The ABINIT Users Mailing List ( CLOSED )

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

Looks like you are right.
xcart and xred redundant and you need to specify
and the default for rprim is the unity matrix.
Note that with plane wave it is hardly possible to do non periodic 
calculations.

On 4/17/07, Hongyi Zhao <hongyi.zhao@gmail.com> wrote:
> On Mon, Apr 16, 2007 at 08:14:42PM +0200, Anglade Pierre-Matthieu wrote:
> > The simulation cell is periodical. This means that there is "six"
> > distances between you atoms. And only one of them is fixed by xcart.
>
> Do you mean that the parameter *rprim* will be treat as the Default
> 3x3 unity matrix if I don't explicitly setting it?  Because of the
> symmetry in the H_2 molecular, so we can only compute the one direction
> belong to X, Y or Z in order to determine the optimum interatomic
> distance of H atoms, is this right?
>
> Furthmore, I again read the tutorial of the Abinit, and obtain the
> following knowledge about my question I posted in the last mail, I don't
> know whether it is just the case that happened in the abinit's
> processing, so I also posted here for your further advices:
>
> 1.  I think that if I give the parameters *xcart* and *acell* but not the
> *rprim*, than the abinit will treat the *rprim* as the default 3x3 unity
> matrix.
>
> 2.  In the tutorial of Abinit, I find the following explains:
>
> -------------
> If xred and xangst are ABSENT from the input file and xcart is provided,
> then the values of xred will be computed from the provided xcart
> -------------
>
> So I think that in the following input file, the abint will process it
> as I described here:
>
> ------------------
> ndtset 21
> xcart:  -0.5  0.0 0.0
>         0.5  0.0 0.0
> xcart+  -0.025 0.0 0.0
>         0.025 0.0 0.0
> getwfk  -1
> nband    1
> acell 10 10 10
> ntypat 1
> znucl 1
> natom 2
> typat  1 1
> ecut 10.0
> nkpt 1
> nstep 10
> toldfe 1.0d-6
> diemac 1.0
> diemix 0.5
> ------------------
>
> According to my understanding, when the abinit is started, it will do
> the following things:
>
> a)  It will convert the xcart into its xred counterpart, at the same
> time, the rprim will be set as its default value, that is, a 3x3 unity
> matrix.
>
> b)  Then, the cartesian coordinates of atoms are given through the
>  following formula for each *xcart*:
>     t_cartesian = t1*r1*a1+t2*r2*a2+t3*r3*a3,
>     where (t1,t2,t3) are the reduced coordinates" given in columns of
>     "xred", (r1,r2,r3) are the columns of dimensionless array "rprim",
>     and (a1,a2,a3) are the elements of the array "acell" giving length
>     scales in Bohr.
>
> c)  Base on each *xcart*, the optimum value is computed and obtained.
>
> Can anyone tell me whether my understanding is true or not?
>
>
> Thanks in advance.
>
> Hongyi Zhao
> GnuPG DSA: 0xD108493B


--
Pierre-Matthieu Anglade