Re: [Wien] wien2wannier with SOC but without sp

[Gavin Abo]

Regarding your question 1) below, yes, it is necessary to add the -up 
and -dn to the "x w2w -so" command as the Wien2Wannier 2.0 User’s Guide 
says on page 5:


"Note: it does not make sense to run x w2w -so without either -up or -dn;"

Regarding your question 2) below, in the thread of posts for a 
wien2wannier spin polarized with spin orbit coupling calculation [2], it 
looks like they did a converged scf calculation with


init -b -sp
runsp
initso
runsp -so

before doing the wien2wannier calculation [3].  This implies using a 
counterpart scf calculation for a non-spin polarized with spin orbit 
coupling calculation.


As you probably know, a non-spin polarized with spin orbit coupling scf 
calculation could be done with something like


init -b
run
initso
run -so

However, the wien2wannier article [4] seems to suggest that spin orbit 
calculations done with it must be spin polarized:


"Spin-orbit calculation must be run as formally spin-polarized even if 
the polarization is zero."


However, while it might be possible to “mimick” a spin polarized 
calculation similar to what is described for a optic calculation in 
section "8.17.1 Execution" on page 177 in the WIEN2k 17.1 usersguide [6].


The accepted scf procedure for wien2wannier [7] seems to be to use 
runsp_c instead:


init -b -sp
runsp_c
initso
runsp_c -so

[1] 
https://github.com/wien2wannier/wien2wannier/releases/download/v2.0.0/wien2wannier_userguide.pdf
[2] 
http://wien.zeus.theochem.tuwien.ac.narkive.com/lEYFiNqR/error-in-wien2wannier-for-spin-orbit-coupling-case

[3] https://github.com/wien2wannier/wien2wannier/wiki/wien2wannier-with-SOC
[4] https://arxiv.org/abs/1004.3934v1
[5] 
http://wien.zeus.theochem.tuwien.ac.narkive.com/uipUZHQ4/wien2wannier-for-non-spinpolarized-soc-cases

[6] http://susi.theochem.tuwien.ac.at/reg_user/textbooks/usersguide.pdf
[7] 
http://zeus.theochem.tuwien.ac.at/pipermail/wien/2014-December/022086.html


On 1/2/2018 1:50 AM, Sahra Sahraii wrote:

Dear wien2k users and developer

I have a question regarding to wien2wannier with soc, but without sp.

I found the  work flow in Wien2wannier user guide  for spin-orbit 
coupling :


|$ prepare_w2wdir W $ init_w2w -up ... > findbands -so -all -1 1 
(13:30:56) > write_inwf -f W (13:31:01) ... > minimal and maximal band 
indices [Nmin Nmax]? 41 46 > next proj. (6 to go; Ctrl-D if done)? 
1:dt2g added 3 projections: 2:dxy,dxz,dyz > next proj. (3 to go; 
Ctrl-D if done)? 1:dt2g added 3 projections: 2:dxy,dxz,dyz --> 6 
bands, 6 initial projections ... $ x lapw1; x lapwso $ x w2w -so -up; 
w2w -so -dn $ x wannier90 -so|
 I wonder that for considering spin orbit coupling I also should 
consider spin up and down.

1) Is it necessary to add up and dn in the above commands.

2) I also wants to know  if   I should do a converged  so Wien2k 
calculation  before running wien2wannier  or I  should do a non so 
calculation?

Thank you in advance
Best
Sahraii


___
Wien mailing list
Wien@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

[Wien] wien2wannier with SOC but without sp

[Sahra Sahraii]

Dear wien2k users and developer
I have a question regarding to wien2wannier with soc, but without sp. 
I found the  work flow in Wien2wannier user guide  for spin-orbit coupling :
$ prepare_w2wdir W
$ init_w2w -up
...
> findbands -so -all -1 1 (13:30:56)
> write_inwf -f W (13:31:01)
...
> minimal and maximal band indices [Nmin Nmax]? 41 46
> next proj. (6 to go; Ctrl-D if done)? 1:dt2g
added 3 projections: 2:dxy,dxz,dyz
> next proj. (3 to go; Ctrl-D if done)? 1:dt2g
added 3 projections: 2:dxy,dxz,dyz

--> 6 bands, 6 initial projections
...
$ x lapw1; x lapwso
$ x w2w -so -up; w2w -so -dn
$ x wannier90 -so I wonder that for considering spin orbit coupling I also 
should consider spin up and down.Is it necessary to add up and dn in the above 
commands.
I also wants to know  if   I should do a converged  so Wien2k calculation  
before running wien2wannier  or I  should do a non so calculation? 
Thank you in advanceBestSahraii







___
Wien mailing list
Wien@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

Re: [Wien] wien2wannier with SOC in WIEN2k 16.1

[Peter Blaha]

I've corrected the version on the web. Just download SRC.tar.gz.

It also contains a small fix for all run*_lapw scripts (VERSION in scf 
files corrected).


On 12/15/2016 08:44 PM, Elias Assmann wrote:

Dear Kefeng Wang,

On 12/15/2016 06:01 PM, Kefeng Wang wrote:

2.   Then I run “init_w2w –up”, after k mesh (10 10 10) and find
band, “write_inwf” shows “error: unrecognized arguments: -up” and
init_w2w exit.  I also found the help file did not include [-up/dn]
option for write_inwf command. So I manually run write_inwf and then
copy case.inwf to case.inwfup and case.inwfdn.  Is this right?


Oops, this isn't right!  For some reason, the latest version of that
script did not make it into the 16.1 distribution.  You can get the
correct version from .

Otherwise, here are step-by-step instructions for wien2wannier with SOC:
.


1.   Then I run *prepare_w2wdir WANN, but this command cannot
recognize Case.spaghettiup_ene and Case.Spaghettidn_ene file,
instead it shows “cannot find spaghetti_ene file”. I copy them
manually. Is something wrong?


I don't know.  Help me help you by showing the precise error message.


3.   After that, x lapw1 –up; x lapw1 –dn; x lapwso –up; x w2w
–so –up; xw2w –so –dn; and finally x wannier90 –so. Everything goes
smooth, I got reasonable band structure except that there is only one
Case_band.dat file. I tried run “x wannier90 –so  -up”, but it shows
“wannier90: -so should be used without –up/-dn”. Shall we expect two
band data file for two spin channels or somehow I am stupid?


This seems right.  Why do you expect separate bands for up and down?
With SOC, that is not a good quantum number anymore.


Also, I found a recipe for SOC case without spin polarization, but it
seems outdated. What should I do with this new version for the SOC
case without spin polarization?


Which one?  Why do you think it is outdated?


Elias



--

  P.Blaha
--
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300 FAX: +43-1-58801-165982
Email: bl...@theochem.tuwien.ac.atWIEN2k: http://www.wien2k.at
WWW:   http://www.imc.tuwien.ac.at/TC_Blaha
--
___
Wien mailing list
Wien@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

Re: [Wien] wien2wannier with SOC in WIEN2k 16.1

[Fecher, Gerhard]

With SO each wave function will still have  a "up" and a ""down" part, but they 
are connected and cannot be separated,
(indeed not s(1/2,+1/2) or s(1/2,-1/2), at least in the 2 component version 
they are purely up or down, as well as all others with j=l+s, mj=+-j).
how does Wannier deal with the |j, m_j> or say better with the relativistic 
|kappa, mu> wave functions ?

Ciao
Gerhard

DEEP THOUGHT in D. Adams; Hitchhikers Guide to the Galaxy:
"I think the problem, to be quite honest with you,
is that you have never actually known what the question is."


Dr. Gerhard H. Fecher
Institut of Inorganic and Analytical Chemistry
Johannes Gutenberg - University
55099 Mainz
and
Max Planck Institute for Chemical Physics of Solids
01187 Dresden

Von: Wien [wien-boun...@zeus.theochem.tuwien.ac.at] im Auftrag von Elias 
Assmann [elias.assm...@gmail.com]
Gesendet: Freitag, 16. Dezember 2016 06:56
An: A Mailing list for WIEN2k users
Betreff: Re: [Wien] wien2wannier with SOC in WIEN2k 16.1

On 12/15/2016 10:13 PM, Kefeng Wang wrote:
> Thanks a lot for your help and explanation. With the new script
> write_inwf_lapw, everything works fine now.

Great.

> ->For this one, the exact error message is  “recommended file
> ‘GaAs.spaghetti_ene’ not found (will continue)”. But it does not affect
> the calculation since spaghetti file is only for reference. I believe
> the command did not recognize the “spaghettiup_ene” and
> “spaghettidn_ene” file.

If I am not mistaken (certainly people on this list can correct me if I
am), with SOC you should have called spaghetti as ‘x spaghetti -so’,
which would have produced GaAs.spaghetti_ene.  In the presence of SOC, a
priori it does not make sense to talk of separate “↑” and “↓” states.

Elias


___
Wien mailing list
Wien@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

Re: [Wien] wien2wannier with SOC in WIEN2k 16.1

[Elias Assmann]

On 12/15/2016 10:13 PM, Kefeng Wang wrote:
> Thanks a lot for your help and explanation. With the new script
> write_inwf_lapw, everything works fine now.

Great.

> ->For this one, the exact error message is  “recommended file
> ‘GaAs.spaghetti_ene’ not found (will continue)”. But it does not affect
> the calculation since spaghetti file is only for reference. I believe
> the command did not recognize the “spaghettiup_ene” and
> “spaghettidn_ene” file.

If I am not mistaken (certainly people on this list can correct me if I
am), with SOC you should have called spaghetti as ‘x spaghetti -so’,
which would have produced GaAs.spaghetti_ene.  In the presence of SOC, a
priori it does not make sense to talk of separate “↑” and “↓” states.

Elias




signature.asc
Description: OpenPGP digital signature
___
Wien mailing list
Wien@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

Re: [Wien] wien2wannier with SOC in WIEN2k 16.1

[Kefeng Wang]

Dear Elias, 

 

Thanks a lot for your help and explanation. With the new script
write_inwf_lapw, everything works fine now.

 

> 1.   Then I run *prepare_w2wdir WANN, but this command cannot 

> recognize Case.spaghettiup_ene and Case.Spaghettidn_ene file,

> instead it shows "cannot find spaghetti_ene file". I copy them

> manually. Is something wrong?

 

I don't know.  Help me help you by showing the precise error message.

 

 

->For this one, the exact error message is  "recommended file
'GaAs.spaghetti_ene' not found (will continue)". But it does not affect the
calculation since spaghetti file is only for reference. I believe the
command did not recognize the "spaghettiup_ene" and "spaghettidn_ene" file.

 

 
> 3.   After that, x lapw1 -up; x lapw1 -dn; x lapwso -up; x w2w 
> -so -up; xw2w -so -dn; and finally x wannier90 -so. Everything goes 
> smooth, I got reasonable band structure except that there is only one
> Case_band.dat file. I tried run "x wannier90 -so  -up", but it shows
> "wannier90: -so should be used without -up/-dn". Shall we expect two
> band data file for two spin channels or somehow I am stupid?
 
This seems right.  Why do you expect separate bands for up and down?
With SOC, that is not a good quantum number anymore.
 
->I did not realize that. Thanks for pointing out. But if so, what happens
to the wannier function? After getting reasonable band structure, I tried to
plot wannier function. But It seems messed up.
I copied case.inwplot to case.inwplotup, and could ran "x wplot -up -so -wf
1" successfully ("x wplot -so -wf 1" did not work since there is ony *.vspup
file instead of *.vsp).
After that, if I try to run "wplot2xsf -up", it shows "wplot2xsf: Could not
open file `GaAs-WANN_centres.xyzup'.  Will proceed without shift." And if I
ran "wplot2xsf", it shows "No Wannier functions given and no `psink' files
found.  Nothing to do.".
 
Thanks again for your help.
 
Best
Kefeng
 

 

___
Wien mailing list
Wien@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

Re: [Wien] wien2wannier with SOC in WIEN2k 16.1

[Elias Assmann]

On 12/15/2016 06:01 PM, Kefeng Wang wrote:
> 2.   Then I run “init_w2w –up”, after k mesh (10 10 10) and find
> band, “write_inwf” shows “error: unrecognized arguments: -up” and
> init_w2w exit.  I also found the help file did not include [-up/dn]
> option for write_inwf command. So I manually run write_inwf and then
> copy case.inwf to case.inwfup and case.inwfdn.  Is this right?

PS: You did the right thing (provided you want to have the same trial
orbitals for both spins, of course).

Elias




signature.asc
Description: OpenPGP digital signature
___
Wien mailing list
Wien@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

Re: [Wien] wien2wannier with SOC in WIEN2k 16.1

[Elias Assmann]

Dear Kefeng Wang,

On 12/15/2016 06:01 PM, Kefeng Wang wrote:
> 2.   Then I run “init_w2w –up”, after k mesh (10 10 10) and find 
> band, “write_inwf” shows “error: unrecognized arguments: -up” and 
> init_w2w exit.  I also found the help file did not include [-up/dn] 
> option for write_inwf command. So I manually run write_inwf and then 
> copy case.inwf to case.inwfup and case.inwfdn.  Is this right?

Oops, this isn't right!  For some reason, the latest version of that
script did not make it into the 16.1 distribution.  You can get the
correct version from .

Otherwise, here are step-by-step instructions for wien2wannier with SOC:
.

> 1.   Then I run *prepare_w2wdir WANN, but this command cannot 
> recognize Case.spaghettiup_ene and Case.Spaghettidn_ene file,
> instead it shows “cannot find spaghetti_ene file”. I copy them
> manually. Is something wrong?

I don't know.  Help me help you by showing the precise error message.

> 3.   After that, x lapw1 –up; x lapw1 –dn; x lapwso –up; x w2w 
> –so –up; xw2w –so –dn; and finally x wannier90 –so. Everything goes 
> smooth, I got reasonable band structure except that there is only one
> Case_band.dat file. I tried run “x wannier90 –so  -up”, but it shows
> “wannier90: -so should be used without –up/-dn”. Shall we expect two
> band data file for two spin channels or somehow I am stupid?

This seems right.  Why do you expect separate bands for up and down?
With SOC, that is not a good quantum number anymore.

> Also, I found a recipe for SOC case without spin polarization, but it
> seems outdated. What should I do with this new version for the SOC 
> case without spin polarization?

Which one?  Why do you think it is outdated?


Elias

-- 
Elias Assmann

Wien2Wannier: maximally localized Wannier functions
  from linearized augmented plane waves

 http://wien2wannier.github.io/
  https://github.com/wien2wannier/wien2wannier/
___
Wien mailing list
Wien@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

[Wien] wien2wannier with SOC in WIEN2k 16.1

[Kefeng Wang]

Dear developers and user community,

 

I am trying to run wien2wannier in the new WIEN2k 16.1 on a simple case GaAs
with spin polarization and spin orbital coupling. I met some problems.

 

My platform and other information:

WIEN2k 16.1, x86_64, linux, intel compiler ifort and mkl (2016.1)

 

After the spin polarized scf run with SOC 

*runsp_c_lapw -so -ec 0.0001

 

I got the band structure for both spin channel.  Following is the procedure
I run wien2annier and the problem I met:

 

1.   Then I run *prepare_w2wdir WANN, but this command cannot recognize
Case.spaghettiup_ene and Case.Spaghettidn_ene file, instead it shows "cannot
find spaghetti_ene file". I copy them manually. Is something wrong?

 

2.   Then I run "init_w2w -up", after k mesh (10 10 10) and find band,
"write_inwf" shows "error: unrecognized arguments: -up" and init_w2w exit.
I also found the help file did not include [-up/dn] option for write_inwf
command. So I manually run write_inwf and then copy case.inwf to case.inwfup
and case.inwfdn.  Is this right?

 

3.   After that, x lapw1 -up; x lapw1 -dn; x lapwso -up; x w2w -so -up;
xw2w -so -dn; and finally x wannier90 -so. Everything goes smooth, I got
reasonable band structure except that there is only one Case_band.dat file.
I tried run "x wannier90 -so  -up", but it shows "wannier90: -so should be
used without -up/-dn". Shall we expect two band data file for two spin
channels or somehow I am stupid?

 

 

Also, I found a recipe for SOC case without spin polarization, but it seems
outdated. What should I do with this new version for the SOC case without
spin polarization?

 

Thanks a lot for your help in advance.

 

Best

Kefeng Wang

 

 

 

 

 

 

___
Wien mailing list
Wien@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

Re: [Wien] wien2wannier with SOC

[Elias Assmann]

Dear Jianxin,

Normally, wien2wannier usage questions should go to the Wien2k mailing
list for the possible benefit of other users.  I am copying to the list now.

On 09/16/2016 09:38 PM, Zhu, Jianxin wrote:
> x lapw1 –p
> x lapwso –p
> cp fccYb-W.fermi fccYb-W.fermiup
> cp fccYb-W.fermi fccYb-W.fermidn
> cp fccYb-W.vsp fccYb-W.vspup
> cp fccYb-W.vsp fccYb-W.vspdn
> x w2w –so –up –p
> x w2w –so –dn –p
> x wannier90 -so

This sounds like the “old” prescription -- copying vsp files should no
longer be necessary.  See also this page

(I just edited that to make it hopefully a little easier to understand).

> (I am using the
> wien2wannier coming with wien2k 14.2. I notice that you also post the
> program on the Github but I hope there is no severe change.)

In fact, that is the first thing I have to recommend: please get the
latest “master“ version from GitHub and try again.  There have been
important changes especially with regards to SOC.

If you have the same problems (“wiggly” Wannier bands) with the master
version and the procedure from the Wiki, please report back.


Elias






signature.asc
Description: OpenPGP digital signature
___
Wien mailing list
Wien@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