I have a question concerning surface calculations with elk. What I am trying to do is to calculate a slab of hcp Co see input below. This calculation with 5 layers of Co works fine. However if I add another two layers, increasing the c lattice constant accordingly to keep the vacuum constant, the calculation does not converge anymore. To be more precise elk fails to find the linearisation energies and the potential convergence becomes highly oscillation. I tried the same for Fe and found the same problem. I also checked the typical parameters for these cases namely gmaxvr,rgkmax lmaxvr lmaxapw ... but no improvement could be made. So is there any "trick" or does I do something very wrong ?
tasks
0
xctype
21
autolinengy
.true.
nempty
10
reducebf
0.9
bfieldc
0 0 -1
mixtype
1
ngridk
9 9 1
scale
1.8897261329
avec
2.5053999424 0 0
-1.2526999712 2.1697399968 0
0 0 16.3572006224
atoms
1 !NUMBER OF SPECIES
'Co.in' !KIND
5 !NUMBER OF ATOMS
0.333333343 0.666666687 0.0625
0.666666627 0.333333313 0.1875
0.333333343 0.666666687 0.3125
0.666666627 0.333333313 0.4375
0.333333343 0.666666687 0.5625
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Dear users & developers,
I have a question concerning surface calculations with elk. What I am trying to do is to calculate a slab of hcp Co see input below. This calculation with 5 layers of Co works fine. However if I add another two layers, increasing the c lattice constant accordingly to keep the vacuum constant, the calculation does not converge anymore. To be more precise elk fails to find the linearisation energies and the potential convergence becomes highly oscillation. I tried the same for Fe and found the same problem. I also checked the typical parameters for these cases namely gmaxvr,rgkmax lmaxvr lmaxapw ... but no improvement could be made. So is there any "trick" or does I do something very wrong ?
tasks
0
xctype
21
autolinengy
.true.
nempty
10
reducebf
0.9
bfieldc
0 0 -1
mixtype
1
ngridk
9 9 1
scale
1.8897261329
avec
2.5053999424 0 0
-1.2526999712 2.1697399968 0
0 0 16.3572006224
atoms
1 !NUMBER OF SPECIES
'Co.in' !KIND
5 !NUMBER OF ATOMS
0.333333343 0.666666687 0.0625
0.666666627 0.333333313 0.1875
0.333333343 0.666666687 0.3125
0.666666627 0.333333313 0.4375
0.333333343 0.666666687 0.5625
These mixing parameters work fine for a Fe/Pt[001] slabs.
!------------------------------
! linear mixing
mixtype
3
betamax
0.2
beta0
0.01
!-------------------------------
Good luck