[QE-users] Raman spectra for metallic system in QE

Shubham Patel spatelphy at iitkgp.ac.in
Sun Nov 9 13:05:13 CET 2025


Dear Professor Paolo,

Thank you for your email. I recompiled the latest QE v.7.5 with a change in the routine 'solve_e.f90' as suggested here https://gitlab.com/QEF/q-e/-/merge_requests/2715/diffs?commit_id=842aafabf257567c991c1b5abc0e01284a340740 :


from  this      CALL davcio_drho(dvscfin(1,1,ipol),lrdrho, iudrho,ipol,+1)

to
                        CALL davcio_drho(drhop(1,1,ipol),lrdrho, iudrho,ipol,+1)

The issue is still there. I am still getting the astronomical numbers. 

===================================
     Electric Fields Calculation

      iter #   1 total cpu time :    24.0 secs   av.it.:  13.7
      thresh= 1.000E-02 alpha_mix =  0.700 |ddv_scf|^2 =  7.320E+02

      iter #   2 total cpu time :    37.3 secs   av.it.:  36.7
      thresh= 1.000E-02 alpha_mix =  0.700 |ddv_scf|^2 =  8.521E+11

      iter #   3 total cpu time :    50.9 secs   av.it.:  37.7
      thresh= 1.000E-02 alpha_mix =  0.700 |ddv_scf|^2 =  9.918E+12

      iter #   4 total cpu time :    65.6 secs   av.it.:  41.3
      thresh= 1.000E-02 alpha_mix =  0.700 |ddv_scf|^2 =  2.886E+13

      iter #   5 total cpu time :    81.2 secs   av.it.:  43.7
      thresh= 1.000E-02 alpha_mix =  0.700 |ddv_scf|^2 =  6.668E+13

      iter #   6 total cpu time :    96.8 secs   av.it.:  44.7
      thresh= 1.000E-02 alpha_mix =  0.700 |ddv_scf|^2 =  1.312E+16

      iter #   7 total cpu time :   115.9 secs   av.it.:  52.7
      thresh= 1.000E-02 alpha_mix =  0.700 |ddv_scf|^2 =  1.242E+17

      iter #   8 total cpu time :   136.3 secs   av.it.:  57.0
      thresh= 1.000E-02 alpha_mix =  0.700 |ddv_scf|^2 =  3.581E+17

      iter #   9 total cpu time :   157.0 secs   av.it.:  64.7
      thresh= 1.000E-02 alpha_mix =  0.700 |ddv_scf|^2 =  3.226E+21

      iter #  10 total cpu time :   180.5 secs   av.it.:  66.3
      thresh= 1.000E-02 alpha_mix =  0.700 |ddv_scf|^2 =  3.783E+22

      iter #  11 total cpu time :   206.6 secs   av.it.:  74.3
      thresh= 1.000E-02 alpha_mix =  0.700 |ddv_scf|^2 =  1.192E+25

      iter #  12 total cpu time :   233.6 secs   av.it.:  77.7
      thresh= 1.000E-02 alpha_mix =  0.700 |ddv_scf|^2 =  1.454E+26
                                                                                                                                         
...
==================================================

Sincerely,
Shubham

----- Original Message -----
From: "paolo giannozzi" <paolo.giannozzi at uniud.it>
To: "Quantum ESPRESSO users Forum" <users at lists.quantum-espresso.org>
Cc: "Shubham Patel" <spatelphy at iitkgp.ac.in>
Sent: Saturday, 8 November, 2025 18:48:21
Subject: Re: [QE-users] Raman spectra for metallic system in QE

If you are using v.7.4.1 and later, please see this:
    https://gitlab.com/QEF/q-e/-/merge_requests/2715
Paolo

On 11/5/2025 4:27 PM, Shubham Patel via users wrote:
> Dear Quantum Espresso users,
> 
> I am doing phonon calculations to calculate the dielectric constant and Raman spectra, but in the output I am getting the dielectric constant in form of '*****', is it because the value is too large?
> More specifically, is it true that calculating Raman spectra for metals is not implemented in Quantum Espresso. How does that work in VASP (for metals)?
> 
> ###########################################
> 
> iter #  30 total cpu time :   130.1 secs   av.it.:  21.3
>        thresh= 6.211E-10 alpha_mix =  0.700 |ddv_scf|^2 =  1.347E-18
> 
>        iter #  31 total cpu time :   134.9 secs   av.it.:  22.7
>        thresh= 1.161E-10 alpha_mix =  0.700 |ddv_scf|^2 =  5.013E-19
> 
>       End of electric fields calculation
> 
>            Dielectric constant in cartesian axis
> 
>            (******************      -0.381280124       0.000000000 )
>            (       0.381280154******************       0.000000000 )
>            (       0.000000000       0.000000000    -284.777139614 )
> ##########################################################
> 
> Also, there is this scf cycle for dv_scf, the value of which is keep increasing.
> 
> #######################################################
> 
> Derivative coefficient:  0.001000    Threshold: 1.00E-12
>       kpoint   1 ibnd**** pcgreen: root not converged 2.723E-08
>       kpoint   1 ibnd**** pcgreen: root not converged 2.007E-09
>       kpoint   1 ibnd**** pcgreen: root not converged 8.698E-09
>       kpoint   1 ibnd**** pcgreen: root not converged 9.726E-11
>       kpoint   1 ibnd**** pcgreen: root not converged 5.332E-11
>       kpoint   1 ibnd**** pcgreen: root not converged 1.593E-12
>       kpoint   1 ibnd**** pcgreen: root not converged 1.676E-12
>       kpoint   1 ibnd**** pcgreen: root not converged 1.638E-12
>       kpoint   1 ibnd**** pcgreen: root not converged 1.803E-12
>       Non-scf  u_k: avg # of iterations = 33.0
>       Non-scf Du_k: avg # of iterations = 71.8
> 
>            Dielectric constant from finite-differences
> 
>            (******************      -0.000005722       0.000000000 )
>            (      -0.000003815******************       0.000000000 )
>            (      -0.000000000      -0.000000000  -31779.441608902 )
> 
>       Computing Second order response
>       kpoint   1 ibnd**** pcgreen: root not converged 3.031E+01
>       kpoint   1 ibnd**** pcgreen: root not converged 6.080E+01
>       kpoint   1 ibnd**** pcgreen: root not converged 6.610E+01
> 
> 
>        iter #   1   av.it.:  65.3
>        thresh= 1.000E-02 alpha_mix =  0.700 |ddv_scf|^2 =  1.221E+30
>       kpoint   1 ibnd**** pcgreen: root not converged 2.161E+07
>       kpoint   1 ibnd**** pcgreen: root not converged 2.800E+06
>       kpoint   1 ibnd**** pcgreen: root not converged 9.143E+06
> 
> 
>        iter #   2   av.it.:  82.3
>        thresh= 1.000E-02 alpha_mix =  0.700 |ddv_scf|^2 =  2.786E+39
>       kpoint   1 ibnd**** pcgreen: root not converged 2.487E+08
>       kpoint   1 ibnd**** pcgreen: root not converged 1.773E+10
>       kpoint   1 ibnd**** pcgreen: root not converged 1.710E+09
> 
> 
>        iter #   3   av.it.:  74.0
>        thresh= 1.000E-02 alpha_mix =  0.700 |ddv_scf|^2 =  6.365E+40
>       kpoint   1 ibnd**** pcgreen: root not converged 2.377E+08
>       kpoint   1 ibnd**** pcgreen: root not converged 1.246E+08
>       kpoint   1 ibnd**** pcgreen: root not converged 1.434E+08
> 
> 
>        iter #   4   av.it.:  89.0
>        thresh= 1.000E-02 alpha_mix =  0.700 |ddv_scf|^2 =  4.085E+41
>       kpoint   1 ibnd**** pcgreen: root not converged 6.110E+07
>       kpoint   1 ibnd**** pcgreen: root not converged 4.883E+08
>       kpoint   1 ibnd**** pcgreen: root not converged 5.098E+07
> 
> 
>        iter #   5   av.it.:  81.7
>        thresh= 1.000E-02 alpha_mix =  0.700 |ddv_scf|^2 =  3.790E+41
> ...
> 
> ################################################
> 
> Here is my input for the ph.x
> 
> &inputph
>    tr2_ph=1.0d-18,
>    prefix='conbse',
>    amass(2)= 92.90638,
>    amass(3)= 78.96,
>    outdir='./out/',
>    fildyn='conbse.dynG',
>    lraman=.true.,
>    epsil=.true.,
>    trans=.true.,
>    asr=.true.
>   /
> 0.0 0.0 0.0
> 
> Could somebody tell me why this cycle for dv_scf which increases infinitely? I am doing a Gamma point calculation for pw.x and ph.x.
> 
> Thank you for your time.
> 
> Sincerely,
> Shubham
> _______________________________________________________________________________
> The Quantum ESPRESSO Foundation stands in solidarity with all civilians worldwide who are victims of terrorism, military aggression, and indiscriminate warfare.
> --------------------------------------------------------------------------------
> Quantum ESPRESSO is supported by MaX (www.max-centre.eu)
> users mailing list users at 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 206, 33100 Udine Italy, +39-0432-558216


More information about the users mailing list