ABINIT Discussion Forums

Dear Abinit users,

I am trying to optimize a molecule using the abinit code, so far I couldn't get it to converge the SCF in the first optimization step using 200 SCF cycles. I was runing the calculation parallel on 4 procs.

I would like to know what am I doing wrong in the input file? Does it need some key words that I am missing? Following is the input file that I have been using to optimize the geometry..

# optimize the geometry / interatomic distance
# optimizing after finding the optimal ecut as 20.0 and
# acell to be 8 8 8 angstroms from ecut-test and acell-test runs

ionmov 3 # Use the modified Broyden algorithm
ntime 20 # Maximum number of Broyden "timesteps"
tolmxf 5.0d-3 # Stopping criterion for the geometry optimization : when
# the residual forces are less than tolmxf, the Broyden
# algorithm can stop
#nsym=1 #nosymmetry

#starting value of the
#atomic coordinates
[ xangst ]

toldff 5.0d-4 # Will stop the SCF cycle when, twice in a row,
# the difference between two consecutive evaluations of
# forces differ by less than toldff (in Hartree/Bohr)
#toldfe 1.0d-6
#------------------------------------------------------------------------------
#Definition of the unit cell
acell 20 20 20 angstroms # The keyword "acell" refers to theI
# lengths of the primitive vectors (in Bohr)
#rprim 1 0 0 0 1 0 0 0 1 # This line, defining orthogonal primitive vectors,
# is commented, because it is precisely the default value of rprim
#Definition of the atom types

ntypat 4 #2 types of atom
znucl 79 16 6 1
#Au S C H

#Definition of the atoms
natom 115 # There are 115 atoms

typat 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2
2 2 2 2 3 4 4 3 4 4 3 4 4 3 4 4 3 4 4 3 4 4 3 4 4 3 4 4 3 4 4 3 4 4 3 4 4 3 4 4
3 4 4 3 4 4 3 4 4 3 4 4 3 4 4 3 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4

#Definition of the planewave basis set
ecut 20.0 # Maximal kinetic energy cut-off, in Hartree

#Definition of the k-point grid
#kptopt 1 # Enter the k points manually
#nshiftk 1
#shiftk 0.5 0.5 0.5
#nkpt 1 # Only one k point is needed for isolated system,
# # taken by default to be 0.0 0.0 0.0

#ngkpt 2 2 2
#Chksymbreak 0

#Definition of the SCF procedure
nstep 200 # Maximal number of SCF cycles
#toldfe is no more defined, as toldff is used above...
diemac 2.0

----
In the above another question that Arieses is , I want to treat it as a single molecule, but if used the keyword nkpt=1 the calculation halts. To my knowledge I must use nkpt=1 but the program does not allow me to use it.

I would also like to know is there any additional keywords that I need to use when running it on parallel ?
Thanks.


Neranjan

Try with:

kptopt 0
nkpt 1
kpt 0.0 0.0 0.0

Also, have you successfully optimized a simpler molecule, just as a test case? For example, N2 or something like that? It would help you get the kpt related variables correct, before moving on to a big system.

Thank you very much Sir, I was able to run my calculation successfully using the given keywords. I really appreciate the help given.
Thanks

Hello,

I am using the following grid (3 x 3 x5) how to calculate the number of kpoints? as I am always getting following warning:
" distrb2: WARNING
nkpt*nsppol ( 68) is not a multiple of nproc_kpt ( 16) The k-point parallelisation is not efficient."

how to avoid this warning?

One more question please, I have the follwing message:
"-P-0000 Wavefunctions not converged for nnsclo,ikpt= 50 1 max resid= 2.77067E-27 starting lobpcg, with nblockbd,mpi_enreg%nproc_band 96 1
WARNING in zpotrf, info= 1"
I am assuming that the SCF iterations (50) is not enough and I should increase that number to get the energy converged! Is that correct? if so what do you suggest?


warm regards,