Roman N. - 2014-01-06

Dear Kay, Lars and all others,

thank you very much for very prompt replies – this is exactly that one can call “very well maintained forum”!

You are completely right that there are more semi-core local-orbitals in the default Elk species file compared to Wien2k (4p is brought to valence). Therefore, I have already calculated Elk eos with 4p in the core (equivalent to setting “ecvcut -3.0”) – there is small difference (~0.1% in lattice constant) compared to 4p in valence, but the disagreement of lattice constant with Wien2k is still more than 1%. In both Elk and Wien2k calculations core states were computed fully relativistically (i.e. with spin-orbit coupling) while valence states were computed scalar relativistically.
To compare apple-to-apple we also run Elk & Wien2k calculations with matching parameters: PBE, kp-mesh – 16x16x16 (gamma-centered), rgkmax=9.0, gmaxvr=14.0, lmaxapw=10, Gauss smearing with swidth=0.001 Ha. Furthermore, all energy values in species file were fixed and matching (see them below). But even this set of calculations with matching input parameters gave difference in lattice constant of 0.5%.

Could you please look on the input files given below and tell me that is done wrong in such calculations.

Thank you very much in advance!

Roman

Wien2k output file at alat=11.6au (results at final iteration) (all energy values are in Ry):
…
ATOMIC SPHERE DEPENDENT PARAMETERS FOR ATOM Sn1
:e__0001: OVERALL ENERGY PARAMETER IS 0.1015
OVERALL BASIS SET ON ATOM IS LAPW
:E0_0001: E( 0)= -0.2220
APW+lo
:E1_0001: E( 1)= 0.1170
APW+lo
:E1_0001: E( 1)= -5.8320 E(BOTTOM)= -5.843 E(TOP)= -5.821
LOCAL ORBITAL
:E2_0001: E( 2)= 0.0440
APW+lo
:E2_0001: E( 2)= -1.2700
LOCAL ORBITAL
…
:FER : F E R M I - ENERGY(GAUSS-.M.)= 0.47495
…

Elk Fermi energy output file (all energy values are in Ha) – EFERMI.OUT:
0.2546340335

Elk species file (energy values do not deviate more than by 0.0002 Ha from Wien2k ones, see them above; I fixed all energy values in Elk, so that the linearization energies do not move from Wien2k final ones) – Sn.in:
'Sn' : spsymb
'tin' : spname
-50.0000 : spzn
216395.0920 : spmass
0.282843E-06 2.5000 45.4746 600 : sprmin, rmt, sprmax, nrmt
17 : spnst
1 0 1 2.00000 T : spn, spl, spk, spocc, spcore
2 0 1 2.00000 T
2 1 1 2.00000 T
2 1 2 4.00000 T
3 0 1 2.00000 T
3 1 1 2.00000 T
3 1 2 4.00000 T
3 2 2 4.00000 T
3 2 3 6.00000 T
4 0 1 2.00000 T
4 1 1 2.00000 F
4 1 2 4.00000 F
4 2 2 4.00000 F
4 2 3 6.00000 F
5 0 1 2.00000 F
5 1 1 1.00000 F
5 1 2 1.00000 F
1 : apword
0.0680 0 F : apwe0, apwdm, apwve
0 : nlx
5 : nlorb
0 2 : lorbl, lorbord
-0.0937 0 F : lorbe0, lorbdm, lorbve
-0.0937 1 F
1 2 : lorbl, lorbord
0.0758 0 F : lorbe0, lorbdm, lorbve
0.0758 1 F
2 2 : lorbl, lorbord
0.0393 0 F : lorbe0, lorbdm, lorbve
0.0393 1 F
1 3 : lorbl, lorbord
0.0758 0 F : lorbe0, lorbdm, lorbve
0.0758 1 F
-2.8987 0 F
2 3 : lorbl, lorbord
0.0393 0 F : lorbe0, lorbdm, lorbve
0.0393 1 F
-0.6177 0 F

Elk input file (most of parameters match Wien2k ones) – elk.in:
tasks
0

avec
1.0 1.0 0.0
1.0 0.0 1.0
0.0 1.0 1.0

scale
5.8

sppath
'./'

atoms
1 : n species
'Sn.in' : spfname
2 : natoms
0.0 0.0 0.0 0.0 0.0 0.0 : atposl, bfcmt
0.25 0.25 0.25 0.0 0.0 0.0 : atposl, bfcmt

ngridk
16 16 16

beta0
0.05

rgkmax
9.0

gmaxvr
14.0

lmaxapw
10

lmaxvr
7

lradstp
4

nempty
4.0

stype
0

swidth
0.001

xctype
20