Re: [QE-users] using ESM with boundary condition "bc1" for charged slab

[Ghosh, Prasenjit]

Thanks a lot Paolo for the clarification.

With regards,
Prasenjit

- Original Message -
From: "Paolo Giannozzi" 
To: "users" , "pghosh" 

Sent: Wednesday, February 14, 2024 9:46:36 PM
Subject: Re: [QE-users] using ESM with boundary condition "bc1" for charged slab

I inquired with the author of ESM and this is what he told me:
"Since BC1 has no screening medium in the system, we need to keep charge 
neutral in the DFT cell. Otherwise, the total energy is diverging. In 
the earlier versions, the manual included a note for such charge 
neutrality condition, but for added safety, a routine to check this has 
been incorporated into the program.

In an ESM-RISM calculation, even when using BC1, there is no need to 
impose the charge neutrality condition in the DFT cell, as the RISM part 
automatically generates a compensating charge to maintain the charge 
neutrality condition throughout the whole system."

Paolo

On 2/9/24 06:07, Ghosh, Prasenjit wrote:
>   
> Non si ricevono spesso messaggi di posta elettronica da 
> pgh...@iiserpune.ac.in. Informazioni sul perché è importante 
> 
>   
> 
> Dear all,
> 
> I am performing a calculation for a negatively charged slab using
> assume_isolated = 'esm'
> esm_bc = 'bc1'
> 
> Till qe-7.0 the scf calculation proceeds. However, from version 7.1 the 
> calculation stops with the following error:
> "cannot have charge, when ESM-BC1".
> 
> This is the consequence of a check by the "esm_check" subroutine. What I 
> fail to understand is that this check is also present in qe-6.8, 
> however, there the calculation doesn't stop.
> Am I missing something?
> Also why is this stopped from version 7.1. Are there any issues in the 
> previous versions regarding the implementations of the same? I thought 
> that this is an effective way to perform charged slab calculations.
> 
> Any help in clarifying this will be highly appreciated.
> 
> With regards,
> Prasenjit
> 
> 
> ___
> The Quantum ESPRESSO community stands by the Ukrainian
> people and expresses its concerns about the devastating
> effects that the Russian military offensive has on their
> country and on the free and peaceful scientific, cultural,
> and economic cooperation amongst peoples
> ___
> Quantum ESPRESSO is supported by MaX 
> (http://secure-web.cisco.com/1wg-Sn_ZUMtTyRMZsJKyKizaSqviZGjuik7-lGVJ130wHC5m66P1pvQ-ZUMUrYeXMA3qovdkUqGYUVjeTA3Mtg8obh8En3oN8CaZYpId-5ugTnWdq_I3lPum3Z5dVkPCsx--ir1MrwJldToxCEhBi8JOASttShakfuwRQsrHgG1vXd8luY3SSSwp3WjTDfNaWMAcQmKCxEWh5k67PcPvMMSqvvWl5Ky4RllnaHPcD1sveD9rSkht8dgXkN0Pths4Vvy0lzUAxnVmnPiJUJrWnnwkgrRp5-bL5nzsis57bBSuGRHvuKSB2q76txY6-ehxL/http%3A%2F%2Fwww.max-centre.eu)
> users mailing list users@lists.quantum-espresso.org
> https://lists.quantum-espresso.org/mailman/listinfo/users

-- 
Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
Univ. Udine, via delle Scienze 208, 33100 Udine Italy, +39-0432-558216
___
The Quantum ESPRESSO community stands by the Ukrainian
people and expresses its concerns about the devastating
effects that the Russian military offensive has on their
country and on the free and peaceful scientific, cultural,
and economic cooperation amongst peoples
___
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] [EXTERNAL] Re: Error in routine init_us_2 (1): internal error: dimension of interpolation table

[Karkee, Rijan via users]

Hi Paolo,

I am using the latest version of QE (7.3).

-Rijan

From: Paolo Giannozzi 
Sent: Wednesday, February 14, 2024 9:39 AM
To: Quantum ESPRESSO users Forum 
Cc: Karkee, Rijan 
Subject: [EXTERNAL] Re: [QE-users] Error in routine init_us_2 (1): internal 
error: dimension of interpolation table

The calculation of a numerical derivative requires to compute some
quantities at k+dk and k-dk. It might happen in some unfortunate cases
that q_1=|k+dk+G| or q_2=|k-dk+G| exceed the maximum value set in the
interpolation tables for \beta(q) (nonlocal projectors).

I think that in the latest version of the code out-of-bounds values are
skipped, so there is no crash (but a tiny error is introduced). I cannot
guarantee it, though. Alternatively, one has to locate where the tables
are allocated and increase their size a bit

Paolo

On 2/12/24 23:30, Karkee, Rijan via users wrote:
> Hi QE community,
>
> While running the ph.x calculation for Bi2Te3, I ran into this error
>
>   
> %%
>   Error in routine init_us_2 (1):
>   internal error: dimension of interpolation table
>   
> %%
>
> It completes the Electri Field Calculations, shows dielectric constant,
> also calculates Effective charges (d Force / dE) in cartesian axis with
> asr applied:
>atom  1BiMean Z*:7.08468
>E*x (8.24990   -0.0   -0.0 )
>E*y (0.08.24990   -0.0 )
>E*z (   -0.0   -0.04.75424 )
> atom  2BiMean Z*:7.08468
>E*x (8.24990   -0.00.0 )
>E*y (0.08.24990   -0.0 )
>E*z (   -0.0   -0.04.75424 )
> atom  3BiMean Z*:7.08489
>E*x (8.25032   -0.0   -0.0 )
>E*y (0.08.25032   -0.0 )
>E*z (   -0.0   -0.04.75403 )
> .
> And then stops after
>   Calling punch_plot_e
>   Writing on file  drho
>
>   Computing Pc [DH,Drho] |psi>
>
>   Derivative coefficient:  0.001000Threshold: 1.00E-12
>
>
>   
> %%
>   Error in routine init_us_2 (1):
>   internal error: dimension of interpolation table
>   
> %%
>
> My 
> https://urldefense.com/v3/__http://scf.in__;!!Bt8fGhp8LhKGRg!B8qGM3eY7I6ls08ee3azXoXsWGBhGUcnLyT08T_ihlrTcFSuroOiAJoTPFKH9MHT0OAalXdN2fafPyByMzkVmCYBrlQ$
>   ran with no error  (shown below the input)
> 
>  calculation = 'scf'
>  pseudo_dir  = '/lustre/scratch4/turquoise/rkarkee/pseudo'
>  outdir   = './'
>  prefix = 'BiTe'
>  tprnfor = .true.
>  tstress = .true.
> /
>
> 
>  ecutwfc =  90,
>  ibrav   = 0,
>  nat = 15,
>  ntyp= 2,
> !  occupations='tetrahedra'
> !   noncolin = .true.
> !lspinorb = .true.
>
> /
>
> 
>  conv_thr =  1.0e-10
>  electron_maxstep = 200
>  mixing_beta  =  7.0e-01
>  diagonalization  = "david"
>
>
> /
> 
> ion_dynamics='fire'
> /
> 
> press_conv_thr=0.01
> /
>
> K_POINTS {automatic}
> 8 8 2 1 1 1
>
>
> ATOMIC_SPECIES
> Bi 208.9804  Bi.upf
> Te  127.60  Te.upf
>
>
> CELL_PARAMETERS (angstrom)
> 4.354840746   0.0   0.0
>-2.177420373   3.771402716   0.0
>-0.0   0.0  29.863328832
>
> ATOMIC_POSITIONS (crystal)
> Bi  -0.000.000.4002380211
> Bi  -0.000.000.5997619789
> Bi   0.6668700.3334300.7335713665
> Bi   0.6668700.3334300.9330952695
> Bi   0.3334300.6668700.0669047085
> Bi   0.3334300.6668700.2664286035
> Te   0.00   -0.000.2088757429
> Te  -0.00   -0.000.7911242721
> Te   0.6668700.3334300.5422091044
> Te   0.6668700.3334300.1244576093
> Te   0.3334300.6668700.8755423987
> Te   0.3334300.6668700.4577909256
> Te  -0.000.00   -0.00
> Te   0.6668700.3334300.42
> Te   0.3334300.6668700.666958
>
> Then ph.x input is as follow
> Strain Raman Calculation
> 
>   tr2_ph=1.0d-16,
>   prefix='BiTe',
>   trans=.true.,
>   epsil=.true.
>   

Re: [QE-users] Error in routine init_us_2 (1): internal error: dimension of interpolation table

[Paolo Giannozzi]

The calculation of a numerical derivative requires to compute some 
quantities at k+dk and k-dk. It might happen in some unfortunate cases 
that q_1=|k+dk+G| or q_2=|k-dk+G| exceed the maximum value set in the 
interpolation tables for \beta(q) (nonlocal projectors).


I think that in the latest version of the code out-of-bounds values are 
skipped, so there is no crash (but a tiny error is introduced). I cannot 
guarantee it, though. Alternatively, one has to locate where the tables 
are allocated and increase their size a bit


Paolo

On 2/12/24 23:30, Karkee, Rijan via users wrote:

Hi QE community,

While running the ph.x calculation for Bi2Te3, I ran into this error

  %%
      Error in routine init_us_2 (1):
      internal error: dimension of interpolation table
  %%

It completes the Electri Field Calculations, shows dielectric constant, 
also calculates Effective charges (d Force / dE) in cartesian axis with 
asr applied:

       atom      1Bi    Mean Z*:        7.08468
       E*x (        8.24990       -0.0       -0.0 )
       E*y (        0.0        8.24990       -0.0 )
       E*z (       -0.0       -0.0        4.75424 )
            atom      2Bi    Mean Z*:        7.08468
       E*x (        8.24990       -0.0        0.0 )
       E*y (        0.0        8.24990       -0.0 )
       E*z (       -0.0       -0.0        4.75424 )
            atom      3Bi    Mean Z*:        7.08489
       E*x (        8.25032       -0.0       -0.0 )
       E*y (        0.0        8.25032       -0.0 )
       E*z (       -0.0       -0.0        4.75403 )
.
And then stops after
  Calling punch_plot_e
      Writing on file  drho

      Computing Pc [DH,Drho] |psi>

      Derivative coefficient:  0.001000    Threshold: 1.00E-12


  %%
      Error in routine init_us_2 (1):
      internal error: dimension of interpolation table
  %%

My scf.in ran with no error  (shown below the input)

     calculation = 'scf'
     pseudo_dir  = '/lustre/scratch4/turquoise/rkarkee/pseudo'
         outdir       = './'
         prefix = 'BiTe'
         tprnfor = .true.
     tstress = .true.
/


     ecutwfc     =  90,
     ibrav       = 0,
     nat         = 15,
     ntyp        = 2,
!  occupations='tetrahedra'
!       noncolin = .true.
!    lspinorb = .true.

/


     conv_thr         =  1.0e-10
     electron_maxstep = 200
     mixing_beta      =  7.0e-01
     diagonalization  = "david"


/

ion_dynamics='fire'
/

press_conv_thr=0.01
/

K_POINTS {automatic}
8 8 2 1 1 1


ATOMIC_SPECIES
Bi     208.9804  Bi.upf
Te      127.60  Te.upf


CELL_PARAMETERS (angstrom)
    4.354840746   0.0   0.0
   -2.177420373   3.771402716   0.0
   -0.0   0.0  29.863328832

ATOMIC_POSITIONS (crystal)
Bi              -0.00        0.00        0.4002380211
Bi              -0.00        0.00        0.5997619789
Bi               0.666870        0.333430        0.7335713665
Bi               0.666870        0.333430        0.9330952695
Bi               0.333430        0.666870        0.0669047085
Bi               0.333430        0.666870        0.2664286035
Te               0.00       -0.00        0.2088757429
Te              -0.00       -0.00        0.7911242721
Te               0.666870        0.333430        0.5422091044
Te               0.666870        0.333430        0.1244576093
Te               0.333430        0.666870        0.8755423987
Te               0.333430        0.666870        0.4577909256
Te              -0.00        0.00       -0.00
Te               0.666870        0.333430        0.42
Te               0.333430        0.666870        0.666958

Then ph.x input is as follow
Strain Raman Calculation

  tr2_ph=1.0d-16,
  prefix='BiTe',
  trans=.true.,
  epsil=.true.
  lraman=.true.
  outdir='./',
  fildyn='HfTe5.dynG',
  alpha_mix=0.4,
  nmix_ph=8
  max_seconds=169200,
/
0 0 0

Thanks

Best
Rijan



--
Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
Univ. Udine, via delle Scienze 208, 33100 Udine Italy, +39-0432-558216
___
The Quantum ESPRESSO community stands by the Ukrainian
people and expresses its concerns about the devastating
effects that the Russian military offensive has on their
country and on the free and peaceful scientific, cultural,
and economic cooperation amongst peoples
___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users 

Re: [QE-users] Different results for charged cluster in PHonon 5.1 and 7.2

[Lorenzo Paulatto]

Hello,

On top of what Paola says, I would recommend checking the output of pw.x 
first. If the total energy calculation is not identical, the phonons 
results will be widely different. And if there is a difference in the 
total energy calculation, it could be traced back more easily to a 
change in the code.


kind regards.


On 14/02/2024 11:14, RABONE Jeremy wrote:


Hello,


  I have re-run a Phonon calculation for a charged cluster that was 
done using version 5.1 with version 7.2 and the results are completely 
different; there is a large positive shift in the frequencies. Can 
somebody tell me what was done differently in 5.1 for charged cells ?



Kind regards,


Jeremy Rabone


___
The Quantum ESPRESSO community stands by the Ukrainian
people and expresses its concerns about the devastating
effects that the Russian military offensive has on their
country and on the free and peaceful scientific, cultural,
and economic cooperation amongst peoples
___
Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
users mailing listus...@lists.quantum-espresso.org
https://lists.quantum-espresso.org/mailman/listinfo/users

--
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://anharmonic.github.io/
23-24/423 B115, 4 place Jussieu 75252 Paris CX 05___
The Quantum ESPRESSO community stands by the Ukrainian
people and expresses its concerns about the devastating
effects that the Russian military offensive has on their
country and on the free and peaceful scientific, cultural,
and economic cooperation amongst peoples
___
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] using ESM with boundary condition "bc1" for charged slab

[Paolo Giannozzi]

I inquired with the author of ESM and this is what he told me:
"Since BC1 has no screening medium in the system, we need to keep charge 
neutral in the DFT cell. Otherwise, the total energy is diverging. In 
the earlier versions, the manual included a note for such charge 
neutrality condition, but for added safety, a routine to check this has 
been incorporated into the program.


In an ESM-RISM calculation, even when using BC1, there is no need to 
impose the charge neutrality condition in the DFT cell, as the RISM part 
automatically generates a compensating charge to maintain the charge 
neutrality condition throughout the whole system."


Paolo

On 2/9/24 06:07, Ghosh, Prasenjit wrote:


Non si ricevono spesso messaggi di posta elettronica da 
pgh...@iiserpune.ac.in. Informazioni sul perché è importante 




Dear all,

I am performing a calculation for a negatively charged slab using
assume_isolated = 'esm'
esm_bc = 'bc1'

Till qe-7.0 the scf calculation proceeds. However, from version 7.1 the 
calculation stops with the following error:

"cannot have charge, when ESM-BC1".

This is the consequence of a check by the "esm_check" subroutine. What I 
fail to understand is that this check is also present in qe-6.8, 
however, there the calculation doesn't stop.

Am I missing something?
Also why is this stopped from version 7.1. Are there any issues in the 
previous versions regarding the implementations of the same? I thought 
that this is an effective way to perform charged slab calculations.


Any help in clarifying this will be highly appreciated.

With regards,
Prasenjit


___
The Quantum ESPRESSO community stands by the Ukrainian
people and expresses its concerns about the devastating
effects that the Russian military offensive has on their
country and on the free and peaceful scientific, cultural,
and economic cooperation amongst peoples
___
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


--
Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
Univ. Udine, via delle Scienze 208, 33100 Udine Italy, +39-0432-558216
___
The Quantum ESPRESSO community stands by the Ukrainian
people and expresses its concerns about the devastating
effects that the Russian military offensive has on their
country and on the free and peaceful scientific, cultural,
and economic cooperation amongst peoples
___
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] Different results for charged cluster in PHonon 5.1 and 7.2

[Paolo Giannozzi]

On 2/14/24 11:14, RABONE Jeremy wrote:

   I have re-run a Phonon calculation for a charged cluster that was 
done using version 5.1 with version 7.2 and the results are completely 
different; there is a large positive shift in the frequencies. Can 
somebody tell me what was done differently in 5.1 for charged cells ?


no, nobody can: v.5.1 is almost 10 years old. All that can be done is to 
investigate whether the results for a specific case are correct for the 
latest version are correct, and if not, why


Paolo
--
Paolo Giannozzi, Dip. Scienze Matematiche Informatiche e Fisiche,
Univ. Udine, via delle Scienze 208, 33100 Udine Italy, +39-0432-558216
___
The Quantum ESPRESSO community stands by the Ukrainian
people and expresses its concerns about the devastating
effects that the Russian military offensive has on their
country and on the free and peaceful scientific, cultural,
and economic cooperation amongst peoples
___
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

[QE-users] Different results for charged cluster in PHonon 5.1 and 7.2

[RABONE Jeremy]

Hello,


  I have re-run a Phonon calculation for a charged cluster that was done using 
version 5.1 with version 7.2 and the results are completely different; there is 
a large positive shift in the frequencies. Can somebody tell me what was done 
differently in 5.1 for charged cells ?


Kind regards,


Jeremy Rabone
___
The Quantum ESPRESSO community stands by the Ukrainian
people and expresses its concerns about the devastating
effects that the Russian military offensive has on their
country and on the free and peaceful scientific, cultural,
and economic cooperation amongst peoples
___
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