Page 1 of 1

Optimization of tetragonal perovskit???

Posted: Mon Feb 24, 2020 11:47 am
by new_986
Dear all
Greetings
I am trying to do optimization for tetragonal phase of a perovskite specifically CsPbCl3
I give spgroup input variable which 140 due to papers but my run stops and gives me the following error:

--- !ERROR
src_file: symatm.F90
src_line: 201
mpi_rank: 13
message: |
Largest error (above) is so large (0.01) that either input atomic coordinates (xred)
are wrong or space group symmetry data is wrong.
Action : correct your input file.
By the way atomic coordinates are correct and I try in reduced and in angstrom coordinates...

I have tries to give nsym and symrel, the structure turns to cubic
here is my input

ndtset 2

#Definition of the atomic positions without relaxing the lattice parameters and fixed Pb
optcell1 0
ionmov1 2

#Definition of the basis of the geometry determined by dataset 1 (so using getxcart 1)
getxcart2 1
optcell2 3
ionmov2 2
dilatmx2 1.5
ecutsm2 0.5

# COMMON INPUT DATA

#Definition of the unit cell
acell 5.64 5.64 5.77 Angstr
#rprim 1.0 0.0 0.0
0.0 1.0 0.0
0.0 0.0 1.0

#Definition of the atom types
ntypat 3
znucl 55 82 17

#Definition of the atoms
natom 5
typat 1 2 3 3 3
xred 0.0 0.0 0.0
0.5 0.5 0.5
0.5 0.5 0.0
0.5 0.0 0.5
0.0 0.5 0.5

#xangst
0.0 0.0 0.0
2.82 2.82 2.885
2.82 2.82 0.0
0.0 2.82 2.885
2.82 0.0 2.885

#Definition of the planewave basis set
ecut 45.0

# find the optimal total energy
ntime 200
tolmxf 1.0d-5
toldff 1.0d-6
nband 40
spgroup 140
brvltt -1
chksymbreak 0
chkprim 0
#Definition of the k-point grid
kptopt 1
ngkpt 4 4 4
nshiftk 1
shiftk
0.5 0.5 0.5
getwfk -1
prtcif 1
prtgeo 1
#Definition of the SCF procedure
nstep 100
#toldfe 1.0d-6
diemac 12.0


Regards

Nawzad

Re: Optimization of tetragonal perovskit???

Posted: Mon Feb 24, 2020 11:04 pm
by ebousquet
Dear Nawzad,
You indeed break the symmetries with the cell parameters but you kept the atomic positions to high-symmetry ones and it sounds your system goes back to cubic if all the atomic positions are in high-sym (and I would say as it should be for perovskites without epitaxial constraints). Although, if I'm right, space group 140 does not exist if you have only 5-atoms unit cell perovskites.
Best wishes,
Eric

Re: Optimization of tetragonal perovskit???

Posted: Tue Feb 25, 2020 9:32 am
by new_986
ebousquet wrote:
Mon Feb 24, 2020 11:04 pm
Dear Nawzad,
You indeed break the symmetries with the cell parameters but you kept the atomic positions to high-symmetry ones and it sounds your system goes back to cubic if all the atomic positions are in high-sym (and I would say as it should be for perovskites without epitaxial constraints). Although, if I'm right, space group 140 does not exist if you have only 5-atoms unit cell perovskites.
Best wishes,
Eric
Dear Eric
Thanks for replying me
Tetragonal phase is like cubic when it comes to the number of atoms but it has different due to the reference below

http://sciold.ui.ac.ir/~sjalali/papers/P2017.6.pdf

And for the space group I have tried some other possible space groups, it doesn't work

140 or p4/mbm is the most possible one

Please see the paper
https://iopscience.iop.org/article/10.1 ... 14/012/pdf

regards

Re: Optimization of tetragonal perovskit???

Posted: Sun Mar 08, 2020 10:29 pm
by ebousquet
Dear Nawzad,
The paper you sent confirm what I'm saying. Please have a look on papers listing all possible space groups of distorted perovskites like this one:
https://scripts.iucr.org/cgi-bin/paper?bk0118
The space group 140 contains octahedra rotations, which requires extending the unit cell in some directions.
Best wishes,
Eric

Re: Optimization of tetragonal perovskit???

Posted: Mon Mar 09, 2020 11:44 am
by new_986
Dear Eric
I have already tried extensions but still no relaxation happens...
Thanks a lot

Nawzad

Re: Optimization of tetragonal perovskit???

Posted: Mon Mar 09, 2020 9:44 pm
by ebousquet
Dear Nawzad,
You need to extend the cell and break the symmetry by moving the atoms accordingly, is it what you did?
Best wishes,
Eric

Re: Optimization of tetragonal perovskit???

Posted: Mon Mar 16, 2020 11:08 am
by new_986
ebousquet wrote:
Mon Mar 09, 2020 9:44 pm
Dear Nawzad,
You need to extend the cell and break the symmetry by moving the atoms accordingly, is it what you did?
Best wishes,
Eric
Dear Eric
I have already these following extensions for perovskite
0.0 0.0 0.0
0.5 0.5 0.51
0.5 0.5 0.0
0.5 0.0 0.5
0.0 0.5 0.5
and

0.0 0.0 0.0
0.5 0.5 0.51
0.5 0.5 0.0
0.5 0.0 0.51
0.0 0.5 0.51

but the relaxation did not happen

best rggards

Nawzad

Re: Optimization of tetragonal perovskit???

Posted: Mon Mar 30, 2020 9:44 am
by ebousquet
Dear Nawzad,
The positions you sent are not for doubled cell, right?
Best wishes,
Eric

Re: Optimization of tetragonal perovskit???

Posted: Sat Apr 11, 2020 6:43 pm
by new_986
ebousquet wrote:
Mon Mar 30, 2020 9:44 am
Dear Nawzad,
The positions you sent are not for doubled cell, right?
Best wishes,
Eric
Dear Eric
No, it is for one unit cell, a=b not equal to c

Regards

Nawzad

Re: Optimization of tetragonal perovskit???

Posted: Tue Apr 14, 2020 3:22 pm
by new_986
ebousquet wrote:
Mon Mar 30, 2020 9:44 am
Dear Nawzad,
The positions you sent are not for doubled cell, right?
Best wishes,
Eric
Dear Eric
I have tried to get fractional coordinates using software SPUDS, following are the resultant coordinates:

Cs 4b 0.0 0.5 0.25
Pb 4c 0.0 0.0 0.0
Cl 4a 0.0 0.0 0.25
Cl 8h 0.3428 0.8428 0.0

How to translate these to xred ??

Thansk

Nawzad

Re: Optimization of tetragonal perovskit???

Posted: Wed Apr 15, 2020 6:16 pm
by ebousquet
Dear Nawzad,
OK, I see, one of the solution is that you need to learn about the basics of crystallography, this is going to be important for you (and for anyone wanting to do research in crystals at the microscopic level). I don't have a specific link to send you but I guess you can find something online by searching on google. The best would be to follow a summer school on crystallography (from Bilbao Crystallography for example) but this is not the best time to go around right now :| ...

For example, here to reconstruct your crystal you need to know what are Wyckoff sites (the 4b, 4c, 8h, etc labels) and that's it. This is something you have to understand a bit before going ahead. Here I can tell you that the unit cell contains 20 atoms according to what you sent.

Now, there are softwares (online like Biblao or that you can download) that you can use to do so. The best interface with Abinit is qAgate, which, among others, can build an abinit input file from a cif file or other position format.

You can install it like that on linux:

Code: Select all

sudo add-apt-repository ppa:piti-diablotin/abiout
sudo apt-get update
sudo apt install -y qagate
Then you can run it by typing qAgate command and the grapical interface is intuitive enough for just inputing/outputing structure files. I let you play with it.

But it is better you understand a bit about Wyckoff positions before using the codes...

Best wishes,
Eric

Re: Optimization of tetragonal perovskit???

Posted: Mon Apr 27, 2020 4:02 pm
by new_986
Dear Eric
I have found the correct wyckoff positions and the output file tells me the space group is correct,, everything work well except the c/a ratio becomes 1 while it suppose to keep the structure tetragonal (c/a=constant),,
how to keep c/a ratio constant in abinit please?

Best regards

Nawzad

Re: Optimization of tetragonal perovskit???

Posted: Tue Apr 28, 2020 8:47 am
by ebousquet
new_986 wrote:
Mon Apr 27, 2020 4:02 pm
I have found the correct wyckoff positions and the output file tells me the space group is correct,, everything work well except the c/a ratio becomes 1 while it suppose to keep the structure tetragonal (c/a=constant),,
how to keep c/a ratio constant in abinit please?
Excellent, great!
For the c/a, there is not optcell option to fix this ratio constant. However, it is strange that it stays at 1 exactly, either it relaxed back to cubic or the optcell value is not the good one. During the relaxation, is this ratio always equal to 1? If so this means that the optcell used does not allow for cell relaxation.

Best wishes,
Eric

Re: Optimization of tetragonal perovskit???

Posted: Tue Apr 28, 2020 11:07 am
by new_986
ebousquet wrote:
Tue Apr 28, 2020 8:47 am
new_986 wrote:
Mon Apr 27, 2020 4:02 pm
I have found the correct wyckoff positions and the output file tells me the space group is correct,, everything work well except the c/a ratio becomes 1 while it suppose to keep the structure tetragonal (c/a=constant),,
how to keep c/a ratio constant in abinit please?
Excellent, great!
For the c/a, there is not optcell option to fix this ratio constant. However, it is strange that it stays at 1 exactly, either it relaxed back to cubic or the optcell value is not the good one. During the relaxation, is this ratio always equal to 1? If so this means that the optcell used does not allow for cell relaxation.

Best wishes,
Eric
Dear Eric
c/a ratio is not 1 at the beginning. it is more than one due to the literature and as i used them, it decreases until the unit cell turns to cubic.
after my previous post, I have fixed some atoms using iatfix and natfix. Now the unit cell has not turned to cubic but lattice parameters are not correct (a/=b <c)... I need to fix atoms what keep c/a more than one and a=b. any suggestion please?

Regards

Nawzad

Re: Optimization of tetragonal perovskit???

Posted: Tue May 05, 2020 9:24 am
by ebousquet
Dear Nawzad,
Without seeing your input it'll be difficult to say if there is something wrong or if it DFT-exchange correlation that does not work for your system (i.e. DFT predict cubic while experimentally it is not, but I would be surprised it does not work for such systems).
Best wishes,
Eric

Re: Optimization of tetragonal perovskit???

Posted: Sat May 16, 2020 1:52 pm
by new_986
Dear Eric
I was following a specific reference for what is related to the Wyckoff position, Unfortunately the reference was not quit correct. I feel sorry for bothering you...My input file is working correctly now
Thanks a lot

Nawzad

Re: Optimization of tetragonal perovskit???

Posted: Mon May 18, 2020 11:14 am
by ebousquet
OK, very good then :-)
Eric