[Wien] correspondence between QTL and LMAX parameters

[Peter Blaha]

Yes, reducing LMAX in case.inso is a valid option (similar as switching off SO
for certain atoms) for analysis. On the other hand, in most cases it will be
pretty clear (or just check the qtl or help files), which character dominates
for a certain band (eigenvalue).

> Thank you for your answer. I have one more question concerning the topic.
> Is the manipulation of the LMAX parameter for spin-orbit calculations only
> (case.inso) senseless too? Is it valid to say that for LMAX=1 (in case.inso)
> the results are not affected by spin-orbit coupling effects due to 
> d-orbitals?
> Or in other words, can one study spin-orbit coupling effects as a function
> of LMAX? With the many thanks in advance,
> Martin Gmitra
> 
> 
> On Sun, Jan 18, 2009 at 10:06 AM, Peter Blaha 
> <pblaha at theochem.tuwien.ac.at <mailto:pblaha at theochem.tuwien.ac.at>> wrote:
> 
>     If LMAX=2 is the value of l-max defined in case.in1; your calculations
>     are complete nonsense.
> 
>     An APW method is NOT a "LCAO" calculation. With atomic orbitals (LCAO)
>     you may restrict yourself to the "chemical l" values, since
>     wavefunctions are extended and overlap with those on other sites, i.e. a
>     s-function on site 1 will have some weight on neughboring sits 2,3,...
> 
>     In APW we have a spatial decomposition. Our "s-function" is truncated at
>     RMT, and they are connected to plane waves outside. Thus there is NO
>     (direct) overlap between wave functions from different sites. Instead,
>     the tails of an s or p function on site 1 will be reexpanded on site 2.
>     This "reexpansion" results in eg. f-character (and higher ones), since
>     the tails of an s function on site 1 can equally well be represented by
>     f-functions centered on site 2.
> 
>     Martin Gmitra schrieb:
>      > Dear Wien2k users,
>      >
>      > I have considered LMAX=2 and LNSMAX=2 for Aluminium. The DOS contains
>      > non zero f character. Why there is a non zero contribution to the
>     f-like
>      > QTL?
>      > The wave function should have s,p and d character only, is not it?
>      >
>      > Thanks,
>      > Martin Gmitra
>      >
>      >
>      >
>      >
>     ------------------------------------------------------------------------
>      >
>      > _______________________________________________
>      > Wien mailing list
>      > Wien at zeus.theochem.tuwien.ac.at
>     <mailto:Wien at zeus.theochem.tuwien.ac.at>
>      > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>     _______________________________________________
>     Wien mailing list
>     Wien at zeus.theochem.tuwien.ac.at <mailto:Wien at zeus.theochem.tuwien.ac.at>
>     http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien

-- 

                                       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/
--------------------------------------------------------------------------