[SOLVED] Problem with group definition for magnetic structu

[Thomas]

Hello

I would like to obtain the total energy of the Hematite (Fe2O3) and compare it with its non-magnetic case. In this type of calculations, the symmetry is really important, and remains the most important problem in input file.
The Hematite is built from the space group n°167 (trigonal structure), so with a=b and c, and angles equal to 90, 90 and 120.

Here my input file :

#Hematite

ndtset 2
spgroup 167
brvltt 7

#spin related quantities (only second dataset)
spinat2 0.0 0.0 4.0
0.0 0.0 2.0
nsppol2 2
nspden2 2
#generate the total density of states in both cases
prtdos 1

iscf 5
kptopt 1
prtden 1
tolvrs 1.0d-10
acell 5.032 5.032 13.733 angstr
ecut 20.0
natom 2
nband 15
ngkpt 4 4 4
nshiftk 1

nstep 100
ntypat 2
occopt 3
rprim 0.000 -1.000 1.000
0.866 0.500 1.000
-0.866 0.500 1.000

angdeg 90.0 90.0 120.0
tsmear 0.01
typat 1 2
xred 0.0000 0.0000 0.3553
0.3059 0.0000 0.2500
znucl 26 8

The calculation stops almost immediatly... The log file is giving this as ERROR message :

chkorthsy: ERROR -
The symmetry operation number 2 does not preserve
vector lengths and angles.
The value of the residual is 1.6945E+05.
Action : modify rprim, acell and/or symrel so that
vector lengths and angles are preserved.
Beware, the tolerance on symmetry operations is very small.

Thanks for the help!

[gonze]

Dear Thomas,

Even without considering the magnetic characteristics, your input file
is wrong :
- With natom 2, defining one Fe atom and one O atom, you will describe FeO,
not Fe2O3 . If you want to have ABINIT generate the full primitive cell from its irreducible
part, you should rely on the use of the input variable natrd (in addition to using natom,
that should be a multiple of 5 .
- You specify both rprim and angdeg, very strange
- You likely misuse acell, because the values of acell are the scaling factors for the
vectors defined by rprim, while you likely expect them to work together with your definition
provided by angdeg.
May I suggest you have a look at the input file for Al2O3,
v1/Input/t96.in
in which you will find different ways of specifying the Al2O3 structure : this might give
you a hint on how to generate your own input file.

Best wishes,
X

[Thomas]

Thank you gonze for the first and complete corrections!

The file t96.in has been very useful! It corrects all my major problems of the last days (not only on the magnetic structure).
Thank you once again!