[QE-users] QE7.0 Berry phase polarization calculation inconsistency and strange numeric results
Aykut Turfanda
a11aykutturfanda at gmail.com
Mon Oct 31 17:04:34 CET 2022
I need to post it again because I forgot to add input files and the version
number.
Dear QE Users,
I am a PhD student working with Berry phase polarization calculations of
single layer 2D material hBN in quantum espresso (applying vacuum in the
z-direction).
I found -0.0000002 C/m^2 polarization in the z-direction in Quantum
ESPRESSO 7.0 without electric field; however, total phases belonging to
electronic and ionic contribution found as equal to each other and given as
0.40381 and -0.40381 with 5 significant digits.
Is it OK to take -0.0000002 C/m^2 as the polarization of the system? or Is
this a numerical error?
The material in the z-direction is not polar, is it possible that the
polarization operator for a non-polar material is ill defined for this
calculation? Should I use berry phase polarization for the polar materials
only?
Also, when I change the dimension of the unit cell in the z-direction and
position of the material inside the cell, this polarization value changes
even though I use the same material.
Most importantly, this polarization value changes with the number of
processors used with the mpirun -np. This is very strange, should I stick
to -np 1. I should not make it parallel?
I also saw that this polarization is given with microC/cm^2 in the papers.
Does this mean the last two digits from " P = -0.0000002 (mod
2.9386420) C/m^2" are not considered from the results obtained in the
Quantum ESPRESSO run?
Are all those numerical errors or physical phenomena because I know that
there are vacuum fluctuations and Lamb modes, and so on? Can we probe those
with QE?
I look forward to hearing from you soon.
Best regards,
Aykut Turfanda
PhD student at Istanbul Technical University
###############################
SUMMARY OF PHASES
~~~~~~~~~~~~~~~~~
Ionic Phase: -0.40381
Electronic Phase: 0.40381
TOTAL PHASE: -0.00000 MOD_TOT: 1
VALUES OF POLARIZATION
~~~~~~~~~~~~~~~~~~~~~~
The calculation of phases done along the direction of vector 3
of the reciprocal lattice gives the following contribution to
the polarization vector (in different units, and being Omega
the volume of the unit cell):
P = -0.0000030 (mod 40.7880378) (e/Omega).bohr
P = -0.0000000 (mod 0.0514003) e/bohr^2
P = -0.0000002 (mod 2.9386420) C/m^2
The polarization direction is: ( 0.00000 , 0.00000 , 1.00000 )
################################
&control
calculation = 'nscf',
restart_mode='from_scratch'
prefix = 'bn'
outdir = './tempdir'
pseudo_dir = '~/pseudo_my/'
etot_conv_thr = 1d-05
forc_conv_thr = 1d-04
lberry=.true.
gdir = 3
nppstr = 10
/
&system
ibrav = 0,
nat = 2,
ntyp = 2,
ecutwfc = 80
occupations='fixed'
smearing='gaussian'
degauss=0.02
/
&electrons
conv_thr = 1d-08
/
&ions
/
&cell
/
ATOMIC_SPECIES
B 10.811 B.pbe.nc.sr.upf
N 14.0067 N.pbe.nc.sr.upf
ATOMIC_POSITIONS angstrom
B -0.0000000000 1.4480800000 1.6085249900
N 1.2540740000 0.7240400000 1.6085249900
K_POINTS automatic
10 10 10 0 0 0
CELL_PARAMETERS angstrom
2.5081477165 0.0000000000 0.0000000000
-1.2540738583 2.1721196390 0.0000000000
0.0000000000 0.0000000000 21.5841000986
##############################
&control
calculation = 'scf',
restart_mode='from_scratch'
prefix = 'bn'
outdir = './tempdir'
pseudo_dir = '~/pseudo_my/'
etot_conv_thr = 1d-05
forc_conv_thr = 1d-04
/
&system
ibrav = 0,
nat = 2,
ntyp = 2,
ecutwfc = 80
occupations='fixed'
smearing='gaussian'
degauss=0.02
/
&electrons
conv_thr = 1d-08
/
&ions
/
&cell
/
ATOMIC_SPECIES
B 10.811 B.pbe.nc.sr.upf
N 14.0067 N.pbe.nc.sr.upf
ATOMIC_POSITIONS angstrom
B -0.0000000000 1.4480800000 1.6085249900
N 1.2540740000 0.7240400000 1.6085249900
K_POINTS automatic
10 10 1 0 0 0
CELL_PARAMETERS angstrom
2.5081477165 0.0000000000 0.0000000000
-1.2540738583 2.1721196390 0.0000000000
0.0000000000 0.0000000000 21.5841000986
#################################
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20221031/2cea2985/attachment.html>
More information about the users
mailing list