Regarding very high Re [Sigma] and Im[Sigma]

General discussion around the EPW software

Moderator: stiwari

Post Reply
vinpandey
Posts: 18
Joined: Thu Jul 13, 2017 6:23 am
Affiliation:

Regarding very high Re [Sigma] and Im[Sigma]

Post by vinpandey »

Dear all,
I am doing the EPW calculation for Fe2VGa to calculate the scattering rate. Fe2VGa is a semimetallic system. Generally, the scattering rate for such system s is of the order of 1e-14 sec. However, I am finding the scattering rate very small, i.e., 1e-18 sec. I further looked into the standard EPW output file and found that the Re[sigma] and Im[Sigma] has very high value for example in some cases it is of the order of 200 eV and 642 eV respectively. The maximum phonon frequency for Fe2VGa is around 320 cm^-1. The Z values are very small like 0.000120. Following is my input file for EPW calculation. I have converged my calculation with respect to fine q grid.
Can anybody let me know what else to look into to understand my results better ? Thanks in advance

EPW input file

--
&inputepw
prefix = 'fe2vga'
amass(1) = 55.845
amass(2) = 50.9415
amass(3) = 69.723
outdir = './'
dvscf_dir = './save'
elph = .true.
kmaps = .true.
epbwrite = .false.
epbread = .true.
epwwrite = .false.
epwread = .true.
restart=.true.
iverbosity=0
!!lpolar = .true.
!!selecqread=.true.
wannierize = .false.
nbndsub = 24
nbndskip = 11
efermi_read= .true.
fermi_energy= 18.26

dis_win_max = 30.0d0
dis_froz_max = 18.0d0
dis_win_min = 0.0d0
iprint = 2

num_iter = 50000

proj(1)= 'Fe:d'
proj(2)= 'V:d'
proj(3)= 'Fe:p'
proj(4)= 'V:p'
wdata(1) = 'dis_num_iter = 50000'
wdata(2) = 'dis_mix_ratio = 0.1d0'
wdata(3) = 'num_print_cycles = 5000'
wdata(4) = 'exclude_bands = 1-9,34, 35'
wdata(5) = 'bands_plot = .true.'
wdata(6) = 'begin kpoint_path'
wdata(7) = 'W 0.25000000 0.75000000 0.500000000 L 0.50000000 0.50000000 0.500000000'
wdata(8) = 'L 0.50000000 0.50000000 0.500000000 G 0.00000000 0.00000000 0.000000000'
wdata(9) = 'G 0.00000000 0.00000000 0.000000000 X 0.000000 0.50000000 0.500000000'
wdata(10) = 'X 0.000000 0.50000000 0.500000000 W 0.25000000 0.75000000 0.500000000'
wdata(11)= 'W 0.25000000 0.75000000 0.500000000 K 0.3750000 0.75000000 0.3750000000'
wdata(12) = 'end kpoint_path'
wdata(13) = 'bands_num_points = 100'


elecselfen = .true.
phonselfen = .false.


a2f = .false.
!parallel_k = .true.
!parallel_q = .false.
fsthick = 1.5 ! eV
eptemp = 300 ! K
degaussw = 0.01 ! eV
filkf = 'path.dat'

!nkf1 = 10
!nkf2 = 10
!nkf3 = 1
nqf1 = 35
nqf2 = 35
nqf3 = 35
nk1 = 12
nk2 = 12
nk3 = 12
nq1 = 6
nq2 = 6
nq3 = 6
/

16 cartesian
0.000000000000000E+00 0.000000000000000E+00 0.000000000000000E+00
-0.168126614636308E+00 0.168126614636308E+00 -0.168126614636308E+00
-0.336253229272616E+00 0.336253229272616E+00 -0.336253229272616E+00
0.504379843908924E+00 -0.504379843908924E+00 0.504379843908924E+00
0.000000000000000E+00 0.336253229272616E+00 0.000000000000000E+00
-0.168126614636308E+00 0.504379843908924E+00 -0.168126614636308E+00
0.672506458545232E+00 -0.336253229272616E+00 0.672506458545232E+00
0.504379843908924E+00 -0.168126614636308E+00 0.504379843908924E+00
0.336253229272616E+00 0.277555756156289E-16 0.336253229272616E+00
0.000000000000000E+00 0.672506458545232E+00 0.000000000000000E+00
0.840633073181540E+00 -0.168126614636308E+00 0.840633073181540E+00
0.672506458545232E+00 -0.555111512312578E-16 0.672506458545232E+00
0.000000000000000E+00 -0.100875968781785E+01 0.000000000000000E+00
0.672506458545232E+00 -0.336253229272616E+00 0.100875968781785E+01
0.504379843908924E+00 -0.168126614636308E+00 0.840633073181540E+00
-0.336253229272616E+00 -0.100875968781785E+01 0.000000000000000E+00

hlee
Posts: 415
Joined: Thu Aug 03, 2017 12:24 pm
Affiliation: The University of Texas at Austin

Re: Regarding very high Re [Sigma] and Im[Sigma]

Post by hlee »

Dear vinpandey:

It seems you are using the old version of EPW.

So, the first thing I would like to ask you is to use the most-recent version of EPW, for instance, QE v7.1 with EPW v5.5 which can be downloaded at https://gitlab.com/QEF/q-e/-/releases .

For your question on transport, you need to wait for other developers.

Sincerely,

H. Lee

vinpandey
Posts: 18
Joined: Thu Jul 13, 2017 6:23 am
Affiliation:

Re: Regarding very high Re [Sigma] and Im[Sigma]

Post by vinpandey »

Hi hlee
Thanks for your reply,
I am using the new EPW v5.6 now.
I am doing the electron self energy calculation for Fe2VGa.
Now I am facing a new issue, i.e., the epw.x code stops just before it starts the electron self energy calculation and gives segmentation fault error.
Following is the epw input file for Fe2VGa.

--
&inputepw
prefix = 'fe2vga'
amass(1) = 55.845
amass(2) = 50.9415
amass(3) = 69.723
outdir = './'
dvscf_dir = './save'
etf_mem = 0
elph = .true.
epbwrite = .true.
epbread = .false.
epwwrite = .true.
epwread = .false.
restart=.true.
!iverbosity=0
!!lpolar = .true.
wannierize = .true.
nbndsub = 24
!!nbndskip = 11
efermi_read= .true.
fermi_energy= 18.20

dis_win_max = 30.0d0
dis_froz_max = 18.0d0
dis_win_min = 0.0d0
iprint = 2

num_iter = 50000

proj(1)= 'Fe:d'
proj(2)= 'V:d'
proj(3)= 'Fe:p'
proj(4)= 'V:p'
wdata(1) = 'dis_num_iter = 50000'
wdata(2) = 'dis_mix_ratio = 0.1d0'
wdata(3) = 'num_print_cycles = 5000'
wdata(4) = 'exclude_bands = 1-9,34, 35'
wdata(5) = 'bands_plot = .true.'
wdata(6) = 'begin kpoint_path'
wdata(7) = 'W 0.25000000 0.75000000 0.500000000 L 0.50000000 0.50000000 0.500000000'
wdata(8) = 'L 0.50000000 0.50000000 0.500000000 G 0.00000000 0.00000000 0.000000000'
wdata(9) = 'G 0.00000000 0.00000000 0.000000000 X 0.000000 0.50000000 0.500000000'
wdata(10) = 'X 0.000000 0.50000000 0.500000000 W 0.25000000 0.75000000 0.500000000'
wdata(11)= 'W 0.25000000 0.75000000 0.500000000 K 0.3750000 0.75000000 0.3750000000'
wdata(12) = 'end kpoint_path'
wdata(13) = 'bands_num_points = 100'


elecselfen = .true.

fsthick = 1.5 ! eV
temps = 300 ! K
degaussw = 0.01 ! eV
filkf = 'path.dat'

nqf1 = 40
nqf2 = 40
nqf3 = 40

nk1 = 8
nk2 = 8
nk3 = 8

nq1 = 2
nq2 = 2
nq3 = 2
/

Below is my error file:
forrtl: severe (174): SIGSEGV, segmentation fault occurred
Image PC Routine Line Source
epw.x 000000000120B051 Unknown Unknown Unknown
epw.x 000000000120918B Unknown Unknown Unknown
epw.x (deleted) 0000000001193DA4 Unknown Unknown Unknown
epw.x (deleted) 0000000001193BB6 Unknown Unknown Unknown
epw.x (deleted) 0000000001114379 Unknown Unknown Unknown
epw.x 000000000111D396 Unknown Unknown Unknown
libpthread-2.17.s 00007FCAB2805100 Unknown Unknown Unknown
epw.x 00000000004E04F1 wan2bloch_mp_ephw 1834 wan2bloch.f90
epw.x (deleted) 00000000004474DF Unknown Unknown Unknown
epw.x 000000000042D31F elphon_shuffle_wr 1001 elphon_shuffle_wrap.f90
epw.x 0000000000407A1F MAIN__ 148 epw.f90
epw.x 0000000000406D9E Unknown Unknown Unknown
libc-2.17.so 00007FCAB2154B15 __libc_start_main Unknown Unknown
epw.x 0000000000406CA9 Unknown Unknown Unknown

Can anybody help me out?

Post Reply