Warning(linengy) and calculated charge errors with Lu2O3

[Andrew Shyichuk]

Hi,

I am trying to calculate DOS of lutetium oxide Lu2O3 with elk 4.0.15 and stock species.
The input is database geometry, not optimized.

I've tried different settings and functionals and anyway I am getting the following issues:
1. Warning(linengy): could not find 7 linearisation energies in s.c. loop, always 7 and every loop.
2. Total charge error is normaly about 1, but it happened to be up to dozens once.
3. DOS is weird, bandgap is basically absent.

Also, I've noticed that checkmt reduces the MT radii to about 1.3 for Lu and about 0.84 for O.

So far I've tried adding spingpol, spinorb, autolinengy=true or nxoapwlo=1 - same problems, although nxoapwlo=1 resulted in completely different DOS.

I will be very much grateful for any help.

Please consider the simplest input of a primitive cell of cubic Lu2O3:

tasks
0

sppath
'/home/shyichuk/elk-4.0.15/species/'

autokpt
.false.

ngridk
3 3 3

avec
5.198000000 5.198000000 5.198000000
5.198000000 5.198000000 -5.198000000
5.198000000 -5.198000000 5.198000000

atoms
2 : nspecies
'Lu.in' : spfname
16 : natoms; atposl, bfcmt below
0.50000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000
0.50000000 0.50000000 0.50000000 0.00000000 0.00000000 0.00000000
0.00000000 0.00000000 0.50000000 0.00000000 0.00000000 0.00000000
0.00000000 0.50000000 0.00000000 0.00000000 0.00000000 0.00000000
0.25000000 0.72000000 0.97000000 0.00000000 0.00000000 0.00000000
0.75000000 0.28000000 0.03000000 0.00000000 0.00000000 0.00000000
0.75000000 0.78000000 0.53000000 0.00000000 0.00000000 0.00000000
0.25000000 0.22000000 0.47000000 0.00000000 0.00000000 0.00000000
0.97000000 0.25000000 0.28000000 0.00000000 0.00000000 0.00000000
0.03000000 0.75000000 0.72000000 0.00000000 0.00000000 0.00000000
0.53000000 0.75000000 0.22000000 0.00000000 0.00000000 0.00000000
0.47000000 0.25000000 0.78000000 0.00000000 0.00000000 0.00000000
0.22000000 0.03000000 0.75000000 0.00000000 0.00000000 0.00000000
0.78000000 0.97000000 0.25000000 0.00000000 0.00000000 0.00000000
0.28000000 0.47000000 0.25000000 0.00000000 0.00000000 0.00000000
0.72000000 0.53000000 0.75000000 0.00000000 0.00000000 0.00000000
'O.in' : spfname
24 : natoms; atposl, bfcmt below
0.52500000 0.00500000 0.24000000 0.00000000 0.00000000 0.00000000
0.47500000 0.99500000 0.76000000 0.00000000 0.00000000 0.00000000
0.73500000 0.23500000 0.26000000 0.00000000 0.00000000 0.00000000
0.26500000 0.76500000 0.74000000 0.00000000 0.00000000 0.00000000
0.97500000 0.26500000 0.53000000 0.00000000 0.00000000 0.00000000
0.02500000 0.73500000 0.47000000 0.00000000 0.00000000 0.00000000
0.76500000 0.49500000 0.97000000 0.00000000 0.00000000 0.00000000
0.23500000 0.50500000 0.03000000 0.00000000 0.00000000 0.00000000
0.53000000 0.23500000 0.99500000 0.00000000 0.00000000 0.00000000
0.47000000 0.76500000 0.00500000 0.00000000 0.00000000 0.00000000
0.97000000 0.02500000 0.76500000 0.00000000 0.00000000 0.00000000
0.03000000 0.97500000 0.23500000 0.00000000 0.00000000 0.00000000
0.24000000 0.26500000 0.73500000 0.00000000 0.00000000 0.00000000
0.76000000 0.73500000 0.26500000 0.00000000 0.00000000 0.00000000
0.26000000 0.47500000 0.50500000 0.00000000 0.00000000 0.00000000
0.74000000 0.52500000 0.49500000 0.00000000 0.00000000 0.00000000
0.76500000 0.76000000 0.76500000 0.00000000 0.00000000 0.00000000
0.23500000 0.24000000 0.23500000 0.00000000 0.00000000 0.00000000
0.99500000 0.74000000 0.97500000 0.00000000 0.00000000 0.00000000
0.00500000 0.26000000 0.02500000 0.00000000 0.00000000 0.00000000
0.50500000 0.47000000 0.73500000 0.00000000 0.00000000 0.00000000
0.49500000 0.53000000 0.26500000 0.00000000 0.00000000 0.00000000
0.73500000 0.03000000 0.52500000 0.00000000 0.00000000 0.00000000
0.26500000 0.97000000 0.47500000 0.00000000 0.00000000 0.00000000

[Lars Nordström]

Are you sure you are using a.u. and not Ångström as length units?

Best wishes,
Lars

[Andrew Shyichuk]

I've checked the geometry, yes, I've made a mistake. Cell vectros are in Angstroms.