Bader calculation not converging for a geometry optimized structure
Posted: Thu Aug 04, 2016 5:15 am
Dear All,
I have been unable to obtain a Bader charge calculation convergence for one of my systems consisting of a single-layer MoS2 on a sapphire substrate, following the PAW calculations by adding the LAECHG=.TRUE. approach.
Of all the structures that I have simulated for this project, only this system structure is not giving me the electronic convergence (please refer to the "stdout" file attached).
INCAR:
System = MoS2 on Al2O3(0001)
NWRITE = 2
ISTART = 0
ICHARG = 2
INIWAV = 1
ENCUT = 400.0
PREC = ACCURATE
ISPIN = 2
NELMDL = -100
NELM = 1000
EDIFF = 1E-05
LREAL = AUTO
IALGO = 48
LCHARG = .TRUE.
LWAVE = .TRUE.
LVDW = .TRUE.
LAECHG = .TRUE.
ADDGRID = .FALSE.
NSW = 0
NELMIN = 1
IBRION = 2
EDIFFG = -0.013
ISIF = 1
ISYM = 0
POTIM = 0.10
ISMEAR = -5 (-1-Fermi, 1-Methfessel/Paxton)
SIGMA = 0.05 broadening in eV
NPAR = 4
LPLANE = .TRUE.
NCORE = 24
Additionally, I have tried the following steps, to no avail:
i) Increasing the NELM value from 120 up to 1400 (some of my structures did not converge successfully using NELM=120, but did eventually converge when NELM was increased to 400)
ii) Using the Davidson algorithm (IALGO=38), instead of RMM-DIIS (IALGO=48)
iii) increasing the NELMDL steps (from NELMDL= -10 to NELMDL= -100)
iv) using ADDGRID=TRUE
v Re-optimizing the structure using higher EDIFF, EDIFFG, and ENCUT
vi) Using VASP 5.2.12 and VASP 5.3
I would really appreciate any help if you could guide me in the right direction.
I could not upload the zipped file (~82mb) here, but please find the following link:
https://drive.google.com/file/d/0B7ym8d ... sp=sharing
I have been unable to obtain a Bader charge calculation convergence for one of my systems consisting of a single-layer MoS2 on a sapphire substrate, following the PAW calculations by adding the LAECHG=.TRUE. approach.
Of all the structures that I have simulated for this project, only this system structure is not giving me the electronic convergence (please refer to the "stdout" file attached).
INCAR:
System = MoS2 on Al2O3(0001)
NWRITE = 2
ISTART = 0
ICHARG = 2
INIWAV = 1
ENCUT = 400.0
PREC = ACCURATE
ISPIN = 2
NELMDL = -100
NELM = 1000
EDIFF = 1E-05
LREAL = AUTO
IALGO = 48
LCHARG = .TRUE.
LWAVE = .TRUE.
LVDW = .TRUE.
LAECHG = .TRUE.
ADDGRID = .FALSE.
NSW = 0
NELMIN = 1
IBRION = 2
EDIFFG = -0.013
ISIF = 1
ISYM = 0
POTIM = 0.10
ISMEAR = -5 (-1-Fermi, 1-Methfessel/Paxton)
SIGMA = 0.05 broadening in eV
NPAR = 4
LPLANE = .TRUE.
NCORE = 24
Additionally, I have tried the following steps, to no avail:
i) Increasing the NELM value from 120 up to 1400 (some of my structures did not converge successfully using NELM=120, but did eventually converge when NELM was increased to 400)
ii) Using the Davidson algorithm (IALGO=38), instead of RMM-DIIS (IALGO=48)
iii) increasing the NELMDL steps (from NELMDL= -10 to NELMDL= -100)
iv) using ADDGRID=TRUE
v Re-optimizing the structure using higher EDIFF, EDIFFG, and ENCUT
vi) Using VASP 5.2.12 and VASP 5.3
I would really appreciate any help if you could guide me in the right direction.
I could not upload the zipped file (~82mb) here, but please find the following link:
https://drive.google.com/file/d/0B7ym8d ... sp=sharing