el-ph coupling is zero

[nikhitha79]

I am calculating the el-ph coupling strength, lambda via the EPC method (scf-ph-q2r-matdyn) and the EPW method. My system is a semiconductor CsGeI3. In both these methods the lambda is printing as zero. I am attaching the inputs.
&CONTROL
calculation = 'scf'
etot_conv_thr = 5.0000000000d-08
forc_conv_thr = 1.0000000000d-08
outdir = './'
prefix = 'csgei3'
pseudo_dir = './'
tprnfor = .true.
tstress = .true.
verbosity = 'high'
/
&SYSTEM
degauss = 2.0000000000d-02
ecutrho = 400
ecutwfc = 100
ibrav = 0
nat = 5
nosym = .false.
input_dft = 'PBESOL'
la2F = .true.
ntyp = 3
occupations = 'smearing'
smearing = 'gaussian'
degauss = 0.05
/
&ELECTRONS
conv_thr = 1.0000000000d-09
electron_maxstep = 80
mixing_beta = 4.0000000000d-01
/
ATOMIC_SPECIES
Cs 132.9054519 Cs.upf
Ge 72.64 Ge.upf
I 126.90447 I.upf
ATOMIC_POSITIONS crystal
Cs 0.0000250000 0.0000250000 0.0000250000
Ge 0.4684260000 0.4684260000 0.4684260000
I 0.5042230000 0.5042230000 0.0111030000
I 0.0111030000 0.5042230000 0.5042230000
I 0.5042230000 0.0111030000 0.5042230000
K_POINTS automatic
6 6 6 0 0 0
CELL_PARAMETERS angstrom
6.0991431900 0.0000000000 0.0000000000
0.1638917687 6.0969408018 0.0000000000
0.1638917687 0.1595453990 6.0948529437
&inputph
tr2_ph = 1.0d-20,
prefix = 'csgei3',
fildvscf = 'csgei3.dvscf',
fildyn = 'csgei3.dyn',
outdir = './',
ldisp = .true.
nmix_ph=20
alpha_mix(1)=0.9
electron_phonon = 'interpolated',
el_ph_sigma = 0.001,
el_ph_nsigma = 10,
nq1 = 2, nq2 = 2, nq3 = 2,
trans=.true.
/


&input
fildyn='csgei3.dyn',
flfrc='csgei3.fc',
la2f = .true.
zasr = 'crystal'
loto_2d = .true.
/

&input
asr = 'simple',
flfrc = 'csgei3.fc',
flfrq = 'csgei3.freq',
la2F = .true.,
dos = .false.,
fldos = 'phonon.dos'
ndos = 50
/
251
0.000000 0.000000 0.000000 1
0.010000 0.010000 0.010000 1
0.020000 0.020000 0.020000 1


The output for the ph calculation is as such:

Mode symmetry, C_1 (1) point group:

freq ( 1- 1) = 5.1 [cm-1] --> A
freq ( 2- 2) = 10.6 [cm-1] --> A
freq ( 3- 3) = 18.0 [cm-1] --> A
freq ( 4- 4) = 20.4 [cm-1] --> A
freq ( 5- 5) = 25.9 [cm-1] --> A
freq ( 6- 6) = 27.5 [cm-1] --> A
freq ( 7- 7) = 35.5 [cm-1] --> A
freq ( 8- 8) = 41.3 [cm-1] --> A
freq ( 9- 9) = 49.3 [cm-1] --> A
freq ( 10- 10) = 52.9 [cm-1] --> A
freq ( 11- 11) = 62.0 [cm-1] --> A
freq ( 12- 12) = 77.4 [cm-1] --> A
freq ( 13- 13) = 108.6 [cm-1] --> A
freq ( 14- 14) = 153.8 [cm-1] --> A
freq ( 15- 15) = 156.0 [cm-1] --> A
electron-phonon interaction ...

Gaussian Broadening: 0.001 Ry, ngauss= 0
DOS = 0.000000 states/spin/Ry/Unit Cell at Ef= 3.886122 eV
lambda( 1)= 0.0000 gamma= 0.00 GHz
lambda( 2)= 0.0000 gamma= 0.00 GHz
lambda( 3)= 0.0000 gamma= 0.00 GHz
lambda( 4)= 0.0000 gamma= 0.00 GHz
lambda( 5)= 0.0000 gamma= 0.00 GHz
lambda( 6)= 0.0000 gamma= 0.00 GHz
lambda( 7)= 0.0000 gamma= 0.00 GHz
lambda( 8)= 0.0000 gamma= 0.00 GHz
lambda( 9)= 0.0000 gamma= 0.00 GHz
lambda( 10)= 0.0000 gamma= 0.00 GHz
lambda( 11)= 0.0000 gamma= 0.00 GHz
lambda( 12)= 0.0000 gamma= 0.00 GHz
lambda( 13)= 0.0000 gamma= 0.00 GHz
lambda( 14)= 0.0000 gamma= 0.00 GHz
lambda( 15)= 0.0000 gamma= 0.00 GHz
Gaussian Broadening: 0.002 Ry, ngauss= 0
DOS = 0.000000 states/spin/Ry/Unit Cell at Ef= 3.979661 eV
lambda( 1)= 0.0000 gamma= 0.00 GHz
lambda( 2)= 0.0000 gamma= 0.00 GHz
lambda( 3)= 0.0000 gamma= 0.00 GHz
lambda( 4)= 0.0000 gamma= 0.00 GHz
lambda( 5)= 0.0000 gamma= 0.00 GHz
lambda( 6)= 0.0000 gamma= 0.00 GHz
lambda( 7)= 0.0000 gamma= 0.00 GHz
lambda( 8)= 0.0000 gamma= 0.00 GHz
lambda( 9)= 0.0000 gamma= 0.00 GHz
lambda( 10)= 0.0000 gamma= 0.00 GHz
lambda( 11)= 0.0000 gamma= 0.00 GHz
lambda( 12)= 0.0000 gamma= 0.00 GHz
lambda( 13)= 0.0000 gamma= 0.00 GHz
lambda( 14)= 0.0000 gamma= 0.00 GHz
lambda( 15)= 0.0000 gamma= 0.00 GHz

[xiaozha]

Hi,

I would be helpful to see the inputs of your EPW calculations and your band structure.

Just by looking at the description and the snapshots of the output you attached (DOS=0), my guess would be that your Fermi level is set to a somewhat mid-gap position (so there is no state available for the scattering to happen).

Best!
Xiao

[Shashi]

Hi Nikitha,

As Xiao pointed out, if DOS is zero there are no states for scattering. If this is indeed what you intend to do, then you need to shift the Fermi energy in EPW by setting efermi_read=.true. and fermi_energy=xx.

Also you have not showed EPW input only pasted QE input. In QE, you can not shift Fermi, you can use jellium model or you can use EPW with the above tag I mentioned.

Regards,
Shashi