[Wien] Bandstructure calculation of Bi

[Jakob B. Wagner]

Thank you all. It turned out that it was the missing -so option in spaghetti
run via w2web, which was the problem. The bandstructure now showing the
right trends. The rest is just fine-tuning.

Jakob

----- Original Message -----
From: "Carrier, Pierre" <Pierre_Carrier at nrel.gov>
To: <wien at zeus.theochem.tuwien.ac.at>
Sent: Thursday, June 03, 2004 6:06 PM
Subject: RE: [Wien] Bandstructure calculation of Bi


> Hi,
>
> So, in order to get a spin-orbit band structure using the command line, at
least the following procedure works (but probably not optimal) :
>
> run_lapw -so
> (( generate the bands in <case>.in1[c], and change INPUT file from 4 to 5,
or use -band ))
> x lapw1 [-c] -so [-band]
> x lapwso -so               (I'm not totally sure about the necessity to
put -so here, but it generates properly the <case>.energyso file you want)
> x spaghetti -so           (don't forget the -so flag here, because it will
read the <case>.energy file that also exists (because of the second step; x
lapw1) and
>                                  you will not get the SO bands from the
<case>.energyso file you want to read ).
>
> Good luck,
> Ciao!
> Pierre
> -----Original Message-----
> From: Jorissen Kevin [mailto:wien-admin at zeus.theochem.tuwien.ac.at]On
Behalf Of Jorissen Kevin
> Sent: Thursday, June 03, 2004 7:01 AM
> To: wien at zeus.theochem.tuwien.ac.at
> Subject: RE: [Wien] Bandstructure calculation of Bi
>
>
>
> A possible stupid mistake :
> w2web doesn't know that you've been using -so.  It remembers things like
parallel and spin polarization, but *not* spin orbit coupling.
> So if you just click the buttons in w2web to make the spaghetti, you're
probably still using vector-files instead of vectorso-files.
> Solution : go to the commandline, and use the -so switch.
>
> Kevin.
>
>
>
> -----Oorspronkelijk bericht-----
> Van: Jakob B. Wagner [mailto:wagner_j at fhi-berlin.mpg.de]
> Verzonden: wo 6/2/2004 2:06
> Aan: wien at zeus.theochem.tuwien.ac.at
> CC:
> Onderwerp: [Wien] Bandstructure calculation of Bi
>
>
>
> Hi all
>
> I'm working on bandstructure calculation on semimetallic Bi with the A7
> structure.
> According to X. Gonze (PRB v41, p11827 (1990)) is the spin-orbit coupling
in
> Bi significant and band-splitting is observed around the Gamma-point.
> I have tried to add a relativitic local orbital and running the scf-cycle
> including the spin-orbit interaction, but without any observable changes
in
> the bandstructure compared to the one cacluated without spin-orbit
> interaction.
>
> Does anyone have any experience with calculations on bismuth or knowledge
of
> a step by step tutorial dealing with spin-orbit coupling calculations in
> Wien2k to rule out some of the (stupid) mistakes I'm fighting with?
>
> Jakob
>
> ------------------------------------------
>
> Jakob B. Wagner, Ph.D.
> FHI-MPG Berlin
>
> ----------------------key:1-0.0735-11600-23.05:fhi--------------------
>
> _______________________________________________
> Wien mailing list
> Wien at zeus.theochem.tuwien.ac.at
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
>
>
>
>