Relaxation in case of Spin Orbit Coupling

jh502125 · Post by **jh502125** » Mon Nov 14, 2011 11:59 pm

Dear all:

I am trying to relax MnBi hexagonal structure ( spgroup 194) using Broyden-Fletcher-Goldfarb-Shanno minimization (BFGS).
I used HGH pseudopotential and initialized the spin regarding spin-orbit coupling.
Ultimately what I want to do is observing structural relaxation of MnBi in finite temperature(ionmov=13).

However, my convergence test seems not to be convergend at all. Total energy is oscillating and also for the ucvol(unit cell volume.)
I tried the same test with Trollier-Martin pseudopotential, and it converged well.
Is there any problem in my input file? Or is it for the pseudo-potential?
**I am attaching my input file.

Code: Select all

# MnBin Relaxation
#spin orbit coupling
 nsppol   1
 nspinor 2 
 nspden 4
 kptopt 4
#optimization of the lattice parameters
optcell 2
ionmov  3
ntime  30
dilatmx 1.5
ecutsm  0.5

nshiftk 1
shiftk 0 0 0.5
ngkpt 4 4 4

#Structure
 iscf   5
 toldfe   1.0d-6
  acell   8.050232711 8.050232711 11.432842231
spgroup  194
angdeg   90 90 120
brvltt   -1
  ecut   27
  natom  4
  nband  48
occopt 3
tsmear 0.001
getwfk -1
 nstep   50
ntypat   2
  typat   1 1 2 2 
xred 0.0  0.0  0.0
     0.0 0.0 0.5
     1/3 2/3 0.25
     2/3 1/3 0.75
 
 znucl    25 83


#conserve old < 6.7.2 behavior for calculating forces at each SCF step
 optforces 1

And, This is the structure of unitcell:
[img]http://cst-www.nrl.navy.mil/lattice/struk.picts/b8_1.s.png[/img]

Also, can I get the relaxed structure from trollier-martin psp, and use it for the spin-orbit coupling calculation which is the calculation only available for HGH pseudo-potential? In other words, Is the structure relaxing in different way for the different kinds of pseudo-potential?

FYI, I am learning alot from this forum. Thanks so much.

Thanks!
Soo

ekt · Post by **ekt** » Fri Sep 14, 2012 2:14 am

HI. How did your story with MnBi finish?
Did you calculate magnetic anisotropy?

maryamsajedi · Post by **maryamsajedi** » Thu Sep 20, 2012 4:23 pm

Dear all,
I came across the error below when I was trying to caculate the relaxation:

At line 209 of file ioarr.F90 (unit = 9, file = '')
Fortran runtime error: File 'reli_DS11_DEN' does not exist

The input file was under these conditions:
ecut 60
ecutsm 0.5
dilatmx 1.05
kptopt 1
ngkpt 6 6 6
nshiftk 4

optcell 0
ionmov 2
tolmxf 1.0d-5
ntime 100

What shall I do to remove the error?

maxim · Post by **maxim** » Thu Sep 20, 2012 10:56 pm

Dear maryamsajedi,

Pls give me your input and *files files so I can look at it and try to solve this problem.

Best wishes,
Maxim

maryamsajedi · Post by **maryamsajedi** » Sat Sep 22, 2012 9:40 am

Dear Maxim,
I`m trying to do a structural relaxation but the code stops with the following error message about k-point:

berryphase_new : BUG -
For k-point # 3,
the determinant of the overlap matrix is found to be 0.
Action : contact ABINIT group.

And in the log file this warning was shown:

WARNING - These should obey zval-nelect_occ=charge to better than 0.10E-10
Action : check input file for occ,wtk, and charge.
Note that wtk is NOT automatically normalized when occopt=2,
but IS automatically normalized otherwise.

I have attached the input file so that you can figure out what am I doing wrong in this calculation.
Thanks for your time,
Maryam

si_relaxedion.in: (3.11 KiB) Downloaded 249 times

maxim · Post by **maxim** » Sat Sep 22, 2012 12:15 pm

Dear maryamsajedi,

You should try a denser k-point grid in the z direction as your c is pretty small (only 2.604612042 bohr - it is only 1.378301435 angstroms - check that you have not confused bohr and angstrom units). Please describe the structure (supercell, etc.). If you do not watched the forum for similar problems, then again, that there is the only one correct density of k-point grid, so try different grids with different k-point grid density.

Best wishes,
Maxim

maryamsajedi · Post by **maryamsajedi** » Sun Sep 23, 2012 7:28 am

Dear Maxim,
Thank you so much for your help I will get back to the structral relaxation after correction but I have another problem with SCF density computation a non-SCF band structure calculation.The code stops following this error message below

rwwf.F90:579:ERROR
Reading option of rwwf. Trying to read
a RF wf record of a wf file, unit= 12
gave iostat=5016. Your file is likely not correct.
Action: check your input wf file.
t2x_WF2

I would really appreciate if you could help me since Iam a beginner in Abinit.I am using Abinit-6.12.3.
Best regards,
Maryam

t5.in: (2.55 KiB) Downloaded 247 times

ABINIT Discussion Forums

Relaxation in case of Spin Orbit Coupling

Relaxation in case of Spin Orbit Coupling

Re: Relaxation in case of Spin Orbit Coupling

Error in Relaxation

Re: Relaxation in case of Spin Orbit Coupling

Relaxation

Re: Relaxation in case of Spin Orbit Coupling

Re: Relaxation in case of Spin Orbit Coupling