Page 1 of 1

Dimer run...good to 0.03 eV/A, but drifted away

Posted: Sat Aug 28, 2021 5:13 pm
by kai
Hi, everyone,

I came across a problem while running dimer simulation. The TS was first pre-converged to 0.05 ev/A by CINEB and thereafter I switch to dimer. While it is efficient to go 0.03 eV/A, but drifted away to force as high as 0.26 ev/A.

I suspect some tweak is needed on DFNMin=0.01 and DFNMax=1.0 away from default value , but I do not know how.

Any ideas? (full dimer run as in zip)

Many thanks for any comments and suggestions

Kai

DIMCAR as below:

Step Force Torque Energy Curvature Angle
1 0.12331 1.94742 -1389.04553 -0.21645 8.97463
1 0.12331 1.48693 -1389.04553 -0.39059 13.03547
1 0.12331 1.60718 -1389.04553 -0.41003 10.73649
1 0.12331 1.27927 -1389.04553 -0.53769 6.42823
2 0.08478 2.10841 -1389.04598 -0.43819 8.15178
2 0.08478 1.71687 -1389.04598 -0.48350 6.35525
2 0.08478 1.83578 -1389.04598 -0.47015 8.23059
2 0.08478 1.90027 -1389.04598 -0.49310 9.24203
3 0.08745 2.65405 -1389.04617 0.19139 5.81595
3 0.08745 2.62321 -1389.04617 0.13412 6.25249
3 0.08745 2.53722 -1389.04617 0.28306 2.23943
3 0.08745 2.56331 -1389.04617 0.29625 1.40707
4 0.29122 2.96916 -1389.03963 0.91757 -2.42456
4 0.29122 2.95360 -1389.03963 0.89837 1.42922
4 0.29122 2.95158 -1389.03963 0.90372 -0.13041
4 0.29122 2.95176 -1389.03963 0.90288 0.15200
5 0.51717 3.20758 -1389.02717 0.69767 -2.54584
5 0.51717 3.25006 -1389.02717 0.66158 -0.29625
5 0.51717 3.25377 -1389.02717 0.66135 0.10185
5 0.51717 3.25251 -1389.02717 0.66130 -0.02287
6 0.76045 2.98488 -1389.00658 1.06063 -3.16157
6 0.76045 2.99967 -1389.00658 1.05649 0.48768
6 0.76045 2.99218 -1389.00658 1.05399 -0.09928
6 0.76045 2.99343 -1389.00658 1.05471 0.02623

INCAR as below:

PREC = Normal
ENCUT = 400.000
IBRION = 3 # NEB needs 1 or 3, not 2
NSW = 60
ISIF = 2
NELMIN = 6
EDIFF = 1.0e-08 # -5
EDIFFG = -0.02
VOSKOWN = 1
NBLOCK = 1
NWRITE = 1
NELM = 60
ALGO = Very Fast (blocked Davidson)
ISPIN = 1 # no spin polarized calculation
INIWAV = 1
ISTART = 1
ICHARG = 1
LWAVE = .FALSE.
LCHARG = .FALSE.
ADDGRID = .FALSE.
ISMEAR = 0 # recommended for semiconductor
SIGMA = 0.05 # recommended for semiconductor
LREAL = A # Manual recommends A for mixed ions or more than 20 atoms
POTIM = 0

# NEB
#ICHAIN = 0 # 0 switches on NBE
#IMAGES = 3 # number of images, exclude is and fs
#SPRING = -5
#LCLIMB = .T.

# DIMER PARAMETERS
ICHAIN=2 # swithes on dimer
DdR=0.005
DRotMax=4
DFNMin=0.01
DFNMax=1.0


# optimizer
IOPT = 2 #7 # fire
MAXMOVE = 0.1
#TIMESTEP = 0.1

Re: Dimer run...good to 0.03 eV/A, but drifted away

Posted: Tue Aug 31, 2021 11:05 pm
by graeme
I did a couple of subsequent calculations and I believe that the dimer calculation has originated near a shoulder in the potential. On a shoulder, there is a negative mode, but when you follow it, it can disappear. Try getting closer with the NEB calculation. You can also post that if you want and I can take a look at it.

Re: Dimer run...good to 0.03 eV/A, but drifted away

Posted: Wed Jul 13, 2022 7:08 am
by madmax
[quote=graeme post_id=21486 time=1630451100 user_id=2]
I did a couple of subsequent calculations and I believe that the dimer calculation has originated near a shoulder in the potential. On a shoulder, there is a negative mode, but when you follow it, it can disappear. Try getting closer with the NEB calculation. You can also post that if you want and I can take a look at it.
[/quote]Dear Professor Graeme, I had a really similar problem like this subject. This is DIMCAR below:
Step Force Torque Energy Curvature Angle
1 0.05495 4.30704 -242.71845 -0.05476 14.45845
1 0.05495 5.10473 -242.71845 0.47393 9.37700
1 0.05495 2.84423 -242.71845 -0.38317 8.57867
1 0.05495 3.92778 -242.71845 -0.23363 14.50821
1 0.05495 4.18419 -242.71845 -0.32989 14.50717
1 0.05495 3.99605 -242.71845 -0.51007 15.75826
1 0.05495 4.72609 -242.71845 -0.65781 15.82688
1 0.05495 4.46858 -242.71845 -1.07019 13.93441
1 0.05495 3.92453 -242.71845 -1.21093 12.58067
1 0.05495 3.21794 -242.71845 -1.60503 8.32319
2 0.05102 4.36537 -242.71854 -1.02708 11.56266
2 0.05102 5.16830 -242.71854 -1.07041 12.87411
2 0.05102 4.78400 -242.71854 -1.14484 7.88905
2 0.05102 4.92546 -242.71854 -1.03381 6.45873
2 0.05102 5.08156 -242.71854 -0.89390 8.26185
2 0.05102 4.46747 -242.71854 -1.09772 6.06845
2 0.05102 3.60024 -242.71854 -1.36295 5.47838
2 0.05102 3.28798 -242.71854 -1.48405 5.46714
2 0.05102 3.35067 -242.71854 -1.60201 7.07189
2 0.05102 3.42452 -242.71854 -1.81804 7.76488
3 0.05159 5.64462 -242.71861 -0.36262 12.05881
3 0.05159 4.60952 -242.71861 -0.58603 11.95875
3 0.05159 5.68314 -242.71861 -0.56289 4.43614
3 0.05159 5.61829 -242.71861 -0.46353 4.31606
3 0.05159 5.30574 -242.71861 -0.48150 3.62016
3 0.05159 5.66401 -242.71861 -0.43319 0.89803
3 0.05159 5.68880 -242.71861 -0.43255 1.01982
3 0.05159 5.62664 -242.71861 -0.12714 0.71893
3 0.05159 5.57724 -242.71861 -0.18437 0.20305
3 0.05159 5.56807 -242.71861 -0.18428 0.20421
4 0.03376 6.62146 -242.71866 0.61661 0.90185
4 0.03376 6.61795 -242.71866 0.61751 0.41978
4 0.03376 6.60820 -242.71866 0.61886 0.17588
4 0.03376 6.59241 -242.71866 0.62932 -0.01274
4 0.03376 6.59264 -242.71866 0.62934 -0.00280
4 0.03376 6.59269 -242.71866 0.62932 -0.00050
4 0.03376 6.59270 -242.71866 0.62931 0.00005
4 0.03376 6.59269 -242.71866 0.62931 0.00011
4 0.03376 6.59269 -242.71866 0.62931 0.00008
4 0.03376 6.59269 -242.71866 0.62931 0.00005
5 1.26711 6.04536 -242.67147 2.99653 0.31485
5 1.26711 6.05487 -242.67147 2.99125 0.47936
5 1.26711 6.04480 -242.67147 2.99465 0.11510
5 1.26711 6.04255 -242.67147 2.99582 0.02895
5 1.26711 6.04201 -242.67147 2.99612 0.00717
5 1.26711 6.04188 -242.67147 2.99620 0.00176
5 1.26711 6.04185 -242.67147 2.99622 0.00042
5 1.26711 6.04184 -242.67147 2.99622 0.00010
5 1.26711 6.04184 -242.67147 2.99622 0.00002
5 1.26711 6.04184 -242.67147 2.99623 0.00000
6 2.42486 6.77587 -242.53977 2.30672 1.34993
6 2.42486 6.75370 -242.53977 2.25585 0.11406
6 2.42486 6.75238 -242.53977 2.25788 0.03848
6 2.42486 6.75198 -242.53977 2.25843 0.01109
6 2.42486 6.75187 -242.53977 2.25859 0.00323
6 2.42486 6.75184 -242.53977 2.25864 0.00094
6 2.42486 6.75183 -242.53977 2.25866 0.00027
6 2.42486 6.75183 -242.53977 2.25866 0.00007
6 2.42486 6.75182 -242.53977 2.25866 0.00002
6 2.42486 6.75182 -242.53977 2.25866 0.00000
7 3.62620 7.44144 -242.32035 2.72579 -7.36930
7 3.62620 7.45612 -242.32035 2.61547 1.48637
And what do you mean "Try getting closer with the NEB calculation.", I am confused about your suggections. Thanks a lot.

Re: Dimer run...good to 0.03 eV/A, but drifted away

Posted: Wed Jul 13, 2022 12:46 pm
by madmax
Here is the parameters of cNBE and DIMER calculations:
cNEB Parameters
ISTART = 0 (Read existing wavefunction; if there)
ISPIN = 2 (Non-Spin polarised DFT)
MAGMOM = 41*3
ICHARG = 2 (Non-self-consistent: GGA/LDA band structures)
LREAL = Auto (Projection operators: automatic)
ENCUT = 520 (Cut-off energy for plane wave basis set, in eV)
PREC = Normal (Precision level)
LWAVE = .FALSE. (Write WAVECAR or not)
LCHARG = .False. (Write CHGCAR or not)
ADDGRID= .TRUE. (Increase grid; helps GGA convergence)
# LVTOT = .TRUE. (Write total electrostatic potential into LOCPOT or not)
# LVHAR = .TRUE. (Write ionic + Hartree electrostatic potential into LOCPOT or not)
# NELECT = (No. of electrons: charged cells; be careful)
# LPLANE = .TRUE. (Real space distribution; supercells)
# NPAR = 4 (Max is no. nodes; don't set for hybrids)
# Nwrite = 2 (Medium-level output)
# KPAR = 2 (Divides k-grid into separate groups)
# NGX = 500 (FFT grid mesh density for nice charge/potential plots)
# NGY = 500 (FFT grid mesh density for nice charge/potential plots)
# NGZ = 500 (FFT grid mesh density for nice charge/potential plots)

Electronic Relaxation
ISMEAR = 0 (Gaussian smearing; metals:1)
SIGMA = 0.05 (Smearing value in eV; metals:0.2)
NELM = 90 (Max electronic SCF steps)
NELMIN = 6 (Min electronic SCF steps)
EDIFF = 1E-4 (SCF energy convergence; in eV)
# GGA = PS (PBEsol exchange-correlation)

Ionic Relaxation
NSW = 300 (Max ionic steps)
#IBRION = 2 (Algorithm: 0-MD; 1-Quasi-New; 2-CG)
ISIF = 2 (Stress/relaxation: 2-Ions, 3-Shape/Ions/V, 4-Shape/Ions)
EDIFFG = -0.03 (Ionic convergence; eV/AA)
# ISYM = 2 (Symmetry: 0=none; 2=GGA; 3=hybrids)
IVDW = 11
IBRION = 3
POTIM = 0
IOPT = 1
ICHAIN = 0
LCLIMB = TRUE
SPRING = -5
IMAGES = 5 #point_num

DIMER Parameters
ISTART = 0 (Read existing wavefunction; if there)
ISPIN = 2 (Non-Spin polarised DFT)
MAGMOM = 41*3
ICHARG = 2 (Non-self-consistent: GGA/LDA band structures)
LREAL = Auto (Projection operators: automatic)
ENCUT = 520 (Cut-off energy for plane wave basis set, in eV)
PREC = Normal (Precision level)
LWAVE = .FALSE. (Write WAVECAR or not)
LCHARG = .False. (Write CHGCAR or not)
ADDGRID= .TRUE. (Increase grid; helps GGA convergence)
# LVTOT = .TRUE. (Write total electrostatic potential into LOCPOT or not)
# LVHAR = .TRUE. (Write ionic + Hartree electrostatic potential into LOCPOT or not)
# NELECT = (No. of electrons: charged cells; be careful)
# LPLANE = .TRUE. (Real space distribution; supercells)
# NPAR = 4 (Max is no. nodes; don't set for hybrids)
# Nwrite = 3 (Medium-level output)
# KPAR = 2 (Divides k-grid into separate groups)
# NGX = 500 (FFT grid mesh density for nice charge/potential plots)
# NGY = 500 (FFT grid mesh density for nice charge/potential plots)
# NGZ = 500 (FFT grid mesh density for nice charge/potential plots)

Electronic Relaxation
ISMEAR = 0 (Gaussian smearing; metals:1)
SIGMA = 0.05 (Smearing value in eV; metals:0.2)
NELM = 90 (Max electronic SCF steps)
NELMIN = 6 (Min electronic SCF steps)
EDIFF = 1E-6 (SCF energy convergence; in eV)
# GGA = PS (PBEsol exchange-correlation)

Ionic Relaxation
NSW = 300 (Max ionic steps)
IBRION = 3 (Algorithm: 0-MD; 1-Quasi-New; 2-CG)
POTIM = 0
ISIF = 2 (Stress/relaxation: 2-Ions, 3-Shape/Ions/V, 4-Shape/Ions)
EDIFFG = -0.01 (Ionic convergence; eV/AA)
# ISYM = 2 (Symmetry: 0=none; 2=GGA; 3=hybrids)
IVDW = 11
LORBIT = 11
# DIMER PARAMETERS
ICHAIN = 2
DdR = 0.005
DRotMax = 10
DFNMin = 0.01
DFNMax = 1.0
IOPT = 2