Muffin-tin radius too small. What should I do?

[Min Choi]

Hello, I am new to Elk and am trying to calculate SHG from wurtzite CdS along with BSE dielectric function.

I have created geometry with spacegroup utility with the following spacegroup.in file:
'P63mc' : hrmg
7.81402 7.81402 12.75376 : a, b, c
90.0 90.0 120.0 : bc, ac, ab
1 1 1 : ncell
.true. : primcell
2 : nspecies
'Cd' : spsymb, spfname
2 : nwpos
0.0000 0.0000 0.0000 : wpos
0.3333 0.6666 0.3333
'S'
2
0.0000 0.0000 0.3750
0.3333 0.6666 0.8750

Then I basically just replaced the information in GEOMETRY.OUT to an input file in GaAs-NLO example:

tasks
0 ! ground-state calculation
120
125
188

nempty
8

! matrix elements are expanded to this angular momentum cut-off
lmaxvr
4

! number of valence states to use in BSE Hamiltonian
nvbse
3

! number of conduction states
ncbse
4

! use smallest possible maximum |G|
gmaxvr
0.0

! a scissor correction is required to match the gap to experiment
scissor
0.0331

swidth
0.005

! components of the optical tensor
optcomp
1 2 3

wplot
800 100 0 : nwplot, ngrkf, nswplot
0.0 1.0 : wplot

mixtype
2

avec
3.907010000 6.767139826 0.000000000
7.814020000 0.000000000 0.000000000
0.000000000 0.000000000 12.75376000

sppath
'../species/'

atoms
2 : nspecies
'Cd.in' : spfname
8 : natoms; atposl, bfcmt below
0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000
0.00000000 0.00000000 0.50000000 0.00000000 0.00000000 0.00000000
0.66660000 0.66670000 0.33330000 0.00000000 0.00000000 0.00000000
0.33330000 0.33340000 0.83330000 0.00000000 0.00000000 0.00000000
0.66670000 0.66670000 0.33330000 0.00000000 0.00000000 0.00000000
0.33340000 0.33330000 0.83330000 0.00000000 0.00000000 0.00000000
0.66670000 0.66660000 0.33330000 0.00000000 0.00000000 0.00000000
0.33330000 0.33330000 0.83330000 0.00000000 0.00000000 0.00000000
'S.in' : spfname
8 : natoms; atposl, bfcmt below
0.00000000 0.00000000 0.37500000 0.00000000 0.00000000 0.00000000
0.00000000 0.00000000 0.87500000 0.00000000 0.00000000 0.00000000
0.66660000 0.66670000 0.87500000 0.00000000 0.00000000 0.00000000
0.33330000 0.33340000 0.37500000 0.00000000 0.00000000 0.00000000
0.66670000 0.66670000 0.87500000 0.00000000 0.00000000 0.00000000
0.33340000 0.33330000 0.37500000 0.00000000 0.00000000 0.00000000
0.66670000 0.66660000 0.87500000 0.00000000 0.00000000 0.00000000
0.33330000 0.33330000 0.37500000 0.00000000 0.00000000 0.00000000

ngridk
8 8 8

vkloff
0.25 0.5 0.625

I instantly get the following error when I run ./elk

Error(checkmt): muffin-tin radius too small for species 1 (Cd)
Radius : -0.2460929900E-01

Could someone guide me through what I am doing incorrectly?
Thank you.

[Lars Nordström]

Hi!

You should use more digits in atomic positions in spacegroup.in,
instead of 0.3333 and 0.6666 you could use 0.33333333 and 0.66666666.
Then it will not produce two almost identical positions like
0.66660000 0.66670000 0.33330000 0.00000000 0.00000000 0.00000000
0.66670000 0.66670000 0.33330000 0.00000000 0.00000000 0.00000000

Best regads,
Lars

[Min Choi]

Thanks Lars.

It is surprisingly simple, but it works!
I would have never noticed this.
Thank you so much!

Min.

[Anonymous]

Hi, i have the same error messages

Error(checkmt): muffin-tin radius too small for species 1 (Fe)
Radius : -0.2460929900E-01

Have you an idea? Thanks!

tasks
0

plot3d
0 0 0
0 0 1
0 1 0
1 0 0
60 60 60

xctype
20

autolinengy
true

mixtype
3

maxscl
100

nempty
10

sppath
'/home/musr/Downloads/elk-4.3.6/species/'

avec
12.89605801 0.000000000 0.000000000
-6.448029005 11.16831385 0.000000000
0.000000000 0.000000000 17.27285270

atoms
4 : nspecies
'Fe.in' : spfname
4 : natoms; atposl, bfcmt below
0.00000000 0.00000000 0.00000000 0.00000000 0.00000000 0.00000000
0.50000000 0.00000000 0.50000000 0.00000000 0.00000000 0.00000000
0.00000000 0.50000000 0.50000000 0.00000000 0.00000000 0.00000000
0.50000000 0.50000000 0.50000000 0.00000000 0.00000000 0.00000000
'Sn.in' : spfname
9 : natoms; atposl, bfcmt below
0.50000000 0.00000000 0.50000000 0.00000000 0.00000000 0.00000000
0.00000000 0.50000000 0.50000000 0.00000000 0.00000000 0.00000000
0.50000000 0.50000000 0.50000000 0.00000000 0.00000000 0.00000000
0.16517000 0.83483000 0.83924000 0.00000000 0.00000000 0.00000000
0.83483000 0.16517000 0.16076000 0.00000000 0.00000000 0.00000000
0.16517000 0.33034000 0.83924000 0.00000000 0.00000000 0.00000000
0.83483000 0.66966000 0.16076000 0.00000000 0.00000000 0.00000000
0.66966000 0.83483000 0.83924000 0.00000000 0.00000000 0.00000000
0.33034000 0.16517000 0.16076000 0.00000000 0.00000000 0.00000000
'Si.in' : spfname
2 : natoms; atposl, bfcmt below
0.33333333 0.66666667 0.23520000 0.00000000 0.00000000 0.00000000
0.66666667 0.33333333 0.76480000 0.00000000 0.00000000 0.00000000
'O.in' : spfname
16 : natoms; atposl, bfcmt below
0.33333333 0.66666667 0.41540000 0.00000000 0.00000000 0.00000000
0.66666667 0.33333333 0.58460000 0.00000000 0.00000000 0.00000000
0.33333333 0.66666667 0.87290000 0.00000000 0.00000000 0.00000000
0.66666667 0.33333333 0.12710000 0.00000000 0.00000000 0.00000000
0.15330000 0.84670000 0.61510000 0.00000000 0.00000000 0.00000000
0.84670000 0.15330000 0.38490000 0.00000000 0.00000000 0.00000000
0.15330000 0.30660000 0.61510000 0.00000000 0.00000000 0.00000000
0.84670000 0.69340000 0.38490000 0.00000000 0.00000000 0.00000000
0.69340000 0.84670000 0.61510000 0.00000000 0.00000000 0.00000000
0.30660000 0.15330000 0.38490000 0.00000000 0.00000000 0.00000000
0.46360000 0.53640000 0.18250000 0.00000000 0.00000000 0.00000000
0.53640000 0.46360000 0.81750000 0.00000000 0.00000000 0.00000000
0.46360000 0.92720000 0.18250000 0.00000000 0.00000000 0.00000000
0.53640000 0.07280000 0.81750000 0.00000000 0.00000000 0.00000000
0.07280000 0.53640000 0.18250000 0.00000000 0.00000000 0.00000000
0.92720000 0.46360000 0.81750000 0.00000000 0.00000000 0.00000000
ngridk
1 1 1

[mfechner]

Dear Anonymous,

you may check your structure: your last three iron atoms and the first three Sn atoms occupy the same positions. This is not really a healthy situation. You may use the feature to put out a xrysden file and have a look at the structure to check.

best
Michael

[Anonymous]

Hi,

Thank you very much for your answer and in general very nice work.

You have solved my problem.

I have another question which is not really answered by the manual:

One gets two files plotting the 3D potential (43) using GGA : VCL3D and VXC3D. It is written in the manual (Page 35) one gets the exchange-correlation potential and Coulomb potential plots.

Is VCL3D the Coulomb potential file?

Because in some detailed potplot.f90:

"Outputs the exchange, correlation and Coulomb potentials, read in from STATE.OUT, for
1D, 2D or 3D plotting."
let appear the question that 3 files should be gerarted?

Thanks for your answer.