The ABINIT Users Mailing List ( CLOSED )

["D. R. Hamann" <drhamann@mat-simresearch.com>]

Dear P. Ganesh,

It is extremely difficult to compute dielectric and piezoelectric 
constants by finite-difference methods, especially when atomic 
relaxations are involved (ie., in anything but the simplest unit 
cells).  I suggest you use the density functional perturbation theory 
capabilities of Abinit instead, reading and working through the 
"Response Function 1," "Response Function 2," and "Elastic" tutorials.

Don Hamann

P. Ganesh wrote:
> Dear All,
>
> I am trying to compute polarization vs electric-field.  If I start 
> with the ground state structure and apply a very small electric-field 
> (~0.00001 a.u) and relax the ions and cell shape only (i.e optcell = 
> 3, ionmmov = 2) then the zeroth broyden step converges with a 
> polarisation along [x0z] direction.  But right after the 1st SCF cycle 
> in broyden step 1 I get the electronic, ionic and total polarisation 
> to be exactly zero.  Further, at SCF 7 the polarization agains become 
> non-zero, and very close to the previous non-zero value and the 
> program aborts with the following warning message:
>
> scfcv (electric field calculation) : WARNING -
>   The difference between pel (electronic Berry phase updated
>   at each SCF cycle)
>   and pel_cg (electronic Berryphase computed using the berryphase 
> routine) is
>   pdif_mod =  0.178600608E+01
>
> pel_cg(1) =  0.000000000E+00
> pel_cg(2) =  0.000000000E+00
> pel_cg(3) =  0.000000000E+00
> pel(1) = -0.886982093E+00
> pel(2) =  0.000000000E+00
> pel(3) = -0.999634356E+00
>
> I dont understand why the polarization suddenly goes to zero at SCF 1 
> and how it suddenly came back to be non-zero? I am using v5.4.4.
>
> Thank you very much.
>
> PGanesh

--
D. R. Hamann
Mat-Sim Research LLC    | Deptartment of Physics
P.O. Box 742            |  and Astronomy
Murray Hill, NJ 07974   | Rutgers University
phone: 908-370-8079     | 732-445-4381

email: drhamann@mat-simresearch.com