Question About Cell shape changing During NEB

[callyli00]

Dear admin,

While my normal NEB was performing, I cannot help but noticed that the cell volume as well as the cell parameter are not changing even if the ISIF tag in INCAR was set to 3.

This problem obviously resulted a convergent diffculty for my system.

Is there any way to correct this problem or it just against the basic algorithm of NEB?

Looking for your kind help,

Best regards,

PS, the INCAR file was listed as follows:


ISTART = 0

EDIFF = 0.00010
EDIFFG = -0.050000

NSW = 1000
IBRION = 3
POTIM = 0
SIGMA = 0.02

IOPT = 3
IMAGES=8
SPRING=-5
ICHAIN=0
LCLIMB=.T.
NPAR=4

LDAU = .TRUE.
LDAUTYPE = 2
LDAUL = 2 -1 -1
LDAUU = 5 0 0
LDAUJ = 1 0 0

ISMEAR = 0
ENCUT = 600
ALGO = Fast
ISPIN = 2

LWAVE= .FALSE.
LCHARG = .FALSE.
LPLANE=.TRUE.

ISIF=3

LREAL=A
PREC=Normal

Thank you

[graeme]

The proper way to do this is to use the solid state NEB algorithm by setting LNEBCELL=.TRUE. This will allow the cell variables to be included in the NEB calculation. For this, use IOPT=3

It is also possible to have the the cell variables relax, if you se one of the built-in vasp optimizers with the NEB. Be warned, however, that the cell will be relaxed for every image and so in this way, not included in the reaction coordinate.

[callyli00]

Thanks for you patient answering!

I've changed the setting refering to your suggestion, but the problem still exists.

The information concerning to VTST code in OUTCAR is like this:
VTST: version 2.03d, (02/18/09)

CHAIN: initializing optimizer

OPT: Using Quick-Min optimizer
OPT: QM, Init
OPT: QM, MAXMOVE 0.200000
OPT: QM, TIMESTEP 0.100000

CHAIN: Read ICHAIN 0
CHAIN: Running the NEB
NEB: SPRING -5.000000
NEB: LCLIMB T
NEB: LTANGENTOLD F
NEB: LDNEB F
NEB: LDNEBORG F
NEB: EFIRST 0.000000
NEB: ELAST 0.000000

There is no imply about I turning on the LNEBCELL tag. Is this caused by the low version of VTSTCODE?
Thank you.

[graeme]

Yes, the solid-state NEB was introduced in v2.04.