[Pw_forum] Negative phonon frequency for 2D material of Sn (stanene)
Yuma Nakamura
zcym15 at mails.tsinghua.edu.cn
Wed Apr 27 11:04:56 CEST 2016
Dear All
I started Quantum Espresso half years ago so I am relatively beginner user.
Now I am trying to reproduce the phonon dispersion of Stanene(graphene-like tin mono-layer) according to the article[S. Jomehpour Zaveh et al.,Superlattices and Microstructures 91 (2016) 383].
With the reported set of thresholds(etot_conv_thr, forc_conv_thr, conv_thr and tr2_ph),the negative frequency appears at Gamma point.
It seems that this negative frequency is not small enough to be modified by acoustic sum rule. (please check the output below)
I understand this is because its structure is not stable.
In the final relaxed state, the stress is -0.02 kbar in xx(yy) component and -0.24 kbar in zz component.
Is it optimization problem or do I miss some important control tags?
I used in Quantum Espresso 5.2.1
Here is my trial set of thresholds and result.
————————————————optimization & scf————————————
calculation mode = vc-relax
ion_dynamics = bfgs
ecutwfc = 65 *
etot_conv_thr =1.0D-12 *
forc_conv_thr =1.0D-6 *
conv_thr =1.0D-14
k-point =21x21x1 *
————————————————ph run————————————————————————
tr2_ph =1.0D-16
q-point =5x5x1 *
——————————————————————————————————————————————
* theses values are referred from the literature, the other values are not mentioned and determined by myself.
Diagonalizing the dynamical matrix
q = ( 0.000000000 0.000000000 0.000000000 )
**************************************************************************
freq ( 1) = -0.551779 [THz] = -18.405359 [cm-1]
freq ( 2) = -0.551779 [THz] = -18.405359 [cm-1]
freq ( 3) = 0.352400 [THz] = 11.754783 [cm-1]
freq ( 4) = 3.460541 [THz] = 115.431227 [cm-1]
freq ( 5) = 5.407571 [THz] = 180.377167 [cm-1]
freq ( 6) = 5.407571 [THz] = 180.377167 [cm-1]
**************************************************************************
Mode symmetry, C_3v (3m) point group:
freq ( 1 - 2) = -18.4 [cm-1] --> E L_3 I+R
freq ( 3 - 3) = 11.8 [cm-1] --> A_1 L_1 I+R
freq ( 4 - 4) = 115.4 [cm-1] --> A_1 L_1 I+R
freq ( 5 - 6) = 180.4 [cm-1] --> E L_3 I+R
In those setups, I got converged results(even phonon frequency is negative).
The complete input files are following. The output files are attached.
—————————————————————pw.x————————————————————————————
&control
calculation = 'vc-relax'
restart_mode = 'from_scratch'
outdir = './'
prefix = 'stanene'
tstress = .true.
tprnfor = .true.
pseudo_dir = './'
verbosity = 'high'
etot_conv_thr = 1.0E-12
forc_conv_thr = 1.0D-6
/
&system
ibrav = 0
celldm(1) = 8.83069022
nat = 2
ntyp = 1
ecutwfc = 65
ecutrho = 600
degauss = 0.02
smearing ='mp'
occupations ='smearing',
/
&electrons
diagonalization = 'david'
mixing_mode = 'plain'
mixing_beta = 0.7
conv_thr = 1.0d-14
/
&ions
ion_dynamics = 'bfgs'
/
&cell
cell_dofree = '2Dxy'
/
ATOMIC_SPECIES
Sn 118.710 Sn.pbe-dn-rrkjus_psl.0.2.UPF
CELL_PARAMETERS (alat= 8.83069022)
1.002153232 0.000000000 0.000000000
-0.501076616 0.867890158 0.000000000
0.000000000 0.000000000 4.061049000
ATOMIC_POSITIONS {crystal}
Sn 0.666666666 0.333333333 -0.000019757
Sn 0.333333333 0.666666667 0.044809919
K_POINTS automatic
36 36 1 0 0 0
—————————————————————ph.x————————————————————————————
phonons of Sn
&inputph
tr2_ph=1.0d-16,
prefix='stanene',
ldisp=.true.,
nq1=5, nq2=5, nq3=1
start_q=1
last_q=1
amass(1)=118.710,
outdir=“.//1”,
fildyn='stanene.dyn',
/
Best regards,
+++++++++++++++++
Yuma Nakamura
Double Master Degree candidate
Dept of Chemistry in Tsinghua University, China
Dept of Physics in Tohoku University, Japan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20160427/7aa84cf1/attachment.html>
More information about the users
mailing list