Page 1 of 1

the problem of optimization of the lattice parameter ??

Posted: Sun May 01, 2011 3:48 pm
by hjli
Respected All,
I have recently learned abinit 6.4.3 for windows, there are three examples about optimization of the lattice parameter :
t34.# Crystalline silicon : computation of the optimal lattice parameter
# Convergence with respect to the number of k points.
t41..# Crystalline aluminum : optimization of the lattice parameter
# at fixed number of k points and broadening.
t42.# Crystalline silicon : computation of the optimal lattice parameter
# Convergence with respect to the number of k points.

Can anybody tell me what is the difference between them ?

the second problem:
could I use this .in file to optimize the lattice parameter of Si?

ndtset 15 # Number of data sets
acell: 9.6 9.6 9.6 # Initial acell
acell+ 0.1 0.1 0.1 # Interval of acell
kptopt 1 # Kpoint option ( default 0.0 )
ngkpt 6 6 6 # Number of k-points
toldfe 1.0d-6 # Tolerance on the difference of total energy
iscf 5 # SCF cycle, CG based on the minim. of the energy
nstep 15 # Number of self-consistent field steps
diemac 12.0 # Macroscopic dielectric constant diemac of the system
# is a useful help to speed-up the SCF procedure
ecut 9.0 # Energy cut off
enunit 2 # Energy unit of output
nband 5 # Number of bands
ixc 1 # Exchange Correlation choice
typat 1 1 # Number of types of atoms
znucl 14 # Atomic number of the atoms
natom 2
rprim 0.0 0.5 0.5 # Lattice vector
0.5 0.0 0.5
0.5 0.5 0.0
xred 0.00 0.00 0.00 # Basic vector
0.25 0.25 0.25

with best regards

Re: the problem of optimization of the lattice parameter ??

Posted: Mon May 02, 2011 5:13 pm
by Boris
Hi

If you want to compute the optimal cell parameter, it is better to let abinit find it on its own. Therefore, you don't need to use any dataset involving a change in the cell parameter (namely, acell+ 0.1 0.1 0.1).

Instead, you should use the following keywords: IONMOV, NTIME, OPTCELL and TOLMXF

If you use these keywords, abinit will perform a geometry optimization and will provide you with the optimum cell parameters.

For your first question, I guess the differences should be explained in the tutorials.

Regards

Re: the problem of optimization of the lattice parameter ??

Posted: Tue May 03, 2011 7:32 am
by hjli
Thanks,
I have another problem,Before I optimize the lattice parameters(eg. ZnO), I must define the number of k-points(ngkpt) and the energy cut off (ecut), in which .in files the the lattice parameters are experimental results??

Re: the problem of optimization of the lattice parameter ??

Posted: Tue May 03, 2011 7:07 pm
by Boris
Sorry but I don't understand your question

Boris