ZnO calculation

GW, Bethe-Salpeter …

Moderators: maryam.azizi, bruneval

Locked
Lantao
Posts: 4
Joined: Wed Apr 08, 2015 4:29 pm

ZnO calculation

Post by Lantao » Wed Apr 08, 2015 4:39 pm

Hello everyone,

I'm running the calculation of ZnO. The gap energy calculated is around 0.66eV which is much lower than the experimental result.

So I'm trying to make a correction of the gap by the GW calculations. But I've got a problem.

With a plasmon-pole model calculation, I have the following result:

k = 0.000 0.000 0.000
Band E0 <VxcLDA> SigX SigC(E0) Z dSigC/dE Sig(E) E-E0 E
20 5.768 -16.078 -13.765 0.524 0.757 -0.320 -13.930 2.148 7.917
21 6.536 -16.306 -12.081 -1.608 0.758 -0.318 -14.321 1.985 8.521

E^0_gap 0.768
E^GW_gap 0.604
DeltaE^GW_gap -0.163

I have two questions about these results:

1. From the potential files, I can read that 12 electrons are taken into account for Zn and 6 electrons for O. The direct gap is always computed between the band whose number is equal to the number of electrons in the cell divided by two and the next one. So I think the gap is between the 18th and 19th band ( (12*2+6*2)/2=18). But it seems that the programme found the gap between the 20th and 21st band. I don’t understand this. And I check the eigenvalues of different bands, I got:

k = 0.000 0.000 0.000
Band E0 <VxcLDA> SigX SigC(E0) Z dSigC/dE Sig(E) E-E0 E
16 4.172 -14.901 -12.290 1.462 0.693 -0.442 -12.077 2.824 6.996
17 4.172 -14.989 -12.321 1.438 0.693 -0.444 -12.145 2.844 7.016
18 5.465 -14.532 -14.659 2.944 0.714 -0.401 -12.522 2.011 7.476
19 5.465 -14.679 -12.502 0.705 0.712 -0.404 -12.626 2.053 7.519
20 5.768 -16.078 -13.765 0.524 0.757 -0.320 -13.930 2.148 7.917
21 6.536 -16.306 -12.081 -1.608 0.759 -0.318 -14.321 1.985 8.521
22 9.704 -10.396 -5.997 -3.114 0.824 -0.213 -9.337 1.060 10.764

2. I work on the GW calculations in order to correct the band gap. But after this calculation, the gap is even lower. I wonder if I made a mistake in the input file. Then I checked my file but I didn’t find anything.

Maybe someone can help me by figuring our my mistake?

Thanks a lot.

Have a good day.

Best,

Lantao

saurinrawal
Posts: 3
Joined: Tue Jan 20, 2015 7:28 pm

Re: ZnO calculation

Post by saurinrawal » Wed May 06, 2015 8:12 pm

Hey can you post your input file.

Thanks

Lantao
Posts: 4
Joined: Wed Apr 08, 2015 4:29 pm

Re: ZnO calculation

Post by Lantao » Wed May 06, 2015 11:14 pm

Finally, someone come to reply me. Thanks.

Actually, with more empty bands taken into account, the gap shows between the 18th and 19th band. This is a good news.

On the other hand, even I did a careful convergence analysis, I still can not have the good results.

I can even have a negative gap which is not at all logic.

Help me please. And here's the input file.

# ZnO : Calculation of the GW corrections
# Dataset 1 : ground state calculation
# Dataset 2 : calculation of the KSS file
# Dataset 3 : calculation of the screening (epsilon^-1 matrix for W)
# Dataset 4 : calculation of the self-Energy matrix elements (GW corrections)

# In the first tutorial of GW calculation, we checked the convergence so that we know
# for the Self-Energy computation, we need ecutwfn = 5.0 ecutsigx = 6.0 nband = 100
# for screening computation, we need ecutwfn = 4.0 ecuteps = 6.0 nband = 100

# The program broke after 3 datasets
#ndtset 4

# restartxf -2 # RESTART from (X,F) history
# restartxf=-2 :Read the HIST file and select the atomic
# positions and cell parameters with the lowest energy
# use restartxf=-1 or -2 to restart a calculation that was not completed

############# Usual self-consistent ground-state calculation ###############

#nshiftk1 1
#shiftk1 0.0 0.0 0.5
#istwfk1 *0
#prtden1 1

################################################################################

######## Definition of parameters for the calculation of the KSS file ########

#iscf2 -2
#getden2 -1
#nbandkss2 -1 # This lead to the generation (full diagonalization of the KS
# hamiltonian) and storage of the maximum possible number of states
# (or bands) common to all points.
#kssform2 1 # The value 1 corresponds to ask a KSS through a diagonalization
# of the KS hamiltonian.
#nband2 150
#nbdbuf2 120

################################################################################

############## Calculation of the screening (epsilon^-1 matrix) ############

#optdriver3 3 # Screening calculation
#getkss3 -1 # Obtain KSS file from previous dataset
#nband3 150
#ecutwfn3 30.0 # Cut-off energy of the planewave set to represent
# the wavefunctions. It would be more convenient to keep
# the default ecut value.
#ecuteps3 40.0 # Cut-off energy of the planewave set to represent
# the dielectric matrix.
#ppmfrq3 16.7 eV # Imaginary frequency where to calculate the screening
#nbdbuf3 120

################################################################################

##### Calculation of the Self-Energy matrix elements (GW corrections) ######

optdriver 4 # Self-Energy calculation
getkss -1 # Obtain KSS file from dataset 1
getscr -1 # Obtain SCR file from previous dataset
nband 150 # Bands to be used in the Self-Energy calculation
nbdbuf 120
ppmfrq 16.7 eVecutwfn 30.0 # Planwaves to be used to present the wavefunctions.
# It would be more convenient to keep the default ecut value.ecutsigx 40.0 # Dimension of the G sum in Sigma-x.
# It would be better to keep the default ecut value.
nkptgw 1 # number of k-point where to calculate the GW correction
kptgw # k-point
0.000 0.000 0.000
bdgw 18 19 # calculate GW corrections for bands from 4 to 5

################################################################################

# Data common to the three different datasets
##################### Definition of the atom types #####################

ntypat 2 # There are two types of atoms
znucl 30 8 # The keyword "znucl" refers to the atomic number of the
# possible type(s) of atom. The pseudopotential(s)
# mentioned in the "files" file must correspond
# to the type(s) of atom.
natom 4 # There are 4 atoms per cell
typat 1 2 # 1 Zn atoms (first type), 1 O atom (second type)
spgroup 186
spgaxor 1
spgorig 1
brvltt -1 # assign brvltt from the space group information
natrd 2 # number of atom read
xred # Definition of the primitive cell
1./3 2./3 1.0951595455E-04 # Position of Zn
1./3 2./3 3.7989048405E-01 # O

acell 6.2633392923E+00 6.2633392923E+00 1.0083776580E+01 Bohr #and scale of Cartisien Coordinate
#In unit of angstrom 1Bohr=0.5291772108 Anstrong
rprim
1.0000000000E+00 4.2710751043E-37 -7.9295776162E-37
-5.0000000000E-01 8.6602540378E-01 0.0000000000E+00
6.1232339957E-17 1.0605752387E-16 1.0000000000E+00


############################################################################

##################### Definition of the k-point grid ###################

ngkpt 5 5 3
nshiftk 1
shiftk
0.0 0.0 0.0
# 0.0 0.0 0.5

#kptrlatt
# 17 0 0
# 0 17 0
# 0 0 11
#shiftk 0.0 0.0 0.0 # When the primitive vectors of the lattice do NOT form a FCC
# or a BCC lattice, we usually use nshiftk = 1 and
# shiftk 0.5 0.5 0.5.
istwfk *1 # This is mandatory in all the GW steps
symmorphi 1
############################################################################

################ Exchange-correlation functional ##############

ixc 11 # GGA, Perdew-Burke-Ernzerhof GGA function
# pawecutdg 80 # PAW-Energy Cutoff for the double grid
timopt 2

##################################################################

############### Definition of the SCF procedure ################

nstep 100 # Maximal number of SCF cycles
iscf 17
ecut 60.0 # Maximal kinetic energy cut-off, in Hartree
tolwfr 1.0d-10

##################################################################

And here's the result in my output file:
=== KS Band Gaps ===
>>>> For spin 1
Minimum optical gap = 0.6651 [eV], located at k-point : 0.0000 0.0000 0.0000
Fundamental gap = 0.6651 [eV], Top of valence bands at : 0.0000 0.0000 0.0000
Bottom of conduction at : 0.0000 0.0000 0.0000
SIGMA fundamental parameters:
PLASMON POLE MODEL 1
number of plane-waves for SigmaX 4127
number of plane-waves for SigmaC and W 4127
number of plane-waves for wavefunctions 2719
number of bands 150
number of independent spin polarizations 1
number of spinorial components 1
number of k-points in IBZ 10
number of q-points in IBZ 10
number of symmetry operations 12
number of k-points in BZ 75
number of q-points in BZ 75
number of frequencies for dSigma/dE 9
frequency step for dSigma/dE [eV] 0.25
number of omega for Sigma on real axis 0
max omega for Sigma on real axis [eV] 0.00
zcut for avoiding poles [eV] 0.10

EPSILON^-1 parameters (SCR file):
dimension of the eps^-1 matrix on file 4127
dimension of the eps^-1 matrix used 4127
number of plane-waves for wavefunctions 2719
number of bands 150
number of q-points in IBZ 10
number of frequencies 2
number of real frequencies 1
number of imag frequencies 1

matrix elements of self-energy operator (all in [eV])

Perturbative Calculation

k = 0.000 0.000 0.000
Band E0 <VxcLDA> SigX SigC(E0) Z dSigC/dE Sig(E) E-E0 E
18 -2.174 -22.729 -26.479 4.905 0.741 -0.349 -21.873 0.856 -1.318
19 -1.509 -12.023 -8.581 -3.198 0.798 -0.253 -11.829 0.194 -1.315

E^0_gap 0.665
E^GW_gap 0.003
DeltaE^GW_gap -0.662

Lantao
Posts: 4
Joined: Wed Apr 08, 2015 4:29 pm

Re: ZnO calculation

Post by Lantao » Wed May 06, 2015 11:16 pm

Finally, someone come to reply me. Thanks.

Actually, with more empty bands taken into account, the gap shows between the 18th and 19th band. This is a good news.

On the other hand, even I did a careful convergence analysis, I still can not have the good results.

I can even have a negative gap which is not at all logic.

Help me please. And here's the input file.

# ZnO : Calculation of the GW corrections
# Dataset 1 : ground state calculation
# Dataset 2 : calculation of the KSS file
# Dataset 3 : calculation of the screening (epsilon^-1 matrix for W)
# Dataset 4 : calculation of the self-Energy matrix elements (GW corrections)

# In the first tutorial of GW calculation, we checked the convergence so that we know
# for the Self-Energy computation, we need ecutwfn = 5.0 ecutsigx = 6.0 nband = 100
# for screening computation, we need ecutwfn = 4.0 ecuteps = 6.0 nband = 100

ndtset 4

# restartxf -2 # RESTART from (X,F) history
# restartxf=-2 :Read the HIST file and select the atomic
# positions and cell parameters with the lowest energy
# use restartxf=-1 or -2 to restart a calculation that was not completed

############# Usual self-consistent ground-state calculation ###############

nshiftk1 1
shiftk1 0.0 0.0 0.5
istwfk1 *0
prtden1 1

################################################################################

######## Definition of parameters for the calculation of the KSS file ########

iscf2 -2
getden2 -1
nbandkss2 -1 # This lead to the generation (full diagonalization of the KS
# hamiltonian) and storage of the maximum possible number of states
# (or bands) common to all points.
kssform2 1 # The value 1 corresponds to ask a KSS through a diagonalization
# of the KS hamiltonian.
nband2 150
nbdbuf2 120

################################################################################

############## Calculation of the screening (epsilon^-1 matrix) ############

optdriver3 3 # Screening calculation
getkss3 -1 # Obtain KSS file from previous dataset
nband3 150
ecutwfn3 30.0 # Cut-off energy of the planewave set to represent
# the wavefunctions. It would be more convenient to keep
# the default ecut value.
ecuteps3 40.0 # Cut-off energy of the planewave set to represent
# the dielectric matrix.
ppmfrq3 16.7 eV # Imaginary frequency where to calculate the screening
nbdbuf3 120

################################################################################

##### Calculation of the Self-Energy matrix elements (GW corrections) ######

optdriver4 4 # Self-Energy calculation
getkss4 -1 # Obtain KSS file from dataset 1
getscr4 -1 # Obtain SCR file from previous dataset
nband4 150 # Bands to be used in the Self-Energy calculation
nbdbuf4 120
ppmfrq4 16.7 eVecutwfn 30.0 # Planwaves to be used to present the wavefunctions.
# It would be more convenient to keep the default ecut value.ecutsigx 40.0 # Dimension of the G sum in Sigma-x.
# It would be better to keep the default ecut value.
nkptgw4 1 # number of k-point where to calculate the GW correction
kptgw4 # k-point
0.000 0.000 0.000
bdgw4 18 19 # calculate GW corrections for bands from 4 to 5

################################################################################

# Data common to the three different datasets
##################### Definition of the atom types #####################

ntypat 2 # There are two types of atoms
znucl 30 8 # The keyword "znucl" refers to the atomic number of the
# possible type(s) of atom. The pseudopotential(s)
# mentioned in the "files" file must correspond
# to the type(s) of atom.
natom 4 # There are 4 atoms per cell
typat 1 2 # 1 Zn atoms (first type), 1 O atom (second type)
spgroup 186
spgaxor 1
spgorig 1
brvltt -1 # assign brvltt from the space group information
natrd 2 # number of atom read
xred # Definition of the primitive cell
1./3 2./3 1.0951595455E-04 # Position of Zn
1./3 2./3 3.7989048405E-01 # O

acell 6.2633392923E+00 6.2633392923E+00 1.0083776580E+01 Bohr #and scale of Cartisien Coordinate
#In unit of angstrom 1Bohr=0.5291772108 Anstrong
rprim
1.0000000000E+00 4.2710751043E-37 -7.9295776162E-37
-5.0000000000E-01 8.6602540378E-01 0.0000000000E+00
6.1232339957E-17 1.0605752387E-16 1.0000000000E+00


############################################################################

##################### Definition of the k-point grid ###################

ngkpt 5 5 3
nshiftk 1
shiftk
0.0 0.0 0.0
# 0.0 0.0 0.5

#kptrlatt
# 17 0 0
# 0 17 0
# 0 0 11
#shiftk 0.0 0.0 0.0 # When the primitive vectors of the lattice do NOT form a FCC
# or a BCC lattice, we usually use nshiftk = 1 and
# shiftk 0.5 0.5 0.5.
istwfk *1 # This is mandatory in all the GW steps
symmorphi 1
############################################################################

################ Exchange-correlation functional ##############

ixc 11 # GGA, Perdew-Burke-Ernzerhof GGA function
# pawecutdg 80 # PAW-Energy Cutoff for the double grid
timopt 2

##################################################################

############### Definition of the SCF procedure ################

nstep 100 # Maximal number of SCF cycles
iscf 17
ecut 60.0 # Maximal kinetic energy cut-off, in Hartree
tolwfr 1.0d-10

##################################################################

And here's the result in my output file:
=== KS Band Gaps ===
>>>> For spin 1
Minimum optical gap = 0.6651 [eV], located at k-point : 0.0000 0.0000 0.0000
Fundamental gap = 0.6651 [eV], Top of valence bands at : 0.0000 0.0000 0.0000
Bottom of conduction at : 0.0000 0.0000 0.0000
SIGMA fundamental parameters:
PLASMON POLE MODEL 1
number of plane-waves for SigmaX 4127
number of plane-waves for SigmaC and W 4127
number of plane-waves for wavefunctions 2719
number of bands 150
number of independent spin polarizations 1
number of spinorial components 1
number of k-points in IBZ 10
number of q-points in IBZ 10
number of symmetry operations 12
number of k-points in BZ 75
number of q-points in BZ 75
number of frequencies for dSigma/dE 9
frequency step for dSigma/dE [eV] 0.25
number of omega for Sigma on real axis 0
max omega for Sigma on real axis [eV] 0.00
zcut for avoiding poles [eV] 0.10

EPSILON^-1 parameters (SCR file):
dimension of the eps^-1 matrix on file 4127
dimension of the eps^-1 matrix used 4127
number of plane-waves for wavefunctions 2719
number of bands 150
number of q-points in IBZ 10
number of frequencies 2
number of real frequencies 1
number of imag frequencies 1

matrix elements of self-energy operator (all in [eV])

Perturbative Calculation

k = 0.000 0.000 0.000
Band E0 <VxcLDA> SigX SigC(E0) Z dSigC/dE Sig(E) E-E0 E
18 -2.174 -22.729 -26.479 4.905 0.741 -0.349 -21.873 0.856 -1.318
19 -1.509 -12.023 -8.581 -3.198 0.798 -0.253 -11.829 0.194 -1.315

E^0_gap 0.665
E^GW_gap 0.003
DeltaE^GW_gap -0.662

hisandeep88
Posts: 1
Joined: Sun Feb 04, 2018 8:22 pm

Re: ZnO calculation

Post by hisandeep88 » Sun Feb 04, 2018 8:31 pm

Hello Lantao

I would suggest you to use wurtzite structure(basis) in place of FCC or BCC as you have used in the simulation

Regards
Sandeep

Locked