Carbon Dimer Using Wavelets

Total energy, geometry optimization, DFT+U, spin....

Moderator: bguster

Locked
asmartins
Posts: 4
Joined: Sat Mar 06, 2010 12:48 pm

Carbon Dimer Using Wavelets

Post by asmartins » Mon Apr 12, 2010 7:11 pm

Dear all,

I'm trying to do a calculation of a C-C dimer (just two carbon atoms) using wavelets. I generated the input file (in attach) and when I run the calculations the following message appears:

" STOP nl1: projector region outside cell"

I tried to modify ACELL, in order to generate a big supercell, or change the pseudopotential, but it didn't work!

Do you have some suggestion?
Attachments
dimer1.in
(1.18 KiB) Downloaded 311 times

dcaliste
Posts: 44
Joined: Thu Oct 22, 2009 9:15 am
Contact:

Re: Carbon Dimer Using Wavelets

Post by dcaliste » Tue Apr 27, 2010 3:33 pm

There is an issue with the spatial expansion of the basis set. The projectors for your carbon atoms expand further than the cell. I suggest to decrease the size of the projectors expansion: decrease wvl_cpmult and wvl_fpmult to 8 for instance.

Some points about the wavelet implementation:
#Definition of occupation numbers
occopt 1
tsmear 0.004
#Options to improve the convergence
diemac 5.5
nband 4

The wavelet code is made for gap systems. So the occupancy will be set with 2 for all bands (no spin case) anyway. The number of bands is also computed automatically and additional bands are not possible.

#Definition of the k-point grid
kptopt 1 # Option for the automatic generation of k points, taking
# into account the symmetry
ngkpt 2 2 1
nshiftk 1
shiftk 0.5 0.5 0.0

The k point implemntation is not currently available inside ABINIT + BigDFT. By the way, it is not useful for isolated systems.

Locked