The ABINIT Users Mailing List ( CLOSED )

[pganesh@ciw.edu]

I am trying to compute the polarisation (say PbTiO3 in P4mm) for which I 
guess I need to use the Berry phase method.  I first compute the 
wavefunctions using kptopt=1 and then read-in the wavefunctions to get the 
polarisation.  My input file (for polarisation) has the following parameters 
apart from the usual variables for GS calculations:

berryopt -1  #to use the berry_new routine (is there a way to do otherwise?)
rfdir 1 1 1  # to write the cartesian coordinates of polarisation
kptopt 3     # to use the full kpt in the BZ 

Now is the final "Polarisation" as mentioned in the output file the right 
answer or is there an additive constant to it since polarisation forms a 
lattice in the modern theory?

Is there a tutorial to the older way of computing polarisation in some older 
version of abinit? (using berryopt =1 and nberry and kberry and bdberry etc.)