Page 1 of 1

WARNING: stress and forces are not correct

Posted: Mon Jan 23, 2006 12:18 pm
by Damien
Hi,

Actually, I am trying to calculate the Density of States of a carbon nanotube. However, I always have this WARNING (stress and forces are not correct (second dervivative of E(xc) not defined) !!!).

What I do not understand is which parameter I have to change not to have this Warning again?

Please, let me know if someone have an idea about my problem,

Thank you,

Damien

My Incar file is the following:

Startparameter for this Run:
NWRITE = 2;! LPETIM=F write-flag & timer
ISTART = 0;! job : 0-new 1-cont 2-samecut
ICHARG = 11;
LCHARG = T;

Electronic Relaxation 10
PREC = HIGH
NELM = 15 ;
EDIFF = 1E-05 ! stopping-criterion for ELM
BMIX = 3.00 ;

Ionic Relaxation
EDIFFG = -1E-04
NSW = 0 ! number of steps for IOM
NBLOCK = 1;! KBLOCK = 5 inner block; outer block
IBRION = 1 ! ionic relax: 0-MD 1-quasi-New 2-CG
ISIF = 2 ! stress and relaxation
ISYM = 2
IWAVPR = 2 ! prediction: 0-non 1-charg 2-wave 3-comb
LCORR = T ! Harris-correction to forces
POTIM = 0.20 ! time-step for ion-motion
TEBEG = 0.001 ; ! TEEND=1873.0 initial temperature
SMASS = 3.00 ! Nose mass-parameter (am)
GGA = 91;
VOSKOWN = 1;

DOS related values:
ISMEAR = 0;
SIGMA = 0.1;
RWIGS = 1.3;

ectronic Relaxation 2
ALGO = N
LDIAG = T ! sub-space diagonalisation
LREAL = T ! real-space projection
NPAR=1;

WARNING: stress and forces are not correct

Posted: Thu Jan 26, 2006 2:53 pm
by admin
It makes no sense to keep the charge density constant (ICHARG-11)
and relax the ions (ISIF=2, IBRION=1) at one time. Actually, as you set NSW=0,
the positions are not updated anyway, and hence you can ignore that warning.
In your case I would prepare 2 separate INCAR files,
1) for the ionic relaxations (ICHARG not set)
2) for the post-processing after the structure is relaxed (ICHARG=11, NSW=0, but IBRION and ISIF not set )