Exercise3: spectral function of n -doped MgO

Running EPW
1

Dear Members,

I tried the above mentioned exercise 3 for which I have got the details from the web page:
https://docs.epw-code.org/_downloads/80c451e841eff22049941c0fd8bd6063/Wed.4.Verdi.pdf

I completed the scf step and the next phonon calculations.
Here are the input files.

input file of mgo.scf.in

–
&control
calculation=‘scf’,
prefix=‘mgo’,
pseudo_dir= ‘./’,
outdir = ‘./’,
/
&system
ibrav=2,
celldm(1) = 7.9595
nat=2,
ntyp=2,
ecutwfc = 30,
/
&electrons
conv_thr=1.d-10,
/
ATOMIC_SPECIES
Mg 24.305 Mg_ONCV_PBE-1.0.upf
O 15.999 O_ONCV_PBE-1.0.upf
ATOMIC_POSITIONS crystal
Mg 0.0 0.0 0.0
O 0.5 0.5 0.5
K_POINTS {automatic}
6 6 6 0 0 0


input file of mgo.ph.in


–mgo.ph.in
&inputph
prefix = ‘mgo’
fildvscf = ‘dvscf’
ldisp = .true
fildyn = ‘mgo.dyn’
nq1=4,
nq2=4,
nq3=4,
tr2_ph = 1.0d-12
/


I am stuck at the following step.

python /home/krish/q-e-qe-6.7MaX-Release/EPW/bin/pp.py
Enter the prefix used for PH calculations (e.g. diam)
mgo
Traceback (most recent call last):
File “/home/krish/q-e-qe-6.7MaX-Release/EPW/bin/pp.py”, line 116, in
XML = hasXML(prefix)
File “/home/krish/q-e-qe-6.7MaX-Release/EPW/bin/pp.py”, line 93, in hasXML
“No dyn0 file found cannot tell if xml format was used.”)
main.FileNotFoundError: No dyn0 file found cannot tell if xml format was used.


I am not sure how to get this work. Can someone help me?
I am using the recent release of q-e-qe-6.7MaX-Release and the corresponding EPW available within it.

Thank you.

Best regards,
KRISH

2

By the way, I have manually copied the required files to the ‘save’ folder and tried to run the epw using the command line

mpirun -np 4 /home/krish/q-e-qe-6.7MaX-Release/bin/epw.x <mgo.epw.in> mgo.epw.out
application called MPI_Abort(MPI_COMM_WORLD, 1) - process 0

The CRASH file detail

task # 0
from epw_readin : error # 2
Could not find namelist &inputepw

The mgo.epw.out file (last few lines . . . )

Reading input from standard input

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Error in routine epw_readin (2):
Could not find namelist &inputepw
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

stopping …

Here is the mgo.epw.in file

–
&inputepw
prefix = ‘mgo’
amass(1) = 24.305
amass(2) = 15.999
outdir = ‘./’
dvscf_dir = ‘./save/’

elph = .true.
kmaps = .false.
epbwrite = .true.
epbread = .false.
epwwrite = .true.
epwread = .false.
lpolar = .true.

wannierize = .true.
nbndsub = 4
bands_skipped = 5
num_iter = 300
dis_win_min = 0
dis_froz_max= 11.0
dis_win_max = 21.8
proj(1) = ‘O:p’
proj(2) = ‘Mg:s’

elecselfen = .false.
phonselfen = .false.
a2f = .false.
band_plot = .true.

fsthick = 2.0
nstemp = 20
degaussw = 0.05

filqf = ‘path.dat’
filkf = 'path.dat

nkf1 = 8
nkf2 = 8
nkf3 = 8

nqf1 = 8
nqf2 = 8
nqf3 = 8

nk1 = 4
nk2 = 4
nk3 = 4

nq1 = 4
nq2 = 4
nq3 = 4
/

Is there any error present in the input file? I am quite knew to the epw, but I have cross verified with the epw input keywords and I am unable to figure out what is really going on. When I checked the qe/test-suite using ‘make run-tests-epw-parallel’ all test have been passed without any error. Yet, I am stuck with this tutorial. Could you please help me out?

Thank you.

Best regards,
KRISH

3

Dear KRISH:

python /home/krish/q-e-qe-6.7MaX-Release/EPW/bin/pp.py
Enter the prefix used for PH calculations (e.g. diam)
mgo
Traceback (most recent call last):
File “/home/krish/q-e-qe-6.7MaX-Release/EPW/bin/pp.py”, line 116, in
XML = hasXML(prefix)
File “/home/krish/q-e-qe-6.7MaX-Release/EPW/bin/pp.py”, line 93, in hasXML
“No dyn0 file found cannot tell if xml format was used.”)
main.FileNotFoundError: No dyn0 file found cannot tell if xml format was used.

Some of tutorials in 2018 school are not compatible with the recent version of EPW.
Regarding the error above, you can use pp.py in the directory of EPW’s examples, for example, q-e/EPW/examples/mgb2/phonons .


Regarding your second error:

mpirun -np 4 /home/krish/q-e-qe-6.7MaX-Release/bin/epw.x <mgo.epw.in> mgo.epw.out
application called MPI_Abort(MPI_COMM_WORLD, 1) - process 0



task # 0
from epw_readin : error # 2
Could not find namelist &inputepw

The error message of “Could not find namelist &inputepw” is a little strange to me, but I can identify a few issues with your mpirun command and epw.in .

First I would like to ask you to issue the following command:

mpirun -np 4 /home/krish/q-e-qe-6.7MaX-Release/bin/epw.x -nk 4 -in mgo.epw.in > mgo.epw.out

It is more safe to use “-in” instead of “<” for input redirection and when running EPW, you should use the same number of pools as that of processes, that is, you should specify “-nk 4” if you want to use 4 cores (-np 4).

Second, please use the following epw.in since there are several changes in epw inputs in the recent version of EPW.
(please check the Releases page at https://docs.epw-code.org/doc/Releases.html and pay attention to the differences between the previous mgo.epw.in and the new one below)

--
&inputepw
  prefix      = 'mgo'
  amass(1)    = 24.305
  amass(2)    = 15.999
  outdir      = './'
  dvscf_dir   = './save'

  elph        = .true.
  kmaps       = .false.
  epbwrite    = .true.
  epbread     = .false.
  epwwrite    = .true.
  epwread     = .false.
  lpolar      = .true.

  wannierize  = .true.
  nbndsub     =  4
  bands_skipped = 'exclude_bands = 1-5'
  num_iter    = 300
  dis_win_min = 0
  dis_froz_max= 11.0
  dis_win_max = 21.8
  proj(1)     = 'O:p'
  proj(2)     = 'Mg:s'

  elecselfen  = .false.
  phonselfen  = .false.
  a2f         = .false.
  band_plot   = .true.

  fsthick     = 2.0
  temps       = 20 ! K
  degaussw    = 0.05

  filqf       = 'path2.dat'
  filkf       = 'path2.dat'

  nk1         = 4
  nk2         = 4
  nk3         = 4
  nq1         = 4
  nq2         = 4
  nq3         = 4
 /

Sincerely,

H. Lee

4

Dear H. Lee,

Thank you for your help.
I have followed your instructions and used the updated mgp.epw.in.
However, this time I am getting the following error.

CARSH file
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
task # 0
from epw_readin : error # 29
opening file path2.dat
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

mgo.epw.out file

MPI processes distributed on 1 nodes
K-points division: npool = 4
Reading input from mgo.epw.in

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Error in routine epw_readin (29):
opening file path2.dat
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

stopping …

I am unable to find this file, path2.dat in the working directory. Is this causing the error? How can I fix this issue?

Thank you.
Best regards,
KRISH

5

Dear KRISH

I am unable to find this file, path2.dat in the working directory. Is this causing the error? How can I fix this issue?

In the exercise3 directory contained in the file of Wed.4.Verdi.tar (https://docs.epw-code.org/_downloads/94cb40b72d91bd20d2b0c441a74ffa1f/Wed.4.Verdi.tar),
the file of path2.dat exists.

Sincerely,

H. Lee

6

Dear H Lee,

Thank you for your help.
I have copied the path2.dat file from the tutorial.
I have managed to get the epw run work. However, I am unable to get the doping of small charge using the steps given in the tutorial.
I have introduced a small charge doping of 0.00037 and modified the value of 16 in the crystal.fmt file to 16.00037 and executed the epw run.

Here is the input file the mgo.epw.in2

–
&inputepw
prefix = ‘mgo’
amass(1) = 24.305
amass(2) = 15.999
outdir = ‘./’
dvscf_dir = ‘./save’

elph = .true.
kmaps = .false.
epbwrite = .true.
epbread = .false.
epwwrite = .true.
epwread = .false.
lpolar = .true.

wannierize = .true.
nbndsub = 4
bands_skipped = ‘exclude_bands = 1-5’
num_iter = 300
dis_win_min = 0
dis_froz_max= 11.0
dis_win_max = 21.8
proj(1) = ‘O:p’
proj(2) = ‘Mg:s’

elecselfen = .false.
phonselfen = .false.
a2f = .false.
band_plot = .false.

fsthick = 2.0
temps = 20 ! K
degaussw = 0.002

! filqf = ‘path2.dat’
! filkf = ‘path2.dat’

nkf1 = 100
nkf2 = 100
nkf3 = 100
nqf1 = 1
nqf2 = 1
nqf3 = 1

nk1 = 4
nk2 = 4
nk3 = 4
nq1 = 4
nq2 = 4
nq3 = 4
/


Fermi energy coarse grid = 9.174856 eV

Skipping the first 5 bands:

The Fermi level will be determined with 6.00000 electrons

Fermi energy is calculated from the fine k-mesh: Ef = 12.210050 eV


The Fermi level should be calculated for 6.00037 electrons, but it is not. Could you help me?

Thank you.
KRISH

7

Dear KRISH:

As I already said, some of tutorials in 2018 school don’t match the current version of EPW.

In any case, first I would suggest you the following changes since you need to load crystal.fmt to use the new number of electrons:

(BEFORE)

kmaps = .false.
epbwrite = .true.
epbread = .false.
epwwrite = .true.
epwread = .false.

(AFTER)

kmaps = .true.
epbwrite = .false.
epbread = .false.
epwwrite = .false.
epwread = .true.

Sincerely,

H. Lee