The ABINIT Users Mailing List ( CLOSED )

[kong <kongl@physics.rutgers.edu>]

Hi,

I am doing relaxation of CO2 dimer in abinit 5.3.4. The displacement is less
than 0.5A away 
from equilibrium. The first ionic step is fine. It crashed after the first SCF
in the second ionic step.
The error message is " -P-0000  fxphas : BUG  The eigenvector number  18 has
zero norm."

The program runs fine if I copy the atomic coordinates after the first ionic
step from the crashecd 
output file and put them into a new input file,  all the other input variables
are the same as before. See below for the details.
Thanks for any suggestions!

Lingzhu


ixc 11
iscf 7
ionmov 2
ntime   30        
tolmxf  5.0d-4    
prtwf 0

# Parallelization 
wfoptalg 4
nloalg 4
fftalg 401
iprcch 0
intxc 0
istwfk 1
fft_opt_lob 2
npband 8
npfft 1
# Parallelization 

acell 30  30 20
ecut 25.0         

ntypat 2       
nkpt 1         
nstep 50          
toldff  5.0d-5   
diemac 2.0     

xangst
         -1.9993 0.00000  0.00000 # C
          1.8993 0.00000  0.00000 # C
         -1.1993 1.13923048 0.00000    # O
         -2.3993 -1.03923048 0.00000   # O
          2.5993 1.23923048 0.00000    # O
          1.2993 -1.13923048 0.00000   # O

znucl  6 8        
natom 6
typat 2*1 4*2
nband 32