Results containing 0 as Raman intensities

[eamouj78]

Dear all,

I have been using the QERAMAN code for some time now. Recently, I am want to calculate the Raman intensities of a 72 atom Carbon structure. Due to this large system, the 'dvec' and 'elph' files are large (To have an idea, the 'elph' file is ~ 1.1 GB). Everything ends perfectly well. However when I execute the Raman, I get everything '0'

"Raman tensor
iel = 1
imode 1
0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
imode 2
0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
imode 3
0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
imode 4
"

and
" 0.0300 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
4.9562 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
9.8823 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
14.8085 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
19.7347 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
24.6608 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
29.5870 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
34.5132 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
39.4393 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
44.3655 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
49.2917 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
54.2178 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
59.1440 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
64.0701 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
68.9963 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
73.9225 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000 0.00000000E+000
"

What is the Reason for this behavior? Is it memory?

Thanks in advance

[nguyen-group]

Hi,

I also did not try with such a big system. How much of the k-points did you use for scf.in?
If it is small, it might not cover the energy eigenvalue at the resonance condition; then, everything might become zero even if we have non-zero for electron-phonon and electron-photon matrix elements.

Best all,
Nguyen

[Aldi-Saraiva]

Dear @eamouj78, I’ve experienced a similar issue when calculating a semiconductor without a smearing parameter. Although the SCF cycle converges, QERaman seems to treat the system as non-insulating, resulting in a zero Raman tensor. While I’m not certain if this is the case for your system, it might be worth checking.

[dapshenay]

Dear developers,

Thank you for very helpful and valuable code!

I have similar issue when I tryed to reproduce mos2-example2 for qe-7.5 and latest QERaman, while mos2-example1 was OK. The error is following

At line 244 of file raman_read.f90 (unit = 102, file = './mos2.elph')
Fortran runtime error: Bad real number in item 5 of list input

Error termination. Backtrace:
#0  0x7f8489e232ba in ???
#1  0x7f8489e23db9 in ???
#2  0x7f8489e2491f in ???
#3  0x7f848a090fc4 in ???
#4  0x7f848a09226d in ???
#5  0x7f848a09300a in ???
#6  0x556a20e9bc23 in __raman_read_MOD_read_elph
        at /home/dmitry/nfs_sm/_prog_/QERaman-test/qe-7.5/QERaman/src/raman_read.f90:244
#7  0x556a20e93bd0 in raman
        at /home/dmitry/nfs_sm/_prog_/QERaman-test/qe-7.5/QERaman/src/raman.f90:52
#8  0x556a20e9325e in main
        at /home/dmitry/nfs_sm/_prog_/QERaman-test/qe-7.5/QERaman/src/raman.f90:4

I looked inside mos2.elph and find the following:

      Electron-phonon matrix elements Mij(k,q) = sqrt(hbar/2*omega)<psi_{k+q,j}|dvscf_q*psi_{k,i}>
      nbnd   nmodes   nqs   nkstot   nksqtot
       15        9        1      576      576
      kx   ky   kz   weight_k   ik   ik+q
      ibnd  jbnd  imode  enk[eV]  enk+q[eV]  omega(q)[meV]   |M|[meV]   Re(M)[meV]   Im(M)[meV]
     ------------------------------------------------------------------------------
      0.00000000      0.00000000      0.00000000      0.00347222        1        1
     1     1     1    -20.41176032    -20.41176032        0.0000000000    0.0000000000E+00    0.0000000000E+00    0.0000000000E+00
     1     1     2    -20.41176032    -20.41176032        0.0000000000    0.0000000000E+00    0.0000000000E+00    0.0000000000E+00
     1     1     3    -20.41176032    -20.41176032********************    0.0000000000E+00    0.0000000000E+00    0.0000000000E+00
     1     1     4    -20.41176032    -20.41176032********************    0.0000000000E+00    0.0000000000E+00    0.0000000000E+00
     1     1     5    -20.41176032    -20.41176032********************    0.0000000000E+00    0.0000000000E+00    0.0000000000E+00
...........

I tried to change formatting to get rid of asterisks, but, anyway in this case I obtain zero matrix elements and Raman intensity.

Could you, please, help with this issue, as the new method of mos2-example2 can really improve performance?

Best regards,
Dmitry