The fermi-level of metal in the eph_task with inputfile

Hello

I am learning using the new eph_task to conculate the phono-linewidth of material.
When I conculate the semiconductor everything was fine.
But when I conculate the metal ,the fermi-level is confusing.
There are more than one fermi-level in the output file.
In the *.abo file .
I find the fisrt Fermi level is set by me .

 Fermi level set by the user at: 3.07  (eV)
 Old fermi level:     1.259802E-01, with nelect:     9.400000E+01
 New fermi level:     1.129354E-01, with nelect:     9.019173E+01
 Real(R)+Recip(G) space primitive vectors, cartesian coordinates (Bohr,Bohr^-1):
 R(1)= -8.5043173  8.5043173  8.5043173  G(1)=  0.0000000  0.0587937  0.0587937
 R(2)=  8.5043173 -8.5043173  8.5043173  G(2)=  0.0587937  0.0000000  0.0587937
 R(3)=  8.5043173  8.5043173 -8.5043173  G(3)=  0.0587937  0.0587937  0.0000000
 Unit cell volume ucvol=  2.4602450E+03 bohr^3

But later when the program ready to do the eph_task or maybe should called the interpolation. There comes to a new fermi level that is quit different to which I set.

 Linear tetrahedron method.
 Mesh step:  10.0 (meV) with npts: 1799
 From emin: -10.4 to emax:   7.6 (eV)
 Number of k-points in the IBZ: 396
 Fermi level:   2.93312829E+00 (eV)
 Total electron DOS at Fermi level in states/eV:   2.79399221E+01
 Total number of electrons at eF:   90.1

I want to know which one is the real fermi-level the eph_task used.
Here is my inputfile.

###
optdriver 7
eph_task 1
occopt 3

ngkpt  20 20 20
nshiftk 1
shiftk 0.0 0.0 0.0

eph_ngqpt_fine 20 20 20
ddb_ngqpt 4 4 4

! ph_ngqpt 25 25 25


eph_intmeth  1

ph_nqpath 3

ph_ndivsm 40

ph_qpath  0 0 0
          0.5 0.5 0.5
          1 1 1
tmesh 300 1 1
 

eph_fermie 3.07312829 eV 
eph_fsewin 0.02


mixprec 1

boxcutmin 1.1
rprim  -0.5  0.5  0.5
        0.5  -0.5  0.5
        0.5  0.5  -0.5



ntypat 3

znucl 56  26 51


pp_dirpath  "/public1/home/sc60786/zhuzhenyu/BaFe4Sb12/"

pseudos   "56ba.2.hgh, 26fe.8.hgh, 51sb.5.hgh"


acell  1.7008634522E+01  1.7008634522E+01  1.7008634522E+01

natom 17
typat 1 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3
xred
                       0.0000000000E+00  0.0000000000E+00  0.0000000000E+00
                       5.0000000000E-01  5.0000000000E-01  5.0000000000E-01
                       5.0000000000E-01  0.0000000000E+00  0.0000000000E+00
                       0.0000000000E+00  5.0000000000E-01  0.0000000000E+00
                       0.0000000000E+00  0.0000000000E+00  5.0000000000E-01
                       1.6226951463E-01  3.3468219511E-01  4.9695170974E-01
                       6.6531780489E-01  8.2758731952E-01  1.6226951463E-01
                       1.7241268048E-01  8.3773048537E-01  3.3468219511E-01
                       6.6531780489E-01  5.0304829026E-01  8.3773048537E-01
                       8.2758731952E-01  1.6226951463E-01  6.6531780489E-01
                       3.3468219511E-01  1.7241268048E-01  8.3773048537E-01
                       5.0304829026E-01  8.3773048537E-01  6.6531780489E-01
                       4.9695170974E-01  1.6226951463E-01  3.3468219511E-01
                       3.3468219511E-01  4.9695170974E-01  1.6226951463E-01
                       1.6226951463E-01  6.6531780489E-01  8.2758731952E-01
                       8.3773048537E-01  3.3468219511E-01  1.7241268048E-01
                       8.3773048537E-01  6.6531780489E-01  5.0304829026E-01



nbdbuf 2
ecut 40
nband 60


nstep 300
getwfk_filepath "/public1/home/sc60786/zhuzhenyu/BaFe4Sb12/SCF/scfo_DS2_WFK"
getddb_filepath "/public1/home/sc60786/zhuzhenyu/BaFe4Sb12/PH/ph.ddb.out"
getdvdb_filepath "/public1/home/sc60786/zhuzhenyu/BaFe4Sb12/PH/ph_DVDB"

Best wish

zhu zhenyu

Postgraduate student of MGI

Shanghai University

now I find I can updata my attachment.
So here is my inputflie and outputfliephline.abi (2.2 KB)
phline.abo (1.3 MB)
thanks

Hello picksas,

you do not seem to be using any smearing or occopt so I am not sure what it does with your Fermi level. What is your intention?

In the tetrahedron integration it indeed comes up with a slightly lower Fermi energy, and not the one you impose either.

@gmatteo do you know if the latter EF is used anywhere?

MV