How to start MD calculation?

Message

Jiaye Li · #1 Post by **Jiaye Li** » Sat Jun 30, 2007 7:11 am

Dear support

I want to run MD calculations with vasp under coarse criterion. The pressure, number and temperature need to be fixed. This is my INCAR:

Startparameter for this run:
NWRITE = 2 write-flag & timer
PREC = low medium, high low
ISTART = 0 job : 0-new 1-cont 2-samecut
ICHARG = 2 charge: 1-file 2-atom 10-const
ISPIN = 1 spin polarized calculation?
Electronic Relaxation 1
ENCUT = 250.0 eV
NELM = 60; NELMIN= 8; NELMDL=-8 # of ELM steps
EDIFF = 0.1E-04 stopping-criterion for ELM
LREAL = A real-space projection
Ionic relaxation
EDIFFG = 0.1E-03 stopping-criterion for IOM
NSW = 500 ------- Is it usefull for MD???
IBRION = 0 ionic relax: 0-MD 1-quasi-New 2-CG
NBLOCK = 1
NFREE = 2 steps in history (QN), initial steepest desc. (CG)
ISIF = 2 stress and relaxation
ISYM = 0 0-nonsym 1-usesym 2-fastsym

POTIM = 3.00 ------------ Does it mean 3 fs???
TEBEG = 300.0; TEEND = 300.0 temperature during run --------------------Does it mean the temperature will be 300K during calculation???
SMASS = 2.00 Nose mass-parameter (am)
! PSTRESS= 0.0 ---------How to control the pressure???

DOS related values:
EMIN = 10.00; EMAX =-10.00 energy-range for DOS
ISMEAR = 0; SIGMA = 0.20 broadening in eV -4-tet -1-fermi 0-gaus

Electronic relaxation 2 (details)
GGA= 91
IALGO = 48 algorithm
BMIX = 2.0
LDIAG = T sub-space diagonalisation
! IMIX = 4 mixing-type and parameters
! AMIX = 0.20; BMIX = 0.0001
! AMIX_MAG = 0.80; BMIX_MAG = 0.0001
! AMIN = 0.10
! WC = 1000; INIMIX= 1; MIXPRE= 1
! TIME = 0.10 timestep for ELM

Write flags
LWAVE = T write WAVECAR
LCHARG = T write CHGCAR
LVTOT = T write LOCPOT, local potential
LORBIT = 10
Dipole corrections
IDIPOL = 0 1-x, 2-y, 3-z
LDIPOL = F correct potential

Is my INCAR correct?
Thanks!

#2 Post by **admin** » Mon Jul 02, 2007 7:29 am

1) MD can only be run for constant V
2) your input:
--) NSW=500 is reasonable, you can continue with a second MD run if you keep WAVECAR and re-start with CONTCAR (out, run N) as new POSCAR (in, run N+1)
--) it makes no sense to choose ISIF=2 for a low-PREC calculation, please use the dafault value (ISIF=0 for IBRION=0)
--) in general, it is safer to set ALGO=Fast instead of IALGO=48
--) it makes no sense to set LVTOT in a MD run
--) the instantaneous T is calculated from the kinetic energies of the atoms, so it may fluctuate around 300 K.
--) POTIM is the time step in fs, you are right.
all other parameters seem reasonable as far as one can say without knowing the details of your system.

Jiaye Li · #3 Post by **Jiaye Li** » Wed Jul 04, 2007 9:02 am

Thank you very much. I made changes accordingly before submitting the jobs.