subject:"Re\: \[QE\-users\] Crystal structure changes while doing vc\-relax calculation."

Re: [QE-users] Crystal structure changes while doing vc-relax calculation.

2022-02-09 Thread Husak Michal

If you use space group for QE setting, you can enter ONLY the symmetry
independent atoms ... Othervise QE will generate the atoms based on symetry 
multiple times = overlap.

If your supercell keep the original symetry it will work ...
If your supercell does not have the original symmetry and it realy have
P1 group, QE nor other code can keep the lattice parameters symmetric,
becouse they must differ  There is no forces/sg symmetry forcing the 
lattice parameters to keep original symmetry ... P1 is triclinic ...

I am primary a crystallographer (developing just crystallographic GUI for QE), 
I do not 100% understand QE internal processing - I belive QE try to find 
symmetry itself and keep it. If none found, none restraines used 

Can you please:
1) Send me the starting structure in CIF format ?
2) Send me the supercell you need to process in CIF format ?

In such case I can try our CIF -> QE conversion engine and see whatever I can 
generete what you need 

Michal Husak
hus...@vscht.cz

From: users  on behalf of Vivek 
Christhunathan 
Sent: Wednesday, February 9, 2022 12:44:22 PM
To: Quantum ESPRESSO users Forum
Subject: Re: [QE-users] Crystal structure changes while doing vc-relax 
calculation.

Hi Michal,

Thank you very much for your suggestion.

>crystal_sg  and correct space gropu nuber (space_group)
>Correct ibrav will be generated automatically , no need to eneter ...


The space group number of CaWO4 unit-cell (ICSD no: 18135) is 88 since it has 
I41/a space-group, but the Space group number of CaWO4 supercell (2x1x1) is 1 
as it has P1 space-group name (Note: I used VESTA tool make supercell). I tried 
with both the sapace-group numbers. I received an error called "Error in 
routine check_atoms (1): atoms # 1 and # 62 overlap!" when I used 
space_group=88. I think this is because of the space-group changes when we make 
supercell from the CaWO4 unit-cell (using VESTA tool).

In v6.4.1, I used space_group=1 for the calculation but again symmetry of the 
supercell changes (a≠b). I have included my input for your reference.

&control
calculation = 'vc-relax'
restart_mode = 'from_scratch'
prefix = 'CaWO4'
pseudo_dir = './'
outdir = './outCaWO4'
max_seconds = 3500,
/

&system
celldm(1) = 19.820415
celldm(2) = 0.5
celldm(3) = 1.084549393
space_group = 1
nat = 48
ntyp = 3
ecutwfc = 65.0
ecutrho = 780.0
occupations= 'smearing'
smearing= 'gaussian'
degauss= 0.02
/

&electrons
mixing_beta = 0.2,
conv_thr = 1.0d-6,
electron_maxstep = 5,
/

&IONS
/

&cell
cell_factor = 3.0d0
/

ATOMIC_SPECIES
 Ca 40.08   Ca.pbe-spn-rrkjus_psl.1.0.0.UPF
 W  183.84  W.pbe-spn-rrkjus_psl.1.0.0.UPF
 O  16.00   O.pbe-n-rrkjus_psl.1.0.0.UPF

ATOMIC_POSITIONS (crystal_sg)
Ca 0.0 0.25000 0.62500
Ca 0.5 0.25000 0.62500
Ca 0.0 0.75000 0.37500
Ca 0.5 0.75000 0.37500
Ca 0.25000 0.75000 0.12500
Ca 0.75000 0.75000 0.12500
Ca 0.25000 0.25000 0.87500
Ca 0.75000 0.25000 0.87500
 W 0.0 0.25000 0.12500
 W 0.5 0.25000 0.12500
 W 0.0 0.75000 0.87500
 W 0.5 0.75000 0.87500
 W 0.25000 0.75000 0.62500
 W 0.75000 0.75000 0.62500
 W 0.25000 0.25000 0.37500
 W 0.75000 0.25000 0.37500
 O 0.07485 0.00930 0.20973
 O 0.574850023 0.00930 0.20973
 O 0.425150007 0.99076 0.790300012
 O 0.925150037 0.99076 0.790300012
 O 0.175150007 0.99076 0.709699988
 O 0.675150037 0.99076 0.709699988
 O 0.324849993 0.00930 0.290300012
 O 0.824849963 0.00930 0.290300012
 O 0.370350003 0.399699986 0.459699988
 O 0.870350003 0.399699986 0.459699988
 O 0.129649997 0.600300014 0.540300012
 O 0.629649997 0.600300014 0.540300012
 O 0.379649997 0.600300014 0.959699988
 O 0.879649997 0.600300014 0.959699988
 O 0.120350003 0.399699986 0.04027
 O 0.620350003 0.399699986 0.04027
 O 0.324849993 0.50924 0.709699988
 O 0.824849963 0.50924 0.709699988
 O 0.175150007 0.49076 0.290300012
 O 0.675150037 0.49076 0.290300012
 O 0.42

Re: [QE-users] Crystal structure changes while doing vc-relax calculation.

2022-02-09 Thread Vivek Christhunathan

Hi Michal,

Thank you very much for your suggestion.


*>crystal_sg  and correct space gropu nuber (space_group)>Correct ibrav
will be generated automatically , no need to eneter ...*


The space group number of CaWO4 unit-cell (ICSD no: 18135) is 88 since it
has I41/a space-group, but the Space group number of CaWO4 supercell (2x1x1)
is 1 as it has P1 space-group name (Note: I used VESTA tool make supercell).
I tried with both the sapace-group numbers. I received an error called "Error
in routine check_atoms (1): atoms # 1 and # 62 overlap!" when I used
space_group=88. I think this is because of the space-group changes when we
make supercell from the CaWO4 unit-cell (using VESTA tool).

In v6.4.1, I used space_group=1 for the calculation but again symmetry of
the supercell changes (a≠b). I have included my input for your reference.





























































































*&controlcalculation = 'vc-relax'restart_mode = 'from_scratch'prefix =
'CaWO4'pseudo_dir = './'outdir = './outCaWO4'max_seconds =
3500,/&systemcelldm(1) = 19.820415celldm(2) = 0.5celldm(3) =
1.084549393space_group = 1nat = 48ntyp = 3ecutwfc = 65.0ecutrho =
780.0occupations= 'smearing'smearing= 'gaussian'degauss=
0.02/&electronsmixing_beta = 0.2,conv_thr = 1.0d-6,electron_maxstep =
5,/&IONS/&cellcell_factor = 3.0d0/ATOMIC_SPECIES Ca 40.08
Ca.pbe-spn-rrkjus_psl.1.0.0.UPF W  183.84  W.pbe-spn-rrkjus_psl.1.0.0.UPF O
 16.00   O.pbe-n-rrkjus_psl.1.0.0.UPFATOMIC_POSITIONS (crystal_sg)Ca
0.0 0.25000 0.62500Ca 0.5
  0.25000 0.62500Ca 0.0 0.75000
0.37500Ca 0.5 0.75000 0.37500Ca
0.25000 0.75000 0.12500Ca 0.75000
  0.75000 0.12500Ca 0.25000 0.25000
0.87500Ca 0.75000 0.25000
0.87500 W 0.0 0.25000 0.12500 W
0.5 0.25000 0.12500 W 0.0
  0.75000 0.87500 W 0.5 0.75000
0.87500 W 0.25000 0.75000 0.62500 W
0.75000 0.75000 0.62500 W 0.25000
  0.25000 0.37500 W 0.75000 0.25000
0.37500 O 0.07485 0.00930
0.20973 O 0.574850023 0.00930 0.20973 O
0.425150007 0.99076 0.790300012 O 0.925150037
  0.99076 0.790300012 O 0.175150007 0.99076
0.709699988 O 0.675150037 0.99076 0.709699988 O
0.324849993 0.00930 0.290300012 O 0.824849963
  0.00930 0.290300012 O 0.370350003 0.399699986
0.459699988 O 0.870350003 0.399699986
0.459699988 O 0.129649997 0.600300014 0.540300012 O
0.629649997 0.600300014 0.540300012 O 0.379649997
  0.600300014 0.959699988 O 0.879649997 0.600300014
0.959699988 O 0.120350003 0.399699986 0.04027 O
0.620350003 0.399699986 0.04027 O 0.324849993
  0.50924 0.709699988 O 0.824849963 0.50924
0.709699988 O 0.175150007 0.49076
0.290300012 O 0.675150037 0.49076 0.290300012 O
0.425150007 0.49076 0.20973 O 0.925150037
  0.49076 0.20973 O 0.07485 0.50924
0.790300012 O 0.574850023 0.50924 0.790300012 O
0.120350003 0.899699986 0.959699988 O 0.620350003
  0.899699986 0.959699988 O 0.379649997 0.10029
0.04027 O 0.879649997 0.10029
0.04027 O 0.129649997 0.10029 0.459699988 O
0.629649997 0.10029 0.459699988 O 0.370350003
  0.899699986 0.540300012 O 0.870350003 0.899699986
0.540300012K_POINTS AUTOMATIC2 2 2 1 1 1 *

Can anyone please explain why my system 'CaWO4 supercell' keep on changing
its symmetry and why I am not able to control it?

Thanks in advance.

Kind Regards,

Vivek C


__

Vivek Christhunathan

PhD researcher

Mechanical Engineering

College of Engineering and Informatics

Alice Perry Engineering Building

National University of Ireland Galway, Ireland
E-mail: v.christhunath...@nuigalway.ie
Phone: +353 899811181, +91 9600752742



On Mon, Feb 7, 2022 at 3:47 PM Vivek Christhunathan 
wrote:

> Hi Lorenzo,
>
> >$scan_ibrav.x
>
> I tried it on my machine, and it printed exactly the same as your prior
> message.
>
> I received the following cell parameter values when I run the calculation
> with these tags such as ibrav = 8, celldm( 1) = 19.

Re: [QE-users] Crystal structure changes while doing vc-relax calculation.

2022-02-07 Thread Husak Michal

Suggestion:

Try to describe your system by the help of
crystal_sg  and correct space gropu nuber (space_group)
Correct ibrav will be generated automatically , no need to eneter ...

And you will need to ente ronly the symmetry idedpendent atoms = less mistakes.

And you do not risk QE will fidn and made rstrains to other symetry than you 
have in mind ...

Michal Husak





___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

Re: [QE-users] Crystal structure changes while doing vc-relax calculation.

2022-02-07 Thread Vivek Christhunathan

Hi Lorenzo,

>$scan_ibrav.x

I tried it on my machine, and it printed exactly the same as your prior
message.

I received the following cell parameter values when I run the calculation
with these tags such as ibrav = 8, celldm( 1) = 19.820415000, celldm( 2) =
0.5, and celldm( 3) = 1.084549393.

*CELL_PARAMETERS (alat= 19.82041500)*

*   1.013661333   0.11180  -0.00334*

*   0.05590   0.506839078   0.00525*
*  -0.00363   0.01138   1.097952283*

And also I received the following cell parameter values when I run the
calculation with these tags such as ibrav = 8, celldm( 1) =
19.820415000, celldm(
2) = 0.5, celldm( 3) = 1.084549393, and *cell_dofree = 'ibrav' *(Note:
I included cell_dofree here because ibrav>0).

*ibrav =  8*

* celldm(1) = 20.00638452*

* celldm(2) =  0.50013749*

* celldm(3) =  1.08713582*

*Input lattice vectors:*

* 1.00938273 0. 0.*

* 0. 0.50483015 0.*

* 0. 0. 1.09733611*

*New lattice vectors in INITIAL alat:*

* 1.00938273 0. 0.*

* 0. 0.50483015 0.*

* 0. 0. 1.09733611*

*New lattice vectors in NEW alat (for information only):*

* 1. 0. 0.*

* 0. 0.50013749 0.*

* 0. 0. 1.08713582*

*Discrepancy in bohr = 0.000.000.00*


* bfgs converged in  11 scf cycles and  10 bfgs steps*

* (criteria: energy <  1.0E-04 Ry, force <  1.0E-03Ry/Bohr, cell <
5.0E-01kbar)*


* End of BFGS Geometry Optimization*


* Final enthalpy =   -2905.7085616081 Ry*

*Begin final coordinates*

* new unit-cell volume =   4353.90605 a.u.^3 (   645.18230 Ang^3 )*

* density =  5.92828 g/cm^3*


*CELL_PARAMETERS (alat= 19.82041500)*

*   1.009382726   0.0   0.0*

*   0.0   0.504830148   0.0*

*   0.0   0.0   1.097336114*

Still I don't get the Tetragonal crystal structure of CaWO4, it only gives
the lattice parameter values a=5.291688 ang and b=5.293143 ang (a≠b).
Note: Tetrahedra
CaWO4 Unit-cell was used to make this CaWO4 supercell. Is there any other
way to relax CaWO4's structure without affecting its symmetry?

And also I'm not sure why ibrav = 8 is required if CaWO4 has a tetragonal
crystal structure. Because ibrav = 8 for Orthorhombic is described in the
QE-input description. It would be great if you could explain this to me.

Thank you very much in advance.

Kind Regards,

Vivek C


__

Vivek Christhunathan

PhD researcher

Mechanical Engineering

College of Engineering and Informatics

Alice Perry Engineering Building

National University of Ireland Galway

Ireland
E-mail: v.christhunath...@nuigalway.ie
Phone: +353 899811181, +91 9600752742




On Thu, Feb 3, 2022 at 3:55 PM Vivek Christhunathan 
wrote:

> Hi Lorenzo,
>
> Thank you very much for your response.
>
> Sorry for the earlier message. I did not get notified on this.
>
> >$ ~/espresso.BASE/PW/tools/scan_ibrav.x
>
> Yes, I will get used to it on this.
>
> As per your suggestion, I will use the following values for my calculation
> and let you know the results.
>
> ibrav = 8
> celldm( 1) = 19.820415000
> celldm( 2) = 0.5
> celldm( 3) = 1.084549393
> at1 19.820415 0.00 0.00
> at2 0.00 9.910207 0.00
> at3 0.00 0.00 21.496219
>
>
> Thanks a million.
>
> Kind Regards,
>
> Vivek C
>
>
> __
>
> Vivek Christhunathan
>
> PhD researcher
>
> Mechanical Engineering
>
> College of Engineering and Informatics
>
> Room 2053
>
> Alice Perry Engineering Building
>
> National University of Ireland Galway
>
> Ireland
> E-mail: v.christhunath...@nuigalway.ie
> Phone: +353 899811181, +91 9600752742
>
>
>
> On Thu, Feb 3, 2022 at 2:05 PM Vivek Christhunathan 
> wrote:
>
>> Hello Everyone,
>>
>> I could not solve this issue even after trying with v6.8 (the older
>> version was v6.4.1). Still, the CaWO4 (2x1x1 supercell) changes its
>> symmetry from Tetragonal to Orthorhombic while doing structural
>> relaxation. Even I tried with multiple different input parameters.
>>
>> Can I get any suggestions regarding this, please?
>>
>> Kind Regards,
>>
>> Vivek C
>>
>>
>> __
>>
>> Vivek Christhunathan
>>
>> PhD researcher
>>
>> Mechanical Engineering
>>
>> College of Engineering and Informatics
>>
>> Room 2053
>>
>> Alice Perry Engineering Building
>>
>> National University of Ireland Galway
>>
>> Ireland
>> E-mail: v.christhunath...@nuigalway.ie
>> Phone: +353 899811181, +91 9600752742
>>
>>
>>
>> On Tue, Feb 1, 2022 at 12:52 PM Vivek Christhunathan 
>> wrote:
>>
>>> Hello Everyone,
>>>
>>> I'm bringing attention to a topic that has previously been discussed in
>>> this forum. I too faced the same type of problem when I try to do
>>> structural re

Re: [QE-users] Crystal structure changes while doing vc-relax calculation.

2022-02-03 Thread Vivek Christhunathan

Hi Lorenzo,

Thank you very much for your response.

Sorry for the earlier message. I did not get notified on this.

>$ ~/espresso.BASE/PW/tools/scan_ibrav.x

Yes, I will get used to it on this.

As per your suggestion, I will use the following values for my calculation
and let you know the results.

ibrav = 8
celldm( 1) = 19.820415000
celldm( 2) = 0.5
celldm( 3) = 1.084549393
at1 19.820415 0.00 0.00
at2 0.00 9.910207 0.00
at3 0.00 0.00 21.496219


Thanks a million.

Kind Regards,

Vivek C


__

Vivek Christhunathan

PhD researcher

Mechanical Engineering

College of Engineering and Informatics

Room 2053

Alice Perry Engineering Building

National University of Ireland Galway

Ireland
E-mail: v.christhunath...@nuigalway.ie
Phone: +353 899811181, +91 9600752742



On Thu, Feb 3, 2022 at 2:05 PM Vivek Christhunathan 
wrote:

> Hello Everyone,
>
> I could not solve this issue even after trying with v6.8 (the older
> version was v6.4.1). Still, the CaWO4 (2x1x1 supercell) changes its
> symmetry from Tetragonal to Orthorhombic while doing structural
> relaxation. Even I tried with multiple different input parameters.
>
> Can I get any suggestions regarding this, please?
>
> Kind Regards,
>
> Vivek C
>
>
> __
>
> Vivek Christhunathan
>
> PhD researcher
>
> Mechanical Engineering
>
> College of Engineering and Informatics
>
> Room 2053
>
> Alice Perry Engineering Building
>
> National University of Ireland Galway
>
> Ireland
> E-mail: v.christhunath...@nuigalway.ie
> Phone: +353 899811181, +91 9600752742
>
>
>
> On Tue, Feb 1, 2022 at 12:52 PM Vivek Christhunathan 
> wrote:
>
>> Hello Everyone,
>>
>> I'm bringing attention to a topic that has previously been discussed in
>> this forum. I too faced the same type of problem when I try to do
>> structural relaxation for CaWO4 (2x1x1 supercell), that its crystal
>> structure changed from Tetragonal to Orthorhombic. I tried all the
>> following forum's suggestions to get rid of this problem.
>>
>> As per their suggestion, I tried using the following tags in my input
>> file,
>>
>>
>>
>>
>> *>ion_dynamics = 'damp'>and>cell_dynamics = "damp-pr">which should
>> respect the constraint.*
>>
>> And also I tried using *cell_dofree=‘ibrav'*. but still lattice
>> parameter values a and b are not the same. Still, its structure changes
>> from Tetragonal to Orthorhombic.
>>
>>
>>
>>
>>
>> *>You can open Modules/cell_base.f90 and at line 84 change>LOGICAL   ::
>> enforce_ibrav = .FALSE.! True if ibrav ...>to>LOGICAL   :: enforce_ibrav =
>> .TRUE.! True if ibrav ...>(do not forget to recompile pw.x) to use
>> cell_dofree="ibrav" together with any >other cell_dofree*
>>
>> Since I'm utilizing the Government's supercomputer cluster, I'm not sure
>> how I'd make these adjustments to the QE code. Can I get your suggestion
>> regarding this?
>>
>>
>>
>> *>Should be present in the available patches for v.6.4.1:
>> filebackports-6.4.1.diff*
>>
>> 
>> in https://github.com/QEF/q-e/releases
>>
>>
>> Could you please advise me on how to proceed with this link?
>>
>> I have included calculations input and output for your perusal.
>>
>> If you could share your thoughts on this issue, that would be quite
>> helpful.
>>
>> *Input:*
>>
>> &control
>> calculation = 'vc-relax'
>> restart_mode = 'restart'
>> prefix = 'CaWO4'
>> tstress = .true.
>> tprnfor = .true.
>> pseudo_dir = './'
>> outdir = './outCaWO4'
>> wf_collect=.true.
>> forc_conv_thr = 1d-4
>> etot_conv_thr = 1D-4
>> !disk_io = 'medium'
>> max_seconds = 7200,
>> /
>>
>> &system
>> ibrav = 0
>> *celldm(1) = 19.820415*
>> nat = 48
>> ntyp = 3
>> ecutwfc = 65.0
>> ecutrho = 780.0
>> occupations= 'smearing'
>> smearing= 'gaussian'
>> degauss= 0.02
>> /
>>
>> &electrons
>> mixing_beta = 0.2,
>> conv_thr = 1.0d-6,
>> electron_maxstep = 5,
>> /
>>
>> &IONS
>> *ion_dynamics = 'damp'*
>>
>> /
>>
>> &cell
>> cell_factor = 3.0d0
>> *cell_dynamics = 'damp-pr'*
>> !cell_dofree= 'ibrav'
>> /
>>
>> ATOMIC_SPECIES
>>  Ca 40.08   Ca.pbe-spn-rrkjus_psl.1.0.0.UPF
>>  W  183.84  W.pbe-spn-rrkjus_psl.1.0.0.UPF
>>  O  16.00   O.pbe-n-rrkjus_psl.1.0.0.UPF
>>
>> *CELL_PARAMETERS (alat)*
>> *1.00 0.00 0.00*
>> *0.00 0.49 0.00*
>> *0.00 0.00 1.0845493931*
>>
>> ATOMIC_POSITIONS (crystal)
>> Ca 0.0 0.25000 0.62500
>> Ca 0.5 0.25000 0.62500
>> Ca 0.0 0.75000 0.37500
>> Ca 0.5 0.75000 0.37500
>> Ca 0.25000 0.75000 0.12500
>> Ca 0.75000 0.75000 0.12500
>> Ca 0.25000 0.25000 0.87500
>> Ca 0.75000 0.25000

Re: [QE-users] Crystal structure changes while doing vc-relax calculation.

2022-02-03 Thread Vivek Christhunathan

Hello Everyone,

I could not solve this issue even after trying with v6.8 (the older version
was v6.4.1). Still, the CaWO4 (2x1x1 supercell) changes its symmetry
from Tetragonal
to Orthorhombic while doing structural relaxation. Even I tried with
multiple different input parameters.

Can I get any suggestions regarding this, please?

Kind Regards,

Vivek C


__

Vivek Christhunathan

PhD researcher

Mechanical Engineering

College of Engineering and Informatics

Room 2053

Alice Perry Engineering Building

National University of Ireland Galway

Ireland
E-mail: v.christhunath...@nuigalway.ie
Phone: +353 899811181, +91 9600752742



On Tue, Feb 1, 2022 at 12:52 PM Vivek Christhunathan 
wrote:

> Hello Everyone,
>
> I'm bringing attention to a topic that has previously been discussed in
> this forum. I too faced the same type of problem when I try to do
> structural relaxation for CaWO4 (2x1x1 supercell), that its crystal
> structure changed from Tetragonal to Orthorhombic. I tried all the
> following forum's suggestions to get rid of this problem.
>
> As per their suggestion, I tried using the following tags in my input file,
>
>
>
>
> *>ion_dynamics = 'damp'>and>cell_dynamics = "damp-pr">which should respect
> the constraint.*
>
> And also I tried using *cell_dofree=‘ibrav'*. but still lattice parameter
> values a and b are not the same. Still, its structure changes from
> Tetragonal to Orthorhombic.
>
>
>
>
>
> *>You can open Modules/cell_base.f90 and at line 84 change>LOGICAL   ::
> enforce_ibrav = .FALSE.! True if ibrav ...>to>LOGICAL   :: enforce_ibrav =
> .TRUE.! True if ibrav ...>(do not forget to recompile pw.x) to use
> cell_dofree="ibrav" together with any >other cell_dofree*
>
> Since I'm utilizing the Government's supercomputer cluster, I'm not sure
> how I'd make these adjustments to the QE code. Can I get your suggestion
> regarding this?
>
>
>
> *>Should be present in the available patches for v.6.4.1:
> filebackports-6.4.1.diff*
>
> 
> in https://github.com/QEF/q-e/releases
>
>
> Could you please advise me on how to proceed with this link?
>
> I have included calculations input and output for your perusal.
>
> If you could share your thoughts on this issue, that would be quite
> helpful.
>
> *Input:*
>
> &control
> calculation = 'vc-relax'
> restart_mode = 'restart'
> prefix = 'CaWO4'
> tstress = .true.
> tprnfor = .true.
> pseudo_dir = './'
> outdir = './outCaWO4'
> wf_collect=.true.
> forc_conv_thr = 1d-4
> etot_conv_thr = 1D-4
> !disk_io = 'medium'
> max_seconds = 7200,
> /
>
> &system
> ibrav = 0
> *celldm(1) = 19.820415*
> nat = 48
> ntyp = 3
> ecutwfc = 65.0
> ecutrho = 780.0
> occupations= 'smearing'
> smearing= 'gaussian'
> degauss= 0.02
> /
>
> &electrons
> mixing_beta = 0.2,
> conv_thr = 1.0d-6,
> electron_maxstep = 5,
> /
>
> &IONS
> *ion_dynamics = 'damp'*
>
> /
>
> &cell
> cell_factor = 3.0d0
> *cell_dynamics = 'damp-pr'*
> !cell_dofree= 'ibrav'
> /
>
> ATOMIC_SPECIES
>  Ca 40.08   Ca.pbe-spn-rrkjus_psl.1.0.0.UPF
>  W  183.84  W.pbe-spn-rrkjus_psl.1.0.0.UPF
>  O  16.00   O.pbe-n-rrkjus_psl.1.0.0.UPF
>
> *CELL_PARAMETERS (alat)*
> *1.00 0.00 0.00*
> *0.00 0.49 0.00*
> *0.00 0.00 1.0845493931*
>
> ATOMIC_POSITIONS (crystal)
> Ca 0.0 0.25000 0.62500
> Ca 0.5 0.25000 0.62500
> Ca 0.0 0.75000 0.37500
> Ca 0.5 0.75000 0.37500
> Ca 0.25000 0.75000 0.12500
> Ca 0.75000 0.75000 0.12500
> Ca 0.25000 0.25000 0.87500
> Ca 0.75000 0.25000 0.87500
>  W 0.0 0.25000 0.12500
>  W 0.5 0.25000 0.12500
>  W 0.0 0.75000 0.87500
>  W 0.5 0.75000 0.87500
>  W 0.25000 0.75000 0.62500
>  W 0.75000 0.75000 0.62500
>  W 0.25000 0.25000 0.37500
>  W 0.75000 0.25000 0.37500
>  O 0.07485 0.00930 0.20973
>  O 0.574850023 0.00930 0.20973
>  O 0.425150007 0.99076 0.790300012
>  O 0.925150037 0.99076 0.790300012
>  O 0.175150007 0.99076 0.709699988
>  O 0.675150037 0.99076 0.709699988
>  O 0.324849993 0.00930 0.290300012
>  O 0.824849963 0.00930 0.290300012
>  O 0.370350003 0.399699986 0.459699988
>  O 0.87035000

Re: [QE-users] Crystal structure changes while doing vc-relax calculation.

2022-02-01 Thread Lorenzo Paulatto

Dear Vivek,
cell_dofree='ibrav' only works if you specified a value of ibrav>0 to begin 
with. If you don't know it, you can try to find it reading the description in 
INPUT_PW.html or with the scan_ibrav.x tool:

$ ~/espresso.BASE/PW/tools/scan_ibrav.x
Enter the unit of measur (angstrom, bohr) or alat in bohr units, or alat in Ang
strom units, followed by ' A'
19.820415
alat (bohr) 19.820415
Enter the cell basis vectors (one per line)
1.00 0.00 0.00
0.00 0.49 0.00
0.00 0.00 1.0845493931
Requested axes in Bohr units:
at1 19.820415 0.00 0.00
at2 0.00 9.910207 0.00
at3 0.00 0.00 21.496219

Scanning ibrav 1
The best cell with this ibrav is not good enough (chisq=0.8E+02)

Scanning ibrav 2
The best cell with this ibrav is not good enough (chisq=0.1E+04)

Scanning ibrav 3
The best cell with this ibrav is not good enough (chisq=0.1E+03)

Scanning ibrav -3
The best cell with this ibrav is not good enough (chisq=0.1E+04)

Scanning ibrav 4
The best cell with this ibrav is not good enough (chisq=0.8E+02)

Scanning ibrav 5
The best cell with this ibrav is not good enough (chisq=0.8E+02)

Scanning ibrav -5
The best cell with this ibrav is not good enough (chisq=0.1E+04)

Scanning ibrav 6
The best cell with this ibrav is not good enough (chisq=0.5E+02)

Scanning ibrav 7
The best cell with this ibrav is not good enough (chisq=0.1E+04)

Scanning ibrav 8
Minimization succeeded (chisq= 0. )
ibrav = 8
celldm( 1) = 19.820415000
celldm( 2) = 0.5
celldm( 3) = 1.084549393
at1 19.820415 0.00 0.00
at2 0.00 9.910207 0.00
at3 0.00 0.00 21.496219

...

Dr. Lorenzo Paulatto
IdR @ IMPMC - CNRS UMR 7590 & Sorbonne Université
phone: +33 (0)1 442 79822 / skype: paulatz
http://www.impmc.upmc.fr/~paulatto/ 
(https://link.getmailspring.com/link/78f32a0f-3925-4a88-b3c2-dce9b9225...@getmailspring.com/0?redirect=http%3A%2F%2Fwww.impmc.upmc.fr%2F~paulatto%2F&recipient=dXNlcnNAbGlzdHMucXVhbnR1bS1lc3ByZXNzby5vcmc%3D)
 - https://anharmonic.github.io/ 
(https://link.getmailspring.com/link/78f32a0f-3925-4a88-b3c2-dce9b9225...@getmailspring.com/1?redirect=https%3A%2F%2Fanharmonic.github.io%2F&recipient=dXNlcnNAbGlzdHMucXVhbnR1bS1lc3ByZXNzby5vcmc%3D)
23-24/423 B115, 4 place Jussieu 75252 Paris CX 05
On Feb 1 2022, at 1:52 pm, Vivek Christhunathan  wrote:
> Hello Everyone,
>
> I'm bringing attention to a topic that has previously been discussed in this 
> forum. I too faced the same type of problem when I try to do structural 
> relaxation for CaWO4 (2x1x1 supercell), that its crystal structure changed 
> from Tetragonal to Orthorhombic. I tried all the following forum's 
> suggestions to get rid of this problem.
>
> As per their suggestion, I tried using the following tags in my input file,
>
> >ion_dynamics = 'damp'
> >and
> >cell_dynamics = "damp-pr"
> >which should respect the constraint.
>
> And also I tried using cell_dofree=‘ibrav'. but still lattice parameter 
> values a and b are not the same. Still, its structure changes from Tetragonal 
> to Orthorhombic.
> >You can open Modules/cell_base.f90 and at line 84 change
> >LOGICAL :: enforce_ibrav = .FALSE.! True if ibrav ...
> >to
> >LOGICAL :: enforce_ibrav = .TRUE.! True if ibrav ...
> >(do not forget to recompile pw.x) to use cell_dofree="ibrav" together with 
> >any >other cell_dofree
>
> Since I'm utilizing the Government's supercomputer cluster, I'm not sure how 
> I'd make these adjustments to the QE code. Can I get your suggestion 
> regarding this?
> >Should be present in the available patches for v.6.4.1: file
> backports-6.4.1.diff
>  (https://link.getmailspring.com/link/78f32a0f-3925-4a88-b3c2-dce9b9225...@getmailspring.com/2?redirect=https%3A%2F%2Fgithub.com%2FQEF%2Fq-e%2Freleases%2Fdownload%2Fqe-6.4.1%2Fbackports-6.4.1.diff&recipient=dXNlcnNAbGlzdHMucXVhbnR1bS1lc3ByZXNzby5vcmc%3D)>
> in https://github.com/QEF/q-e/releases 
> (https://link.getmailspring.com/link/78f32a0f-3925-4a88-b3c2-dce9b9225...@getmailspring.com/3?redirect=https%3A%2F%2Fgithub.com%2FQEF%2Fq-e%2Freleases&recipient=dXNlcnNAbGlzdHMucXVhbnR1bS1lc3ByZXNzby5vcmc%3D)
> Could you please advise me on how to proceed with this link?
>
> I have included calculations input and output for your perusal.
>
> If you could share your thoughts on this issue, that would be quite helpful.
>
> Input:
>
> &control
> calculation = 'vc-relax'
> restart_mode = 'restart'
> prefix = 'CaWO4'
> tstress = .true.
> tprnfor = .true.
> pseudo_dir = './'
> outdir = './outCaWO4'
> wf_collect=.true.
> forc_conv_thr = 1d-4
> etot_conv_thr = 1D-4
> !disk_io = 'medium'
> max_seconds = 7200,
> /
>
> &system
> ibrav = 0
> celldm(1) = 19.820415
> nat = 48
> ntyp = 3
> ecutwfc = 65.0
> ecutrho = 780.0
> occupations= 'smearing'
> smearing= 'gaussian'
> degauss= 0.02
> /
>
> &electrons
> mixing_beta = 0.2,
> conv_thr = 1.0d-6,
> electron_maxstep = 5,
> /
>
> &

Re: [QE-users] Crystal structure changes while doing vc-relax calculation.

Re: [QE-users] Crystal structure changes while doing vc-relax calculation.

Re: [QE-users] Crystal structure changes while doing vc-relax calculation.

Re: [QE-users] Crystal structure changes while doing vc-relax calculation.

Re: [QE-users] Crystal structure changes while doing vc-relax calculation.

Re: [QE-users] Crystal structure changes while doing vc-relax calculation.

Re: [QE-users] Crystal structure changes while doing vc-relax calculation.

7 matches

Site Navigation

Mail list logo

Footer information