentials with semicore electrons as valence electrons. > However, you may create a supercell (or better a slab) and apply an > external electric filed (zig-zag). Do I correctly understand your suggestion: Apply an external electric filed and find the strain it induces by optimizing lattice parameters and atomic coordinates simultaneously. With this I can access strain coefficients (strain developed per electric field applied). There is no way I can extract the polarization induced by this field. Correct? Thank you once again Oleg
