General discussion around the EPW software
Moderator: stiwari
wqx
Posts: 4 Joined: Mon Sep 19, 2022 1:10 am
Affiliation: nanjing nomal university
Post
by wqx » Tue Oct 04, 2022 12:52 pm
Dear all,
I am using QE to calculate phonon dispersion of copper. But I found my result is different with reference. Could you find some problems in my calculation?
Code: Select all
&CONTROL
calculation='scf'
pseudo_dir='./'
outdir='./'
restart_mode='from_scratch'
prefix='Cu'
/
&SYSTEM
ibrav= 2
nat= 1
ntyp= 1
celldm(1)=6.834030
ecutwfc = 90
occupations='smearing'
degauss=0.01
/
&ELECTRONS
conv_thr = 1.0d-10
/
ATOMIC_SPECIES
Cu 63.546 Cu_pbe_v1.2.uspp.F.UPF
ATOMIC_POSITIONS (crystal)
Cu 0.0000000000 0.0000000000 0.0000000000
K_POINTS {automatic}
11 11 11 1 1 1
Code: Select all
phonon calculation
&inputph
outdir='./'
prefix='Cu'
tr2_ph=1.0d-14
ldisp=.true.
amass(1)=63.546
nq1=4,nq2=4,nq3=4
fildyn='Cu.dyn'
fildvscf='dvscf'
/
Code: Select all
&input
fildyn='Cu.dyn'
zasr='simple'
flfrc='Cu.fc'
/
Code: Select all
&input
asr='simple'
amass(1)=63.546
flfrc='Cu.fc'
flfrq='Cu.freq'
q_in_band_form=.true.
q_in_cryst_coord=.true.
/
6
0.0000 0.0000 0.0000 30
0.0000 0.5000 0.5000 30
0.2500 0.5000 0.7500 30
0.0000 0.5000 0.5000 30
0.0000 0.0000 0.0000 30
0.5000 0.5000 0.5000 30
The figure at the bottom is my result.
2.jpg (91.41 KiB) Viewed 3986 times
Attachments
1.jpg (42.21 KiB) Viewed 3986 times
hlee
Posts: 415 Joined: Thu Aug 03, 2017 12:24 pm
Affiliation: The University of Texas at Austin
Post
by hlee » Wed Oct 05, 2022 4:49 pm
Dear wqx:
Basically, the question on the phonon code in QE should be asked on the QE forum.
In any case, I think the issue with your inputs is related to your chosen path; for instance, the coordinate of the 2nd (1st) point is different from that of the 4th (5th) point.
Code: Select all
6
0.0000 0.0000 0.0000 30
0.0000 0.5000 0.5000 30
0.2500 0.5000 0.7500 30
0.0000 0.5000 0.5000 30
0.0000 0.0000 0.0000 30
0.5000 0.5000 0.5000 30
You can confirm it by comparing the distance between the 1st and the 2nd point with that between the 4th and the 5th point in your reference figure.
Sincerely,
H. Lee
wqx
Posts: 4 Joined: Mon Sep 19, 2022 1:10 am
Affiliation: nanjing nomal university
Post
by wqx » Thu Oct 06, 2022 3:04 am
hlee wrote: ↑ Wed Oct 05, 2022 4:49 pm
Dear wqx:
Basically, the question on the phonon code in QE should be asked on the QE forum.
In any case, I think the issue with your inputs is related to your chosen path; for instance, the coordinate of the 2nd (1st) point is different from that of the 4th (5th) point.
Code: Select all
6
0.0000 0.0000 0.0000 30
0.0000 0.5000 0.5000 30
0.2500 0.5000 0.7500 30
0.0000 0.5000 0.5000 30
0.0000 0.0000 0.0000 30
0.5000 0.5000 0.5000 30
You can confirm it by comparing the distance between the 1st and the 2nd point with that between the 4th and the 5th point in your reference figure.
Sincerely,
H. Lee
Dear H. Lee,
Thank you for your help! I checked my reference and my codes. K path is the key issue of phonon dispersion.
Best wishes.
wqx