EPW Forum

Hello Everyone,

I am using EPW_v2.3.5. I am trying to calculate electron self-energy. I have successfully done it for bulk silicon and graphene. Now I am trying for Mos2, I am facing the following error

from readmat : error # 1
wrong qpoint

which means mismatch of q-points between the epw.in and prfix.dyn_q files. But I checked this is not the case in my calculations. Below is the epw input file

--
&inputepw
prefix = 'mos2'
amass(1) = 32.065,
amass(2) = 95.94,
outdir = '../epw_output/'

iverbosity = 0

elph = .true.
epbwrite = .true.
epbread = .false.

epwwrite = .true.
epwread = .false.

nbndsub = 9
nbndskip = 0
kmaps = .false.



wannierize = .true.
num_iter = 5000
iprint = 2
dis_win_max = 30
dis_froz_max= 2.5
proj(1) = 'random'

elinterp = .true.
phinterp = .true.

tshuffle2 = .true.
tphases = .false.
epstrict = .true.

elecselfen = .true.
phonselfen = .false.
a2f = .false.
fly = .true.
parallel_k = .true.
parallel_q = .false.
ephwrite = .true.
fsthick = 8.d10

degaussw = 0.025

filukk = './mos2.ukk'
dvscf_dir = './save'
max_memlt = 5.0d0
nkf1 = 10
nkf2 = 10
nkf3 = 1
nqf1 = 10
nqf2 = 10
nqf3 = 1

nk1 = 16
nk2 = 16
nk3 = 1

nq1 = 8
nq2 = 8
nq3 = 1
/
10 cartesian
0.0000000 0.0000000 0.0000000 0.0312500
0.0000000 0.1461865 0.0000000 0.1875000
0.0000000 0.2923730 0.0000000 0.1875000
0.0000000 0.4385595 0.0000000 0.1875000
0.0000000 -0.5847460 0.0000000 0.0937500
0.1266012 0.2192798 0.0000000 0.1875000
0.1266012 0.3654663 0.0000000 0.3750000
0.1266012 0.5116528 0.0000000 0.3750000
0.2532025 0.4385595 0.0000000 0.1875000
0.2532025 0.5847460 0.0000000 0.1875000

I can interpolate the phonon band structure using "q2r.x and matdyn.x" binaries. I dont know why epw complain about the q-point.

And one thing I noticed in this case is that code does not say anything about reading q-points in star. I mean
===================================================================
irreducible q point # 2
===================================================================


q( 2 ) = ( 0.0000000 0.1461865 0.0000000 )


===================================================================
irreducible q point # 3
===================================================================

The 2nd q-point has further 6 q-points in star.
Is this problem related with the symmetry in the system?

Many thanks in advace.

Dear dinesh169,

Can you use the latest EPW v4 as we do not really support such old version.

The installation is very easy. You can follow the steps http://epw.org.uk/Main/DownloadAndInstall

Bugs related to symmetry have been fixed in this new version.

Note however that the version 6 of QE and version v4.1 of EPW will be available very soon (first week of October if everything goes well).

Best,

Samuel

Hello Dr. Samuel Ponc,

Thank you very much for your reply. I am using the new EPW 4.0, and the problem mentioned previously is solved.
Now I am trying the same calculations for bulk MoS2. I came across this error:

Error in routine calbec (3):
size mismatch

I am using the uniform k-grid for wannierization. I read in the forum that it may be because of GGA functional and i am also using the same.
Is there any way to avoid it? because i would not be good if I have to change the functional, then i might have to run all the for the single layer.

Below is the nscf and epw files respectively:
&CONTROL
calculation = 'nscf' ,
restart_mode ='from_scratch' ,
outdir = './' ,
pseudo_dir = './' ,
prefix = 'mos2' ,
wf_collect=.false.
nstep= 80
etot_conv_thr = 1.0D-7
forc_conv_thr = 1.0D-5
tstress = .true.
tprnfor = .true.
verbosity = 'high'

/
&SYSTEM
ibrav = 0,
A = 3.16
nat = 6,
ntyp = 2,
ecutwfc = 70 ,
ecutrho = 300 ,
occupations = 'smearing' ,
degauss = 0.03 ,
smearing = 'mp'
nbnd = 30
/
&ELECTRONS
electron_maxstep = 100
conv_thr = 1.0d-10
mixing_mode = 'plain'
mixing_beta = 0.5
mixing_ndim = 8
diagonalization = 'david'
diago_david_ndim = 4
diago_full_acc = .true.
/
&IONS
ion_dynamics = 'bfgs'
trust_radius_max = 0.5
trust_radius_min = 1.0D-16
/
&CELL
cell_dynamics = 'bfgs'
press = 0.00D0
cell_dofree = 'all'
cell_factor = 1.8D0
/
ATOMIC_SPECIES
S 32.065 16-S.GGA.txt.UPF
Mo 95.94 42-Mo.GGA.txt.UPF

CELL_PARAMETERS (alat)
0.999060846 0.000000000 0.000000000
-0.499530423 0.865212072 -0.000000000
0.000000000 0.000000000 3.840581648

ATOMIC_POSITIONS (crystal)
Mo 0.333333000 0.666667000 0.249999899
Mo 0.666667000 0.333333000 0.750000101
S 0.333333000 0.666667000 0.619509115
S 0.666667000 0.333333000 0.380490885
S 0.666667000 0.333333000 0.119509061
S 0.333333000 0.666667000 0.880490939

K_POINTS crystal
144
0.00000000 0.00000000 0.00000000 6.944444e-03
0.00000000 0.00000000 0.25000000 6.944444e-03



--
&inputepw
prefix = 'mos2'
amass(1) = 32.065,
amass(2) = 95.94,
outdir = './'



iverbosity = 0

elph = .true.
epbwrite = .true.
epbread = .false.

epwwrite = .true.
epwread = .false.

nbndsub = 30
nbndskip = 0

wannierize = .true.
num_iter = 10000
iprint = 2
dis_win_max = 15
dis_froz_max= 5
proj(1) = 'random'
kmaps = .false.

elinterp = .true.
phinterp = .true.

tshuffle2 = .true.
tphases = .false.

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

parallel_k = .true.
parallel_q = .false.

fsthick = 136056981 ! eV
eptemp = 300 ! K (same as PRB 76, 165108)
degaussw = 0.1 ! eV

dvscf_dir = './save'
filukk = './mos2.ukk'

nkf1 = 12
nkf2 = 12
nkf3 = 4
nqf1 = 6
nqf2 = 6
nqf3 = 2

nk1 = 6
nk2 = 6
nk3 = 2

nq1 = 6
nq2 = 6
nq3 = 2
/
14 cartesian
0.0000000 0.0000000 0.0000000 0.0277778
0.0000000 0.0000000 -0.1301886 0.0277778
0.0000000 0.1926310 0.0000000 0.1666667
0.0000000 0.1926310 -0.1301886 0.1666667
0.0000000 0.3852620 0.0000000 0.1666667

Have a nice evening.

Dear Dinesh

Unfortunately the error seems related to the pseudo. It looks like EPW experiences problems when using GGA in some cases - I have personally used PBE for some system and it was working fine. We need to investigate this further - if you want, it could be useful if you send us the pseudo so to reproduce the problem.

Best
Carla

Hello Carla,

Thank you very much for your reply. Here is the link to pseudopotential files

ftp://ftp.abinit.org/pub/abinitio/Psps/GGA_FHI/42-Mo.GGA.fhi

I converted this file to UPF for using in quantum espresso.
Actually I solved the problem. I was using ecutrho = 4*ecutwfc+X(X= some integer). When i specified ecutrho =4*ecutwfc default value, it worked out. It was just an intuitive guess. In the source code this error was because of size mismatch between betapsi and number of k-points, number of bands. So I tried default values and It worked out.

Have a nice day.
Dinesh

That's great to know Dinesh!
I think it's best to always keep ecutrho at the default 4*ecutwfc for NC pseudos with QE.

Best
Carla

Dear Dinesh,

Thanks for letting us know. Happy that it now works for you.

Best,

Samuel

Dear Admins,

Thank you very much for your replies. I want to ask you one question.
Phonon calculations are also very time consuming. I tried to do phonon calculation on every q point separately and then put the prefix.dyn_q#, prefix.dvscf_q# and patterns.#.xml ( in prefix.phsave directory) together in save directory. I face the the following error

Error in routine read_modes (1):
problems with current_iq

Though when I do the phonon calculations on a grid at one time and copy the dyn, dvscf, patterns ( in prefix.phsave directory) then everything works fine. Am I missing some files or is this not the right way to do it?

have a nice evening.

Best wishes,

Dear Dinesh

The procedure is correct as long as you are copying every file correctly - please double check again, you should not encounter any problem.

Best
Carla

Hi,

If I remember correctly when you calculate each q-point separately you may need to change the q-point number in the patterns.*.xml files. At the beginning of each file you have the following lines.

<?iotk binary="F"?>
<?iotk qe_syntax="F"?>
<Root>
<IRREPS_INFO>
<QPOINT_NUMBER type="integer" size="1">
2

The line after QPOINT_NUMBER should be the nr of the respective q-point. In other words, 2 if this is q-point nr 2. This number will be the correct value if you calculated all q-points at once on a mesh, otherwise it will be 1.

Best,
Roxana