[Wien] BAND-PLOT BUG

[Peter Blaha]

After you have done a scf calculation (or use restore_lapw XXX) you 
should in general NEVER again run

x lapw0   (why would you create a new potential ???
x orb -up/dn   (why would you create new vorbup/dn files ??

For a band structure just do:

x lapw1 -p -band -up (dn) -orb
x lapw2 -p -band -up (dn) -qtl   # but NOT -orb !!!
x spaghetti -p -up (dn)

for lapw1 the -orb switch adds the LDA+U potential to the hamiltonian. 
So you clearly need it.
for lapw2 -orb directs lapw2 to calculate a new dmatup (dn) file. But 
since you have -qtl, this is done for all states, not just the occupied 
ones and for a k-mesh along the symmetry lines and not a uniform mesh in 
the BZ. Thus you "destroy" your density matrices.

And if you rerun    x orb   afterwards, a new orbital potential (vorb) 
is calculated from the wrong density matrix (dmat); clearly the results 
must be wrong.

It has NOTHING to do with   clean_lapw

Hopefully, you have issued a   save_lapw xxx   before that, than you can 
recover using restore xxx; otherwise you have destroyed your scf 
solution (there might be a case.vorbup_old and case.dmatup_old (dn), 
which you probably could also copy back.


Am 15.12.2017 um 21:23 schrieb 徐远骥:
> Dear all:
> 
>     I am doing GGA+U calculations using the newest WIEN2K_17 version 
> now. After I get the covered GGA+U results, I am using the following 
> commands to plot the band:
> 
>     x lapw0 -p
>     x orb -up -p
>     x orb -dn -p
>     x lapw1 -orb -band -up -p
>     x lapw1 -orb -band -dn -p
>     x lapw2 -orb -band -qtl -up -p
>     x lapw2 -orb -band -qtl -dn -p
>     x spaghetti -up -p
>     x spaghetti -dn -p
> 
>    and I am get the right results!
> 
> 
>    Then, I using "clean -s" to clean the results and running the 
> following commands:
> 
>     x lapw0 -p
>     x orb -up -p
>     x orb -dn -p
>     x lapw1 -orb -band -up -p
>     x lapw1 -orb -band -dn -p
>     x lapw2  -band -qtl -up -p
>     x lapw2  -band -qtl -dn -p
>     x spaghetti -up -p
>     x spaghetti -dn -p
> 
>    sure, I am get the wrong results.(This means the new version 
> WIEN2K_17 is different with WIEN2K_14 when plotting gga+u bands).
> 
>    Then I using "clean -s" to clean the results again and running the 
> following commands again in the same directory:
> 
>     x lapw0 -p
>     x orb -up -p
>     x orb -dn -p
>     x lapw1 -orb -band -up -p
>     x lapw1 -orb -band -dn -p
>     x lapw2  -orb -band -qtl -up -p
>     x lapw2  -orb -band -qtl -dn -p
>     x spaghetti -up -p
>     x spaghetti -dn -p
> 
> But the results is not change again and can't go back to the right results.
> 
> I don't know if this situation appears in your calculations ?  If it 
> appears in your calculations, you should be careful.
> 
> Best wishes!
> 
> Yours
> 
> 
> 
> 
> 
> --
> Xu yuanji(徐远骥)
> T03,Institute of Physics
> University of Chinese Academy of Sciences (UCAS)
> Mobile: 0086-153 1372 6987
> E-mail: yuanjixu at iphy.ac.cn <mailto:yuanjixu at iphy.ac.cn>
> 
> 
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
> SEARCH the MAILING-LIST at:  http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html
> 

-- 
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300             FAX: +43-1-58801-165982
Email: blaha at theochem.tuwien.ac.at    WIEN2k: http://www.wien2k.at
WWW: 
http://www.imc.tuwien.ac.at/tc_blaha-------------------------------------------------------------------------