I have a strange problem: I have a metallic slab (Ag 100). When I calculate the electron-phonon coupling and phonon linewidth the latter is always constant 0 at q1=Gamma. Away from gamma (even a small step is enough q2=0.01 |Gamma-X|) this is no longer the case as you can see below.
Code: Select all
# Phonon frequency and phonon lifetime in meV
# Q-point Mode Phonon freq (meV) Phonon linewidth (meV)
1 1 0.14018096E-05 0.0000000000E+00
1 2 0.15037978E-05 0.0000000000E+00
1 3 0.15678200E-05 0.0000000000E+00
1 4 0.32037347E+01 0.0000000000E+00
1 5 0.32037347E+01 0.0000000000E+00
1 6 0.50111038E+01 0.0000000000E+00
1 7 0.50111038E+01 0.0000000000E+00
1 8 0.52040993E+01 0.0000000000E+00
1 9 0.80325952E+01 0.0000000000E+00
1 10 0.80325952E+01 0.0000000000E+00
1 11 0.81943124E+01 0.0000000000E+00
1 12 0.10998395E+02 0.0000000000E+00
1 13 0.10998395E+02 0.0000000000E+00
1 14 0.12627565E+02 0.0000000000E+00
1 15 0.12627565E+02 0.0000000000E+00
1 16 0.12875931E+02 0.0000000000E+00
1 17 0.16689456E+02 0.0000000000E+00
1 18 0.18803622E+02 0.0000000000E+00
1 19 0.25815292E+02 0.0000000000E+00
1 20 0.25815292E+02 0.0000000000E+00
1 21 0.32939073E+02 0.0000000000E+00
1 22 0.52623817E+02 0.0000000000E+00
1 23 0.52623817E+02 0.0000000000E+00
1 24 0.54967801E+02 0.0000000000E+00
1 25 0.54967801E+02 0.0000000000E+00
1 26 0.57559031E+02 0.0000000000E+00
1 27 0.64030663E+02 0.0000000000E+00
2 1 -0.57327299E+00 0.0000000000E+00
2 2 0.50717746E-01 0.0000000000E+00
2 3 0.52337994E+00 0.0000000000E+00
2 4 0.31996073E+01 0.1518331503E-03
2 5 0.32052218E+01 0.1518331503E-03
2 6 0.49762293E+01 0.1529582313E-01
2 7 0.50116566E+01 0.9400167807E-05
2 8 0.52458229E+01 0.1173353266E+00
2 9 0.80309468E+01 0.1160566942E-02
2 10 0.80326033E+01 0.1160566942E-02mode 9 for example has \omega_{q\nu}=8 meV; there is plenty of "DOS" 8 meV above Ef for vertical (q=0) transitions in my system and I think the first equation here: http://epw.org.uk/Documentation/Electron-phononCoupling should definitely yield some non-zero elements;
I have tested different values for degaussq/w but above behaviour does not change.
Your help is very much appreciated!
Best,
Chris
Since a metal has plenty of states below and above Ef I was curious why the linewidth would be constant 0 in this case - did I do something wrong?