[QE-users] geometry outimization under finite homogeneous electric field
Maxim Arsentev
arsentev at isc.nw.ru
Tue Sep 4 15:06:14 CEST 2018
Dear Quantum Espresso users,
I'm trying to perform geometry optimization runs with finite homogeneous
electric field described through the modern theory of the polarization
for the monolayer TiS2
Is it legitimate for such a system (it is not a dielectric)?, if not,
what other methods are there for such a system?
Thank you, here is my code
&control
prefix='TiS3',
calculation = 'relax',
restart_mode = 'from_scratch'
pseudo_dir='$PSEUDO_DIR',
outdir = '$TMP_DIR',
forc_conv_thr = 1.0d-3 ,
!forc_conv_thr = 1.0d-2 ,
nstep = 99,
etot_conv_thr = 1.0d-4 ,
!etot_conv_thr = 2.0d-10 ,
lelfield=.true.,
nberrycyc=3,
/
&system
ibrav= 0, nat= 15, ntyp= 4,
celldm(1)=1.889726,
ecutwfc = 39.0, ecutrho = 300,
!occupations ='smearing', smearing ='gaussian',
!degauss=0.02,
input_dft = 'vdW-DF',
/
&electrons
conv_thr = 1e-6,
mixing_beta=0.3,
electron_maxstep=500,
efield_cart(1)=0.d0,efield_cart(2)=0.d0,efield_cart(3)=0.050d0
/
&ions
/
&cell
cell_factor = 1.5
/
ATOMIC_SPECIES
Ti 47.86700 Ti.pbe-spn-rrkjus_psl.0.3.1.UPF
S 32.0660 S.pbe-n-rrkjus_psl.0.1.UPF
Li 6.941 Li.pbe-s-rrkjus_psl.0.2.1.UPF
H 1.007940 H.pbe-rrkjus_psl.0.1.UPF
ATOMIC_POSITIONS crystal
H 0.213363000 0.178857000 0.243496000
H 0.134546000 0.255217000 0.233180000
Li 0.233954000 0.175579000 0.132762000
Ti 0.980482000 0.008124000 0.995434000
Ti 0.487252000 0.007521000 0.996140000
Ti 0.235353000 0.505413000 0.995123000
Ti 0.735346000 0.515324000 0.995926000
S 0.982820000 0.344132000 0.069262000
S 0.487669000 0.343713000 0.069725000
S 0.235891000 0.838631000 0.069596000
S 0.735943000 0.843450000 0.068669000
S 0.235788000 0.175732000 0.925965000
S 0.735233000 0.176684000 0.925749000
S 0.985574000 0.675253000 0.925626000
S 0.485446000 0.675405000 0.925646000
CELL_PARAMETERS (alat)
6.9649200000 0.0000000000 0.0000000000
0.0000000000 6.0339600000 0.0000000000
0.0000000000 0.0000000000 20.0000000000
K_POINTS {automatic}
6 6 1 0 0 0
More information about the users
mailing list