[Wien] unit in optimization results

[Gerhard Fecher]

Note that a0B= 0.5291772083 Angstroem, so be sure you convert the lattice parameter correctly,
if you compare the calculated in atomic units (a0B) and experimental in Angstroem.

As recognized by Miao, the cubic fcc cell contains 4 atoms (for elements) but the primitive only one. That means that the cubic cell is 4 times as large as the primitive one.
So you find the cubic lattice parameter from (V_min*4)^(1/3) in atomic units. You may wish to convert either first the volume or later the lattice parameter to Angstroem for comparison with the experiment.
Similar equations are found for other non-primitive cells like bcc or bct.

Just to mention, this is not a problem of Wien2k,it is naot a problem at all, as the use of the primitive cell volume is ALWAYS correct !

Ciao
Gerhard


-----Ursprüngliche Nachricht-----
Von: wien-bounces at zeus.theochem.tuwien.ac.at im Auftrag von RK_Thapa
Gesendet: Mo 02.10.2006 11:33
An: A Mailing list for WIEN2k users
Betreff: Re: [Wien] unit in optimization results
 
Dear Dr. Miao,
I had also noted this and posted a querry on 28th Sept. If you divide the
volume of the unit cell you get (by optimization)  by 2 (for bcc) and 4 (for
fcc), one get the lattice constant almost equal to the experimental
value.Otherwise, there is a difference of almost 50 % with the experimental
value of a0.
Regards,
R.K.Thapa
==============
Shu Miao wrote:

> Dear Wien2k users
> I'm reporting a problem in the optimization results. The optimize program
> in the new version will automatically fit the optimization results to find
> out the equilibrium volume. According to previous posts in the mailing
> list, this should be the volume in bohr^3 and the lattice parameter of a
> cubic cell can be obtained by taking the cubic root. However, this seems
> not true. I tried with MgO, a cubic material with space group Fm-3m. The
> a0 is about 7.988 bohr. The optimized volume (V0) is 130.1788. This gives
> a cubic root 5.068 bohr. This is far from what it should be. But this is
> not wrong. I find out that the V0 is not the volume of the unit cell, but
> just the volume of a pair of Mg/O. Since this is a FCC structure, it
> contains 4 pairs of Mg/O. If we multiply 130.1788 by a factor 4, the cubic
> root will give correct lattice parameter.
> Has anyone noted this problem? Is it also true for other systems? If so,
> it can be a lot of trouble to get the optimized lattice parameters in a
> low symmetry, multi-atom structure.
>
> Thanks
>
> Shu Miao
> _______________________________________________
> 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: 3956 bytes
Desc: not available
Url : http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20061003/0a9be925/attachment.bin