I don't think there is a problem with symmetso.
CrSb with AF-structure for Cr reduces the 24 symops to 12.
In this hexagonal lattice, the 2 Sb atoms are connected via inversion
symmetry.
Considering SO in 001 direction does not change this. And this is
definitely ok. (due to hex symmetry you already have a preferred
direction and SO in 001 does not change this.
100 direction, however, make a and b directions inequivalent and
symmetry is reduced. Inversion is still kept, thus the 2 Sb atoms remain
equivalent.
Am 24.02.2026 um 17:28 schrieb Lukasz Plucinski:
Dear Prof. Blaha,
Thank you again for the quick response.
The particular case is with CrSb (NiAs lattice) and M=001.
1. Cr and Sb atoms connected: space group 194, 24 symmetry operations
2. Disconnect Cr atoms into Cr1 and Cr2: space group 164, 12 symmetry
operations
*3. init_so_lapw with M=001: 12 symmetry operations remain (but I think
only 6 should remain)*
4. init_so_lapw with M=100: 4 symmetry operations remain
I might be doing something wrong, but this is how things work for me
with automatic scripts. I can provide the case.struct if needed for testing.
Best,
Lukasz
On Tue, Feb 24, 2026 at 3:28 PM Peter Blaha <[email protected]
<mailto:[email protected]>> wrote:
I don't know what you mean by: "the reduction of symmetries (due to SOC)
did not work properly " ??? Specific example, not just "feelings".
You can always put your symmetry "by hand" into case.struct, but
i) make sure you do not put (or keep) symmetries, which are not present.
ii) eventually, you need to split equivalent positions, if the
corresponding sym.op was removed
iii) The symmetry operations in case.struct, however, must also match
conditions for the local rotation matrix AND the LM list in case.in2.
You CANNOT remove some sym.ops and keep case.in2 the same.
The k-mesh follows automatically from the symmetry operations in your
struct file (and inversion is added, unless -so is specified).
Please note: When comparing total energies with different magnetization
directions in case.inso, it is strongly recommended to use a "common
symmetry" (eventually P1 or P-1) with identical number of inequivalent
atoms.
So test init_so with ALL magn.directions (always using the original
struct file) and compare the symmetries. Then decide on a common
symmetry.
A possible "trick" is always to displace an atom or change a,b,c a bit
until you get the desired symmetry. After init_lapw, change the
positions or a,b,c back to the original values (but do not run
init_lapw
again).
> instgen_lapw -ask
> init_lapw -b -sp -prec 1 ...
> init_so_lapw
> runsp -so -p -ec ... -cc ...
>
> In the init_so_lapw script I can choose the magnetization
quantization
> axis (which becomes the Neel vector) and then symmetso automatically
> reduces the symmetries (if needed) and creates the new
case.struct and
> then a proper klist for SOC.
>
> Sometimes I have a feeling that the reduction of symmetries (due
to SOC)
> did not work properly (for example it was not reduced, even if I
might
> expect some reduction). I would then like to reduce the
symmetries by
> hand (they are listed in the case.struct file). Is there a way to
reduce
> symmetries "by hand" in the case.struct and still run the
calculation
> properly (with reduced symmetries and adequate klist)?
>
> Best,
> Lukasz
>
>
>
>
> On Fri, Feb 20, 2026 at 7:32 PM Peter Blaha
<[email protected] <mailto:[email protected]>
> <mailto:[email protected]
<mailto:[email protected]>>> wrote:
>
> There is a file case.normsoup/dn.
> It contains the up/dn charges for each eigenvalue after x
lapwso -up.
>
> In the next version, x qtl -so -up will provide a
case.qtlup/dn
> file, where the last column contains the total up (dn) charge.
>
> Of course, the interstital may contain spin-polarization, but
usually
> this will be small as compared to the spin-polarization of
the atoms.
>
>
> Am 20.02.2026 um 10:31 schrieb Lukasz Plucinski:
> > Dear All,
> >
> > I often use qtl to calculate "fat bands". With SOC this
allows to
> > attribute spin-polarization to bands, in spaghetti plots.
> >
> > I understand that both qtl and lapw2 only operate on charge
> within MTs.
> > Is there any way to get total charge per E(k) eigenvalue, that
> means a
> > sum of MT and interstitial? Of course then there is no
> > orbital character, but just for spin it should not matter.
> >
> > lapw5 can plot real space charge densities, so it seems the
> information
> > is there.
> >
> > I ask because I think in some materials interstitial spin for
> E(k) may
> > not be negligible and it would be useful if WIEN2k can
plot it.
> >
> > Best,
> > Lukasz
> >
> > _______________________________________________
> > Wien mailing list
> > [email protected]
<mailto:[email protected]>
> <mailto:[email protected]
<mailto:[email protected]>>
> > http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
<http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien> <http://
> zeus.theochem.tuwien.ac.at/mailman/listinfo/wien <http://
zeus.theochem.tuwien.ac.at/mailman/listinfo/wien>>
> > SEARCH the MAILING-LIST at: http://www.mail-archive.com/
<http://www.mail-archive.com/>
> [email protected]/index.html <http://
[email protected]/index.html> <http://www.mail-
<http://www.mail->
> archive.com/[email protected]/index.html <http://
archive.com/[email protected]/index.html>>
>
> --
>
-----------------------------------------------------------------------
> Peter Blaha, Inst. f. Materials Chemistry, TU Vienna, A-1060
Vienna
> Email: [email protected]
<mailto:[email protected]> <mailto:[email protected]
<mailto:[email protected]>>
> WWW: http://www.imc.tuwien.ac.at <http://
www.imc.tuwien.ac.at> <http://www.imc.tuwien.ac.at <http://
www.imc.tuwien.ac.at>>
> WIEN2k: http://www.wien2k.at <http://www.wien2k.at> <http://
www.wien2k.at <http://www.wien2k.at>>
>
-------------------------------------------------------------------------
>
> _______________________________________________
> Wien mailing list
> [email protected]
<mailto:[email protected]>
<mailto:[email protected]
<mailto:[email protected]>>
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien <http://
zeus.theochem.tuwien.ac.at/mailman/listinfo/wien> <http://
> zeus.theochem.tuwien.ac.at/mailman/listinfo/wien <http://
zeus.theochem.tuwien.ac.at/mailman/listinfo/wien>>
> SEARCH the MAILING-LIST at: http://www.mail-archive.com/
<http://www.mail-archive.com/>
> [email protected]/index.html <http://
[email protected]/index.html> <http://www.mail-
<http://www.mail->
> archive.com/[email protected]/index.html <http://
archive.com/[email protected]/index.html>>
>
>
> _______________________________________________
> Wien mailing list
> [email protected]
<mailto:[email protected]>
> http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien <http://
zeus.theochem.tuwien.ac.at/mailman/listinfo/wien>
> SEARCH the MAILING-LIST at: http://www.mail-archive.com/
[email protected]/index.html <http://www.mail-
archive.com/[email protected]/index.html>
--
-----------------------------------------------------------------------
Peter Blaha, Inst. f. Materials Chemistry, TU Vienna, A-1060 Vienna
Email: [email protected] <mailto:[email protected]>
WWW: http://www.imc.tuwien.ac.at <http://www.imc.tuwien.ac.at>
WIEN2k: http://www.wien2k.at <http://www.wien2k.at>
-------------------------------------------------------------------------
_______________________________________________
Wien mailing list
[email protected] <mailto:[email protected]>
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien <http://
zeus.theochem.tuwien.ac.at/mailman/listinfo/wien>
SEARCH the MAILING-LIST at: http://www.mail-archive.com/
[email protected]/index.html <http://www.mail-
archive.com/[email protected]/index.html>
_______________________________________________
Wien mailing list
[email protected]
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:
http://www.mail-archive.com/[email protected]/index.html
--
-----------------------------------------------------------------------
Peter Blaha, Inst. f. Materials Chemistry, TU Vienna, A-1060 Vienna
Email: [email protected]
WWW: http://www.imc.tuwien.ac.at WIEN2k: http://www.wien2k.at
-------------------------------------------------------------------------
_______________________________________________
Wien mailing list
[email protected]
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien
SEARCH the MAILING-LIST at:
http://www.mail-archive.com/[email protected]/index.html
