The ABINIT Users Mailing List ( CLOSED )

[ren <opticalcase@gmail.com>]

Dear Prof Matthieu:

Thank you for your quick reply! My apology, if I interpret correctly
what you are saying, is that for my calculation of a 2x2x2 supercell,
i will get the same result as a 1x1x1 unit cell except that since
2x2x2 supercell is much larger, the computational cost would scale to
more than 8^3?

I am making a 2x2x2 supercell as a starting point as I am considering
vacancy in my material in the future. I read your post in the past in
the forum, suggesting to someone that in considering a supercell (for
a vacancy) e.g, chkprim=0 must be use? Please correct me if I am
wrong?

Regards,
Yee Yan



On Fri, Feb 27, 2009 at 1:23 AM, Anglade Pierre-Matthieu
<anglade@gmail.com> wrote:
> Hi,
>
> The transformation apparently keeps the 2x2x2 supercell.
> In most case it is not appropriate to input chkprim =0 since you will
> get essentially the same results has using a single cell but at
> much more than 8 to the power three times the computational cost. Yet,
> for a converged calculation you'll get the same properties.
>
> regards
>
> PMA
>
>
> On Thu, Feb 26, 2009 at 3:48 PM, ren <opticalcase@gmail.com> wrote:
>> Dear All,
>>
>> I am using Cell2Abinit.py to transform a 2x2x2 supercell of ZnO from
>> Castep to Abinit format. I suppose that the transformation creates a
>> primitive cell? The problem occurs when I am trying to run Abinit, the
>> calculation halt and the log file says
>>
>> " According to the symmetry finder, the unit cell is
>>   NOT primitive. The multiplicity is  8 .
>>   The use of non-primitive unit cells is allowed
>>   only when the input variable chkprim is 0.
>>   Action : either change your unit cell (rprim or angdeg),
>>   or set chkprim to 0. it says that the input structure is not a
>> primitive cell and has a multplicity of 8."
>>
>> While I input chkprim = 0 and the calculation continues run may I know if
>>
>> 1) It is appropriate just to simply use chkprim = 0
>>
>> 2) Is there something i have miss out while using Cell2Abinit that the
>> output is not a primitive cell in the end? My structure input is as
>> below:
>>
>> ####################################################
>> # Atoms Stucture created by CellToAbinit Converter !
>> ####################################################
>>
>> # Definition of the unit cell
>> #The length of the primitive vectors
>> #                    1 Bohr=0.5291772108 Angstroms
>> # acell 5.627901 6.498540 10.410880 Angstroms
>> acell 10.635191 12.280461 19.673712
>> rprim
>>     1.000000  -0.577350  0.000000
>>     0.000000  1.000000  0.000000
>>     0.000000  0.000000  1.000000
>>
>> #Definition of the atom types
>> #      32 kind of atoms
>> #       ['O', 'Zn']
>> ntypat  2
>> znucl  8 30
>>
>> #Definition of tha atoms
>> natom  32
>> typat  1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
>> xred
>>    0.166667 0.333333   0.1913
>>    0.333333 0.166667   0.4413
>>    0.666667 0.333333   0.1913
>>    0.833333 0.166667   0.4413
>>    0.166667 0.833333   0.1913
>>    0.333333 0.666667   0.4413
>>    0.666667 0.833333   0.1913
>>    0.833333 0.666667   0.4413
>>    0.166667 0.333333   0.6913
>>    0.333333 0.166667   0.9413
>>    0.666667 0.333333   0.6913
>>    0.833333 0.166667   0.9413
>>    0.166667 0.833333   0.6913
>>    0.333333 0.666667   0.9413
>>    0.666667 0.833333   0.6913
>>    0.833333 0.666667   0.9413
>>    0.166667 0.333333      0.0
>>    0.333333 0.166667     0.25
>>    0.666667 0.333333      0.0
>>    0.833333 0.166667     0.25
>>    0.166667 0.833333      0.0
>>    0.333333 0.666667     0.25
>>    0.666667 0.833333      0.0
>>    0.833333 0.666667     0.25
>>    0.166667 0.333333      0.5
>>    0.333333 0.166667     0.75
>>    0.666667 0.333333      0.5
>>    0.833333 0.166667     0.75
>>    0.166667 0.833333      0.5
>>    0.333333 0.666667     0.75
>>    0.666667 0.833333      0.5
>>    0.833333 0.666667     0.75
>>
>> ##### End of CellToAbinit Conveter ! #####
>>
>> The calculation runs well for a 1x1x1 and 3x3x3 cell without such
>> these comment though.
>>
>> Would like to hear your advises!
>>
>> Regards,
>>
>> Yee Yan, Tay
>>
>> Nanyang Technological University
>>
>> Singapore
>>
>>
>
>
>
> --
> Pierre-Matthieu Anglade
>
>