Dear Fleur community,

When I calculated the collinear ferromagnetic magnetic system, I set "&soc 1.5707963 0.0 /" (meaning the easy axis is along x-axis) in the inp_system document. However, I cannot run fleur.x (the versions is v26e). And the error message is "Symmetry incompatible with SOC spin-quantization axis , do not perform self-consistent calculations ! dimen7 : symmetry & SOC". This is my inp_system documents,

#######################################
EuCd2As2 ! latest template

&input film=f cartesian=f cal_symm=t /

1.000000000 0.000000000 0.000000000 ! a1
-0.500000000 0.866025404 0.000000000 ! a2
0.000000000 0.000000000 1.000000000 ! a3 and dvac
8.5169890312826 ! lattice constant in bohr. 1 bohr= 0.529177249 ai
1.0 1.0 1.6407942932135 ! scale

5
63.00 0.0000000000000000 -0.0000000000000000 0.0000000000000000
48.01 0.6666666666666667 0.3333333333333333 0.3661496391821320
48.02 0.3333333333333333 0.6666666666666667 0.6338504258178699
33.01 0.6666666666666667 0.3333333333333333 0.7543478004283544
33.02 0.3333333333333333 0.6666666666666667 0.2456521995716456

&soc 1.5707963 0.0 /

&atom element="Eu" id=63.00 bmu=7.0 /
&atom element="Cd" id=48.01 bmu=0.0 /
&atom element="Cd" id=48.02 bmu=0.0 /
&atom element="As" id=33.01 bmu=0.0 /
&atom element="As" id=33.02 bmu=0.0 /

&kpt div1=16 div2=16 div3=9 /

&end /
############################

Can someone tell me where the problem is ?

Thank you very much.

xiuxian

Than you reply.

In my directory, only the input document (inp_system) exists. First, I use the inpgen.x generate the inp and other document; Then running fleur.x and get above error. However, if I set " &soc 0.0 0.0 / " in the inp_system document, there is no error. I do not understand why " &soc 1.5707963 0 / " is wrong ? \

Additionally, there is a Eu atom in my system, it should be a magnetic atom and has a magnetic moment, which is confirmed by VASP. However, FLEUR does not show the magnetc moment, why ?

Dear xiuxian,
there seems to be a problem related to the fact that the coordinate system in inp_system is 60 degrees rotated to the setup chosen in the inp-file. This selects the wrong mirror plane. An easy fix is to rotate the system by specifying:

0.500000000 -0.866025404 0.000000000 ! a1
0.500000000 0.866025404 0.000000000 ! a2
...
or

&lattice latsys='hP' a0=8.5169890312826 a=1.0 c=1.6407942932135 /

for the Bravais Matrix in the inp_system file.

This problem does not occur in the new fleur versions, since the Bravais matrix is stored there unrotated in inp.xml. A similar thing can be done in version 0.26 by specifying it directly in the inp file:

any any ,invs=F,zrfs=F,invs2=F,jspins=2,l_noco=F,l_J=F
     8.51698903     0.00000000     0.00000000
    -4.25849451     7.37592886     0.00000000
     0.00000000     0.00000000    13.97462700    13.97462700     1.00000000

As for the magnetic moment of Eu: also here you get it. Maybe it's good to specify a few local orbitals for Eu like:

&atom element="Eu" id=63.00 lo="5s 5p" bmu=7.0/

Thank you !
Using the new lattice vector, this error is solved.