The ABINIT Users Mailing List ( CLOSED )

["Anglade Pierre-Matthieu" <anglade@gmail.com>]

Hi,

Here are a few comment and pieces of advice. Comment 3 is certainly
most important for your present problem.

1) I can't tell you wether the pseudopotentials are OK or not. This is
you to discover. Usualy, to do that people compare, for each elements
the properties obtained with the pseudo and some experimental or full
electron results. When you download pseudopotentials from the abinit
site you are warn that you use them at your own risk. Then you must
pay attention to all necessary pseuopotentials checkings.

2) ecut=120 hartree is anyway too much for your calculation. You waste
your time with that

3) You must be more carefull with the way you set your kpt mesh. It is
quite certainly not correct now. Try having a look at the abinit
tutorials or at the description of Abinit input variables about kpt.
"prtkpt" will probably be very usefull for you.

4) Be certain that 30 setps are enough to get an accurate convergence
of SCF cycles. Otherwise mind having a look at variables like iprcel
and diemac.

Regards

PMA

On 4/26/07, taoj0003@ntu.edu.sg <taoj0003@ntu.edu.sg> wrote:
> My input file is shown below.
>
> the pseduopotetial i am using are download from the Abinit website. Both the 
> Ti and O are Troullier-Martins pseudopotentials. is there anything wrong?
> thanks
>
>
> # Crystalline TiO2
>
>
>
> kptopt 1 # Option for the automatic generation of k points,
> ecut 120.0 #
>
> toldfe 1.0d-10
>
> acell 4.55 4.55 2.95812 angstrom
> optcell 6
>
> ntypat 2
> znucl 22 8
>
> natom 6
> typat 1 1 2 2 2 2
> xred
>   0.00000000  0.00000000   0.00000000      #Ti
>   0.50000000  0.50000000   0.50000000      #Ti
>   0.30530000  0.30530000   0.00000000      #O
>   -0.30530000 -0.30530000  0.00000000      #O
>   0.80530000  0.19470000   0.50000000      #O
>   0.19470000  0.80530000   0.50000000      #O
>
>
>   #   0.00000000     0.00000000     0.00000000   #Ti
>   #   2.29686500     2.29686500     1.47906000   #Ti
>   #   1.40246577     1.40246577     0.00000000   #O
>   #   -1.40246577    -1.40246577    0.00000000   #O
>   #   3.69933077     0.89439923     1.47906000   #O
>   #   0.89439923     3.69933077     1.47906000   #O
>
>
> nstep 30


--
Pierre-Matthieu Anglade