Re: [QE-users] What's representation 0 in ph.x?

[ykhuang]

Thank you so much.Indeed I used to try to make several single point 
calculations and fit the data to obtain the frequency. However I am not sure 
whether the result is reasonable. I'll read the paper you recommend, I believe 
it will be of help. Best regards. Yike Huang. 
 原始信息 发件人： Marzari Nicola  日期: 
2020/2/20  08:24  (GMT+08:00) 收件人： Quantum ESPRESSO users Forum 
 主题： Re: [QE-users] What's representation 0 
in ph.x? 




You might want to check the 2007 JACS by Dabo et al, where we discuss the 
independence between adsorption energies and CO vibrational shifts; we also 
diagonalize the interatomic force constants obtained by finite differences on a 
local neighborhood.


Nicola

Sent from a tiny keyboard... Contact info:

http://theossrv1.epfl.ch/Main/Contact



On 20 Feb 2020, at 01:08, ykhuang  wrote:





Hello Paulatto,
Thank you for your patient answer. 
Actually in the slab system, the absorption is very weak. Even it is possible 
that the absorption is still overestimated. So I think it makes sense to treat 
CO as a decoupled part.
I'll try to calculate the entire matrix or find other ways.
Thank you again. 



Best regards. 


Yike Huang. 



 原始信息 
发件人： Lorenzo Paulatto  
日期: 2020/2/20 03:33 (GMT+08:00) 
收件人： users@lists.quantum-espresso.org 
主题： Re: [QE-users] What's representation 0 in ph.x? 



> start_irr = 235
> last_irr = 240

Hello,
I did not understant that the molecule was adsorbed on a surface! In 
this case it makes sense to compute several q-points (not orthogonal to 
the surface!) However, if you set start_irr and last_irr, you are only 
computing a few rows of the dynamical matrix. As the phonons are the 
square roots of the eigenvalues of this matrix, you need the entire 
matrix in order to compute them.

You can compute differents irreducible representations (irr) separately 
and then combine the results together as explained in the Grid_example 
of phonon. But you still need all the matrix. If you have reason to 
think that the CO vibration is completely decoupled from the rest of the 
system, you may try to get its vibrational frequency by doing finite 
differences. It would also be possible to modify phonon to only compute 
the minor of the dynamical matrix that involve CO, then diagonalize it, 
but it is not currently implemented.

hth

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


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

Re: [QE-users] What's representation 0 in ph.x?

[Marzari Nicola]

You might want to check the 2007 JACS by Dabo et al, where we discuss the 
independence between adsorption energies and CO vibrational shifts; we also 
diagonalize the interatomic force constants obtained by finite differences on a 
local neighborhood.

Nicola

Sent from a tiny keyboard... Contact info:
http://theossrv1.epfl.ch/Main/Contact

On 20 Feb 2020, at 01:08, ykhuang  wrote:


Hello Paulatto,
Thank you for your patient answer.
Actually in the slab system, the absorption is very weak. Even it is possible 
that the absorption is still overestimated. So I think it makes sense to treat 
CO as a decoupled part.
I'll try to calculate the entire matrix or find other ways.
Thank you again.

Best regards.

Yike Huang.

 原始信息 
发件人： Lorenzo Paulatto 
日期: 2020/2/20 03:33 (GMT+08:00)
收件人： users@lists.quantum-espresso.org
主题： Re: [QE-users] What's representation 0 in ph.x?

> start_irr = 235
> last_irr = 240

Hello,
I did not understant that the molecule was adsorbed on a surface! In
this case it makes sense to compute several q-points (not orthogonal to
the surface!) However, if you set start_irr and last_irr, you are only
computing a few rows of the dynamical matrix. As the phonons are the
square roots of the eigenvalues of this matrix, you need the entire
matrix in order to compute them.

You can compute differents irreducible representations (irr) separately
and then combine the results together as explained in the Grid_example
of phonon. But you still need all the matrix. If you have reason to
think that the CO vibration is completely decoupled from the rest of the
system, you may try to get its vibrational frequency by doing finite
differences. It would also be possible to modify phonon to only compute
the minor of the dynamical matrix that involve CO, then diagonalize it,
but it is not currently implemented.

hth

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

Re: [QE-users] What's representation 0 in ph.x?

[ykhuang]

Hello Paulatto,Thank you for your patient answer. Actually in the slab system, 
the absorption is very weak. Even it is possible that the absorption is still 
overestimated. So I think it makes sense to treat CO as a decoupled part.I'll 
try to calculate the entire matrix or find other ways.Thank you again. Best 
regards. Yike Huang. 
 原始信息 发件人： Lorenzo Paulatto  日期: 2020/2/20  
03:33  (GMT+08:00) 收件人： users@lists.quantum-espresso.org 主题： Re: [QE-users] 
What's representation 0 in ph.x? > start_irr = 235> last_irr = 240Hello,I did 
not understant that the molecule was adsorbed on a surface! In this case it 
makes sense to compute several q-points (not orthogonal to the surface!) 
However, if you set start_irr and last_irr, you are only computing a few rows 
of the dynamical matrix. As the phonons are the square roots of the eigenvalues 
of this matrix, you need the entire matrix in order to compute them.You can 
compute differents irreducible representations (irr) separately and then 
combine the results together as explained in the Grid_example of phonon. But 
you still need all the matrix. If you have reason to think that the CO 
vibration is completely decoupled from the rest of the system, you may try to 
get its vibrational frequency by doing finite differences. It would also be 
possible to modify phonon to only compute the minor of the dynamical matrix 
that involve CO, then diagonalize it, but it is not currently implemented.hth-- 
Lorenzo Paulatto - Paris___Quantum 
ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)users mailing 
list 
users@lists.quantum-espresso.orghttps://lists.quantum-espresso.org/mailman/listinfo/users___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

[QE-users] Psi-k 2020 (EPFL, 14-17 Sep 2020)

[Nicola Marzari]

Dear QE Users,


I'm hijacking a bit the list to announce that we have launched the 
registration and put online the program for Psi-k 2020 - this is the 
largest conference worldwide in electronic-structure simulations, with 
140 invited talks, 170 contributed talks (that you can apply for!) and 
39 symposia in 6 parallel sessions.


Two of the plenary speakers are long-established contributors/developers 
of QE and related packages - hence my excuse for posting here :-).


All info at https://www.psik2020.net - official mail below.

Hope to see you in Lausanne!

nicola




Psi-k 2020: Registration is open! (EPFL 14-17 Sep 2020)

*Psi-k 2020 - 14-17 September 2020 - EPFL, Lausanne (Switzerland)*

Dear Psi-k'ers,

This is to announce that the registration for Psi-k 2020 is now open, 
and the full list of invited speakers and symposia is online.


This is the largest event worldwide in first-principles simulations; it 
is held every 5 years (the last two editions, in Berlin in 2010 and in 
San Sebastian in 2015, had more than 1’000 participants each), and with 
7 plenary talks, 140 invited talks, and 170 contributed talks it 
promises to be the most exciting, defining event in the field.


Early-bird deadline is May 1, 2020 – one can register right away, and 
submit an abstract later (by May 1, 2020, to be considered for a 
contributed talk, or by Aug. 1, 2020, for a poster). All information – 
including registration, program, accommodation, and childcare - is 
available at https://www.psik2020.net .


Looking very much forward to a successful Psi-k 2020, and to seeing you 
all in Lausanne!


Nicola Marzari

Chair, Psi-k Charity and Board of Trustees

*More information: *https://www.psik2020.net 

*The conference*

  * 3.5 days (from 2pm Sept. 14, 2020, to the evening of Sept. 17)
  * 7 plenary talks
  * 39 symposia
  * 6 parallel sessions
  * ~140 invited talks
  * ~170 contributed talks
  * 1 MAD evening to shake it all off!

*Important deadlines*

  * *May 1, 2020* — Early-bird registration deadline *REGISTRATIONS ARE
NOW OPEN! *
  * *May 1, 2020* — Abstract submission deadline to be considered for
contributed oral presentations
  * *May 1, 2020* — Deadline for “Volker Heine Award” applications (to
be advertised soon)
  * *Aug. 1, 2020* — Regular registration deadline
  * *Aug. 1, 2020* — Abstract submission deadline, posters only
  * *Sept. 6, 2020* — Late registration deadline
  * *Sept. 14–17, 2020* — Psi-k 2020 Conference

*Plenary speakers *

  * *Gábor Csányi*, University of Cambridge, UK
  * *Feliciano Giustino*, University of Texas at Austin, USA
  * *Sharon Hammes-Schiffer*, Yale University, USA
  * *Francesco Mauri*, University of Rome Sapienza, Italy
  * *Angelos Michaelides*, University College London, UK
  * *Lucia Reining*, Ecole Polytechnique, France
  * *Nicola Spaldin*, ETH Zurich, Switzerland

*Symposia*

*A | Physical Formalisms*

  * Density and density-matrix functional theories
  * Novel theoretical and computational developments in diagrammatic
many body theory
  * Strong correlations beyond local approximations
  * Open quantum systems and non-adiabatic dynamics
  * Emerging aspects of electron-phonon interactions
  * Strongly correlated materials
  * Exciting core electrons
  * New developments in quantum Monte Carlo simulation
  * Frontiers of DFT and beyond
  * Extreme non-linear optics
  * Excitons in complex materials
  * Non-equilibrium physics
  * New developments in first principles calculations of charge and heat
transport

*B | Algorithms, Software and Data*

  * Recent progress in reduced scaling methods: from quantum chemistry
and DFT to second-principles
  * Statistics and configuration sampling
  * Electrified solid-liquid interfaces
  * Software engineering: libraries for electronic structure applications
  * Machine-learned surrogate models: the quest for ab initio accuracy
at a fraction of the cost
  * Machine learning for many-body quantum physics
  * Structure searching and crystal structure prediction
  * Materials discovery by high-throughput screening and artificial
intelligence
  * Emerging and future computing
  * Current and future challenges associated with data infrastructure
and reproducibility

*C | Applications to Real Materials and Materials Design*

  * Ab initio guided design of structural materials
  * Functional materials and devices
  * Magnetism and spintronics
  * Surfaces and interfaces at realistic conditions
  * New physics in 2D materials and van der Waals multilayers
  * Molecules, macromolecules and biomolecules: challenges and new
opportunities for electronic structure methods
  * Twisted bilayer materials
  * Topological properties in real and momentum space
  * Strong spin-orbit and magnetic systems
  

Re: [QE-users] What's representation 0 in ph.x?

[Lorenzo Paulatto]

start_irr = 235
last_irr = 240


Hello,
I did not understant that the molecule was adsorbed on a surface! In 
this case it makes sense to compute several q-points (not orthogonal to 
the surface!) However, if you set start_irr and last_irr, you are only 
computing a few rows of the dynamical matrix. As the phonons are the 
square roots of the eigenvalues of this matrix, you need the entire 
matrix in order to compute them.


You can compute differents irreducible representations (irr) separately 
and then combine the results together as explained in the Grid_example 
of phonon. But you still need all the matrix. If you have reason to 
think that the CO vibration is completely decoupled from the rest of the 
system, you may try to get its vibrational frequency by doing finite 
differences. It would also be possible to modify phonon to only compute 
the minor of the dynamical matrix that involve CO, then diagonalize it, 
but it is not currently implemented.


hth

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

Re: [QE-users] Counting Basis Functions

[Stefano Baroni]

number of basis functions = number of plane waves in the wave-function expansion
SB

> On 19 Feb 2020, at 18:33, Ben Comer  wrote:
> 
> Thanks for this! I'm not familiar with how this translates into basis 
> functions. How many total basis functions does this result in, and how would 
> I calculate the number? is it just 499x199x55 = 5,461,555?
> 
> On 2/19/20 12:16 PM, Stefano de Gironcoli wrote:
>> On 19/02/20 17:58, Ben Comer wrote:
>>> sticks:   dense  smooth PW G-vecs:dense smooth  PW
>>>  Sum 499 19955 51837 13151  
>>> 1677 
>> 
>> there are 51837 G vectors such that ekin < ecutrho   ... this is used to 
>> expand the density
>> 
>> there are 131151 G-vectorse such that ekin < ecutwfc*4 ... this is used to 
>> apply the potential on a wfc
>> 
>> there are  1677 G-vectrose such that ekin < ecutwfc ...  this used to expand 
>> the wfcs
>> 
>> -
>> 
>> sticks refers to the fact that G-vectors are organized in z-oriented sticks. 
>> this affects the fft distribution
>> 
>>  499 non empty sticks in the FFT up to ecutrho, 199 non empty sticks in the 
>> FFT up to ecutwfc*4, 55 non-empty sticks in the FFT up to Ecutwfc
>> 
>> 
>> ___
>> Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
>> users mailing list users@lists.quantum-espresso.org
>> https://lists.quantum-espresso.org/mailman/listinfo/users
> ___
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
> users mailing list users@lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users

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

Re: [QE-users] Counting Basis Functions

[Ben Comer]

Thanks for this! I'm not familiar with how this translates into basis 
functions. How many total basis functions does this result in, and how 
would I calculate the number? is it just 499x199x55 = 5,461,555?


On 2/19/20 12:16 PM, Stefano de Gironcoli wrote:

On 19/02/20 17:58, Ben Comer wrote:

sticks:   dense  smooth PW G-vecs:    dense smooth  PW
 Sum 499 199    55 51837 
13151      1677 


there are 51837 G vectors such that ekin < ecutrho   ... this is used 
to expand the density


there are 131151 G-vectorse such that ekin < ecutwfc*4 ... this is 
used to apply the potential on a wfc


there are  1677 G-vectrose such that ekin < ecutwfc ...  this used to 
expand the wfcs


-

sticks refers to the fact that G-vectors are organized in z-oriented 
sticks. this affects the fft distribution


 499 non empty sticks in the FFT up to ecutrho, 199 non empty sticks 
in the FFT up to ecutwfc*4, 55 non-empty sticks in the FFT up to Ecutwfc



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

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

Re: [QE-users] Counting Basis Functions

[Stefano de Gironcoli]

On 19/02/20 17:58, Ben Comer wrote:

sticks:   dense  smooth PW G-vecs:    dense smooth  PW
 Sum 499 199    55 51837 
13151      1677 


there are 51837 G vectors such that ekin < ecutrho   ... this is used to 
expand the density


there are 131151 G-vectorse such that ekin < ecutwfc*4 ... this is used 
to apply the potential on a wfc


there are  1677 G-vectrose such that ekin < ecutwfc ...  this used to 
expand the wfcs


-

sticks refers to the fact that G-vectors are organized in z-oriented 
sticks. this affects the fft distribution


 499 non empty sticks in the FFT up to ecutrho, 199 non empty sticks in 
the FFT up to ecutwfc*4, 55 non-empty sticks in the FFT up to Ecutwfc



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

[QE-users] Counting Basis Functions

[Ben Comer]

Hey all,

I'm wondering about how I would count the total number of basis 
functions QE is using in a given calculation. I see some portions in the 
output file that may be related to this, but I'm not an expert so 
reading the output is somewhat challenging. The two portions I think 
might be related to this are below. my best guess is that that it is the 
PW column below, and thus 1677 basis functions are used for this 
calculation. Also, if you have any information on how the number of 
basis functions to be used is decided I'd love to hear it.


Thanks,
Ben Comer
Georgia Tech


 G-vector sticks info
 
 sticks:   dense  smooth PW G-vecs:    dense smooth  PW
 Sum 499 199    55 51837   13151      1677


 Dense  grid:    51837 G-vectors FFT dimensions: (  27, 27, 160)

 Smooth grid:    13151 G-vectors FFT dimensions: (  18, 18, 108)

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

Re: [QE-users] What's representation 0 in ph.x?

[ykhuang]

Hello Paulatto,
Thank you for your reply! Maybe I misunderstand the phonon calculations of 
isolated molecule.

However, in the periodic system, the CO absorbed slab, I can hardly understand 
"representation 0" in ph.x output file.
I copied only a part of output file:
...
 End of self-consistent calculation

 Convergence has been achieved 

 Not diagonalizing because representation0 is not done
...
Phonon calculation terminated normally, but there's no frequency information at 
the end of each q-points iteration.

My input file is:

outdir = './CO/'
prefix = 'scf'
tr2_ph = 1.0d-18
ldisp = .TRUE.
nq1 = 2
nq2 = 2
nq3 = 2
amass(1) = 1.00794
amass(2) = 151.964
amass(3) = 15.9994
amass(4) = 47.867
amass(5) = 12.0107
start_irr = 235
last_irr = 240
/

Thanks.
Yike Huang
PhD candidate
Dalian Institute of chemistry and physics, CAS, China.

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

Re: [QE-users] What's representation 0 in ph.x?

[Lorenzo Paulatto]

1. When I do phonon calculation of isolated CO molecule with q points 
sampling automatically 2x2x2, I find large negative frequency emerges in 
DOS.


Hello Yike,
if it is an isolated molecule, only the phonons at Gamma make sense : 
there is no point in doing a 2x2x2 grid. If the negative frequencies are 
not at Gamma, there is nothing to worry about.


cheers

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

[QE-users] What's representation 0 in ph.x?

[ykhuang]

I'm sorry for attaching Luiz's mail by mistake. I rewrite my mail. Really 
sorry. Hi, QE users.Recently I plan to perform free energy calculations of CO 
absorption using ph.x, matdyn.x, q2r.x and fqha.x. However, I meet mainly 2 
problems.1. When I do phonon calculation of isolated CO molecule with q points 
sampling automatically 2x2x2, I find large negative frequency emerges in DOS. 
Could anyone tell me why ph.x yields negative frequency? Moreover if I want to 
eliminate the negative frequency, should I decrease the tr2_ph value?2. When I 
do phonon calculation of the CO absorbed slab, to simplify the calculation, I 
set start_irr and last_irr parameters including three translational degrees of 
freedom of C and O atom. However, after each q point iteration, ph.x outputs 
one line like:Not Davison diagonalization, representation 0 is not done.As a 
result, no frequency information is yielded, although ph.x job terminates 
normally. However, it is hard to understand representation 0, because 
irreducible representation in ph.x is marked from 1, not 0. I don't know how I 
can solve this problem.Any suggestions will be appreciated. Best regards.Yike 
HuangPhD candidate.Dalian Institute of chemistry and physics, CAS, China.___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

[QE-users] What's representation 0 in ph.x?

[ykhuang]

I'm sorry for attaching Luiz's mail by mistake. I re-write my mail. Really 
sorry.Hi, QE users.Recently I plan to perform free energy calculations of CO 
absorption using ph.x, matdyn.x, q2r.x and fqha.x. However, I meet mainly 2 
problems.1. When I do phonon calculation of isolated CO molecule with q points 
sampling automatically 2x2x2, I find large negative frequency emerges in DOS. 
Could anyone tell me why ph.x yields negative frequency? Moreover if I want to 
eliminate the negative frequency, should I decrease the tr2_ph value?2. When I 
do phonon calculation of the CO absorbed slab, to simplify the calculation, I 
set start_irr and last_irr parameters including three translational degrees of 
freedom of C and O atom. However, after each q point iteration, ph.x outputs 
one line like:Not Davison diagonalization, representation 0 is not done.As a 
result, no frequency information is yielded, although ph.x job terminates 
normally. However, it is hard to understand representation 0, because 
irreducible representation in ph.x is marked from 1, not 0. I don't know how I 
can solve this problem.Any suggestions will be appreciated. Best regards.Yike 
HuangPhD candidate.Dalian Institute of chemistry and physics, CAS, China.___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

[QE-users] What's representation 0 in ph.x?

[ykhuang]

Hi, QE users.Recently I plan to perform free energy calculations of CO 
absorption using ph.x, matdyn.x, q2r.x and fqha.x. However, I meet mainly 2 
problems.1. When I do phonon calculation of isolated CO molecule with q points 
sampling automatically 2x2x2, I find large negative frequency emerges in DOS. 
Could anyone tell me why ph.x yields negative frequency? Moreover if I want to 
eliminate the negative frequency, should I decrease the tr2_ph value?2. When I 
do phonon calculation of the CO absorbed slab, to simplify the calculation, I 
set start_irr and last_irr parameters including three translational degrees of 
freedom of C and O atom. However, after each q point iteration, ph.x outputs 
one line like:Not Davison diagonalization, representation 0 is not done.As a 
result, no frequency information is yielded, although ph.x job terminates 
normally. However, it is hard to understand representation 0, because 
irreducible representation in ph.x is marked from 1, not 0. I don't know how I 
can solve this problem.Any suggestions will be appreciated. Best regards.Yike 
HuangPhD candidate.Dalian Institute of chemistry and physics, CAS, China.
 原始信息 发件人： Luiz Gustavo Davanse da Silveira 
 日期: 2020/2/19  21:03  (GMT+08:00) 收件人： 
users@lists.quantum-espresso.org 主题： [QE-users] Advices on ralaxing perovskite 
supercells Greetings,I am using QE 6.4.1 to relax the atomic positions of a 
BiFeO3 40 atomsupercell, but the calculation didn't converge after 100 scf 
steps (thefinal scf accuracy was about 5E-4). I am a new QE user and I really 
don'tknow what to change in order to the calculation successfully run. 
Anyadvice is very much appreciated.My input follows bellow:   
calculation = 'relax'   prefix = 'bfo'  pseudo_dir = './sssp'   outdir = 
'./outdir' etot_conv_thr = 1.0e-5  forc_conv_thr = 1.0e-4/  ibrav = 
0   nat = 40, ntyp = 3, ecutwfc = 120, ecutrho = 960lda_plus_u = 
.TRUE., Hubbard_U(2) = 4/conv_thr = 1e-6/   ion_dynamics 
= 'bfgs'/ATOMIC_SPECIES Bi 208.98 Bi_pbe_v1.uspp.F.UPF Fe 55.85 
Fe.pbe-spn-kjpaw_psl.0.2.1.UPF O 16 O.pbe-n-kjpaw_psl.0.1.UPFATOMIC_POSITIONS 
(crystal) Bi 0.000 0.000 0.000 Bi 
0.500 0.000 0.000 Bi 
0.000 0.500 0.000 Bi 
0.500 0.500 0.000 Bi 
0.000 0.000 0.500 Bi 
0.500 0.000 0.500 Bi 
0.000 0.500 0.500 Bi 
0.500 0.500 0.500 Fe 
0.250 0.250 0.250 Fe 
0.750 0.250 0.250 Fe 
0.250 0.750 0.250 Fe 
0.750 0.750 0.250 Fe 
0.250 0.250 0.750 Fe 
0.750 0.250 0.750 Fe 
0.250 0.750 0.750 Fe 
0.750 0.750 0.750 O  
0.187667999289205 0.312332000710795 0.000 O  
0.812332000710795 0.187667999289205 0.000 O  
0.312332000710795 0.687667999289205 0.000 O  
0.687667999289205 0.812332000710795 0.000 O  
0.312332000710795 0.187667999289205 0.500 O  
0.687667999289205 0.312332000710795 0.500 O  
0.187667999289205 0.812332000710795 0.500 O  
0.812332000710795 0.687667999289205 0.500 O  
0.000 0.187667999289205 0.312332000710795 O  
0.500 0.312332000710795 0.187667999289205 O  
0.000 0.812332000710795 0.187667999289205 O  
0.500 0.687667999289205 0.312332000710795 O  
0.000 0.312332000710795 0.687667999289205 O  
0.500 0.187667999289205 0.812332000710795 O  
0.000 0.687667999289205 0.812332000710795 O  
0.500 0.812332000710795 0.687667999289205 O  
0.312332000710795 0.000 0.187667999289205 O  
0.687667999289205 0.000 0.312332000710795 O  
0.187667999289205 0.500 0.312332000710795 O  
0.812332000710795 0.500 0.187667999289205 O  
0.187667999289205 0.000 0.812332000710795 O  
0.812332000710795 0.000 0.687667999289205 O  
0.312332000710795 0.500 0.687667999289205 O  
0.687667999289205 0.500 0.812332000710795K_POINTS 
(automatic) 4 4 4 0 0 0CELL_PARAMETERS (angstrom) 7.8 0.0 0.0 0.0 7.8 0.0 0.0 
0.0 7.8Best regards,Luiz G. D. SilveiraUniversidade Federal do Paraná - 

[QE-users] Advices on ralaxing perovskite supercells

[Luiz Gustavo Davanse da Silveira]

Greetings,

I am using QE 6.4.1 to relax the atomic positions of a BiFeO3 40 atom
supercell, but the calculation didn't converge after 100 scf steps (the
final scf accuracy was about 5E-4). I am a new QE user and I really don't
know what to change in order to the calculation successfully run. Any
advice is very much appreciated.

My input follows bellow:


calculation = 'relax'
prefix = 'bfo'
pseudo_dir = './sssp'
outdir = './outdir'
etot_conv_thr = 1.0e-5
forc_conv_thr = 1.0e-4
/

ibrav = 0
nat = 40, ntyp = 3,
ecutwfc = 120, ecutrho = 960
lda_plus_u = .TRUE., Hubbard_U(2) = 4
/

conv_thr = 1e-6
/

ion_dynamics = 'bfgs'
/
ATOMIC_SPECIES
 Bi 208.98 Bi_pbe_v1.uspp.F.UPF
 Fe 55.85 Fe.pbe-spn-kjpaw_psl.0.2.1.UPF
 O 16 O.pbe-n-kjpaw_psl.0.1.UPF
ATOMIC_POSITIONS (crystal)
 Bi 0.000 0.000 0.000
 Bi 0.500 0.000 0.000
 Bi 0.000 0.500 0.000
 Bi 0.500 0.500 0.000
 Bi 0.000 0.000 0.500
 Bi 0.500 0.000 0.500
 Bi 0.000 0.500 0.500
 Bi 0.500 0.500 0.500
 Fe 0.250 0.250 0.250
 Fe 0.750 0.250 0.250
 Fe 0.250 0.750 0.250
 Fe 0.750 0.750 0.250
 Fe 0.250 0.250 0.750
 Fe 0.750 0.250 0.750
 Fe 0.250 0.750 0.750
 Fe 0.750 0.750 0.750
 O  0.187667999289205 0.312332000710795 0.000
 O  0.812332000710795 0.187667999289205 0.000
 O  0.312332000710795 0.687667999289205 0.000
 O  0.687667999289205 0.812332000710795 0.000
 O  0.312332000710795 0.187667999289205 0.500
 O  0.687667999289205 0.312332000710795 0.500
 O  0.187667999289205 0.812332000710795 0.500
 O  0.812332000710795 0.687667999289205 0.500
 O  0.000 0.187667999289205 0.312332000710795
 O  0.500 0.312332000710795 0.187667999289205
 O  0.000 0.812332000710795 0.187667999289205
 O  0.500 0.687667999289205 0.312332000710795
 O  0.000 0.312332000710795 0.687667999289205
 O  0.500 0.187667999289205 0.812332000710795
 O  0.000 0.687667999289205 0.812332000710795
 O  0.500 0.812332000710795 0.687667999289205
 O  0.312332000710795 0.000 0.187667999289205
 O  0.687667999289205 0.000 0.312332000710795
 O  0.187667999289205 0.500 0.312332000710795
 O  0.812332000710795 0.500 0.187667999289205
 O  0.187667999289205 0.000 0.812332000710795
 O  0.812332000710795 0.000 0.687667999289205
 O  0.312332000710795 0.500 0.687667999289205
 O  0.687667999289205 0.500 0.812332000710795
K_POINTS (automatic)
 4 4 4 0 0 0
CELL_PARAMETERS (angstrom)
 7.8 0.0 0.0
 0.0 7.8 0.0
 0.0 0.0 7.8

Best regards,
Luiz G. D. Silveira
Universidade Federal do Paraná - Brazil

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

Re: [QE-users] Number of kpoints for wannier90 calculation

[kagdada hardik]

Thank you for your reply

Using nosym = .true. calculates the nscf for more number of kpoints given
in input file. I have used another tag noinv = .true. together with nosym =
.true. as suggested by Gautam Sharma, which solved my problem.

Only nosym = .true. still consider the time reversal symmetry in k points,
while adding noinv = .true. disable the time reversal symmetry, which
further calculates the band energies only for given kpoints in input.

On Tue, 18 Feb, 2020, 8:42 PM Soumyadeep,  wrote:

> Dear Hardik,
>
>   Use nosym = .true. during nscf then re-run the post processing step,
> pw2wannier90 and wannier90. I think this solved the problem.
>
>   Check the no. of k-points in the scf/nscf output files, are they equal
> to 294 after using this flag?
>
> with best regards
> Soumyadeep
> ---
> Soumyadeep Ghosh,
> Senior Research Fellow,
> Homi Bhabha National Institute (HBNI),
> Raja Ramanna Centre for Advanced Technology, Indore, India-452013
> Mob: (+91)9424664553
> User Lab: 0731244-2580
> Email: soumyadeepghos...@gmail.com, soumyad...@rrcat.gov.in
> ---
>
>
> On 18-02-2020 18:02, kagdada hardik wrote:
> > Dear Users
> >
> > I have to perform Wannier90 calculations and for that nscf must be run
> > with the specified kmesh generated from the wannier90/utility/kmesh.pl
> > [1].
> >
> > I have generated 100 number of kpoints by setting the 10 10 1 in
> > kmesh.pl [1] and for that i have started nscf calculation. However,
> > the output of nscf calculation contains 294 number of kpoints. But for
> > wannier90 calculation, nscf must be performed at 100 number of
> > kpoints.
> >
> > Is there any specific method to perform the nscf having the same
> > number of kpoints in input and output ??? I am working on orthorhombic
> > and hexagonal structure and i have not employed any spin polarization
> > in the calculation.
> >
> > I have tried nosym = .true. and further i also visited this forum :
> > https://lists.quantum-espresso.org/pipermail/users/2016-June/035493.html
> >
> > still the problem is not solved
> >
> > Any suggestions regarding this will be highly appreciated
> >
> > Thank You
> >
> > Hardik L Kagdada
> >
> > Ph.D. Scholar
> > Physics Department
> > Institute of Infrastructure Technology Research and Management
> > Ahmedabad, Gujarat, INDIA
> > ORCID ID: https://orcid.org/-0002-8584-6683
> > Google Scholar:
> > https://scholar.google.com/citations?user=zkez1GAJ=en
> >
> >
> >
> > Links:
> > --
> > [1] http://kmesh.pl
> > ___
> > Quantum ESPRESSO is supported by MaX
> > (www.max-centre.eu/quantum-espresso)
> > users mailing list users@lists.quantum-espresso.org
> > https://lists.quantum-espresso.org/mailman/listinfo/users
>
>
___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

Re: [QE-users] pw error

[janardhan H.L.]

 As the calculations have running normally and an error has appeared during 
saving to the outdir,it seems that all the nodes does not have read-write 
access to outdir. If you are running QE on the cluster, make sure all the nodes 
have read-write access to the outdir via NFS or parallel file system. 


regards
Janardhan
On Monday, 17 February, 2020, 03:13:26 am IST, Sansotta, Stefano 
 wrote:  
 
  
Dear PWSCF users,




I have installed QE version 6.4 in parallel using the following list of 
compilers: 

 



MPIF90         = mpif90F90           = ifortCC             = gccF77            
= ifort
When I run a simple vc-relax simulation at the end of the simulation,  I get 
the following error:
     highest occupied level (ev):    16.8328
!    total energy              =    -674.01889423 Ry     Harris-Foulkes 
estimate   =    -674.01889456 Ry     estimated scf accuracy    <       
0.0034 Ry
     The total energy is the sum of the following terms:
     one-electron contribution =    -318.62056763 Ry     hartree contribution   
   =     199.01850982 Ry     xc contribution           =      -4.61748527 Ry    
 ewald contribution        =    -549.79935115 Ry
     convergence has been achieved in  44 iterations
mkdir fail: [2] No such file or directory
 %% 
    Error in routine create_directory (1):     unable to create directory 
/lxfs/bctc/bctc47/QE/Barium/0H2O/Save/Ba0H2O.save/ 
%%
     stopping ...
 %% 
    Error in routine create_directory (1):     unable to create directory 
/lxfs/bctc/bctc47/QE/Barium/0H2O/Save/Ba0H2O.save/ 
%%

I tried to create the folder before starting the simulation but I get the same 
error. 
I checked some other post and I think it is caused by the library 
"wrappers.f90"Is there anyone that can help me?

I would like to thank you in advance for your help. If you need more details 
about the installation or the input file please let me know. 

Best Regards,
Stefano 












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

Re: [QE-users] non-convergency of DFPT+U calculation

[Timrov Iurii]

Dear Merlin,


Total magnetization must be zero for an antiferromagnetic insulator. Hence, 
some deviation from zero that you obtain is probably a numerical noise (I would 
check the convergence of the results: reduce conv_thr, increase the size of the 
k mesh, etc., and relax well the structure).


> If I wanted to be closer to the smearing calculation, I would have to 
> consider nocolin=.true., is that right?


DFPT+U with noncolin=.true. is not implemented.


Greetings,

Iurii


--
Dr. Iurii Timrov
Postdoctoral Researcher
STI - IMX - THEOS and NCCR - MARVEL
Swiss Federal Institute of Technology Lausanne (EPFL)
CH-1015 Lausanne, Switzerland
+41 21 69 34 881
http://people.epfl.ch/265334

From: users  on behalf of Merlin 
Meheut 
Sent: Wednesday, February 19, 2020 12:38:08 AM
To: Quantum ESPRESSO users Forum
Subject: Re: [QE-users] non-convergency of DFPT+U calculation

Dear Iurii,

Thank you again for your help! So far, the Example02 seems quite clear to me, 
although it will take me some time to apply it to my system and check that 
everything is working. One quick question, though: in my case, the "total 
magnetization" obtained from the first calculation (with smearing) is non-zero 
and non-integer: "total magnetization   =-0.63 Bohr mag/cell" (which makes 
sense since I have two different sites). As I understand it, for the second 
calculation, the variable tot_magnetization should be integer, so I can only 
take it equal to 0 or 1 here (I tried with 0, I don't know which is better). If 
I wanted to be closer to the smearing calculation, I would have to consider 
nocolin=.true., is that right?

Greetings,

Merlin

Le mar. 18 févr. 2020 à 17:32, Timrov Iurii 
mailto:iurii.tim...@epfl.ch>> a écrit :

Dear Merlin,


> Does that mean that magnetic insulators cannot be treated by Quantum Espresso?


It is possible to model magnetic insulators in Quantum ESPRESSO. I am using the 
procedure described in q-e-qe-6.5/HP/examples/example02/README. If the 
description in README file is not clear enough, please let me know.


Greetings,

Iurii


--
Dr. Iurii Timrov
Postdoctoral Researcher
STI - IMX - THEOS and NCCR - MARVEL
Swiss Federal Institute of Technology Lausanne (EPFL)
CH-1015 Lausanne, Switzerland
+41 21 69 34 881
http://people.epfl.ch/265334

From: users 
mailto:users-boun...@lists.quantum-espresso.org>>
 on behalf of Merlin Meheut 
mailto:merlin.meh...@gmail.com>>
Sent: Tuesday, February 18, 2020 5:21:46 PM
To: Quantum ESPRESSO users Forum
Subject: Re: [QE-users] non-convergency of DFPT+U calculation

Hi Iurii,

First thank you very much for this remark, I did not know that one could not 
use tot_magnetization and starting_magnetization together. Indeed, this is 
reported in the Documentation INPUT_PW.def.
At this point, however, I don't see how to make an input for an 
antiferromagnetic insulating material (which is the goal here). Because with 
this input, the state is non magnetic:


ibrav =8, celldm(1)=9.107535, celldm(2)=2.17425,
celldm(3)=1.263056,
nat =28, ntyp = 4, ecutwfc =80.0, ecutrho = 320.0,
nspin=2,  tot_magnetization=0,
 lda_plus_u = .true., U_projection_type = 'atomic',
  Hubbard_U(1)=2.4, Hubbard_U(2)=2.6, ! valeurs Yu et al 2013 =2.4/2.6
/

And if you use this input:


ibrav =8, celldm(1)=9.107535, celldm(2)=2.17425,
celldm(3)=1.263056,
nat =28, ntyp = 4, ecutwfc =${a}.0, ecutrho = ${b}.0,
nspin=2,
 starting_magnetization(1)=0.5, starting_magnetization(2)=-0.5,
lda_plus_u = .true., U_projection_type = 'atomic',
Hubbard_U(1)=2.5, Hubbard_U(2)=2.5,
/

,you  have an error message:

%%
 Error in routine iosys (1):
 fixed occupations and lsda need tot_magnetization

 %%

So as you say, using starting_magnetization imposes using smearing. But if I am 
correct, smearing means that your material is metallic, and therefore you 
cannot compute dielectric quantities or effective charges with phonon. 
Actually, in the INPUT_PW.def file, the word "antiferromagnetic" is only 
present once, in the "starting_magnetization" definitions. Does that mean that 
magnetic insulators cannot be treated by Quantum Espresso?

Would you have any hint as to how to extract myself from such a dead end?

Thanks again for your help,

Greetings,

Merlin

Le dim. 16 févr. 2020 à 23:11, Timrov Iurii 
mailto:iurii.tim...@epfl.ch>> a écrit :

Dear Merlin,


In your input for the PW calculation you have:


> tot_magnetization=0,
> starting_magnetization(1)=0.5, starting_magnetization(2)=-0.5


tot_magnetization and starting_magnetization must not be used together (please 
check the documentation). Moreover, starting_magnetization must be used with 
occupations='smearing', but I do not see smearing in your input. Is the ground 
state metallic 

[QE-users] relaxation under the external saw-like electrical potential

[Sahar Izadi]

Dear all,
I am interested in the paper: https://doi.org/10.1103/PhysRevB.97.054416
I would like to follow its results. At the first step, I need to relax the
crystal structure under the external saw-like electrical potential by QE
package. According to the paper, the bond angles of Cr-I-Cr in the top and
bottom plane of the relaxed structure, referred by theta1 and theta2 have a
difference about 2 degrees at 1 eV/nm. But my DFT calculations show that
the electric field does not change the structure geometry!
Here you can find my input file? If I have forgotten something, please let
me know.

Best regards,
Sahar Izadi
Postdoc researcher at IPM


  calculation='vc-relax',
  outdir='./',
  prefix='relE',
  pseudo_dir='./',
  tefield = .true.,
  dipfield = .true.,
  wf_collect=.true.,
  tprnfor=.true.,
  tstress=.true.,
  etot_conv_thr=1d-6,
  forc_conv_thr=1d-4,
/

  ibrav=4,
  celldm(1)=13.3180905273d0, celldm(3)=5.675666855383d0
  nat=8,
  ntyp=2,
  ecutwfc=80,
  occupations='smearing',
  smearing='mv',
  degauss=0.01d0,
  nspin   = 2
  starting_magnetization(1)=1,
  edir=  3
  eamp= 0.006 ! corresponding to 3.09 eV/nm
  emaxpos=0.95d0,
  eopreg=0.05d0,
/


  conv_thr=1d-10,
  electron_maxstep=500,
  mixing_beta=0.7d0,
/


  ion_dynamics='bfgs',
  upscale=220,
/


  cell_dynamics='bfgs',
  press=0.0d0,
  press_conv_thr=0.05d0,
/


ATOMIC_SPECIES
  Cr 51.996100d0  Cr.pbe-n-nc.UPF
  I  126.904000d0 I.pbe-n-nc.UPF

ATOMIC_POSITIONS {crystal}

  Cr   0.7437072547d0   0.191480d0   0.496536d0
  Cr   0.4103708485d0   0.518526d0   0.496885d0
   I   0.0770490950d0   0.4748440602d0   0.4603865190d0
   I   0.0770444830d0   0.1918398526d0   0.5396142354d0
   I   0.4355397493d0   0.1918537239d0   0.4603871637d0
   I   0.4355307864d0   0.8333206653d0   0.5396123221d0
   I   0.7185253232d0   0.4748317887d0   0.5396124224d0
   I   0.7185332444d0   0.808520d0   0.4603857646d0

K_POINTS {automatic}
  8 8 1 0 0 0
___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu/quantum-espresso)
users mailing list users@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users