Dear Chen, > I am now running a calculation of strained STO and try to calculate > its spontaneous polarization based on Berry phase method. I try two > different ways: > 1) run a self-consistent calculation turning on lelfield=.true. and set > efield=0.0 > 2) first run a self-consistent calculation and then run a > non-self-consistent calculation turning on lberry=.true. > Are these two method equivalent? When I try these two methods on the same > system, method 1) works well but method 2) fails because "probably because > G_par is NOT a reciprocal lattice vector".
I think this is due to the fact that the berry's phase calculation (option lberry=.true.) is working on parallel machines only for the gdir =3 option. The Berry's phase polarization is also calculated through the electric field routines which are working in parallel for any choice of the gdir parameter. For the gdir = 3 case or for calculations on serial machines the two methods are expected to give the same polarization differences. Best regards, Paolo Umari, DEMOCRITOS ---------------------------------------------------------------- This message was sent using IMP, the Internet Messaging Program.
