Inconsistent behavior of quasiparticle DOS with increasing temperature

[simba2828]

Dear EPW experts,

I am using EPW calculations for some 2D materials, and I am facing some issue in the quasiparticle DOS. As we increase the temperature the peak in qDOS plot, Ns(w)/N_F should shift into the left indicating a reduction in the SC-gap. But I am seeing a reverse behavior. The peak is reducing and broadening but it shifting in the right. I am attaching the corressponding figure here. Any help and explanation will be appreciated.

Thank you very much,
Shubham

[hpaudya1]

Hi Shubham,

It looks like your calculations are not converged with respect to number of Mastubara frequency points. What is the value of 'wscut' used (and 'nswi' at different temperature) in your calculations? As the temperature increases, the value of 'nswi' decreases with fixed 'wscut'. I recommend to run a test with fixed 'nsiw' (for example that corresponds to the value at 5 K with the wscut used in the present calculation). This can be done setting 'nswi' in your input file instead of 'wscut'.

I hope it helps.

Happy EPWing,
Hari

[simba2828]

Hello Hari,
Thank you for your email.

The following are my parameters (wscut = 0.4):

Code: Select all

  nsiter   = 500
  conv_thr_iaxis = 1.0d-4
  degaussw    = 0.04 ! eV ~  1/4 of fsthick
  wscut = 0.4   ! eV 10 times of  Upper limit over frequency integration/summation in the Elisashberg eq(1 cm-1 ~ 1/8000 eV)
  fsthick     = 0.5  ! eV ~ 4 times the maximum phonon frequency
  muc     = 0.1
  eps_acustic = 0.1    ! Lowest boundary for the phonon frequency
  ephwrite    = .true. ! Writes .ephmat files used when Eliasberg = .true.

  nsmear      = 1
  delta_smear = 0.01 ! eV 0.04

  degaussq     = 0.15 ! meV
  nqstep       = 500
 

'nswi' is different for different temperatures and it is decreasing with increasing the temperature. As you suggested I will remove 'wscut' and do my calculations with fixed 'nswi'.
As you may see I set ' conv_thr_iaxis = 1.0d-4' and the calculation converges for every temperature value. Could there be still a problem of convergence?


Thank you,
Shubham

[simba2828]

Hi,

I set 'nswi = 150' which is equal to for 5K. The result is not changing, still the peak shifts in the right with increasing temperature. Ia m attaching the input epw.in,

Code: Select all

  eps_acustic = 0.1    ! Lowest boundary for the phonon frequency 
  ephwrite    = .true. ! Writes .ephmat files used when Eliasberg = .true.

  nsmear      = 1
  delta_smear = 0.01 ! eV 0.04

  degaussq     = 0.15 ! meV
  nqstep       = 500

  eliashberg  = .true.
  fermi_plot = .true.

  conv_thr_iaxis = 1.0d-4

  nsiter   = 500
  degaussw    = 0.04 ! eV ~  1/4 of fsthick
  !wscut = 0.4   ! eV 10 times of  Upper limit over frequency integration/summation in the Elisashberg eq(1 cm-1 ~ 1/8000 eV)
  fsthick     = 0.5  ! eV ~ 4 times the maximum phonon frequency 
  muc     = 0.1
  nswi = 150 

  dvscf_dir   = '../ph11/save'


  nk1         = 24
  nk2         = 24
  nk3         = 1

  nq1         = 6
  nq2         = 6
  nq3         = 1

  mp_mesh_k = .true.
  nkf1 = 120
  nkf2 = 120
  nkf3 = 1

  nqf1 = 60
  nqf2 = 60
  nqf3 = 1
 /

The phonon dos (calculated from epw calculations) plot is also attached.

Shubham

[simba2828]

Could someone help with the issue? I am still unable to resolve it.
I used dense fine grids as well 120x120x1 for both k and q points in my epw calculations still same results. Played with other parameters to but no success.

Shubham