[Wien] how to use non-default ROTLOC (local rotation matrix)

[Jorissen Kevin]

Hello Peter,
 
thanks for your explanation (I got it this time :)).
As for the dstart output : mea culpa, I had made an error in the input.  After correcting this, the starting densities look quite the same for both rotloc's, and also SCF runs smoothly.
 
thanks,
 
 
Kevin Jorissen
 
EMAT - Electron Microscopy for Materials Science   (http://webhost.ua.ac.be/emat/)
Dept. of Physics
 
UA - Universiteit Antwerpen
Groenenborgerlaan 171
B-2020 Antwerpen
Belgium
 
tel  +32 3 2653249
fax + 32 3 2653257
e-mail kevin.jorissen at ua.ac.be
 

________________________________

Van: wien-admin at zeus.theochem.tuwien.ac.at namens Peter Blaha
Verzonden: ma 21-3-2005 18:56
Aan: wien at zeus.theochem.tuwien.ac.at
Onderwerp: RE: [Wien] how to use non-default ROTLOC (local rotation matrix)



> What exactly goes wrong in the Brillouin zone integration when the mesh is not symmetrized?
> 
> And why do I get problems already in dstart - a program which doesn't care about the k-mesh?  (case.outputd is identical to the calculation with the default rotloc from symmetry, but running lapw5 on the clmsum created by dstart gives case.rho which are not identical - most numbers are about the same, but some differ by some %).
> 
> Given the case.struct and case.in2 in my previous mail, and using a full k-mesh, should I be able to run the calculation?

If I understood you correctly, than you used a different locrot matrix,
but you also changed the LM list (and included "all" LM) ?

If you did just that, my understanding would have been, that dstart gives
"identical" starting density (when plotting it with lapw5). If it does
not, I'm a bit surprised (not too much since I never really
understood how the nonspherical density components where generated), but
since overlapping spherical densities are still "pretty" spherical, I
never cared to check this.

k-mesh:
When you look at the electron density of say a single X-point, its density
does NOT have the symmetry of the crystal (eg. in cubic case, it may
contain only a px orbital, but no py,pz). Only when adding the equivalent
X-points (100, 010 and 001) you get a density which is symmetric as it
should be.

Alternatively one can "symmetrize" via selection of LM. In cubic we would
only use LM=40, so via the Gaunt selection rules l=1 (p) orbitals do not
contribute to this "non-spherical" term and the resulting density is
spherical as it should be.

Of course, the 3rd method would be that of lapwdm, where you "symmetrize
the wavefunction of his single X-point and "implicitly" sum over all
X-points.


                                      P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------

_______________________________________________
Wien mailing list
Wien at zeus.theochem.tuwien.ac.at
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien


-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/ms-tnef
Size: 7049 bytes
Desc: not available
Url : http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20050322/514c9418/attachment.bin