Wannier interface

[jjmm1974]

Dear all

I am trying to calculate transport properties under the Boltzmann transport equation approximation using the GW levels calculated by yambo. I have seen that ypp interfaces yambo to wannier90, but I did not succeed to use it. I did not find any thread about this either.
I did:
ypp -w
and got

# __ __ ______ __ __ _______ ______
# | \ / \ / \ | \ / \| \ / \
# \$$\ / $$| $$$$$$\| $$\ / $$| $$$$$$$\| $$$$$$\
# \$$\/ $$ | $$__| $$| $$$\ / $$$| $$__/ $$| $$ | $$
# \$$ $$ | $$ $$| $$$$\ $$$$| $$ $$| $$ | $$
# \$$$$ | $$$$$$$$| $$\$$ $$ $$| $$$$$$$\| $$ | $$
# | $$ | $$ | $$| $$ \$$$| $$| $$__/ $$| $$__/ $$
# | $$ | $$ | $$| $$ \$ | $$| $$ $$ \$$ $$
# \$$ \$$ \$$ \$$ \$$ \$$$$$$$ \$$$$$$
#
# GPL Version 4.0.2 Revision 96
# MPI+OpenMP Build
# http://www.yambo-code.org
#
wannier # [R] Wannier 90 I/O

Then I typed "ypp" again and got

| | | || _ || |_| || _ || |
| |_| || |_| || || |_| || _ |
| || || || || | | |
|_ _|| || || _ | | |_| |
| | | _ || ||_|| || |_| || |
|___| |__| |__||_| |_||_______||_______|


<---> [01] CPU structure, Files & I/O Directories
<---> CPU-Threads:1(CPU)-8(threads)
<---> [02] Y(ambo) P(ost)/(re) P(rocessor)
<---> [03] Core DB
<---> :: Electrons : 40.00000
<---> :: Temperature [ev]: 0.000000
<---> :: Lattice factors [a.u.]: 21.84385 7.87270 8.58361
<---> :: K points : 40
<---> :: Bands : 500
<---> :: Symmetries : 2
<---> :: RL vectors : 104389
<---> [04] K-point grid
<---> :: Q-points (IBZ): 40
<---> :: X K-points (IBZ): 40
<---> [05] CORE Variables Setup
<---> [05.01] Unit cells
<01s> [05.02] Symmetries
<01s> [05.03] RL shells
<01s> [05.04] K-grid lattice
<01s> [05.05] Energies [ev] & Occupations
<01s> [06] == Wannier 90 Interface ==At line 58 of file WANNIER_driver.f90 (unit = 10, file = '.nnkp')
Fortran runtime error: End of file

The same results for "ypp -J <name_of_the_folder_with_the_ndb.QP_database>" and "ypp -J <seedname_for_the_nnkp_file>". I checked my distro (I just recompiled it from the svn) and did not find any WANNIER_driver.f90, but WANNIER_driver.F. I had a look to line 58, but was unable to figure out what is going on.

Could you please help me with this?

Thanks in advance

Juanjo

[andrea.ferretti]

Dea Janjo,

you are right, all fortran yambo source have .F extension, which may be converted to .f90 after precompilation.
Now, I am not expert of the yambo-wannier90 interface, but the error you get

<01s> [06] == Wannier 90 Interface ==At line 58 of file WANNIER_driver.f90 (unit = 10, file = '.nnkp')
Fortran runtime error: End of file

(end of file) usually means that app reached the end of file while reading the w90 datafile. In other words it means either that that format is outdated, or that some data that are expected are not there (from the sources I see there is a check on the kit grid, which needs to cover the whole BZ in w90. Is this the case?)

hope it helps
Andrea

[jjmm1974]

Hi Andrea

Thanks a lot for your answer. I think that I misunderstood how the ypp interface works. I will try again and will let you know.

Best

Juanjo

[jjmm1974]

Hi all

I am still facing troubles with the wannier interface. I did the following:

1) I rerun QE (scf + nscf) with an uniform grid of K-points. I used the flags "nosym=.true." and "noinv=.true." for compatibility with wannier90. The nscf run was made with 72 points, compatible with kpoints.x.
2) I rerun yambo under the PPA to get the ndb.QP database. This was fine and in full agreement with my previous calculations.
3) I made a wannier90.x input file with the same uniform K-points grid used in 1) and generated the .nnkp file.

Later I did again "ypp -w" and then "ypp". And I got, as before:

<---> [01] CPU structure, Files & I/O Directories
<---> CPU-Threads:1(CPU)-8(threads)
<---> [02] Y(ambo) P(ost)/(re) P(rocessor)
<---> [03] Core DB
<---> :: Electrons : 40.00000
<---> :: Temperature [ev]: 0.000000
<---> :: Lattice factors [a.u.]: 21.84385 7.87270 8.58361
<---> :: K points : 72
<---> :: Bands : 25
<---> :: Symmetries : 1
<---> :: RL vectors : 104389
<---> [04] K-point grid
<---> :: Q-points (IBZ): 72
<---> :: X K-points (IBZ): 72
<---> [05] CORE Variables Setup
<---> [05.01] Unit cells
<---> [05.02] Symmetries
<---> [WARNING] Spatial Inversion not found among the given symmetry list
<---> [05.03] RL shells
<---> [05.04] K-grid lattice
<---> [05.05] Energies [ev] & Occupations
<---> [06] == Wannier 90 Interface ==At line 58 of file WANNIER_driver.f90 (unit = 10, file = '.nnkp')
Fortran runtime error: End of file

Now ypp realizes that there are not symmetries, but still yields the obscure message about the nnkp file, which is there and looks OK to me. I am attaching it to this post.

I would appreciate any help about how the yambo - wannier90 interface works.

Thanks in advance

Juanjo

[Daniele Varsano]

Dear Juan,
it seems that there is some problem in reading the nnkp files.
If you can't understand what is happening, please post all the files needed to reproduce the error and we will have a look.

Best,
Daniele